Compare commits
431
Commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
97291ffcae | ||
|
|
9a9867e739 | ||
|
|
aa20a5c201 | ||
|
|
600394c366 | ||
|
|
8fd9855b2c | ||
|
|
2a8f1ecdd2 | ||
|
|
6202e95ad8 | ||
|
|
b1fee50718 | ||
|
|
3a59bff246 | ||
|
|
57683d49b6 | ||
|
|
0fd01fa1c3 | ||
|
|
6d66e5dac3 | ||
|
|
446c3e56a8 | ||
|
|
daab1f1b0a | ||
|
|
af70705250 | ||
|
|
f29f8626e9 | ||
|
|
940e6e7b42 | ||
|
|
c08ad75893 | ||
|
|
b8d9b101c9 | ||
|
|
a1ea354225 | ||
|
|
e9f516580e | ||
|
|
b006b8c1ab | ||
|
|
78912adeb6 | ||
|
|
9bbff158c0 | ||
|
|
6e0b343de3 | ||
|
|
5bc9b60623 | ||
|
|
e8b32681ce | ||
|
|
d29b6bf816 | ||
|
|
b6ad1b5e91 | ||
|
|
5f961b0f99 | ||
|
|
89dea4138a | ||
|
|
4fb8b10429 | ||
|
|
f09a4e4d42 | ||
|
|
ab5dc2f46b | ||
|
|
b50899e18c | ||
|
|
30d0fd7def | ||
|
|
23b8d164d2 | ||
|
|
0bfd92846f | ||
|
|
ca3bb600f3 | ||
|
|
5ffef504e4 | ||
|
|
94b7b741f8 | ||
|
|
79ac1105f6 | ||
|
|
87f74c31c6 | ||
|
|
5f67efc8c7 | ||
|
|
fd3fe87d90 | ||
|
|
a85197cda4 | ||
|
|
05884533d1 | ||
|
|
310daa8060 | ||
|
|
7b798ae543 | ||
|
|
3143f87587 | ||
|
|
5fb3626208 | ||
|
|
fb2d1fb2f2 | ||
|
|
cadbf6f275 | ||
|
|
ea5d315dc3 | ||
|
|
46990eb632 | ||
|
|
71e799bb92 | ||
|
|
09069b0095 | ||
|
|
01c44f6877 | ||
|
|
3b94e68af7 | ||
|
|
5f598ff70c | ||
|
|
b8064ce6e3 | ||
|
|
3b8202d21b | ||
|
|
94a75aa100 | ||
|
|
cc62bbc3ec | ||
|
|
78869a13f6 | ||
|
|
c1028e7431 | ||
|
|
aa977c5a32 | ||
|
|
06eb4929a7 | ||
|
|
befd6f9a08 | ||
|
|
053526d2da | ||
|
|
1758dc1333 | ||
|
|
a0d2430319 | ||
|
|
b44e49b4d9 | ||
|
|
ec2e359dd4 | ||
|
|
4d191d7bbd | ||
|
|
930ad5b25e | ||
|
|
ce9ed6de9b | ||
|
|
7be530b445 | ||
|
|
7505766a83 | ||
|
|
328a82835e | ||
|
|
e43e9ace68 | ||
|
|
a7111b0e24 | ||
|
|
6a7e11cd2c | ||
|
|
5444a82c7f | ||
|
|
8b1b9b427d | ||
|
|
ba5eaec957 | ||
|
|
34557c7cff | ||
|
|
b8e4f64520 | ||
|
|
925df6c7f1 | ||
|
|
f84f63cc64 | ||
|
|
23a3d08e85 | ||
|
|
2d890d1189 | ||
|
|
b55e9abd7e | ||
|
|
c71241288c | ||
|
|
158fc71813 | ||
|
|
44900bd5c4 | ||
|
|
031bdb05dd | ||
|
|
5d0d218137 | ||
|
|
755b3b1adc | ||
|
|
ac6bbf8c2f | ||
|
|
896524a6c0 | ||
|
|
9dbd2ac260 | ||
|
|
91766f5b0f | ||
|
|
192fb96552 | ||
|
|
5a5b8625dd | ||
|
|
c5c416dd65 | ||
|
|
9c41888721 | ||
|
|
30323568c4 | ||
|
|
c9c5476cf2 | ||
|
|
44aebe054e | ||
|
|
01ae14522d | ||
|
|
3d81c7fc16 | ||
|
|
8aad7bcdf6 | ||
|
|
f2d86d634e | ||
|
|
8f19bed2b6 | ||
|
|
e1da2802b0 | ||
|
|
72ad7cc53d | ||
|
|
2e60e05fd1 | ||
|
|
d6bff17544 | ||
|
|
82c9884654 | ||
|
|
efc1b8385b | ||
|
|
a2faa84f86 | ||
|
|
4f48b3f5be | ||
|
|
5168f1d8a7 | ||
|
|
4fea65f982 | ||
|
|
b42fae4054 | ||
|
|
7b5c0026f9 | ||
|
|
c315b52cb2 | ||
|
|
b384cacbe3 | ||
|
|
d208769446 | ||
|
|
2423989192 | ||
|
|
de9524d16b | ||
|
|
8e6aa67412 | ||
|
|
b89b90f4ba | ||
|
|
7f2d4b01e8 | ||
|
|
714387ad7c | ||
|
|
e80686fe01 | ||
|
|
90b5d7d055 | ||
|
|
74d37a710c | ||
|
|
c4a9bb3178 | ||
|
|
bc787972cc | ||
|
|
aa8ed82379 | ||
|
|
b0c04ca0ec | ||
|
|
4b84885d19 | ||
|
|
382672bc5e | ||
|
|
4a6f9449d0 | ||
|
|
dc03894ea4 | ||
|
|
c957ed8605 | ||
|
|
ac978048b0 | ||
|
|
f6cc0edded | ||
|
|
efc01f9d20 | ||
|
|
4047b93f89 | ||
|
|
03f43f4408 | ||
|
|
bfb3dea059 | ||
|
|
d7ea126d2e | ||
|
|
17bc14e9c1 | ||
|
|
a09645669c | ||
|
|
60aeb5bddc | ||
|
|
26e50da938 | ||
|
|
5d0b00204f | ||
|
|
5ef410fb70 | ||
|
|
e655ee3bb9 | ||
|
|
b75efedf7b | ||
|
|
1a8fbb97e9 | ||
|
|
db6c027cac | ||
|
|
66ac1429b0 | ||
|
|
0acc731113 | ||
|
|
07a9fbc2e1 | ||
|
|
2358c01c97 | ||
|
|
10484cb0d6 | ||
|
|
5176bda496 | ||
|
|
6f5e0d3491 | ||
|
|
d0f1964dd9 | ||
|
|
3c99e015aa | ||
|
|
775f43ce03 | ||
|
|
b4cb80bc68 | ||
|
|
31c74b12c2 | ||
|
|
51e959e972 | ||
|
|
e871813a2b | ||
|
|
74e7a6fb68 | ||
|
|
5c652df904 | ||
|
|
9c31f04478 | ||
|
|
b18498e880 | ||
|
|
851f9eeb97 | ||
|
|
79e76d688e | ||
|
|
c364a6b9ef | ||
|
|
eceefe363a | ||
|
|
522c514b1a | ||
|
|
d42758b189 | ||
|
|
150cb01649 | ||
|
|
a9daac317e | ||
|
|
cb103a3f83 | ||
|
|
3f9f3857f1 | ||
|
|
12a9b5029c | ||
|
|
f79838687e | ||
|
|
95b880b92e | ||
|
|
b499e10590 | ||
|
|
b1e9f1d5a5 | ||
|
|
20b8db3755 | ||
|
|
71789e12fa | ||
|
|
afc689839a | ||
|
|
4e88bec0e4 | ||
|
|
ad9bf6f846 | ||
|
|
269d72c078 | ||
|
|
3a685e5671 | ||
|
|
a68633091a | ||
|
|
d116a2bea3 | ||
|
|
07b14c3f77 | ||
|
|
7aec2801a1 | ||
|
|
6e6cea89f7 | ||
|
|
5a4947372f | ||
|
|
7eb0abf4da | ||
|
|
67cf5659d4 | ||
|
|
8d8d83bc25 | ||
|
|
2860089712 | ||
|
|
908b3ca4b2 | ||
|
|
54bb79b294 | ||
|
|
2e4ac644f8 | ||
|
|
fbdf360fd9 | ||
|
|
a1245101de | ||
|
|
698244e3f1 | ||
|
|
8c062da5a1 | ||
|
|
f38adabb8c | ||
|
|
3cb97f6e27 | ||
|
|
ddb94e4480 | ||
|
|
eea1857a37 | ||
|
|
a248460e70 | ||
|
|
ead36404c0 | ||
|
|
78add9ad9f | ||
|
|
a2b130f75f | ||
|
|
3d10dd97f9 | ||
|
|
6da7c23fc2 | ||
|
|
77f00914d3 | ||
|
|
ff9c0886ce | ||
|
|
dc07404b48 | ||
|
|
7d0a159aed | ||
|
|
0ace45ca61 | ||
|
|
b4356408b6 | ||
|
|
1f81167f9e | ||
|
|
3c9cb68831 | ||
|
|
db2b663547 | ||
|
|
1d96a747e4 | ||
|
|
025cc19b6c | ||
|
|
a8180ab9d2 | ||
|
|
44eff703e0 | ||
|
|
96d29b9e3a | ||
|
|
e4b1824dd4 | ||
|
|
9fcf02df73 | ||
|
|
e395620e29 | ||
|
|
157f608eaa | ||
|
|
8d55619bb1 | ||
|
|
676a35da56 | ||
|
|
e8fd945879 | ||
|
|
798919690e | ||
|
|
1b96c43fa5 | ||
|
|
64762f064b | ||
|
|
b79dea809a | ||
|
|
8b2453b0bc | ||
|
|
09c6a6e22d | ||
|
|
73c086a5b1 | ||
|
|
8c98d99a8f | ||
|
|
3e4c5a5a28 | ||
|
|
a5a893a417 | ||
|
|
1eff50d243 | ||
|
|
cb82f90590 | ||
|
|
492bb56513 | ||
|
|
d71f06d884 | ||
|
|
d7a1c7c53b | ||
|
|
a282d035aa | ||
|
|
a1b4c22bdd | ||
|
|
44e9e2e759 | ||
|
|
8d07de8630 | ||
|
|
0053fc9fac | ||
|
|
66c4a38aff | ||
|
|
ce580b6285 | ||
|
|
a15e6c2651 | ||
|
|
1b22ef5af3 | ||
|
|
3aab6bfb68 | ||
|
|
a6b82f6813 | ||
|
|
a8ed18bde4 | ||
|
|
871b3cbc8f | ||
|
|
27ec7b6d9b | ||
|
|
5f4b2290e4 | ||
|
|
6dceaf014e | ||
|
|
be0b6fe6dd | ||
|
|
43c69dd544 | ||
|
|
bc20ed4fa8 | ||
|
|
ef4f399315 | ||
|
|
d524adc869 | ||
|
|
251cebfaee | ||
|
|
4d6b5c982e | ||
|
|
fa81aaa5d7 | ||
|
|
4464cd4c92 | ||
|
|
b7d312f4d6 | ||
|
|
68cfde0331 | ||
|
|
9b02c9c2b6 | ||
|
|
4cf03123de | ||
|
|
bb4f6e1c88 | ||
|
|
72c75450e0 | ||
|
|
b543952c4a | ||
|
|
6031d7c576 | ||
|
|
82f255a486 | ||
|
|
a07f77574c | ||
|
|
5c8ff74107 | ||
|
|
dfe01741e7 | ||
|
|
ebb16c9110 | ||
|
|
04e618d2e9 | ||
|
|
39b754e5b1 | ||
|
|
c91c136005 | ||
|
|
b3af87e932 | ||
|
|
636066e8e7 | ||
|
|
75d677942a | ||
|
|
ba212a525b | ||
|
|
e410383fb6 | ||
|
|
9ca47631a9 | ||
|
|
27c058707d | ||
|
|
28c4fa5c87 | ||
|
|
d5944cef3f | ||
|
|
a3024cb838 | ||
|
|
40da9d9c5f | ||
|
|
2878cacad8 | ||
|
|
fcc04ed585 | ||
|
|
544a1b98eb | ||
|
|
59a45ede62 | ||
|
|
f72d3972a3 | ||
|
|
15df495507 | ||
|
|
e21515979b | ||
|
|
f16378bded | ||
|
|
a57d1662c4 | ||
|
|
b9c9d76442 | ||
|
|
08662410f7 | ||
|
|
f81f7bd788 | ||
|
|
141c64cde3 | ||
|
|
ecbc656365 | ||
|
|
14c8e44402 | ||
|
|
8bea6bc890 | ||
|
|
23f111995b | ||
|
|
4248d3ea82 | ||
|
|
0dbb01087e | ||
|
|
1c8bb686b5 | ||
|
|
bc68948ed9 | ||
|
|
314ed866a2 | ||
|
|
7221d82e48 | ||
|
|
d5553d8099 | ||
|
|
cb01987c83 | ||
|
|
ae149d0921 | ||
|
|
a6a6ae0e26 | ||
|
|
1c200e268f | ||
|
|
8cc04ee568 | ||
|
|
07787b2379 | ||
|
|
0b178bbed9 | ||
|
|
67eb094629 | ||
|
|
813d1c384a | ||
|
|
8dc789f5af | ||
|
|
68a053308a | ||
|
|
0e33d8e612 | ||
|
|
bbe2d7cc1c | ||
|
|
69d4a63b84 | ||
|
|
eb7c5c3cb8 | ||
|
|
bcda7e4c5c | ||
|
|
bcc64c4fd4 | ||
|
|
cbe6b7dbdc | ||
|
|
932cda4d32 | ||
|
|
b6191ba312 | ||
|
|
3313b7c16d | ||
|
|
09e4d4f9fa | ||
|
|
0b94a746e4 | ||
|
|
9d5d140a79 | ||
|
|
766f8c9801 | ||
|
|
5311612295 | ||
|
|
380d70b8a8 | ||
|
|
81cde25fef | ||
|
|
d19df3efbb | ||
|
|
60ea2b494f | ||
|
|
158dd50a01 | ||
|
|
f2bc10d142 | ||
|
|
a7e7aec432 | ||
|
|
8fb83dc432 | ||
|
|
e31ab7251a | ||
|
|
0e5bb01ad6 | ||
|
|
8107a100d4 | ||
|
|
3533899a57 | ||
|
|
320b2e90eb | ||
|
|
4c125cf0f7 | ||
|
|
6168644730 | ||
|
|
6db3ca0968 | ||
|
|
1dc92c3534 | ||
|
|
18c3989395 | ||
|
|
12fcc6ca6d | ||
|
|
bfc8e16c94 | ||
|
|
ea36f31df4 | ||
|
|
a9c02540c2 | ||
|
|
92137a9ebd | ||
|
|
65c2f2d1e2 | ||
|
|
1669efaa29 | ||
|
|
607f6db67f | ||
|
|
26da18fb26 | ||
|
|
6ebd6704e0 | ||
|
|
12649d99da | ||
|
|
6300e6721e | ||
|
|
4ba8d77d79 | ||
|
|
867c16b6ab | ||
|
|
bd30e301e0 | ||
|
|
9b5ccd14e4 | ||
|
|
cdff1a0bc0 | ||
|
|
8e63be5752 | ||
|
|
3a8668e1ad | ||
|
|
d896276a89 | ||
|
|
1cad97b515 | ||
|
|
8c14b86306 | ||
|
|
141f9e37d4 | ||
|
|
17a04afd1e | ||
|
|
cbdaf76265 | ||
|
|
5ddf7e9fdc | ||
|
|
c8dfd39dad | ||
|
|
26e1ae1ff6 | ||
|
|
1416fc82bc | ||
|
|
43ebba2c09 | ||
|
|
70d45c5b86 | ||
|
|
6b205bf537 | ||
|
|
378fa66a1e | ||
|
|
686a3dad04 | ||
|
|
7ca196e45c | ||
|
|
f29ff06263 | ||
|
|
af9b581b4f | ||
|
|
d9f7b3f516 | ||
|
|
7ad84a0ae1 | ||
|
|
e930a124ba | ||
|
|
8f64331c16 | ||
|
|
02b6c05b14 | ||
|
|
6bc97c7e58 |
+47
@@ -9,6 +9,49 @@
|
||||
**/tmp/
|
||||
**/.apt_generated/
|
||||
|
||||
HELP.md
|
||||
target/
|
||||
!.mvn/wrapper/maven-wrapper.jar
|
||||
!**/src/main/**/target/
|
||||
!**/src/test/**/target/
|
||||
|
||||
### maven build ###
|
||||
*.class
|
||||
/out/
|
||||
/build/
|
||||
/target/
|
||||
**/out/
|
||||
**/build/
|
||||
**/target/
|
||||
|
||||
### STS ###
|
||||
.apt_generated
|
||||
.classpath
|
||||
.factorypath
|
||||
.project
|
||||
.settings
|
||||
.springBeans
|
||||
.sts4-cache
|
||||
.gradle
|
||||
|
||||
### IntelliJ IDEA ###
|
||||
.idea
|
||||
*.iws
|
||||
*.iml
|
||||
*.ipr
|
||||
|
||||
### NetBeans ###
|
||||
/nbproject/private/
|
||||
/nbbuild/
|
||||
/dist/
|
||||
/nbdist/
|
||||
/.nb-gradle/
|
||||
build/
|
||||
!**/src/main/**/build/
|
||||
!**/src/test/**/build/
|
||||
|
||||
### VS Code ###
|
||||
.vscode/
|
||||
|
||||
.factorypath
|
||||
.springBeans
|
||||
@@ -26,3 +69,7 @@
|
||||
**/.DS_Store
|
||||
**/classpath-data.json
|
||||
**/dependencies-and-licenses-overview.txt
|
||||
gradle.properties
|
||||
gradlew
|
||||
gradlew.bat
|
||||
gradle/
|
||||
|
||||
+31
-1
@@ -3,4 +3,34 @@ variables:
|
||||
include:
|
||||
- project: 'gitlab/gitlab'
|
||||
ref: 'main'
|
||||
file: 'ci-templates/maven_java.yml'
|
||||
file: 'ci-templates/gradle_java.yml'
|
||||
|
||||
deploy:
|
||||
stage: deploy
|
||||
tags:
|
||||
- dind
|
||||
script:
|
||||
- echo "Building with gradle version ${BUILDVERSION}"
|
||||
- gradle -Pversion=${BUILDVERSION} publish
|
||||
- gradle bootBuildImage --publishImage -PbuildbootDockerHostNetwork=true -Pversion=${BUILDVERSION}
|
||||
- echo "BUILDVERSION=$BUILDVERSION" >> version.env
|
||||
artifacts:
|
||||
reports:
|
||||
dotenv: version.env
|
||||
rules:
|
||||
- if: $CI_COMMIT_BRANCH == $CI_DEFAULT_BRANCH
|
||||
- if: $CI_COMMIT_BRANCH =~ /^release/
|
||||
- if: $CI_COMMIT_TAG
|
||||
|
||||
pages:
|
||||
stage: build
|
||||
tags:
|
||||
- dind
|
||||
script:
|
||||
- echo "Generating Javadoc..."
|
||||
- gradle generateJavaDoc -PjavadocDestinationDir="javadoc"
|
||||
- mkdir public
|
||||
- mv redaction-service-v1/redaction-service-server-v1/javadoc/* public/
|
||||
artifacts:
|
||||
paths:
|
||||
- public
|
||||
|
||||
@@ -0,0 +1,7 @@
|
||||
plugins {
|
||||
`kotlin-dsl`
|
||||
}
|
||||
|
||||
repositories {
|
||||
gradlePluginPortal()
|
||||
}
|
||||
@@ -0,0 +1,60 @@
|
||||
plugins {
|
||||
`java-library`
|
||||
`maven-publish`
|
||||
pmd
|
||||
checkstyle
|
||||
jacoco
|
||||
}
|
||||
|
||||
group = "com.iqser.red.service"
|
||||
|
||||
java.sourceCompatibility = JavaVersion.VERSION_17
|
||||
java.targetCompatibility = JavaVersion.VERSION_17
|
||||
|
||||
pmd {
|
||||
isConsoleOutput = true
|
||||
}
|
||||
|
||||
tasks.pmdMain {
|
||||
pmd.ruleSetFiles = files("${rootDir}/config/pmd/pmd.xml")
|
||||
}
|
||||
|
||||
tasks.pmdTest {
|
||||
pmd.ruleSetFiles = files("${rootDir}/config/pmd/test_pmd.xml")
|
||||
}
|
||||
|
||||
tasks.named<Test>("test") {
|
||||
useJUnitPlatform()
|
||||
reports {
|
||||
junitXml.outputLocation.set(layout.buildDirectory.dir("reports/junit"))
|
||||
}
|
||||
}
|
||||
|
||||
tasks.test {
|
||||
finalizedBy(tasks.jacocoTestReport) // report is always generated after tests run
|
||||
}
|
||||
|
||||
tasks.jacocoTestReport {
|
||||
dependsOn(tasks.test) // tests are required to run before generating the report
|
||||
reports {
|
||||
xml.required.set(true)
|
||||
csv.required.set(false)
|
||||
html.outputLocation.set(layout.buildDirectory.dir("jacocoHtml"))
|
||||
}
|
||||
}
|
||||
|
||||
java {
|
||||
withJavadocJar()
|
||||
}
|
||||
|
||||
repositories {
|
||||
mavenLocal()
|
||||
mavenCentral()
|
||||
maven {
|
||||
url = uri("https://nexus.knecon.com/repository/gindev/");
|
||||
credentials {
|
||||
username = providers.gradleProperty("mavenUser").getOrNull();
|
||||
password = providers.gradleProperty("mavenPassword").getOrNull();
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -0,0 +1,39 @@
|
||||
<?xml version="1.0" encoding="UTF-8"?>
|
||||
<!DOCTYPE module PUBLIC "-//Puppy Crawl//DTD Check Configuration 1.3//EN"
|
||||
"http://www.puppycrawl.com/dtds/configuration_1_3.dtd">
|
||||
<module name="Checker">
|
||||
<property
|
||||
name="severity"
|
||||
value="error"/>
|
||||
<module name="TreeWalker">
|
||||
<module name="SuppressWarningsHolder"/>
|
||||
<module name="MissingDeprecated"/>
|
||||
<module name="MissingOverride"/>
|
||||
<module name="AnnotationLocation"/>
|
||||
<module name="JavadocStyle"/>
|
||||
<module name="NonEmptyAtclauseDescription"/>
|
||||
<module name="IllegalImport"/>
|
||||
<module name="RedundantImport"/>
|
||||
<module name="RedundantModifier"/>
|
||||
<module name="EmptyBlock"/>
|
||||
<module name="DefaultComesLast"/>
|
||||
<module name="EmptyStatement"/>
|
||||
<module name="EqualsHashCode"/>
|
||||
<module name="ExplicitInitialization"/>
|
||||
<module name="IllegalInstantiation"/>
|
||||
<module name="ModifiedControlVariable"/>
|
||||
<module name="MultipleVariableDeclarations"/>
|
||||
<module name="PackageDeclaration"/>
|
||||
<module name="ParameterAssignment"/>
|
||||
<module name="SimplifyBooleanExpression"/>
|
||||
<module name="SimplifyBooleanReturn"/>
|
||||
<module name="StringLiteralEquality"/>
|
||||
<module name="OneStatementPerLine"/>
|
||||
<module name="FinalClass"/>
|
||||
<module name="ArrayTypeStyle"/>
|
||||
<module name="UpperEll"/>
|
||||
<module name="OuterTypeFilename"/>
|
||||
</module>
|
||||
<module name="FileTabCharacter"/>
|
||||
<module name="SuppressWarningsFilter"/>
|
||||
</module>
|
||||
@@ -0,0 +1,20 @@
|
||||
<?xml version="1.0"?>
|
||||
<ruleset name="Custom ruleset"
|
||||
xmlns="http://pmd.sourceforge.net/ruleset/2.0.0"
|
||||
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||
xsi:schemaLocation="http://pmd.sourceforge.net/ruleset/2.0.0 http://pmd.sourceforge.net/ruleset_2_0_0.xsd">
|
||||
|
||||
<description>
|
||||
Knecon ruleset checks the code for bad stuff
|
||||
</description>
|
||||
|
||||
<rule ref="category/java/errorprone.xml">
|
||||
<exclude name="MissingSerialVersionUID"/>
|
||||
<exclude name="AvoidLiteralsInIfCondition"/>
|
||||
<exclude name="AvoidDuplicateLiterals"/>
|
||||
<exclude name="NullAssignment"/>
|
||||
<exclude name="AssignmentInOperand"/>
|
||||
<exclude name="BeanMembersShouldSerialize"/>
|
||||
</rule>
|
||||
|
||||
</ruleset>
|
||||
@@ -0,0 +1,22 @@
|
||||
<?xml version="1.0"?>
|
||||
<ruleset name="Custom ruleset"
|
||||
xmlns="http://pmd.sourceforge.net/ruleset/2.0.0"
|
||||
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||
xsi:schemaLocation="http://pmd.sourceforge.net/ruleset/2.0.0 http://pmd.sourceforge.net/ruleset_2_0_0.xsd">
|
||||
|
||||
<description>
|
||||
Knecon test ruleset checks the code for bad stuff
|
||||
</description>
|
||||
|
||||
|
||||
<rule ref="category/java/errorprone.xml">
|
||||
<exclude name="MissingSerialVersionUID"/>
|
||||
<exclude name="AvoidLiteralsInIfCondition"/>
|
||||
<exclude name="AvoidDuplicateLiterals"/>
|
||||
<exclude name="NullAssignment"/>
|
||||
<exclude name="AssignmentInOperand"/>
|
||||
<exclude name="TestClassWithoutTestCases"/>
|
||||
<exclude name="BeanMembersShouldSerialize"/>
|
||||
</rule>
|
||||
|
||||
</ruleset>
|
||||
@@ -0,0 +1 @@
|
||||
version = 4.0-SNAPSHOT
|
||||
@@ -1,20 +0,0 @@
|
||||
<?xml version="1.0" encoding="UTF-8"?>
|
||||
<project xmlns="http://maven.apache.org/POM/4.0.0"
|
||||
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
|
||||
|
||||
<modelVersion>4.0.0</modelVersion>
|
||||
|
||||
<artifactId>redaction-service</artifactId>
|
||||
<groupId>com.iqser.red.service</groupId>
|
||||
<version>4.0-SNAPSHOT</version>
|
||||
|
||||
|
||||
<packaging>pom</packaging>
|
||||
|
||||
<modules>
|
||||
<module>redaction-service-v1</module>
|
||||
<module>redaction-service-image-v1</module>
|
||||
</modules>
|
||||
|
||||
</project>
|
||||
Executable
+15
@@ -0,0 +1,15 @@
|
||||
#!/bin/bash
|
||||
dir=${PWD##*/}
|
||||
gradle assemble
|
||||
|
||||
# Get the current Git branch
|
||||
branch=$(git rev-parse --abbrev-ref HEAD)
|
||||
|
||||
# Get the short commit hash (first 5 characters)
|
||||
commit_hash=$(git rev-parse --short=5 HEAD)
|
||||
|
||||
# Combine branch and commit hash
|
||||
buildName="${USER}-${branch}-${commit_hash}"
|
||||
|
||||
gradle bootBuildImage --cleanCache --publishImage -PbuildbootDockerHostNetwork=true -Pversion=$buildName
|
||||
echo "nexus.knecon.com:5001/red/${dir}-server-v1:$buildName"
|
||||
@@ -1,98 +0,0 @@
|
||||
<?xml version="1.0" encoding="UTF-8"?>
|
||||
<project xmlns="http://maven.apache.org/POM/4.0.0"
|
||||
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
|
||||
<parent>
|
||||
<groupId>com.knecon.fforesight</groupId>
|
||||
<artifactId>platform-docker-dependency</artifactId>
|
||||
<version>0.1.0</version>
|
||||
<relativePath/>
|
||||
</parent>
|
||||
<modelVersion>4.0.0</modelVersion>
|
||||
|
||||
<artifactId>redaction-service-image-v1</artifactId>
|
||||
<groupId>com.iqser.red.service</groupId>
|
||||
<version>4.0-SNAPSHOT</version>
|
||||
<packaging>pom</packaging>
|
||||
|
||||
|
||||
<properties>
|
||||
<service.server>redaction-service-server-v1</service.server>
|
||||
<platform.jar>${service.server}.jar</platform.jar>
|
||||
<docker.skip.push>false</docker.skip.push>
|
||||
<docker.image.name>${docker.image.prefix}/${service.server}</docker.image.name>
|
||||
</properties>
|
||||
|
||||
<build>
|
||||
<plugins>
|
||||
<plugin>
|
||||
<groupId>org.apache.maven.plugins</groupId>
|
||||
<artifactId>maven-dependency-plugin</artifactId>
|
||||
</plugin>
|
||||
<plugin>
|
||||
<groupId>org.apache.maven.plugins</groupId>
|
||||
<artifactId>maven-resources-plugin</artifactId>
|
||||
</plugin>
|
||||
<plugin>
|
||||
<groupId>org.codehaus.mojo</groupId>
|
||||
<artifactId>exec-maven-plugin</artifactId>
|
||||
</plugin>
|
||||
<plugin>
|
||||
<groupId>io.fabric8</groupId>
|
||||
<artifactId>docker-maven-plugin</artifactId>
|
||||
</plugin>
|
||||
</plugins>
|
||||
|
||||
<pluginManagement>
|
||||
<plugins>
|
||||
<plugin>
|
||||
<groupId>org.apache.maven.plugins</groupId>
|
||||
<artifactId>maven-dependency-plugin</artifactId>
|
||||
<executions>
|
||||
<execution>
|
||||
<id>download-platform-jar</id>
|
||||
<phase>prepare-package</phase>
|
||||
<goals>
|
||||
<goal>copy</goal>
|
||||
</goals>
|
||||
<configuration>
|
||||
<artifactItems>
|
||||
<dependency>
|
||||
<groupId>${project.groupId}</groupId>
|
||||
<artifactId>${service.server}</artifactId>
|
||||
<version>${project.version}</version>
|
||||
<type>jar</type>
|
||||
<overWrite>true</overWrite>
|
||||
<destFileName>${platform.jar}</destFileName>
|
||||
</dependency>
|
||||
</artifactItems>
|
||||
<outputDirectory>${docker.build.directory}</outputDirectory>
|
||||
</configuration>
|
||||
</execution>
|
||||
</executions>
|
||||
</plugin>
|
||||
<plugin>
|
||||
<groupId>io.fabric8</groupId>
|
||||
<artifactId>docker-maven-plugin</artifactId>
|
||||
<configuration>
|
||||
<images>
|
||||
<image>
|
||||
<name>${docker.image.name}</name>
|
||||
<build>
|
||||
<dockerFileDir>${docker.build.directory}</dockerFileDir>
|
||||
<args>
|
||||
<PLATFORM_JAR>${platform.jar}</PLATFORM_JAR>
|
||||
</args>
|
||||
<tags>
|
||||
<tag>${docker.image.version}</tag>
|
||||
<tag>latest</tag>
|
||||
</tags>
|
||||
</build>
|
||||
</image>
|
||||
</images>
|
||||
</configuration>
|
||||
</plugin>
|
||||
</plugins>
|
||||
</pluginManagement>
|
||||
</build>
|
||||
</project>
|
||||
@@ -1,9 +0,0 @@
|
||||
FROM red/redaction-service-base-v1:2.0.0
|
||||
|
||||
ARG PLATFORM_JAR
|
||||
|
||||
ENV PLATFORM_JAR ${PLATFORM_JAR}
|
||||
|
||||
ENV USES_ELASTICSEARCH false
|
||||
|
||||
COPY ["${PLATFORM_JAR}", "/"]
|
||||
@@ -1,114 +0,0 @@
|
||||
<?xml version="1.0" encoding="UTF-8"?>
|
||||
<project xmlns="http://maven.apache.org/POM/4.0.0"
|
||||
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
|
||||
<parent>
|
||||
<artifactId>platform-dependency</artifactId>
|
||||
<groupId>com.iqser.red</groupId>
|
||||
<version>2.2.0</version>
|
||||
<relativePath/>
|
||||
</parent>
|
||||
<modelVersion>4.0.0</modelVersion>
|
||||
|
||||
<artifactId>redaction-service-v1</artifactId>
|
||||
<groupId>com.iqser.red.service</groupId>
|
||||
<version>4.0-SNAPSHOT</version>
|
||||
|
||||
<packaging>pom</packaging>
|
||||
|
||||
<modules>
|
||||
<module>redaction-service-api-v1</module>
|
||||
<module>redaction-service-server-v1</module>
|
||||
</modules>
|
||||
|
||||
<properties>
|
||||
<pdfbox.version>2.0.24</pdfbox.version>
|
||||
<lombok.version>1.18.26</lombok.version>
|
||||
</properties>
|
||||
|
||||
|
||||
<dependencyManagement>
|
||||
<dependencies>
|
||||
<dependency>
|
||||
<groupId>com.iqser.red</groupId>
|
||||
<artifactId>platform-commons-dependency</artifactId>
|
||||
<version>2.1.0</version>
|
||||
<scope>import</scope>
|
||||
<type>pom</type>
|
||||
</dependency>
|
||||
<dependency>
|
||||
<groupId>org.apache.pdfbox</groupId>
|
||||
<artifactId>pdfbox</artifactId>
|
||||
<version>${pdfbox.version}</version>
|
||||
</dependency>
|
||||
<dependency>
|
||||
<groupId>org.apache.pdfbox</groupId>
|
||||
<artifactId>pdfbox-tools</artifactId>
|
||||
<version>${pdfbox.version}</version>
|
||||
</dependency>
|
||||
</dependencies>
|
||||
</dependencyManagement>
|
||||
|
||||
<build>
|
||||
<pluginManagement>
|
||||
<plugins>
|
||||
<plugin>
|
||||
<groupId>org.sonarsource.scanner.maven</groupId>
|
||||
<artifactId>sonar-maven-plugin</artifactId>
|
||||
</plugin>
|
||||
<plugin>
|
||||
<groupId>org.owasp</groupId>
|
||||
<artifactId>dependency-check-maven</artifactId>
|
||||
<configuration>
|
||||
<format>ALL</format>
|
||||
</configuration>
|
||||
</plugin>
|
||||
<plugin>
|
||||
<groupId>org.jacoco</groupId>
|
||||
<artifactId>jacoco-maven-plugin</artifactId>
|
||||
<version>0.8.8</version>
|
||||
<configuration>
|
||||
<excludes>
|
||||
<exclude>org/drools/**/*</exclude>
|
||||
</excludes>
|
||||
</configuration>
|
||||
<executions>
|
||||
<execution>
|
||||
<id>prepare-agent</id>
|
||||
<goals>
|
||||
<goal>prepare-agent</goal>
|
||||
</goals>
|
||||
</execution>
|
||||
<execution>
|
||||
<id>report</id>
|
||||
<goals>
|
||||
<goal>report</goal>
|
||||
</goals>
|
||||
</execution>
|
||||
</executions>
|
||||
</plugin>
|
||||
</plugins>
|
||||
</pluginManagement>
|
||||
<plugins>
|
||||
<plugin>
|
||||
<groupId>org.projectlombok</groupId>
|
||||
<artifactId>lombok-maven-plugin</artifactId>
|
||||
<version>1.18.20.0</version>
|
||||
<executions>
|
||||
<execution>
|
||||
<id>delombok</id>
|
||||
<phase>generate-sources</phase>
|
||||
<goals>
|
||||
<goal>delombok</goal>
|
||||
</goals>
|
||||
<configuration>
|
||||
<addOutputDirectory>false</addOutputDirectory>
|
||||
<sourceDirectory>src/main/java</sourceDirectory>
|
||||
<outputDirectory>${delomboked.sources}</outputDirectory>
|
||||
</configuration>
|
||||
</execution>
|
||||
</executions>
|
||||
</plugin>
|
||||
</plugins>
|
||||
</build>
|
||||
</project>
|
||||
@@ -0,0 +1,28 @@
|
||||
plugins {
|
||||
id("com.iqser.red.service.java-conventions")
|
||||
id("io.freefair.lombok") version "8.4"
|
||||
}
|
||||
|
||||
description = "redaction-service-api-v1"
|
||||
|
||||
dependencies {
|
||||
implementation("org.springframework:spring-web:6.0.12")
|
||||
implementation("com.iqser.red.service:persistence-service-internal-api-v1:2.351.0")
|
||||
}
|
||||
|
||||
publishing {
|
||||
publications {
|
||||
create<MavenPublication>(name) {
|
||||
from(components["java"])
|
||||
}
|
||||
}
|
||||
repositories {
|
||||
maven {
|
||||
url = uri("https://nexus.knecon.com/repository/red-platform-releases/")
|
||||
credentials {
|
||||
username = providers.gradleProperty("mavenUser").getOrNull();
|
||||
password = providers.gradleProperty("mavenPassword").getOrNull();
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -1,61 +0,0 @@
|
||||
<?xml version="1.0" encoding="UTF-8"?>
|
||||
<project xmlns="http://maven.apache.org/POM/4.0.0"
|
||||
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
|
||||
<modelVersion>4.0.0</modelVersion>
|
||||
<parent>
|
||||
<artifactId>redaction-service-v1</artifactId>
|
||||
<groupId>com.iqser.red.service</groupId>
|
||||
<version>4.0-SNAPSHOT</version>
|
||||
</parent>
|
||||
|
||||
<artifactId>redaction-service-api-v1</artifactId>
|
||||
|
||||
<properties>
|
||||
<persistence-service.version>2.93.0</persistence-service.version>
|
||||
</properties>
|
||||
|
||||
<dependencies>
|
||||
|
||||
<dependency>
|
||||
<groupId>org.springframework</groupId>
|
||||
<artifactId>spring-web</artifactId>
|
||||
<optional>true</optional>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>com.iqser.red.service</groupId>
|
||||
<artifactId>persistence-service-internal-api-v1</artifactId>
|
||||
<version>${persistence-service.version}</version>
|
||||
<exclusions>
|
||||
|
||||
<exclusion>
|
||||
<groupId>com.iqser.red.service</groupId>
|
||||
<artifactId>redaction-service-api-v1</artifactId>
|
||||
</exclusion>
|
||||
<exclusion>
|
||||
<groupId>com.iqser.red.service</groupId>
|
||||
<artifactId>persistence-service-api-v1</artifactId>
|
||||
</exclusion>
|
||||
</exclusions>
|
||||
</dependency>
|
||||
</dependencies>
|
||||
|
||||
<build>
|
||||
<plugins>
|
||||
<plugin>
|
||||
<groupId>org.apache.maven.plugins</groupId>
|
||||
<artifactId>maven-compiler-plugin</artifactId>
|
||||
<configuration>
|
||||
<annotationProcessorPaths>
|
||||
<path>
|
||||
<groupId>org.projectlombok</groupId>
|
||||
<artifactId>lombok</artifactId>
|
||||
<version>${lombok.version}</version>
|
||||
</path>
|
||||
</annotationProcessorPaths>
|
||||
</configuration>
|
||||
</plugin>
|
||||
</plugins>
|
||||
</build>
|
||||
</project>
|
||||
+18
@@ -0,0 +1,18 @@
|
||||
package com.iqser.red.service.redaction.v1.model;
|
||||
|
||||
import java.util.List;
|
||||
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Builder;
|
||||
import lombok.Data;
|
||||
import lombok.NoArgsConstructor;
|
||||
|
||||
@Data
|
||||
@Builder
|
||||
@NoArgsConstructor
|
||||
@AllArgsConstructor
|
||||
public class AnalyzeResponse {
|
||||
|
||||
private String fileId;
|
||||
private List<UnprocessedManualEntity> unprocessedManualEntities;
|
||||
}
|
||||
+20
@@ -0,0 +1,20 @@
|
||||
package com.iqser.red.service.redaction.v1.model;
|
||||
|
||||
import lombok.AccessLevel;
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Builder;
|
||||
import lombok.Data;
|
||||
import lombok.NoArgsConstructor;
|
||||
import lombok.experimental.FieldDefaults;
|
||||
|
||||
@Data
|
||||
@Builder
|
||||
@AllArgsConstructor
|
||||
@NoArgsConstructor
|
||||
@FieldDefaults(level = AccessLevel.PRIVATE)
|
||||
public class DroolsSyntaxDeprecatedWarnings {
|
||||
|
||||
Integer line;
|
||||
Integer column;
|
||||
String message;
|
||||
}
|
||||
+20
@@ -0,0 +1,20 @@
|
||||
package com.iqser.red.service.redaction.v1.model;
|
||||
|
||||
import lombok.AccessLevel;
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Builder;
|
||||
import lombok.Data;
|
||||
import lombok.NoArgsConstructor;
|
||||
import lombok.experimental.FieldDefaults;
|
||||
|
||||
@Data
|
||||
@Builder
|
||||
@AllArgsConstructor
|
||||
@NoArgsConstructor
|
||||
@FieldDefaults(level = AccessLevel.PRIVATE)
|
||||
public class DroolsSyntaxErrorMessage {
|
||||
|
||||
Integer line;
|
||||
Integer column;
|
||||
String message;
|
||||
}
|
||||
+36
@@ -0,0 +1,36 @@
|
||||
package com.iqser.red.service.redaction.v1.model;
|
||||
|
||||
import java.util.LinkedList;
|
||||
import java.util.List;
|
||||
|
||||
import lombok.AccessLevel;
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Builder;
|
||||
import lombok.Data;
|
||||
import lombok.NoArgsConstructor;
|
||||
import lombok.experimental.FieldDefaults;
|
||||
|
||||
@Data
|
||||
@Builder
|
||||
@NoArgsConstructor
|
||||
@AllArgsConstructor
|
||||
@FieldDefaults(level = AccessLevel.PRIVATE)
|
||||
public class DroolsSyntaxValidation {
|
||||
|
||||
@Builder.Default
|
||||
List<DroolsSyntaxErrorMessage> droolsSyntaxErrorMessages = new LinkedList<>();
|
||||
@Builder.Default
|
||||
List<DroolsSyntaxDeprecatedWarnings> droolsSyntaxDeprecatedWarnings = new LinkedList<>();
|
||||
|
||||
|
||||
public void addErrorMessage(int line, int column, String message) {
|
||||
|
||||
getDroolsSyntaxErrorMessages().add(DroolsSyntaxErrorMessage.builder().line(line).column(column).message(message).build());
|
||||
}
|
||||
|
||||
public boolean isCompiled() {
|
||||
|
||||
return droolsSyntaxErrorMessages.isEmpty();
|
||||
}
|
||||
|
||||
}
|
||||
+21
@@ -0,0 +1,21 @@
|
||||
package com.iqser.red.service.redaction.v1.model;
|
||||
|
||||
import com.iqser.red.service.persistence.service.v1.api.shared.model.annotations.ManualRedactions;
|
||||
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Builder;
|
||||
import lombok.Data;
|
||||
import lombok.NoArgsConstructor;
|
||||
|
||||
@Data
|
||||
@Builder
|
||||
@NoArgsConstructor
|
||||
@AllArgsConstructor
|
||||
public class MigrationRequest {
|
||||
|
||||
String dossierTemplateId;
|
||||
String dossierId;
|
||||
String fileId;
|
||||
ManualRedactions manualRedactions;
|
||||
|
||||
}
|
||||
+17
@@ -0,0 +1,17 @@
|
||||
package com.iqser.red.service.redaction.v1.model;
|
||||
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Builder;
|
||||
import lombok.Data;
|
||||
import lombok.NoArgsConstructor;
|
||||
|
||||
@Data
|
||||
@Builder
|
||||
@NoArgsConstructor
|
||||
@AllArgsConstructor
|
||||
public class MigrationResponse {
|
||||
|
||||
String dossierId;
|
||||
String fileId;
|
||||
|
||||
}
|
||||
+14
@@ -0,0 +1,14 @@
|
||||
package com.iqser.red.service.redaction.v1.model;
|
||||
|
||||
public class QueueNames {
|
||||
|
||||
public static final String REDACTION_QUEUE = "redactionQueue";
|
||||
public static final String REDACTION_PRIORITY_QUEUE = "redactionPriorityQueue";
|
||||
public static final String REDACTION_ANALYSIS_RESPONSE_QUEUE = "redactionAnalysisResponseQueue";
|
||||
public static final String REDACTION_DQL = "redactionDQL";
|
||||
|
||||
public static final String MIGRATION_QUEUE = "migrationQueue";
|
||||
public static final String MIGRATION_RESPONSE_QUEUE = "migrationResponseQueue";
|
||||
public static final String MIGRATION_DLQ = "migrationDLQ";
|
||||
|
||||
}
|
||||
+18
@@ -0,0 +1,18 @@
|
||||
package com.iqser.red.service.redaction.v1.model;
|
||||
|
||||
import lombok.AccessLevel;
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Data;
|
||||
import lombok.NoArgsConstructor;
|
||||
import lombok.experimental.FieldDefaults;
|
||||
|
||||
@Data
|
||||
@AllArgsConstructor
|
||||
@NoArgsConstructor
|
||||
@FieldDefaults(level = AccessLevel.PRIVATE)
|
||||
public class RuleValidationModel {
|
||||
|
||||
String ruleFileType;
|
||||
String rulesString;
|
||||
|
||||
}
|
||||
+27
@@ -0,0 +1,27 @@
|
||||
package com.iqser.red.service.redaction.v1.model;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.List;
|
||||
|
||||
import com.iqser.red.service.persistence.service.v1.api.shared.model.analysislog.entitylog.Position;
|
||||
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Builder;
|
||||
import lombok.Data;
|
||||
import lombok.EqualsAndHashCode;
|
||||
import lombok.NoArgsConstructor;
|
||||
|
||||
@Data
|
||||
@AllArgsConstructor
|
||||
@NoArgsConstructor
|
||||
@Builder
|
||||
@EqualsAndHashCode
|
||||
public class UnprocessedManualEntity {
|
||||
|
||||
private String annotationId;
|
||||
private String textBefore;
|
||||
private String textAfter;
|
||||
private String section;
|
||||
@Builder.Default
|
||||
private List<Position> positions = new ArrayList<>();
|
||||
}
|
||||
+4
-1
@@ -4,9 +4,12 @@ import org.springframework.http.MediaType;
|
||||
import org.springframework.web.bind.annotation.PostMapping;
|
||||
import org.springframework.web.bind.annotation.RequestBody;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.model.DroolsSyntaxValidation;
|
||||
import com.iqser.red.service.redaction.v1.model.RuleValidationModel;
|
||||
|
||||
public interface RedactionResource {
|
||||
|
||||
@PostMapping(value = "/rules/test", consumes = MediaType.APPLICATION_JSON_VALUE)
|
||||
void testRules(@RequestBody String rules);
|
||||
DroolsSyntaxValidation testRules(@RequestBody RuleValidationModel rulesValidationModel);
|
||||
|
||||
}
|
||||
|
||||
@@ -0,0 +1,149 @@
|
||||
import org.springframework.boot.gradle.tasks.bundling.BootBuildImage
|
||||
|
||||
plugins {
|
||||
application
|
||||
id("com.iqser.red.service.java-conventions")
|
||||
id("org.springframework.boot") version "3.1.5"
|
||||
id("io.spring.dependency-management") version "1.1.3"
|
||||
id("org.sonarqube") version "4.4.1.3373"
|
||||
id("io.freefair.lombok") version "8.4"
|
||||
}
|
||||
|
||||
description = "redaction-service-server-v1"
|
||||
|
||||
|
||||
val layoutParserVersion = "0.94.0"
|
||||
val jacksonVersion = "2.15.2"
|
||||
val droolsVersion = "9.44.0.Final"
|
||||
val pdfBoxVersion = "3.0.0"
|
||||
val persistenceServiceVersion = "2.360.0"
|
||||
val springBootStarterVersion = "3.1.5"
|
||||
|
||||
configurations {
|
||||
all {
|
||||
exclude(group = "commons-logging", module = "commons-logging")
|
||||
exclude(group = "org.springframework.boot", module = "spring-boot-starter-log4j2")
|
||||
exclude(group = "com.iqser.red.commons", module = "logging-commons")
|
||||
}
|
||||
}
|
||||
|
||||
dependencies {
|
||||
|
||||
implementation(project(":redaction-service-api-v1")) { exclude(group = "com.iqser.red.service", module = "persistence-service-internal-api-v1") }
|
||||
implementation("com.iqser.red.service:persistence-service-internal-api-v1:${persistenceServiceVersion}") { exclude(group = "org.springframework.boot") }
|
||||
implementation("com.knecon.fforesight:layoutparser-service-internal-api:${layoutParserVersion}")
|
||||
|
||||
implementation("com.iqser.red.commons:spring-commons:6.2.0")
|
||||
implementation("com.iqser.red.commons:metric-commons:2.3.0")
|
||||
|
||||
implementation("com.iqser.red.commons:dictionary-merge-commons:1.5.0")
|
||||
implementation("com.iqser.red.commons:storage-commons:2.45.0")
|
||||
implementation("com.knecon.fforesight:tenant-commons:0.21.0")
|
||||
implementation("com.knecon.fforesight:tracing-commons:0.5.0")
|
||||
|
||||
implementation("com.fasterxml.jackson.module:jackson-module-afterburner:${jacksonVersion}")
|
||||
implementation("com.fasterxml.jackson.datatype:jackson-datatype-jsr310:${jacksonVersion}")
|
||||
implementation("org.ahocorasick:ahocorasick:0.6.3")
|
||||
implementation("org.javassist:javassist:3.29.2-GA")
|
||||
|
||||
implementation("org.drools:drools-engine:${droolsVersion}")
|
||||
implementation("org.drools:drools-mvel:${droolsVersion}")
|
||||
implementation("org.kie:kie-spring:7.74.1.Final")
|
||||
|
||||
implementation("org.locationtech.jts:jts-core:1.19.0")
|
||||
|
||||
implementation("org.springframework.cloud:spring-cloud-starter-openfeign:4.0.4")
|
||||
implementation("org.springframework.boot:spring-boot-starter-amqp:${springBootStarterVersion}")
|
||||
implementation("org.springframework.boot:spring-boot-starter-cache:${springBootStarterVersion}")
|
||||
implementation("org.springframework.boot:spring-boot-starter-data-redis:${springBootStarterVersion}")
|
||||
|
||||
implementation("net.logstash.logback:logstash-logback-encoder:7.4")
|
||||
implementation("ch.qos.logback:logback-classic")
|
||||
|
||||
implementation("org.reflections:reflections:0.10.2")
|
||||
|
||||
testImplementation(project(":rules-management"))
|
||||
testImplementation("org.apache.pdfbox:pdfbox:${pdfBoxVersion}")
|
||||
testImplementation("org.apache.pdfbox:pdfbox-tools:${pdfBoxVersion}")
|
||||
|
||||
testImplementation("org.springframework.boot:spring-boot-starter-test:${springBootStarterVersion}")
|
||||
testImplementation("com.knecon.fforesight:viewer-doc-processor:${layoutParserVersion}")
|
||||
testImplementation("com.knecon.fforesight:layoutparser-service-processor:${layoutParserVersion}") {
|
||||
exclude(
|
||||
group = "com.iqser.red.service",
|
||||
module = "persistence-service-shared-api-v1"
|
||||
)
|
||||
}
|
||||
}
|
||||
|
||||
tasks.test {
|
||||
configure<JacocoTaskExtension> {
|
||||
excludes = listOf("org/drools/**/*")
|
||||
}
|
||||
minHeapSize = "512m"
|
||||
maxHeapSize = "1024m"
|
||||
}
|
||||
|
||||
tasks.named<BootBuildImage>("bootBuildImage") {
|
||||
|
||||
environment.put("BPE_DELIM_JAVA_TOOL_OPTIONS", " ")
|
||||
environment.put("BPE_APPEND_JAVA_TOOL_OPTIONS", "-XX:MaxMetaspaceSize=512m -Dfile.encoding=UTF-8")
|
||||
|
||||
imageName.set("nexus.knecon.com:5001/red/${project.name}")// must build image with same name always, otherwise the builder will not know which image to use as cache. DO NOT CHANGE!
|
||||
if (project.hasProperty("buildbootDockerHostNetwork")) {
|
||||
network.set("host")
|
||||
}
|
||||
docker {
|
||||
if (project.hasProperty("buildbootDockerHostNetwork")) {
|
||||
bindHostToBuilder.set(true)
|
||||
}
|
||||
verboseLogging.set(true)
|
||||
|
||||
publishRegistry {
|
||||
username.set(providers.gradleProperty("mavenUser").getOrNull())
|
||||
password.set(providers.gradleProperty("mavenPassword").getOrNull())
|
||||
email.set(providers.gradleProperty("mavenEmail").getOrNull())
|
||||
url.set("https://nexus.knecon.com:5001/")
|
||||
}
|
||||
|
||||
val dockerTag = "nexus.knecon.com:5001/red/${project.name}:${project.version}"
|
||||
tags.set(listOf(dockerTag))
|
||||
}
|
||||
}
|
||||
|
||||
fun parseDroolsImports(droolsFilePath: String): List<String> {
|
||||
val imports = mutableListOf<String>()
|
||||
val importPattern = Regex("^import\\s+(com\\.iqser\\.red\\.service\\.redaction\\.v1\\.[\\w.]+);")
|
||||
val desiredPrefix = "com.iqser.red.service.redaction.v1"
|
||||
|
||||
File(droolsFilePath).forEachLine { line ->
|
||||
importPattern.find(line)?.let { matchResult ->
|
||||
val importPath = matchResult.groupValues[1].trim()
|
||||
if (importPath.startsWith(desiredPrefix)) {
|
||||
val formattedPath = importPath.replace('.', '/')
|
||||
imports.add("$formattedPath.java")
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
return imports
|
||||
}
|
||||
|
||||
val droolsImports = parseDroolsImports("redaction-service-v1/redaction-service-server-v1/src/main/resources/drools/all_rules_documine.drl")
|
||||
|
||||
tasks.register("generateJavaDoc", Javadoc::class) {
|
||||
dependsOn("compileJava")
|
||||
dependsOn("delombok")
|
||||
classpath = project.sourceSets["main"].runtimeClasspath
|
||||
source = fileTree("${buildDir}/generated/sources/delombok/java/main") {
|
||||
include(droolsImports)
|
||||
}
|
||||
destinationDir = file(project.findProperty("javadocDestinationDir")?.toString() ?: "")
|
||||
|
||||
options.memberLevel = JavadocMemberLevel.PUBLIC
|
||||
(options as StandardJavadocDocletOptions).apply {
|
||||
header = "Redaction Service ${project.version}"
|
||||
footer = "Redaction Service ${project.version}"
|
||||
title = "API Documentation for Redaction Service ${project.version}"
|
||||
}
|
||||
}
|
||||
@@ -1,217 +0,0 @@
|
||||
<?xml version="1.0" encoding="UTF-8"?>
|
||||
<project xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
|
||||
xmlns="http://maven.apache.org/POM/4.0.0"
|
||||
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
|
||||
<modelVersion>4.0.0</modelVersion>
|
||||
<parent>
|
||||
<artifactId>redaction-service-v1</artifactId>
|
||||
<groupId>com.iqser.red.service</groupId>
|
||||
<version>4.0-SNAPSHOT</version>
|
||||
</parent>
|
||||
|
||||
<artifactId>redaction-service-server-v1</artifactId>
|
||||
|
||||
<properties>
|
||||
<drools.version>8.37.0.Final</drools.version>
|
||||
<kie.version>7.73.0.Final</kie.version>
|
||||
<locationtech.version>1.19.0</locationtech.version>
|
||||
<javaassist.version>3.29.2-GA</javaassist.version>
|
||||
<ahocorasick.version>0.6.3</ahocorasick.version>
|
||||
<jackson.version>2.14.2</jackson.version>
|
||||
<tennat-commons.version>0.10.0</tennat-commons.version>
|
||||
</properties>
|
||||
|
||||
<dependencies>
|
||||
<dependency>
|
||||
<groupId>com.knecon.fforesight</groupId>
|
||||
<artifactId>tenant-commons</artifactId>
|
||||
<version>${tennat-commons.version}</version>
|
||||
</dependency>
|
||||
<dependency>
|
||||
<groupId>org.springframework.boot</groupId>
|
||||
<artifactId>spring-boot-starter-aop</artifactId>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>com.iqser.red.commons</groupId>
|
||||
<artifactId>dictionary-merge-commons</artifactId>
|
||||
<version>1.3.0</version>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>com.iqser.red.commons</groupId>
|
||||
<artifactId>storage-commons</artifactId>
|
||||
<version>2.1.0</version>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>com.fasterxml.jackson.module</groupId>
|
||||
<artifactId>jackson-module-afterburner</artifactId>
|
||||
<version>${jackson.version}</version>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>com.fasterxml.jackson.datatype</groupId>
|
||||
<artifactId>jackson-datatype-jsr310</artifactId>
|
||||
<version>${jackson.version}</version>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>org.ahocorasick</groupId>
|
||||
<artifactId>ahocorasick</artifactId>
|
||||
<version>${ahocorasick.version}</version>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>org.javassist</groupId>
|
||||
<artifactId>javassist</artifactId>
|
||||
<version>${javaassist.version}</version>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>com.iqser.red.service</groupId>
|
||||
<artifactId>redaction-service-api-v1</artifactId>
|
||||
<version>${project.version}</version>
|
||||
</dependency>
|
||||
<dependency>
|
||||
<groupId>org.drools</groupId>
|
||||
<artifactId>drools-engine</artifactId>
|
||||
<version>${drools.version}</version>
|
||||
</dependency>
|
||||
<dependency>
|
||||
<groupId>org.drools</groupId>
|
||||
<artifactId>drools-mvel</artifactId>
|
||||
<version>${drools.version}</version>
|
||||
</dependency>
|
||||
<dependency>
|
||||
<groupId>org.kie</groupId>
|
||||
<artifactId>kie-spring</artifactId>
|
||||
<version>${kie.version}</version>
|
||||
</dependency>
|
||||
<dependency>
|
||||
<groupId>org.locationtech.jts</groupId>
|
||||
<artifactId>jts-core</artifactId>
|
||||
<version>${locationtech.version}</version>
|
||||
</dependency>
|
||||
<dependency>
|
||||
<groupId>com.google.guava</groupId>
|
||||
<artifactId>guava</artifactId>
|
||||
</dependency>
|
||||
|
||||
<!-- commons -->
|
||||
<dependency>
|
||||
<groupId>com.iqser.red.commons</groupId>
|
||||
<artifactId>spring-commons</artifactId>
|
||||
</dependency>
|
||||
<dependency>
|
||||
<groupId>com.iqser.red.commons</groupId>
|
||||
<artifactId>logging-commons</artifactId>
|
||||
</dependency>
|
||||
<dependency>
|
||||
<groupId>com.iqser.red.commons</groupId>
|
||||
<artifactId>metric-commons</artifactId>
|
||||
</dependency>
|
||||
<!-- other external -->
|
||||
<dependency>
|
||||
<groupId>org.apache.pdfbox</groupId>
|
||||
<artifactId>pdfbox</artifactId>
|
||||
</dependency>
|
||||
<dependency>
|
||||
<groupId>org.apache.pdfbox</groupId>
|
||||
<artifactId>pdfbox-tools</artifactId>
|
||||
</dependency>
|
||||
<!-- spring -->
|
||||
<dependency>
|
||||
<groupId>org.springframework.cloud</groupId>
|
||||
<artifactId>spring-cloud-starter-openfeign</artifactId>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>org.springframework.boot</groupId>
|
||||
<artifactId>spring-boot-starter-amqp</artifactId>
|
||||
</dependency>
|
||||
|
||||
<!-- test dependencies -->
|
||||
<dependency>
|
||||
<groupId>org.springframework.boot</groupId>
|
||||
<artifactId>spring-boot-starter-test</artifactId>
|
||||
<scope>test</scope>
|
||||
</dependency>
|
||||
<dependency>
|
||||
<groupId>com.iqser.red.commons</groupId>
|
||||
<artifactId>test-commons</artifactId>
|
||||
<scope>test</scope>
|
||||
</dependency>
|
||||
</dependencies>
|
||||
|
||||
<build>
|
||||
<plugins>
|
||||
|
||||
<plugin>
|
||||
<groupId>org.apache.maven.plugins</groupId>
|
||||
<artifactId>maven-compiler-plugin</artifactId>
|
||||
<configuration>
|
||||
<annotationProcessorPaths>
|
||||
<path>
|
||||
<groupId>org.projectlombok</groupId>
|
||||
<artifactId>lombok</artifactId>
|
||||
<version>${lombok.version}</version>
|
||||
</path>
|
||||
</annotationProcessorPaths>
|
||||
</configuration>
|
||||
</plugin>
|
||||
|
||||
<plugin>
|
||||
<!-- generate git.properties for exposure in /info -->
|
||||
<groupId>pl.project13.maven</groupId>
|
||||
<artifactId>git-commit-id-plugin</artifactId>
|
||||
<executions>
|
||||
<execution>
|
||||
<goals>
|
||||
<goal>revision</goal>
|
||||
</goals>
|
||||
<configuration>
|
||||
<generateGitPropertiesFile>true</generateGitPropertiesFile>
|
||||
<gitDescribe>
|
||||
<tags>true</tags>
|
||||
</gitDescribe>
|
||||
</configuration>
|
||||
</execution>
|
||||
</executions>
|
||||
</plugin>
|
||||
|
||||
<plugin>
|
||||
<groupId>org.apache.maven.plugins</groupId>
|
||||
<artifactId>maven-jar-plugin</artifactId>
|
||||
<executions>
|
||||
<execution>
|
||||
<id>original-jar</id>
|
||||
<goals>
|
||||
<goal>jar</goal>
|
||||
</goals>
|
||||
<configuration>
|
||||
<classifier>original</classifier>
|
||||
</configuration>
|
||||
</execution>
|
||||
</executions>
|
||||
</plugin>
|
||||
|
||||
<plugin>
|
||||
<!-- repackages the generated jar into a runnable fat-jar and makes it
|
||||
executable -->
|
||||
<groupId>org.springframework.boot</groupId>
|
||||
<artifactId>spring-boot-maven-plugin</artifactId>
|
||||
<executions>
|
||||
<execution>
|
||||
<goals>
|
||||
<goal>repackage</goal>
|
||||
</goals>
|
||||
<configuration>
|
||||
<executable>true</executable>
|
||||
</configuration>
|
||||
</execution>
|
||||
</executions>
|
||||
</plugin>
|
||||
</plugins>
|
||||
</build>
|
||||
</project>
|
||||
+16
-1
@@ -6,19 +6,22 @@ import org.springframework.boot.autoconfigure.ImportAutoConfiguration;
|
||||
import org.springframework.boot.autoconfigure.SpringBootApplication;
|
||||
import org.springframework.boot.autoconfigure.security.servlet.SecurityAutoConfiguration;
|
||||
import org.springframework.boot.context.properties.EnableConfigurationProperties;
|
||||
import org.springframework.cache.annotation.EnableCaching;
|
||||
import org.springframework.cloud.openfeign.EnableFeignClients;
|
||||
import org.springframework.context.annotation.Bean;
|
||||
import org.springframework.context.annotation.Import;
|
||||
|
||||
import com.iqser.red.service.dictionarymerge.commons.DictionaryMergeService;
|
||||
import com.iqser.red.service.redaction.v1.server.client.RulesClient;
|
||||
import com.iqser.red.service.redaction.v1.server.settings.RedactionServiceSettings;
|
||||
import com.iqser.red.storage.commons.StorageAutoConfiguration;
|
||||
import com.knecon.fforesight.tenantcommons.MultiTenancyAutoConfiguration;
|
||||
|
||||
import io.micrometer.core.aop.TimedAspect;
|
||||
import io.micrometer.core.instrument.MeterRegistry;
|
||||
import io.micrometer.observation.ObservationRegistry;
|
||||
import io.micrometer.observation.aop.ObservedAspect;
|
||||
|
||||
@EnableCaching
|
||||
@ImportAutoConfiguration({MultiTenancyAutoConfiguration.class})
|
||||
@Import({MetricsConfiguration.class, StorageAutoConfiguration.class})
|
||||
@EnableFeignClients(basePackageClasses = RulesClient.class)
|
||||
@@ -32,6 +35,10 @@ public class Application {
|
||||
SpringApplication.run(Application.class, args);
|
||||
}
|
||||
|
||||
@Bean
|
||||
public ObservedAspect observedAspect(ObservationRegistry observationRegistry) {
|
||||
return new ObservedAspect(observationRegistry);
|
||||
}
|
||||
|
||||
@Bean
|
||||
public TimedAspect timedAspect(MeterRegistry registry) {
|
||||
@@ -39,10 +46,18 @@ public class Application {
|
||||
return new TimedAspect(registry);
|
||||
}
|
||||
|
||||
|
||||
@Bean
|
||||
public DictionaryMergeService dictionaryMergeService() {
|
||||
|
||||
return new DictionaryMergeService();
|
||||
}
|
||||
|
||||
|
||||
@Bean(initMethod = "findDeprecatedMethods")
|
||||
public DeprecatedElementsFinder deprecatedElementsFinder() {
|
||||
|
||||
return new DeprecatedElementsFinder();
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
+108
@@ -0,0 +1,108 @@
|
||||
package com.iqser.red.service.redaction.v1.server;
|
||||
|
||||
import static java.util.stream.Collectors.groupingBy;
|
||||
|
||||
import java.lang.reflect.Method;
|
||||
import java.lang.reflect.Type;
|
||||
import java.util.Arrays;
|
||||
import java.util.HashMap;
|
||||
import java.util.HashSet;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
import java.util.Set;
|
||||
import java.util.stream.Collectors;
|
||||
|
||||
import org.reflections.Reflections;
|
||||
import org.reflections.scanners.Scanners;
|
||||
import org.reflections.util.ConfigurationBuilder;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.model.dictionary.SearchImplementation;
|
||||
|
||||
import lombok.Getter;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
|
||||
@Slf4j
|
||||
public class DeprecatedElementsFinder {
|
||||
|
||||
public static final String PACKAGE_NAME = "com.iqser.red.service.redaction.v1.server";
|
||||
private Set<Method> deprecatedMethods;
|
||||
@Getter
|
||||
private Map<String, String> deprecatedMethodsSignaturesMap;
|
||||
private Set<Class<?>> deprecatedClasses;
|
||||
@Getter
|
||||
private SearchImplementation classesSearchImplementation;
|
||||
@Getter
|
||||
private SearchImplementation methodsSearchImplementation;
|
||||
|
||||
|
||||
public void findDeprecatedMethods() {
|
||||
|
||||
deprecatedMethods = new HashSet<>();
|
||||
deprecatedClasses = new HashSet<>();
|
||||
deprecatedMethodsSignaturesMap = new HashMap<>();
|
||||
|
||||
Reflections reflections = new Reflections(new ConfigurationBuilder().forPackage(PACKAGE_NAME)
|
||||
.setExpandSuperTypes(true)
|
||||
.setScanners(Scanners.MethodsAnnotated, Scanners.TypesAnnotated, Scanners.SubTypes));
|
||||
|
||||
deprecatedMethods = reflections.get(Scanners.MethodsAnnotated.with(Deprecated.class).as(Method.class));
|
||||
|
||||
deprecatedClasses = reflections.get(Scanners.TypesAnnotated.with(Deprecated.class).asClass());
|
||||
|
||||
if (!deprecatedClasses.isEmpty()) {
|
||||
deprecatedClasses.forEach(deprecatedType -> deprecatedMethods.addAll(Arrays.stream(deprecatedType.getDeclaredMethods())
|
||||
.collect(Collectors.toSet())));
|
||||
classesSearchImplementation = new SearchImplementation(deprecatedClasses.stream()
|
||||
.map(Class::getName)
|
||||
.collect(Collectors.toSet()), false);
|
||||
}
|
||||
|
||||
if (!deprecatedMethods.isEmpty()) {
|
||||
methodsSearchImplementation = new SearchImplementation(deprecatedMethods.stream()
|
||||
.map(Method::getName)
|
||||
.collect(Collectors.toSet()), false);
|
||||
deprecatedMethodsSignaturesMap = new HashMap<>();
|
||||
Map<String, List<Method>> deprecatedMethodsNameMap = deprecatedMethods.stream()
|
||||
.collect(groupingBy(Method::getName));
|
||||
for (String deprecatedMethod : deprecatedMethodsNameMap.keySet()) {
|
||||
deprecatedMethodsSignaturesMap.computeIfAbsent(deprecatedMethod,
|
||||
key -> deprecatedMethodsNameMap.get(deprecatedMethod)
|
||||
.stream()
|
||||
.map(this::getMethodSignature)
|
||||
.collect(Collectors.joining("\n")));
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
log.info("For package: {} found deprecated classes {} deprecated methods {}", PACKAGE_NAME, deprecatedClasses.size(), deprecatedMethods.size());
|
||||
}
|
||||
|
||||
|
||||
public Set<Method> getDeprecatedMethods() {
|
||||
|
||||
if (this.deprecatedMethods == null) {
|
||||
findDeprecatedMethods();
|
||||
}
|
||||
return this.deprecatedMethods;
|
||||
}
|
||||
|
||||
|
||||
public Set<Class<?>> getDeprecatedClasses() {
|
||||
|
||||
if (this.deprecatedClasses == null) {
|
||||
findDeprecatedMethods();
|
||||
}
|
||||
return this.deprecatedClasses;
|
||||
}
|
||||
|
||||
private String getMethodSignature(Method method) {
|
||||
|
||||
String methodName = method.getName();
|
||||
String parameterTypes = Arrays.stream(method.getGenericParameterTypes())
|
||||
.map(Type::getTypeName)
|
||||
.collect(Collectors.joining(", "));
|
||||
|
||||
return method.getGenericReturnType() + " " + methodName + "(" + parameterTypes + ")";
|
||||
}
|
||||
|
||||
}
|
||||
+3
-1
@@ -1,4 +1,4 @@
|
||||
package com.iqser.red.service.redaction.v1.server.settings;
|
||||
package com.iqser.red.service.redaction.v1.server;
|
||||
|
||||
import org.springframework.boot.context.properties.ConfigurationProperties;
|
||||
|
||||
@@ -28,4 +28,6 @@ public class RedactionServiceSettings {
|
||||
|
||||
private int dictionaryCacheExpireAfterAccessDays = 3;
|
||||
|
||||
private int droolsExecutionTimeoutSecs = 300;
|
||||
|
||||
}
|
||||
+31
@@ -0,0 +1,31 @@
|
||||
package com.iqser.red.service.redaction.v1.server;
|
||||
|
||||
import java.time.Duration;
|
||||
|
||||
import org.springframework.boot.autoconfigure.cache.RedisCacheManagerBuilderCustomizer;
|
||||
import org.springframework.context.annotation.Bean;
|
||||
import org.springframework.context.annotation.Configuration;
|
||||
import org.springframework.data.redis.cache.RedisCacheConfiguration;
|
||||
import org.springframework.data.redis.connection.RedisConnectionFactory;
|
||||
import org.springframework.data.redis.core.RedisTemplate;
|
||||
|
||||
@Configuration
|
||||
public class RedisCachingConfiguration {
|
||||
|
||||
@Bean
|
||||
public RedisCacheManagerBuilderCustomizer redisCacheManagerBuilderCustomizer() {
|
||||
|
||||
return (builder) -> builder.withCacheConfiguration("documentDataCache",
|
||||
RedisCacheConfiguration.defaultCacheConfig().entryTtl(Duration.ofMinutes(30)).disableCachingNullValues());
|
||||
}
|
||||
|
||||
|
||||
@Bean
|
||||
public RedisTemplate<String, Object> redisTemplate(RedisConnectionFactory connectionFactory) {
|
||||
|
||||
RedisTemplate<String, Object> template = new RedisTemplate<>();
|
||||
template.setConnectionFactory(connectionFactory);
|
||||
return template;
|
||||
}
|
||||
|
||||
}
|
||||
+2
-2
@@ -1,8 +1,8 @@
|
||||
package com.iqser.red.service.redaction.v1.server.controller;
|
||||
|
||||
import com.iqser.red.commons.spring.ErrorMessage;
|
||||
import com.iqser.red.service.redaction.v1.server.exception.NotFoundException;
|
||||
import com.iqser.red.service.redaction.v1.server.exception.RulesValidationException;
|
||||
import com.iqser.red.service.redaction.v1.server.utils.exception.NotFoundException;
|
||||
import com.iqser.red.service.redaction.v1.server.utils.exception.RulesValidationException;
|
||||
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
|
||||
|
||||
+7
-5
@@ -3,9 +3,11 @@ package com.iqser.red.service.redaction.v1.server.controller;
|
||||
import org.springframework.web.bind.annotation.RequestBody;
|
||||
import org.springframework.web.bind.annotation.RestController;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.model.DroolsSyntaxValidation;
|
||||
import com.iqser.red.service.redaction.v1.model.RuleValidationModel;
|
||||
import com.iqser.red.service.redaction.v1.resources.RedactionResource;
|
||||
import com.iqser.red.service.redaction.v1.server.exception.RulesValidationException;
|
||||
import com.iqser.red.service.redaction.v1.server.redaction.service.DroolsExecutionService;
|
||||
import com.iqser.red.service.redaction.v1.server.service.drools.DroolsSyntaxValidationService;
|
||||
import com.iqser.red.service.redaction.v1.server.utils.exception.RulesValidationException;
|
||||
|
||||
import lombok.RequiredArgsConstructor;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
@@ -15,14 +17,14 @@ import lombok.extern.slf4j.Slf4j;
|
||||
@RequiredArgsConstructor
|
||||
public class RedactionController implements RedactionResource {
|
||||
|
||||
private final DroolsExecutionService droolsExecutionService;
|
||||
private final DroolsSyntaxValidationService droolsSyntaxValidationService;
|
||||
|
||||
|
||||
@Override
|
||||
public void testRules(@RequestBody String rules) {
|
||||
public DroolsSyntaxValidation testRules(@RequestBody RuleValidationModel rulesValidationModel) {
|
||||
|
||||
try {
|
||||
droolsExecutionService.testRules(rules);
|
||||
return droolsSyntaxValidationService.testRules(rulesValidationModel);
|
||||
} catch (Exception e) {
|
||||
throw new RulesValidationException("Could not test rules: " + e.getMessage(), e);
|
||||
}
|
||||
|
||||
+8
-4
@@ -1,10 +1,11 @@
|
||||
package com.iqser.red.service.redaction.v1.server.controller;
|
||||
|
||||
import java.util.Collections;
|
||||
|
||||
import org.springframework.web.bind.annotation.RestController;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.model.RuleBuilderModel;
|
||||
import com.iqser.red.service.redaction.v1.resources.RuleBuilderResource;
|
||||
import com.iqser.red.service.redaction.v1.server.redaction.service.RuleBuilderModelService;
|
||||
|
||||
import lombok.RequiredArgsConstructor;
|
||||
|
||||
@@ -12,13 +13,16 @@ import lombok.RequiredArgsConstructor;
|
||||
@RequiredArgsConstructor
|
||||
public class RuleBuilderController implements RuleBuilderResource {
|
||||
|
||||
private final RuleBuilderModelService ruleBuilderModelService;
|
||||
|
||||
|
||||
@Override
|
||||
public RuleBuilderModel getRuleBuilderModel() {
|
||||
|
||||
return ruleBuilderModelService.getRuleBuilderModel();
|
||||
RuleBuilderModel ruleBuilderModel = new RuleBuilderModel();
|
||||
|
||||
ruleBuilderModel.setWhenClauses(Collections.emptyList());
|
||||
ruleBuilderModel.setThenConditions(Collections.emptyList());
|
||||
|
||||
return ruleBuilderModel;
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
-80
@@ -1,80 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter;
|
||||
|
||||
import java.awt.geom.Rectangle2D;
|
||||
import java.util.ArrayList;
|
||||
import java.util.HashMap;
|
||||
import java.util.List;
|
||||
import java.util.Locale;
|
||||
import java.util.Map;
|
||||
|
||||
import org.springframework.stereotype.Service;
|
||||
|
||||
import com.fasterxml.jackson.databind.ObjectMapper;
|
||||
import com.iqser.red.service.persistence.service.v1.api.shared.model.dossiertemplate.dossier.file.FileType;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image.ImageServiceResponse;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.image.ClassifiedImage;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.document.graph.nodes.ImageType;
|
||||
import com.iqser.red.service.redaction.v1.server.storage.RedactionStorageService;
|
||||
|
||||
import lombok.RequiredArgsConstructor;
|
||||
import lombok.SneakyThrows;
|
||||
|
||||
@Service
|
||||
@RequiredArgsConstructor
|
||||
public class ImageServiceResponseAdapter {
|
||||
|
||||
private final ObjectMapper objectMapper;
|
||||
private final RedactionStorageService redactionStorageService;
|
||||
|
||||
|
||||
@SneakyThrows
|
||||
public Map<Integer, List<ClassifiedImage>> convertImages(String dossierId, String fileId) {
|
||||
|
||||
var imageClassificationStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(dossierId, fileId, FileType.IMAGE_INFO));
|
||||
|
||||
ImageServiceResponse imageServiceResponse = objectMapper.readValue(imageClassificationStream, ImageServiceResponse.class);
|
||||
|
||||
Map<Integer, List<ClassifiedImage>> images = new HashMap<>();
|
||||
imageServiceResponse.getData().forEach(imageMetadata -> {
|
||||
var classification = imageMetadata.getFilters().isAllPassed() ? ImageType.valueOf(imageMetadata.getClassification()
|
||||
.getLabel()
|
||||
.toUpperCase(Locale.ROOT)) : ImageType.OTHER;
|
||||
images.computeIfAbsent(imageMetadata.getPosition().getPageNumber(), x -> new ArrayList<>())
|
||||
.add(new ClassifiedImage(new Rectangle2D.Double(imageMetadata.getPosition().getX1(),
|
||||
imageMetadata.getPosition().getY1(),
|
||||
imageMetadata.getGeometry().getWidth(),
|
||||
imageMetadata.getGeometry().getHeight()), classification, imageMetadata.isAlpha(), imageMetadata.getPosition().getPageNumber()));
|
||||
});
|
||||
|
||||
// Currently This is a copy but, it will be changed later because i don' t think that we should unclassified images.
|
||||
imageServiceResponse.getDataCV().forEach(imageMetadata -> {
|
||||
var classification = imageMetadata.getFilters().isAllPassed() ? ImageType.valueOf(imageMetadata.getClassification()
|
||||
.getLabel()
|
||||
.toUpperCase(Locale.ROOT)) : ImageType.OTHER;
|
||||
images.computeIfAbsent(imageMetadata.getPosition().getPageNumber(), x -> new ArrayList<>())
|
||||
.add(new ClassifiedImage(new Rectangle2D.Double(imageMetadata.getPosition().getX1(),
|
||||
imageMetadata.getPosition().getY1(),
|
||||
imageMetadata.getGeometry().getWidth(),
|
||||
imageMetadata.getGeometry().getHeight()), classification, imageMetadata.isAlpha(), imageMetadata.getPosition().getPageNumber()));
|
||||
});
|
||||
|
||||
return images;
|
||||
}
|
||||
|
||||
|
||||
public void findOcr(ClassificationPage page) {
|
||||
|
||||
page.getImages().forEach(image -> {
|
||||
if (image.getImageType().equals(ImageType.OTHER)) {
|
||||
page.getTextBlocks().forEach(textblock -> {
|
||||
if (image.getPosition().contains(textblock.getMinX(), textblock.getMinY(), textblock.getWidth(), textblock.getHeight())) {
|
||||
image.setImageType(ImageType.OCR);
|
||||
}
|
||||
});
|
||||
}
|
||||
});
|
||||
}
|
||||
|
||||
|
||||
}
|
||||
-63
@@ -1,63 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.Collection;
|
||||
import java.util.HashMap;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
|
||||
import org.springframework.stereotype.Service;
|
||||
|
||||
import com.fasterxml.jackson.databind.ObjectMapper;
|
||||
import com.iqser.red.service.persistence.service.v1.api.shared.model.dossiertemplate.dossier.file.FileType;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table.PdfTableCell;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table.TableCells;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table.TableServiceResponse;
|
||||
import com.iqser.red.service.redaction.v1.server.storage.RedactionStorageService;
|
||||
|
||||
import lombok.RequiredArgsConstructor;
|
||||
import lombok.SneakyThrows;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
|
||||
@Slf4j
|
||||
@Service
|
||||
@RequiredArgsConstructor
|
||||
public class TableServiceResponseAdapter {
|
||||
|
||||
private final ObjectMapper objectMapper;
|
||||
private final RedactionStorageService redactionStorageService;
|
||||
|
||||
|
||||
@SneakyThrows
|
||||
public Map<Integer, List<PdfTableCell>> convertTables(String dossierId, String fileId) {
|
||||
|
||||
var tableClassificationStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(dossierId, fileId, FileType.TABLES));
|
||||
|
||||
TableServiceResponse tableServiceResponse = objectMapper.readValue(tableClassificationStream, TableServiceResponse.class);
|
||||
|
||||
Map<Integer, List<PdfTableCell>> tableCells = new HashMap<>();
|
||||
tableServiceResponse.getData()
|
||||
.forEach(tableData -> tableCells.computeIfAbsent(tableData.getPageInfo().getNumber(), tableCell -> new ArrayList<>())
|
||||
.addAll(convertTableCells(tableData.getTableCells())));
|
||||
|
||||
return tableCells;
|
||||
}
|
||||
|
||||
|
||||
private Collection<? extends PdfTableCell> convertTableCells(List<TableCells> tableCells) {
|
||||
|
||||
List<PdfTableCell> pdfTableCells = new ArrayList<>();
|
||||
|
||||
tableCells.forEach(t -> pdfTableCells.add(PdfTableCell.builder()
|
||||
.y0(t.getY0())
|
||||
.x1(t.getX1())
|
||||
.y1(t.getY1())
|
||||
.x0(t.getX0())
|
||||
.width(t.getWidth())
|
||||
.height(t.getHeight())
|
||||
.build()));
|
||||
|
||||
return pdfTableCells;
|
||||
}
|
||||
|
||||
}
|
||||
-14
@@ -1,14 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
|
||||
|
||||
import java.util.HashMap;
|
||||
import java.util.Map;
|
||||
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
public class Classification {
|
||||
|
||||
private Map<String, Float> probabilities = new HashMap<>();
|
||||
private String label;
|
||||
|
||||
}
|
||||
-11
@@ -1,11 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
|
||||
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
public class FilterGeometry {
|
||||
|
||||
private ImageSize imageSize;
|
||||
private ImageFormat imageFormat;
|
||||
|
||||
}
|
||||
-12
@@ -1,12 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
|
||||
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
public class Filters {
|
||||
|
||||
private FilterGeometry geometry;
|
||||
private Probability probability;
|
||||
private boolean allPassed;
|
||||
|
||||
}
|
||||
-11
@@ -1,11 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
|
||||
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
public class Geometry {
|
||||
|
||||
private float width;
|
||||
private float height;
|
||||
|
||||
}
|
||||
-12
@@ -1,12 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
|
||||
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
public class ImageFormat {
|
||||
|
||||
private float quotient;
|
||||
private boolean tooTall;
|
||||
private boolean tooWide;
|
||||
|
||||
}
|
||||
-14
@@ -1,14 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
|
||||
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
public class ImageMetadata {
|
||||
|
||||
private Classification classification;
|
||||
private Position position;
|
||||
private Geometry geometry;
|
||||
private Filters filters;
|
||||
private boolean alpha;
|
||||
|
||||
}
|
||||
-46
@@ -1,46 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.List;
|
||||
|
||||
import com.fasterxml.jackson.annotation.JsonAlias;
|
||||
import com.fasterxml.jackson.annotation.JsonProperty;
|
||||
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
public class ImageServiceResponse {
|
||||
|
||||
private String dossierId;
|
||||
private String fileId;
|
||||
|
||||
@JsonProperty(value = "imageMetadata")
|
||||
@JsonAlias("data")
|
||||
private List<ImageMetadata> data = new ArrayList<>();
|
||||
|
||||
private List<ImageMetadata> dataCV = new ArrayList<>();
|
||||
|
||||
|
||||
@JsonProperty(value = "imageMetadata")
|
||||
@JsonAlias("data")
|
||||
public void setData(List<ImageMetadata> data) {this.data = data;}
|
||||
|
||||
|
||||
public List<ImageMetadata> getData() {
|
||||
|
||||
if (this.data == null) {
|
||||
this.data = new ArrayList<>();
|
||||
}
|
||||
return data;
|
||||
}
|
||||
|
||||
|
||||
public List<ImageMetadata> getDataCV() {
|
||||
|
||||
if (this.dataCV == null) {
|
||||
this.dataCV = new ArrayList<>();
|
||||
}
|
||||
return dataCV;
|
||||
}
|
||||
|
||||
}
|
||||
-12
@@ -1,12 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
|
||||
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
public class ImageSize {
|
||||
|
||||
private float quotient;
|
||||
private boolean tooLarge;
|
||||
private boolean tooSmall;
|
||||
|
||||
}
|
||||
-14
@@ -1,14 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
|
||||
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
public class Position {
|
||||
|
||||
private float x1;
|
||||
private float x2;
|
||||
private float y1;
|
||||
private float y2;
|
||||
private int pageNumber;
|
||||
|
||||
}
|
||||
-10
@@ -1,10 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
|
||||
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
public class Probability {
|
||||
|
||||
private boolean unconfident;
|
||||
|
||||
}
|
||||
-13
@@ -1,13 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table;
|
||||
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
public class PageInfo {
|
||||
|
||||
private int number;
|
||||
private int rotation;
|
||||
private float width;
|
||||
private float height;
|
||||
|
||||
}
|
||||
-21
@@ -1,21 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table;
|
||||
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Builder;
|
||||
import lombok.Data;
|
||||
import lombok.RequiredArgsConstructor;
|
||||
|
||||
@Data
|
||||
@Builder
|
||||
@AllArgsConstructor
|
||||
@RequiredArgsConstructor
|
||||
public class PdfTableCell {
|
||||
|
||||
private float x0;
|
||||
private float y0;
|
||||
private float x1;
|
||||
private float y1;
|
||||
private float width;
|
||||
private float height;
|
||||
|
||||
}
|
||||
-15
@@ -1,15 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table;
|
||||
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
public class TableCells {
|
||||
|
||||
private float x0;
|
||||
private float y0;
|
||||
private float x1;
|
||||
private float y1;
|
||||
private float width;
|
||||
private float height;
|
||||
|
||||
}
|
||||
-14
@@ -1,14 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.List;
|
||||
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
public class TableData {
|
||||
|
||||
private PageInfo pageInfo;
|
||||
private List<TableCells> tableCells = new ArrayList<>();
|
||||
|
||||
}
|
||||
-19
@@ -1,19 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.List;
|
||||
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
public class TableServiceResponse {
|
||||
|
||||
private String dossierId;
|
||||
private String fileId;
|
||||
private String operation;
|
||||
private String targetFileExtension;
|
||||
private String responseFileExtension;
|
||||
|
||||
private List<TableData> data = new ArrayList<>();
|
||||
|
||||
}
|
||||
-88
@@ -1,88 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
|
||||
|
||||
import com.fasterxml.jackson.annotation.JsonIgnore;
|
||||
import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Rectangle;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
|
||||
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Data;
|
||||
import lombok.NoArgsConstructor;
|
||||
|
||||
@Data
|
||||
@AllArgsConstructor
|
||||
@NoArgsConstructor
|
||||
public abstract class AbstractPageBlock {
|
||||
|
||||
@JsonIgnore
|
||||
protected float minX;
|
||||
@JsonIgnore
|
||||
protected float maxX;
|
||||
@JsonIgnore
|
||||
protected float minY;
|
||||
@JsonIgnore
|
||||
protected float maxY;
|
||||
@JsonIgnore
|
||||
protected PageBlockType classification;
|
||||
@JsonIgnore
|
||||
protected int page;
|
||||
|
||||
int columnIndex;
|
||||
|
||||
@JsonIgnore
|
||||
private Orientation orientation = Orientation.NONE;
|
||||
|
||||
|
||||
public abstract String getText();
|
||||
|
||||
|
||||
public boolean isHeadline() {
|
||||
|
||||
return this instanceof TextPageBlock && this.getClassification() != null && this.getClassification().isHeadline();
|
||||
}
|
||||
|
||||
|
||||
public boolean containsBlock(TextPageBlock other) {
|
||||
|
||||
return this.minX <= other.getPdfMinX() && this.maxX >= other.getPdfMaxX() && this.minY >= other.getPdfMinY() && this.maxY <= other.getPdfMaxY();
|
||||
}
|
||||
|
||||
|
||||
public boolean contains(AbstractPageBlock other) {
|
||||
|
||||
return this.minX <= other.minX && this.maxX >= other.maxX && this.minY >= other.minY && this.maxY <= other.maxY;
|
||||
}
|
||||
|
||||
|
||||
public boolean contains(Rectangle other) {
|
||||
|
||||
return page == other.getPage() && this.minX <= other.getTopLeft().getX() && this.maxX >= other.getTopLeft().getX() + other.getWidth() && this.minY <= other.getTopLeft()
|
||||
.getY() && this.maxY >= other.getTopLeft().getY() + other.getHeight();
|
||||
}
|
||||
|
||||
|
||||
@JsonIgnore
|
||||
public float getHeight() {
|
||||
|
||||
return maxY - minY;
|
||||
}
|
||||
|
||||
|
||||
@JsonIgnore
|
||||
public float getWidth() {
|
||||
|
||||
return maxX - minX;
|
||||
}
|
||||
|
||||
|
||||
public boolean intersectsY(AbstractPageBlock atc) {
|
||||
|
||||
return this.minY <= atc.getMaxY() && this.maxY >= atc.getMinY();
|
||||
}
|
||||
|
||||
|
||||
public boolean intersectsX(AbstractPageBlock atc) {
|
||||
|
||||
return this.minX <= atc.getMaxX() && this.maxX >= atc.getMinX();
|
||||
}
|
||||
|
||||
}
|
||||
-33
@@ -1,33 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.List;
|
||||
|
||||
import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.section.SectionGrid;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.StringFrequencyCounter;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.UnclassifiedText;
|
||||
import com.iqser.red.service.redaction.v1.server.redaction.model.dictionary.DictionaryVersion;
|
||||
|
||||
import lombok.Data;
|
||||
import lombok.NoArgsConstructor;
|
||||
|
||||
@Data
|
||||
@NoArgsConstructor
|
||||
public class ClassificationDocument {
|
||||
|
||||
private List<ClassificationPage> pages = new ArrayList<>();
|
||||
private List<ClassificationSection> sections = new ArrayList<>();
|
||||
private List<ClassificationHeader> headers = new ArrayList<>();
|
||||
private List<ClassificationFooter> footers = new ArrayList<>();
|
||||
private List<UnclassifiedText> unclassifiedTexts = new ArrayList<>();
|
||||
private FloatFrequencyCounter textHeightCounter = new FloatFrequencyCounter();
|
||||
private FloatFrequencyCounter fontSizeCounter = new FloatFrequencyCounter();
|
||||
private StringFrequencyCounter fontCounter = new StringFrequencyCounter();
|
||||
private StringFrequencyCounter fontStyleCounter = new StringFrequencyCounter();
|
||||
private boolean headlines;
|
||||
|
||||
private SectionGrid sectionGrid = new SectionGrid();
|
||||
private DictionaryVersion dictionaryVersion;
|
||||
private long rulesVersion;
|
||||
|
||||
}
|
||||
-16
@@ -1,16 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
|
||||
|
||||
import java.util.List;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
|
||||
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
@AllArgsConstructor
|
||||
public class ClassificationFooter {
|
||||
|
||||
private List<TextPageBlock> textBlocks;
|
||||
|
||||
}
|
||||
-16
@@ -1,16 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
|
||||
|
||||
import java.util.List;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
|
||||
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
@AllArgsConstructor
|
||||
public class ClassificationHeader {
|
||||
|
||||
private List<TextPageBlock> textBlocks;
|
||||
|
||||
}
|
||||
-38
@@ -1,38 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.List;
|
||||
|
||||
import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Rectangle;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.image.ClassifiedImage;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.StringFrequencyCounter;
|
||||
|
||||
import lombok.Data;
|
||||
import lombok.NonNull;
|
||||
import lombok.RequiredArgsConstructor;
|
||||
|
||||
@Data
|
||||
@RequiredArgsConstructor
|
||||
public class ClassificationPage {
|
||||
|
||||
@NonNull
|
||||
private List<AbstractPageBlock> textBlocks;
|
||||
|
||||
private List<ClassifiedImage> images = new ArrayList<>();
|
||||
|
||||
private Rectangle bodyTextFrame;
|
||||
|
||||
private boolean landscape;
|
||||
private int rotation;
|
||||
|
||||
private int pageNumber;
|
||||
|
||||
private FloatFrequencyCounter textHeightCounter = new FloatFrequencyCounter();
|
||||
private FloatFrequencyCounter fontSizeCounter = new FloatFrequencyCounter();
|
||||
private StringFrequencyCounter fontCounter = new StringFrequencyCounter();
|
||||
private StringFrequencyCounter fontStyleCounter = new StringFrequencyCounter();
|
||||
|
||||
private float pageWidth;
|
||||
private float pageHeight;
|
||||
|
||||
}
|
||||
-32
@@ -1,32 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.List;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.image.ClassifiedImage;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.TablePageBlock;
|
||||
|
||||
import lombok.Data;
|
||||
import lombok.NoArgsConstructor;
|
||||
|
||||
@Data
|
||||
@NoArgsConstructor
|
||||
public class ClassificationSection {
|
||||
|
||||
private List<AbstractPageBlock> pageBlocks = new ArrayList<>();
|
||||
private List<ClassifiedImage> images = new ArrayList<>();
|
||||
private String headline;
|
||||
|
||||
|
||||
public List<TablePageBlock> getTables() {
|
||||
|
||||
List<TablePageBlock> tables = new ArrayList<>();
|
||||
pageBlocks.forEach(block -> {
|
||||
if (block instanceof TablePageBlock) {
|
||||
tables.add((TablePageBlock) block);
|
||||
}
|
||||
});
|
||||
return tables;
|
||||
}
|
||||
|
||||
}
|
||||
-14
@@ -1,14 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
|
||||
|
||||
import java.awt.geom.Rectangle2D;
|
||||
|
||||
import lombok.AllArgsConstructor;
|
||||
|
||||
@AllArgsConstructor
|
||||
public class Column {
|
||||
|
||||
int index;
|
||||
ColumnType columnType;
|
||||
Rectangle2D bBox;
|
||||
|
||||
}
|
||||
-6
@@ -1,6 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
|
||||
|
||||
public enum ColumnType {
|
||||
RULING,
|
||||
DISTANCE
|
||||
}
|
||||
-77
@@ -1,77 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.Collections;
|
||||
import java.util.HashMap;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
import java.util.stream.Collectors;
|
||||
|
||||
import lombok.Getter;
|
||||
|
||||
public class FloatFrequencyCounter {
|
||||
|
||||
@Getter
|
||||
Map<Float, Integer> countPerValue = new HashMap<>();
|
||||
|
||||
|
||||
public void add(float value) {
|
||||
|
||||
if (!countPerValue.containsKey(value)) {
|
||||
countPerValue.put(value, 1);
|
||||
} else {
|
||||
countPerValue.put(value, countPerValue.get(value) + 1);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
public void addAll(Map<Float, Integer> otherCounter) {
|
||||
|
||||
for (Map.Entry<Float, Integer> entry : otherCounter.entrySet()) {
|
||||
if (countPerValue.containsKey(entry.getKey())) {
|
||||
countPerValue.put(entry.getKey(), countPerValue.get(entry.getKey()) + entry.getValue());
|
||||
} else {
|
||||
countPerValue.put(entry.getKey(), entry.getValue());
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
public Float getMostPopular() {
|
||||
|
||||
Map.Entry<Float, Integer> mostPopular = null;
|
||||
for (Map.Entry<Float, Integer> entry : countPerValue.entrySet()) {
|
||||
if (mostPopular == null || entry.getValue() >= mostPopular.getValue()) {
|
||||
mostPopular = entry;
|
||||
}
|
||||
}
|
||||
return mostPopular != null ? mostPopular.getKey() : null;
|
||||
}
|
||||
|
||||
|
||||
public List<Float> getHighterThanMostPopular() {
|
||||
|
||||
Float mostPopular = getMostPopular();
|
||||
List<Float> higher = new ArrayList<>();
|
||||
for (Float value : countPerValue.keySet()) {
|
||||
if (value > mostPopular) {
|
||||
higher.add(value);
|
||||
}
|
||||
}
|
||||
|
||||
return higher.stream().sorted(Collections.reverseOrder()).collect(Collectors.toList());
|
||||
}
|
||||
|
||||
|
||||
public Float getHighest() {
|
||||
|
||||
Float highest = null;
|
||||
for (Float value : countPerValue.keySet()) {
|
||||
if (highest == null || value > highest) {
|
||||
highest = value;
|
||||
}
|
||||
}
|
||||
return highest;
|
||||
}
|
||||
|
||||
}
|
||||
-8
@@ -1,8 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
|
||||
|
||||
public enum Orientation {
|
||||
|
||||
NONE,
|
||||
LEFT,
|
||||
RIGHT
|
||||
}
|
||||
-38
@@ -1,38 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
|
||||
|
||||
public enum PageBlockType {
|
||||
H1,
|
||||
H2,
|
||||
H3,
|
||||
H4,
|
||||
H5,
|
||||
H6,
|
||||
HEADER,
|
||||
FOOTER,
|
||||
TITLE,
|
||||
PARAGRAPH,
|
||||
PARAGRAPH_BOLD,
|
||||
PARAGRAPH_ITALIC,
|
||||
PARAGRAPH_UNKNOWN,
|
||||
OTHER,
|
||||
TABLE;
|
||||
|
||||
|
||||
public static PageBlockType getHeadlineType(int i) {
|
||||
|
||||
return switch (i) {
|
||||
case 1 -> PageBlockType.H1;
|
||||
case 2 -> PageBlockType.H2;
|
||||
case 3 -> PageBlockType.H3;
|
||||
case 4 -> PageBlockType.H4;
|
||||
case 5 -> PageBlockType.H5;
|
||||
default -> PageBlockType.H6;
|
||||
};
|
||||
}
|
||||
|
||||
|
||||
public boolean isHeadline() {
|
||||
|
||||
return this.equals(H1) || this.equals(H2) || this.equals(H3) || this.equals(H4) || this.equals(H5) || this.equals(H6);
|
||||
}
|
||||
}
|
||||
-25
@@ -1,25 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.image;
|
||||
|
||||
import java.awt.geom.Rectangle2D;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.document.graph.nodes.ImageType;
|
||||
|
||||
import lombok.Data;
|
||||
import lombok.NonNull;
|
||||
import lombok.RequiredArgsConstructor;
|
||||
|
||||
@Data
|
||||
@RequiredArgsConstructor
|
||||
public class ClassifiedImage {
|
||||
|
||||
@NonNull
|
||||
private Rectangle2D position;
|
||||
@NonNull
|
||||
private ImageType imageType;
|
||||
private boolean isAppendedToSection;
|
||||
@NonNull
|
||||
private boolean hasTransparency;
|
||||
@NonNull
|
||||
private int page;
|
||||
|
||||
}
|
||||
-79
@@ -1,79 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table;
|
||||
|
||||
import java.awt.geom.Point2D;
|
||||
import java.util.ArrayList;
|
||||
import java.util.Iterator;
|
||||
import java.util.List;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
|
||||
import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizationUtilities;
|
||||
|
||||
import lombok.Data;
|
||||
import lombok.EqualsAndHashCode;
|
||||
import lombok.NoArgsConstructor;
|
||||
|
||||
@SuppressWarnings("serial")
|
||||
@Data
|
||||
@EqualsAndHashCode(callSuper = true)
|
||||
@NoArgsConstructor
|
||||
public class Cell extends Rectangle {
|
||||
|
||||
private List<TextPageBlock> textBlocks = new ArrayList<>();
|
||||
|
||||
private List<Cell> headerCells = new ArrayList<>();
|
||||
|
||||
private boolean isHeaderCell;
|
||||
|
||||
private static final int MIN_SIZE = 1;
|
||||
|
||||
private int pageNumber;
|
||||
|
||||
|
||||
public Cell(Point2D topLeft, Point2D bottomRight) {
|
||||
|
||||
super((float) topLeft.getY(), (float) topLeft.getX(), (float) (bottomRight.getX() - topLeft.getX()), (float) (bottomRight.getY() - topLeft.getY()));
|
||||
}
|
||||
|
||||
|
||||
public void addTextBlock(TextPageBlock textBlock) {
|
||||
|
||||
textBlocks.add(textBlock);
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public String toString() {
|
||||
|
||||
StringBuilder sb = new StringBuilder();
|
||||
|
||||
Iterator<TextPageBlock> itty = textBlocks.iterator();
|
||||
TextPositionSequence previous = null;
|
||||
while (itty.hasNext()) {
|
||||
|
||||
TextPageBlock textBlock = itty.next();
|
||||
|
||||
for (TextPositionSequence word : textBlock.getSequences()) {
|
||||
if (previous != null) {
|
||||
if (Math.abs(previous.getMaxYDirAdj() - word.getMaxYDirAdj()) > word.getTextHeight()) {
|
||||
sb.append('\n');
|
||||
} else {
|
||||
sb.append(' ');
|
||||
}
|
||||
}
|
||||
sb.append(word.toString());
|
||||
previous = word;
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
return TextNormalizationUtilities.removeHyphenLineBreaks(sb.toString()).replaceAll("\n", " ").replaceAll(" {2}", " ");
|
||||
}
|
||||
|
||||
|
||||
public boolean hasMinimumSize() {
|
||||
|
||||
return this.getHeight() >= MIN_SIZE && this.getWidth() >= MIN_SIZE;
|
||||
}
|
||||
|
||||
}
|
||||
-22
@@ -1,22 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table;
|
||||
|
||||
import lombok.RequiredArgsConstructor;
|
||||
import lombok.Value;
|
||||
|
||||
@Value
|
||||
@RequiredArgsConstructor
|
||||
public class CellPosition implements Comparable<CellPosition> {
|
||||
|
||||
int row;
|
||||
|
||||
int col;
|
||||
|
||||
|
||||
@Override
|
||||
public int compareTo(CellPosition other) {
|
||||
|
||||
int rowDiff = row - other.row;
|
||||
return rowDiff != 0 ? rowDiff : col - other.col;
|
||||
}
|
||||
|
||||
}
|
||||
-15
@@ -1,15 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table;
|
||||
|
||||
import java.util.List;
|
||||
|
||||
import lombok.Builder;
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
@Builder
|
||||
public class CleanRulings {
|
||||
|
||||
List<Ruling> horizontal;
|
||||
List<Ruling> vertical;
|
||||
|
||||
}
|
||||
-218
@@ -1,218 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table;
|
||||
|
||||
import java.awt.geom.Point2D;
|
||||
import java.awt.geom.Rectangle2D;
|
||||
import java.util.Comparator;
|
||||
import java.util.List;
|
||||
|
||||
@SuppressWarnings("all")
|
||||
public class Rectangle extends Rectangle2D.Float {
|
||||
|
||||
protected static final float VERTICAL_COMPARISON_THRESHOLD = 0.4f;
|
||||
/**
|
||||
* Ill-defined comparator, from when Rectangle was Comparable.
|
||||
* <p>
|
||||
* see https://github.com/tabulapdf/tabula-java/issues/116
|
||||
*
|
||||
* @deprecated with no replacement
|
||||
*/
|
||||
@Deprecated
|
||||
public static final Comparator<Rectangle> ILL_DEFINED_ORDER = new Comparator<Rectangle>() {
|
||||
@Override
|
||||
public int compare(Rectangle o1, Rectangle o2) {
|
||||
|
||||
if (o1.equals(o2)) {
|
||||
return 0;
|
||||
}
|
||||
if (o1.verticalOverlap(o2) > VERTICAL_COMPARISON_THRESHOLD) {
|
||||
return o1.isLtrDominant() == -1 && o2.isLtrDominant() == -1 ? -java.lang.Double.compare(o1.getX(), o2.getX()) : java.lang.Double.compare(o1.getX(), o2.getX());
|
||||
} else {
|
||||
return java.lang.Float.compare(o1.getBottom(), o2.getBottom());
|
||||
}
|
||||
}
|
||||
};
|
||||
|
||||
|
||||
public Rectangle() {
|
||||
|
||||
super();
|
||||
}
|
||||
|
||||
|
||||
public Rectangle(float top, float left, float width, float height) {
|
||||
|
||||
super();
|
||||
this.setRect(left, top, width, height);
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* @param rectangles
|
||||
* @return minimum bounding box that contains all the rectangles
|
||||
*/
|
||||
public static Rectangle boundingBoxOf(List<? extends Rectangle> rectangles) {
|
||||
|
||||
float minx = java.lang.Float.MAX_VALUE;
|
||||
float miny = java.lang.Float.MAX_VALUE;
|
||||
float maxx = java.lang.Float.MIN_VALUE;
|
||||
float maxy = java.lang.Float.MIN_VALUE;
|
||||
|
||||
for (Rectangle r : rectangles) {
|
||||
minx = (float) Math.min(r.getMinX(), minx);
|
||||
miny = (float) Math.min(r.getMinY(), miny);
|
||||
maxx = (float) Math.max(r.getMaxX(), maxx);
|
||||
maxy = (float) Math.max(r.getMaxY(), maxy);
|
||||
}
|
||||
return new Rectangle(miny, minx, maxx - minx, maxy - miny);
|
||||
}
|
||||
|
||||
|
||||
public int compareTo(Rectangle other) {
|
||||
|
||||
return ILL_DEFINED_ORDER.compare(this, other);
|
||||
}
|
||||
|
||||
|
||||
// I'm bad at Java and need this for fancy sorting in
|
||||
// technology.tabula.TextChunk.
|
||||
public int isLtrDominant() {
|
||||
|
||||
return 0;
|
||||
}
|
||||
|
||||
|
||||
public float getArea() {
|
||||
|
||||
return this.width * this.height;
|
||||
}
|
||||
|
||||
|
||||
public float verticalOverlap(Rectangle other) {
|
||||
|
||||
return Math.max(0, Math.min(this.getBottom(), other.getBottom()) - Math.max(this.getTop(), other.getTop()));
|
||||
}
|
||||
|
||||
|
||||
public boolean verticallyOverlaps(Rectangle other) {
|
||||
|
||||
return verticalOverlap(other) > 0;
|
||||
}
|
||||
|
||||
|
||||
public float horizontalOverlap(Rectangle other) {
|
||||
|
||||
return Math.max(0, Math.min(this.getRight(), other.getRight()) - Math.max(this.getLeft(), other.getLeft()));
|
||||
}
|
||||
|
||||
|
||||
public boolean horizontallyOverlaps(Rectangle other) {
|
||||
|
||||
return horizontalOverlap(other) > 0;
|
||||
}
|
||||
|
||||
|
||||
public float verticalOverlapRatio(Rectangle other) {
|
||||
|
||||
float rv = 0, delta = Math.min(this.getBottom() - this.getTop(), other.getBottom() - other.getTop());
|
||||
|
||||
if (other.getTop() <= this.getTop() && this.getTop() <= other.getBottom() && other.getBottom() <= this.getBottom()) {
|
||||
rv = (other.getBottom() - this.getTop()) / delta;
|
||||
} else if (this.getTop() <= other.getTop() && other.getTop() <= this.getBottom() && this.getBottom() <= other.getBottom()) {
|
||||
rv = (this.getBottom() - other.getTop()) / delta;
|
||||
} else if (this.getTop() <= other.getTop() && other.getTop() <= other.getBottom() && other.getBottom() <= this.getBottom()) {
|
||||
rv = (other.getBottom() - other.getTop()) / delta;
|
||||
} else if (other.getTop() <= this.getTop() && this.getTop() <= this.getBottom() && this.getBottom() <= other.getBottom()) {
|
||||
rv = (this.getBottom() - this.getTop()) / delta;
|
||||
}
|
||||
|
||||
return rv;
|
||||
|
||||
}
|
||||
|
||||
|
||||
public float overlapRatio(Rectangle other) {
|
||||
|
||||
double intersectionWidth = Math.max(0, Math.min(this.getRight(), other.getRight()) - Math.max(this.getLeft(), other.getLeft()));
|
||||
double intersectionHeight = Math.max(0, Math.min(this.getBottom(), other.getBottom()) - Math.max(this.getTop(), other.getTop()));
|
||||
double intersectionArea = Math.max(0, intersectionWidth * intersectionHeight);
|
||||
double unionArea = this.getArea() + other.getArea() - intersectionArea;
|
||||
|
||||
return (float) (intersectionArea / unionArea);
|
||||
}
|
||||
|
||||
|
||||
public Rectangle merge(Rectangle other) {
|
||||
|
||||
this.setRect(this.createUnion(other));
|
||||
return this;
|
||||
}
|
||||
|
||||
|
||||
public float getTop() {
|
||||
|
||||
return (float) this.getMinY();
|
||||
}
|
||||
|
||||
|
||||
public void setTop(float top) {
|
||||
|
||||
float deltaHeight = top - this.y;
|
||||
this.setRect(this.x, top, this.width, this.height - deltaHeight);
|
||||
}
|
||||
|
||||
|
||||
public float getRight() {
|
||||
|
||||
return (float) this.getMaxX();
|
||||
}
|
||||
|
||||
|
||||
public void setRight(float right) {
|
||||
|
||||
this.setRect(this.x, this.y, right - this.x, this.height);
|
||||
}
|
||||
|
||||
|
||||
public float getLeft() {
|
||||
|
||||
return (float) this.getMinX();
|
||||
}
|
||||
|
||||
|
||||
public void setLeft(float left) {
|
||||
|
||||
float deltaWidth = left - this.x;
|
||||
this.setRect(left, this.y, this.width - deltaWidth, this.height);
|
||||
}
|
||||
|
||||
|
||||
public float getBottom() {
|
||||
|
||||
return (float) this.getMaxY();
|
||||
}
|
||||
|
||||
|
||||
public void setBottom(float bottom) {
|
||||
|
||||
this.setRect(this.x, this.y, this.width, bottom - this.y);
|
||||
}
|
||||
|
||||
|
||||
public Point2D[] getPoints() {
|
||||
|
||||
return new Point2D[]{new Point2D.Float(this.getLeft(), this.getTop()), new Point2D.Float(this.getRight(), this.getTop()), new Point2D.Float(this.getRight(),
|
||||
this.getBottom()), new Point2D.Float(this.getLeft(), this.getBottom())};
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public String toString() {
|
||||
|
||||
StringBuilder sb = new StringBuilder();
|
||||
String s = super.toString();
|
||||
sb.append(s.substring(0, s.length() - 1));
|
||||
sb.append(String.format(",bottom=%f,right=%f]", this.getBottom(), this.getRight()));
|
||||
return sb.toString();
|
||||
}
|
||||
|
||||
}
|
||||
-437
@@ -1,437 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table;
|
||||
|
||||
import java.awt.geom.Line2D;
|
||||
import java.awt.geom.Point2D;
|
||||
import java.awt.geom.Rectangle2D;
|
||||
import java.util.ArrayList;
|
||||
import java.util.Collections;
|
||||
import java.util.Comparator;
|
||||
import java.util.Formatter;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
import java.util.TreeMap;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.CohenSutherlandClipping;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.Utils;
|
||||
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
|
||||
@Slf4j
|
||||
@SuppressWarnings("all")
|
||||
public class Ruling extends Line2D.Float {
|
||||
|
||||
private static int PERPENDICULAR_PIXEL_EXPAND_AMOUNT = 2;
|
||||
|
||||
|
||||
public Ruling(Point2D p1, Point2D p2) {
|
||||
|
||||
super(p1, p2);
|
||||
}
|
||||
|
||||
|
||||
public static List<Ruling> cropRulingsToArea(List<Ruling> rulings, Rectangle2D area) {
|
||||
|
||||
ArrayList<Ruling> rv = new ArrayList<>();
|
||||
for (Ruling r : rulings) {
|
||||
if (r.intersects(area)) {
|
||||
rv.add(r.intersect(area));
|
||||
}
|
||||
}
|
||||
return rv;
|
||||
}
|
||||
|
||||
|
||||
// log(n) implementation of find_intersections
|
||||
// based on http://people.csail.mit.edu/indyk/6.838-old/handouts/lec2.pdf
|
||||
public static Map<Point2D, Ruling[]> findIntersections(List<Ruling> horizontals, List<Ruling> verticals) {
|
||||
|
||||
class SortObject {
|
||||
|
||||
protected SOType type;
|
||||
protected float position;
|
||||
protected Ruling ruling;
|
||||
|
||||
|
||||
public SortObject(SOType type, float position, Ruling ruling) {
|
||||
|
||||
this.type = type;
|
||||
this.position = position;
|
||||
this.ruling = ruling;
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
List<SortObject> sos = new ArrayList<>();
|
||||
|
||||
TreeMap<Ruling, Boolean> tree = new TreeMap<>(new Comparator<Ruling>() {
|
||||
@Override
|
||||
public int compare(Ruling o1, Ruling o2) {
|
||||
|
||||
return java.lang.Double.compare(o1.getTop(), o2.getTop());
|
||||
}
|
||||
});
|
||||
|
||||
TreeMap<Point2D, Ruling[]> rv = new TreeMap<>(new Comparator<Point2D>() {
|
||||
@Override
|
||||
public int compare(Point2D o1, Point2D o2) {
|
||||
|
||||
if (o1.getY() > o2.getY()) {
|
||||
return 1;
|
||||
}
|
||||
if (o1.getY() < o2.getY()) {
|
||||
return -1;
|
||||
}
|
||||
if (o1.getX() > o2.getX()) {
|
||||
return 1;
|
||||
}
|
||||
if (o1.getX() < o2.getX()) {
|
||||
return -1;
|
||||
}
|
||||
return 0;
|
||||
}
|
||||
});
|
||||
|
||||
for (Ruling h : horizontals) {
|
||||
sos.add(new SortObject(SOType.HLEFT, h.getLeft() - PERPENDICULAR_PIXEL_EXPAND_AMOUNT, h));
|
||||
sos.add(new SortObject(SOType.HRIGHT, h.getRight() + PERPENDICULAR_PIXEL_EXPAND_AMOUNT, h));
|
||||
}
|
||||
|
||||
for (Ruling v : verticals) {
|
||||
sos.add(new SortObject(SOType.VERTICAL, v.getLeft(), v));
|
||||
}
|
||||
|
||||
Collections.sort(sos, new Comparator<SortObject>() {
|
||||
@Override
|
||||
public int compare(SortObject a, SortObject b) {
|
||||
|
||||
int rv;
|
||||
if (Utils.feq(a.position, b.position)) {
|
||||
if (a.type == SOType.VERTICAL && b.type == SOType.HLEFT) {
|
||||
rv = 1;
|
||||
} else if (a.type == SOType.VERTICAL && b.type == SOType.HRIGHT) {
|
||||
rv = -1;
|
||||
} else if (a.type == SOType.HLEFT && b.type == SOType.VERTICAL) {
|
||||
rv = -1;
|
||||
} else if (a.type == SOType.HRIGHT && b.type == SOType.VERTICAL) {
|
||||
rv = 1;
|
||||
} else {
|
||||
rv = java.lang.Double.compare(a.position, b.position);
|
||||
}
|
||||
} else {
|
||||
return java.lang.Double.compare(a.position, b.position);
|
||||
}
|
||||
return rv;
|
||||
}
|
||||
});
|
||||
|
||||
for (SortObject so : sos) {
|
||||
switch (so.type) {
|
||||
case VERTICAL:
|
||||
for (Map.Entry<Ruling, Boolean> h : tree.entrySet()) {
|
||||
try {
|
||||
Point2D i = h.getKey().intersectionPoint(so.ruling);
|
||||
if (i == null) {
|
||||
continue;
|
||||
}
|
||||
rv.put(i, new Ruling[]{h.getKey().expand(PERPENDICULAR_PIXEL_EXPAND_AMOUNT), so.ruling.expand(PERPENDICULAR_PIXEL_EXPAND_AMOUNT)});
|
||||
} catch (UnsupportedOperationException e) {
|
||||
log.info("Some line are oblique, ignoring...");
|
||||
continue;
|
||||
}
|
||||
}
|
||||
break;
|
||||
case HRIGHT:
|
||||
tree.remove(so.ruling);
|
||||
break;
|
||||
case HLEFT:
|
||||
tree.put(so.ruling, true);
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
return rv;
|
||||
|
||||
}
|
||||
|
||||
|
||||
public boolean vertical() {
|
||||
|
||||
return this.length() > 0 && Utils.feq(this.x1, this.x2); //diff < ORIENTATION_CHECK_THRESHOLD;
|
||||
}
|
||||
|
||||
|
||||
public boolean horizontal() {
|
||||
|
||||
return this.length() > 0 && Utils.feq(this.y1, this.y2); //diff < ORIENTATION_CHECK_THRESHOLD;
|
||||
}
|
||||
|
||||
// attributes that make sense only for non-oblique lines
|
||||
// these are used to have a single collapse method (in page, currently)
|
||||
|
||||
|
||||
public boolean oblique() {
|
||||
|
||||
return !(this.vertical() || this.horizontal());
|
||||
}
|
||||
|
||||
|
||||
public float getPosition() {
|
||||
|
||||
if (this.oblique()) {
|
||||
throw new UnsupportedOperationException();
|
||||
}
|
||||
return this.vertical() ? this.getLeft() : this.getTop();
|
||||
}
|
||||
|
||||
|
||||
public float getStart() {
|
||||
|
||||
if (this.oblique()) {
|
||||
throw new UnsupportedOperationException();
|
||||
}
|
||||
return this.vertical() ? this.getTop() : this.getLeft();
|
||||
}
|
||||
|
||||
|
||||
public void setStart(float v) {
|
||||
|
||||
if (this.oblique()) {
|
||||
throw new UnsupportedOperationException();
|
||||
}
|
||||
if (this.vertical()) {
|
||||
this.setTop(v);
|
||||
} else {
|
||||
this.setLeft(v);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
public float getEnd() {
|
||||
|
||||
if (this.oblique()) {
|
||||
throw new UnsupportedOperationException();
|
||||
}
|
||||
return this.vertical() ? this.getBottom() : this.getRight();
|
||||
}
|
||||
|
||||
|
||||
public void setEnd(float v) {
|
||||
|
||||
if (this.oblique()) {
|
||||
throw new UnsupportedOperationException();
|
||||
}
|
||||
if (this.vertical()) {
|
||||
this.setBottom(v);
|
||||
} else {
|
||||
this.setRight(v);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
public void setStartEnd(float start, float end) {
|
||||
|
||||
if (this.oblique()) {
|
||||
throw new UnsupportedOperationException();
|
||||
}
|
||||
if (this.vertical()) {
|
||||
this.setTop(start);
|
||||
this.setBottom(end);
|
||||
} else {
|
||||
this.setLeft(start);
|
||||
this.setRight(end);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
public boolean perpendicularTo(Ruling other) {
|
||||
|
||||
return this.vertical() == other.horizontal();
|
||||
}
|
||||
|
||||
|
||||
public boolean nearlyIntersects(Ruling another, int colinearOrParallelExpandAmount) {
|
||||
|
||||
if (this.intersectsLine(another)) {
|
||||
return true;
|
||||
}
|
||||
|
||||
boolean rv = false;
|
||||
|
||||
if (this.perpendicularTo(another)) {
|
||||
rv = this.expand(PERPENDICULAR_PIXEL_EXPAND_AMOUNT).intersectsLine(another);
|
||||
} else {
|
||||
rv = this.expand(colinearOrParallelExpandAmount).intersectsLine(another.expand(colinearOrParallelExpandAmount));
|
||||
}
|
||||
|
||||
return rv;
|
||||
}
|
||||
|
||||
|
||||
public double length() {
|
||||
|
||||
return Math.sqrt(Math.pow(this.x1 - this.x2, 2) + Math.pow(this.y1 - this.y2, 2));
|
||||
}
|
||||
|
||||
|
||||
public Ruling intersect(Rectangle2D clip) {
|
||||
|
||||
Float clipee = (Float) this.clone();
|
||||
boolean clipped = new CohenSutherlandClipping(clip).clip(clipee);
|
||||
|
||||
if (clipped) {
|
||||
return new Ruling(clipee.getP1(), clipee.getP2());
|
||||
} else {
|
||||
return this;
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
public Ruling expand(float amount) {
|
||||
|
||||
Ruling r = (Ruling) this.clone();
|
||||
try {
|
||||
r.setStart(this.getStart() - amount);
|
||||
r.setEnd(this.getEnd() + amount);
|
||||
} catch (UnsupportedOperationException e) {
|
||||
log.warn("Could not expand ruling!");
|
||||
}
|
||||
return r;
|
||||
}
|
||||
|
||||
|
||||
public Point2D intersectionPoint(Ruling other) {
|
||||
|
||||
Ruling this_l = this.expand(PERPENDICULAR_PIXEL_EXPAND_AMOUNT);
|
||||
Ruling other_l = other.expand(PERPENDICULAR_PIXEL_EXPAND_AMOUNT);
|
||||
Ruling horizontal, vertical;
|
||||
|
||||
if (!this_l.intersectsLine(other_l)) {
|
||||
return null;
|
||||
}
|
||||
|
||||
if (this_l.horizontal() && other_l.vertical()) {
|
||||
horizontal = this_l;
|
||||
vertical = other_l;
|
||||
} else if (this_l.vertical() && other_l.horizontal()) {
|
||||
vertical = this_l;
|
||||
horizontal = other_l;
|
||||
} else {
|
||||
log.warn("lines must be orthogonal, vertical and horizontal");
|
||||
return null;
|
||||
}
|
||||
return new Point2D.Float(vertical.getLeft(), horizontal.getTop());
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public boolean equals(Object other) {
|
||||
|
||||
if (this == other) {
|
||||
return true;
|
||||
}
|
||||
|
||||
if (!(other instanceof Ruling)) {
|
||||
return false;
|
||||
}
|
||||
|
||||
Ruling o = (Ruling) other;
|
||||
return this.getP1().equals(o.getP1()) && this.getP2().equals(o.getP2());
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public int hashCode() {
|
||||
|
||||
return super.hashCode();
|
||||
}
|
||||
|
||||
|
||||
public float getTop() {
|
||||
|
||||
return this.y1;
|
||||
}
|
||||
|
||||
|
||||
public void setTop(float v) {
|
||||
|
||||
setLine(this.getLeft(), v, this.getRight(), this.getBottom());
|
||||
}
|
||||
|
||||
|
||||
public float getLeft() {
|
||||
|
||||
return this.x1;
|
||||
}
|
||||
|
||||
|
||||
public void setLeft(float v) {
|
||||
|
||||
setLine(v, this.getTop(), this.getRight(), this.getBottom());
|
||||
}
|
||||
|
||||
|
||||
public float getBottom() {
|
||||
|
||||
return this.y2;
|
||||
}
|
||||
|
||||
|
||||
public void setBottom(float v) {
|
||||
|
||||
setLine(this.getLeft(), this.getTop(), this.getRight(), v);
|
||||
}
|
||||
|
||||
|
||||
public float getRight() {
|
||||
|
||||
return this.x2;
|
||||
}
|
||||
|
||||
|
||||
public void setRight(float v) {
|
||||
|
||||
setLine(this.getLeft(), this.getTop(), v, this.getBottom());
|
||||
}
|
||||
|
||||
|
||||
public float getWidth() {
|
||||
|
||||
return this.getRight() - this.getLeft();
|
||||
}
|
||||
|
||||
|
||||
public float getHeight() {
|
||||
|
||||
return this.getBottom() - this.getTop();
|
||||
}
|
||||
|
||||
|
||||
public double getAngle() {
|
||||
|
||||
double angle = Math.toDegrees(Math.atan2(this.getP2().getY() - this.getP1().getY(), this.getP2().getX() - this.getP1().getX()));
|
||||
|
||||
if (angle < 0) {
|
||||
angle += 360;
|
||||
}
|
||||
return angle;
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public String toString() {
|
||||
|
||||
StringBuilder sb = new StringBuilder();
|
||||
Formatter formatter = new Formatter(sb);
|
||||
String rv = formatter.format("%s[minX=%f minY=%f maxX=%f maxY=%f]", this.getClass().toString(), this.x1, this.y1, this.x2, this.y2).toString();
|
||||
formatter.close();
|
||||
return rv;
|
||||
}
|
||||
|
||||
|
||||
private enum SOType {
|
||||
VERTICAL,
|
||||
HRIGHT,
|
||||
HLEFT
|
||||
}
|
||||
|
||||
}
|
||||
-342
@@ -1,342 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table;
|
||||
|
||||
import java.awt.geom.Point2D;
|
||||
import java.util.ArrayList;
|
||||
import java.util.Collections;
|
||||
import java.util.HashSet;
|
||||
import java.util.List;
|
||||
import java.util.Set;
|
||||
import java.util.TreeMap;
|
||||
|
||||
import org.apache.commons.collections4.CollectionUtils;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.PageBlockType;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
|
||||
|
||||
import lombok.Getter;
|
||||
import lombok.Setter;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
|
||||
@Slf4j
|
||||
public class TablePageBlock extends AbstractPageBlock {
|
||||
|
||||
private final TreeMap<CellPosition, Cell> cells = new TreeMap<>();
|
||||
|
||||
private final int rotation;
|
||||
@Getter
|
||||
@Setter
|
||||
private String headline;
|
||||
private int unrotatedRowCount;
|
||||
private int unrotatedColCount;
|
||||
private List<List<Cell>> rows;
|
||||
|
||||
|
||||
public TablePageBlock(List<Cell> cells, Rectangle area, int rotation) {
|
||||
|
||||
addCells(cells);
|
||||
minX = area.getLeft();
|
||||
minY = area.getBottom();
|
||||
maxX = area.getRight();
|
||||
maxY = area.getTop();
|
||||
classification = PageBlockType.TABLE;
|
||||
this.rotation = rotation;
|
||||
}
|
||||
|
||||
|
||||
public List<List<Cell>> getRows() {
|
||||
|
||||
if (rows == null) {
|
||||
rows = computeRows();
|
||||
|
||||
// Ignore rows that does not contain any cells and values.
|
||||
List<List<Cell>> rowsToRemove = new ArrayList<>();
|
||||
for (List<Cell> row : rows) {
|
||||
if (row.size() == 1 && row.get(0).getTextBlocks().isEmpty()) {
|
||||
rowsToRemove.add(row);
|
||||
}
|
||||
}
|
||||
rows.removeAll(rowsToRemove);
|
||||
|
||||
computeHeaders();
|
||||
}
|
||||
|
||||
return rows;
|
||||
|
||||
}
|
||||
|
||||
|
||||
public int getRowCount() {
|
||||
|
||||
return getRows().size();
|
||||
}
|
||||
|
||||
|
||||
public int getColCount() {
|
||||
|
||||
return getRows().stream().mapToInt(List::size).max().orElse(0);
|
||||
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* Detect header cells (either first row or first column):
|
||||
* Column is marked as header if cell text is bold and row cell text is not bold.
|
||||
* Defaults to row.
|
||||
*/
|
||||
private void computeHeaders() {
|
||||
|
||||
if (rows == null) {
|
||||
rows = computeRows();
|
||||
}
|
||||
// A bold cell is a header cell as long as every cell to the left/top is bold, too
|
||||
// we move from left to right and top to bottom
|
||||
for (int rowIndex = 0; rowIndex < rows.size(); rowIndex++) {
|
||||
List<Cell> rowCells = rows.get(rowIndex);
|
||||
if (rowCells.size() == 1) {
|
||||
continue;
|
||||
}
|
||||
|
||||
for (int colIndex = 0; colIndex < rowCells.size(); colIndex++) {
|
||||
Cell cell = rowCells.get(colIndex);
|
||||
List<Cell> cellsToTheLeft = rowCells.subList(0, colIndex);
|
||||
Cell lastHeaderCell = null;
|
||||
for (Cell leftCell : cellsToTheLeft) {
|
||||
if (leftCell.isHeaderCell()) {
|
||||
lastHeaderCell = leftCell;
|
||||
} else {
|
||||
break;
|
||||
}
|
||||
}
|
||||
if (lastHeaderCell != null) {
|
||||
cell.getHeaderCells().add(lastHeaderCell);
|
||||
}
|
||||
List<Cell> cellsToTheTop = new ArrayList<>();
|
||||
for (int i = 0; i < rowIndex; i++) {
|
||||
try {
|
||||
cellsToTheTop.add(rows.get(i).get(colIndex));
|
||||
} catch (IndexOutOfBoundsException e) {
|
||||
log.debug("No cell {} in row {}, ignoring.", colIndex, rowIndex);
|
||||
}
|
||||
}
|
||||
for (Cell topCell : cellsToTheTop) {
|
||||
if (topCell.isHeaderCell()) {
|
||||
lastHeaderCell = topCell;
|
||||
} else {
|
||||
break;
|
||||
}
|
||||
}
|
||||
if (lastHeaderCell != null) {
|
||||
cell.getHeaderCells().add(lastHeaderCell);
|
||||
}
|
||||
if (CollectionUtils.isNotEmpty(cell.getTextBlocks()) && cell.getTextBlocks().get(0).getMostPopularWordStyle().equals("bold")) {
|
||||
cell.setHeaderCell(true);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
|
||||
private List<List<Cell>> computeRows() {
|
||||
|
||||
List<List<Cell>> rows = new ArrayList<>();
|
||||
if (rotation == 90) {
|
||||
for (int i = 0; i < unrotatedColCount; i++) { // rows
|
||||
List<Cell> lastRow = new ArrayList<>();
|
||||
for (int j = unrotatedRowCount - 1; j >= 0; j--) { // cols
|
||||
Cell cell = cells.get(new CellPosition(j, i));
|
||||
if (cell != null) {
|
||||
lastRow.add(cell);
|
||||
}
|
||||
}
|
||||
rows.add(lastRow);
|
||||
}
|
||||
} else if (rotation == 270) {
|
||||
for (int i = unrotatedColCount - 1; i >= 0; i--) { // rows
|
||||
List<Cell> lastRow = new ArrayList<>();
|
||||
for (int j = 0; j < unrotatedRowCount; j++) { // cols
|
||||
Cell cell = cells.get(new CellPosition(j, i));
|
||||
if (cell != null) {
|
||||
lastRow.add(cell);
|
||||
}
|
||||
}
|
||||
rows.add(lastRow);
|
||||
}
|
||||
} else {
|
||||
for (int i = 0; i < unrotatedRowCount; i++) {
|
||||
List<Cell> lastRow = new ArrayList<>();
|
||||
for (int j = 0; j < unrotatedColCount; j++) {
|
||||
Cell cell = cells.get(new CellPosition(i, j)); // JAVA_8 use getOrDefault()
|
||||
if (cell != null) {
|
||||
lastRow.add(cell);
|
||||
}
|
||||
}
|
||||
rows.add(lastRow);
|
||||
}
|
||||
}
|
||||
|
||||
return rows;
|
||||
|
||||
}
|
||||
|
||||
|
||||
private void add(Cell chunk, int row, int col) {
|
||||
|
||||
unrotatedRowCount = Math.max(unrotatedRowCount, row + 1);
|
||||
unrotatedColCount = Math.max(unrotatedColCount, col + 1);
|
||||
|
||||
CellPosition cp = new CellPosition(row, col);
|
||||
cells.put(cp, chunk);
|
||||
|
||||
}
|
||||
|
||||
|
||||
private void addCells(List<Cell> cells) {
|
||||
|
||||
if (cells.isEmpty()) {
|
||||
return;
|
||||
}
|
||||
|
||||
cells.removeIf(cell -> cell.getWidth() < 1.1 || cell.getHeight() < 1.1);
|
||||
|
||||
List<List<Cell>> rowsOfCells = calculateStructure(cells);
|
||||
|
||||
for (int i = 0; i < rowsOfCells.size(); i++) {
|
||||
for (int j = 0; j < rowsOfCells.get(i).size(); j++) {
|
||||
add(rowsOfCells.get(i).get(j), i, j);
|
||||
}
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* Calculates the structure of the table. For spanning rows and columns multiple cells with the same values will be inserted.
|
||||
*
|
||||
* @param cells The found cells
|
||||
* @return TablePageBlock Structure
|
||||
*/
|
||||
private List<List<Cell>> calculateStructure(List<Cell> cells) {
|
||||
|
||||
List<List<Cell>> matrix = new ArrayList<>();
|
||||
|
||||
if (cells.isEmpty()) {
|
||||
return matrix;
|
||||
}
|
||||
|
||||
Set<Float> uniqueX = new HashSet<>();
|
||||
Set<Float> uniqueY = new HashSet<>();
|
||||
cells.stream().filter(c -> !c.getTextBlocks().isEmpty() || c.getHeight() > 3 && c.getWidth() > 3).forEach(c -> {
|
||||
uniqueX.add(c.getLeft());
|
||||
uniqueX.add(c.getRight());
|
||||
uniqueY.add(c.getBottom());
|
||||
uniqueY.add(c.getTop());
|
||||
});
|
||||
|
||||
var sortedUniqueX = uniqueX.stream().sorted().toList();
|
||||
var sortedUniqueY = uniqueY.stream().sorted().toList();
|
||||
|
||||
Float prevY = null;
|
||||
for (Float y : sortedUniqueY) {
|
||||
|
||||
List<Cell> row = new ArrayList<>();
|
||||
|
||||
Float prevX = null;
|
||||
for (Float x : sortedUniqueX) {
|
||||
|
||||
if (prevY != null && prevX != null) {
|
||||
var cell = new Cell(new Point2D.Float(prevX, prevY), new Point2D.Float(x, y));
|
||||
|
||||
var intersectionCell = cells.stream().filter(c -> cell.intersects(c) && cell.overlapRatio(c) > 0.1f).findFirst();
|
||||
intersectionCell.ifPresent(value -> cell.getTextBlocks().addAll(value.getTextBlocks()));
|
||||
if (cell.hasMinimumSize()) {
|
||||
row.add(cell);
|
||||
}
|
||||
}
|
||||
prevX = x;
|
||||
}
|
||||
|
||||
if (prevY != null && prevX != null && !row.isEmpty()) {
|
||||
matrix.add(row);
|
||||
}
|
||||
prevY = y;
|
||||
}
|
||||
|
||||
Collections.reverse(matrix);
|
||||
|
||||
return matrix;
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public String getText() {
|
||||
|
||||
StringBuilder sb = new StringBuilder();
|
||||
List<List<Cell>> rows = getRows();
|
||||
|
||||
int i = 0;
|
||||
for (List<Cell> row : rows) {
|
||||
if (i != 0) {
|
||||
sb.append("\n");
|
||||
}
|
||||
if (!row.isEmpty()) {
|
||||
boolean firstColumn = true;
|
||||
for (Cell column : row) {
|
||||
if (!firstColumn) {
|
||||
sb.append(",");
|
||||
}
|
||||
if (column != null && column.getTextBlocks() != null) {
|
||||
boolean first = true;
|
||||
for (TextPageBlock textBlock : column.getTextBlocks()) {
|
||||
if (!first) {
|
||||
sb.append("\n");
|
||||
}
|
||||
sb.append('\"').append(textBlock.getText().replaceAll("\"", "\\\"")).append('\"');
|
||||
first = false;
|
||||
}
|
||||
}
|
||||
firstColumn = false;
|
||||
}
|
||||
}
|
||||
i++;
|
||||
}
|
||||
|
||||
return sb.toString();
|
||||
}
|
||||
|
||||
|
||||
public String getTextAsHtml() {
|
||||
|
||||
StringBuilder sb = new StringBuilder();
|
||||
List<List<Cell>> rows = getRows();
|
||||
|
||||
sb.append("<table border=\"1\">");
|
||||
int i = 0;
|
||||
for (List<Cell> row : rows) {
|
||||
sb.append("\n<tr>");
|
||||
if (!row.isEmpty()) {
|
||||
for (Cell column : row) {
|
||||
sb.append(i == 0 ? "\n<th>" : "\n<td>");
|
||||
if (column != null && column.getTextBlocks() != null) {
|
||||
boolean first = true;
|
||||
for (TextPageBlock textBlock : column.getTextBlocks()) {
|
||||
if (!first) {
|
||||
sb.append("<br />");
|
||||
}
|
||||
sb.append(textBlock.getText().replaceAll("\\n", "<br />"));
|
||||
first = false;
|
||||
}
|
||||
}
|
||||
sb.append(i == 0 ? "</th>" : "</td>");
|
||||
}
|
||||
}
|
||||
sb.append("</tr>");
|
||||
i++;
|
||||
}
|
||||
sb.append("</table>");
|
||||
|
||||
return sb.toString();
|
||||
}
|
||||
|
||||
}
|
||||
-100
@@ -1,100 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
|
||||
|
||||
import org.apache.pdfbox.text.TextPosition;
|
||||
import org.springframework.beans.BeanUtils;
|
||||
|
||||
import com.fasterxml.jackson.annotation.JsonIgnore;
|
||||
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Builder;
|
||||
import lombok.Data;
|
||||
import lombok.NoArgsConstructor;
|
||||
import lombok.SneakyThrows;
|
||||
|
||||
@Data
|
||||
@Builder
|
||||
@NoArgsConstructor
|
||||
@AllArgsConstructor
|
||||
public class RedTextPosition {
|
||||
|
||||
private String textMatrix;
|
||||
private float[] position;
|
||||
|
||||
@JsonIgnore
|
||||
private int rotation;
|
||||
|
||||
@JsonIgnore
|
||||
private float pageHeight;
|
||||
|
||||
@JsonIgnore
|
||||
private float pageWidth;
|
||||
|
||||
private String unicode;
|
||||
|
||||
@JsonIgnore
|
||||
private float dir;
|
||||
|
||||
// not used in reanalysis
|
||||
@JsonIgnore
|
||||
private float widthOfSpace;
|
||||
|
||||
// not used in reanalysis
|
||||
@JsonIgnore
|
||||
private float fontSizeInPt;
|
||||
|
||||
// not used in reanalysis
|
||||
@JsonIgnore
|
||||
private String fontName;
|
||||
|
||||
|
||||
@SneakyThrows
|
||||
public static RedTextPosition fromTextPosition(TextPosition textPosition) {
|
||||
|
||||
var pos = new RedTextPosition();
|
||||
BeanUtils.copyProperties(textPosition, pos);
|
||||
pos.setFontName(textPosition.getFont().getName());
|
||||
|
||||
pos.setFontSizeInPt(textPosition.getFontSizeInPt());
|
||||
|
||||
pos.setTextMatrix(textPosition.getTextMatrix().toString());
|
||||
|
||||
var position = new float[4];
|
||||
|
||||
position[0] = textPosition.getXDirAdj();
|
||||
position[1] = textPosition.getYDirAdj();
|
||||
position[2] = textPosition.getWidthDirAdj();
|
||||
position[3] = textPosition.getHeightDir();
|
||||
|
||||
pos.setPosition(position);
|
||||
return pos;
|
||||
}
|
||||
|
||||
|
||||
@JsonIgnore
|
||||
public float getXDirAdj() {
|
||||
|
||||
return position[0];
|
||||
}
|
||||
|
||||
|
||||
@JsonIgnore
|
||||
public float getYDirAdj() {
|
||||
|
||||
return position[1];
|
||||
}
|
||||
|
||||
|
||||
@JsonIgnore
|
||||
public float getWidthDirAdj() {
|
||||
|
||||
return position[2];
|
||||
}
|
||||
|
||||
|
||||
@JsonIgnore
|
||||
public float getHeightDir() {
|
||||
|
||||
return position[3];
|
||||
}
|
||||
|
||||
}
|
||||
-49
@@ -1,49 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.List;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizationUtilities;
|
||||
|
||||
import lombok.Getter;
|
||||
|
||||
public class SearchableText {
|
||||
|
||||
@Getter
|
||||
private final List<TextPositionSequence> sequences = new ArrayList<>();
|
||||
|
||||
|
||||
public void add(TextPositionSequence textPositionSequence) {
|
||||
|
||||
sequences.add(textPositionSequence);
|
||||
}
|
||||
|
||||
|
||||
public void addAll(List<TextPositionSequence> textPositionSequences) {
|
||||
|
||||
sequences.addAll(textPositionSequences);
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public String toString() {
|
||||
|
||||
return buildString(sequences);
|
||||
}
|
||||
|
||||
|
||||
public static String buildString(List<TextPositionSequence> sequences) {
|
||||
|
||||
StringBuilder sb = new StringBuilder();
|
||||
for (TextPositionSequence word : sequences) {
|
||||
sb.append(word);
|
||||
sb.append(' ');
|
||||
}
|
||||
String text = sb.toString();
|
||||
text = TextNormalizationUtilities.removeHyphenLineBreaks(text);
|
||||
text = TextNormalizationUtilities.removeLineBreaks(text);
|
||||
text = TextNormalizationUtilities.removeRepeatingWhitespaces(text);
|
||||
return text;
|
||||
}
|
||||
|
||||
}
|
||||
-17
@@ -1,17 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
|
||||
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Builder;
|
||||
import lombok.Data;
|
||||
import lombok.NoArgsConstructor;
|
||||
|
||||
@Data
|
||||
@Builder
|
||||
@NoArgsConstructor
|
||||
@AllArgsConstructor
|
||||
public class SimplifiedSectionText {
|
||||
|
||||
private int sectionNumber;
|
||||
private String text;
|
||||
|
||||
}
|
||||
-20
@@ -1,20 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.List;
|
||||
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Builder;
|
||||
import lombok.Data;
|
||||
import lombok.NoArgsConstructor;
|
||||
|
||||
@Data
|
||||
@Builder
|
||||
@NoArgsConstructor
|
||||
@AllArgsConstructor
|
||||
public class SimplifiedText {
|
||||
|
||||
private int numberOfPages;
|
||||
private List<SimplifiedSectionText> sectionTexts = new ArrayList<>();
|
||||
|
||||
}
|
||||
-47
@@ -1,47 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
|
||||
|
||||
import java.util.HashMap;
|
||||
import java.util.Map;
|
||||
|
||||
import lombok.Getter;
|
||||
|
||||
public class StringFrequencyCounter {
|
||||
|
||||
@Getter
|
||||
private final Map<String, Integer> countPerValue = new HashMap<>();
|
||||
|
||||
|
||||
public void add(String value) {
|
||||
|
||||
if (!countPerValue.containsKey(value)) {
|
||||
countPerValue.put(value, 1);
|
||||
} else {
|
||||
countPerValue.put(value, countPerValue.get(value) + 1);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
public void addAll(Map<String, Integer> otherCounter) {
|
||||
|
||||
for (Map.Entry<String, Integer> entry : otherCounter.entrySet()) {
|
||||
if (countPerValue.containsKey(entry.getKey())) {
|
||||
countPerValue.put(entry.getKey(), countPerValue.get(entry.getKey()) + entry.getValue());
|
||||
} else {
|
||||
countPerValue.put(entry.getKey(), entry.getValue());
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
public String getMostPopular() {
|
||||
|
||||
Map.Entry<String, Integer> mostPopular = null;
|
||||
for (Map.Entry<String, Integer> entry : countPerValue.entrySet()) {
|
||||
if (mostPopular == null || entry.getValue() > mostPopular.getValue()) {
|
||||
mostPopular = entry;
|
||||
}
|
||||
}
|
||||
return mostPopular != null ? mostPopular.getKey() : null;
|
||||
}
|
||||
|
||||
}
|
||||
-47
@@ -1,47 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
|
||||
|
||||
import com.fasterxml.jackson.annotation.JsonCreator;
|
||||
import com.fasterxml.jackson.annotation.JsonValue;
|
||||
|
||||
import lombok.Getter;
|
||||
|
||||
@Getter
|
||||
public enum TextDirection {
|
||||
ZERO(0f),
|
||||
QUARTER_CIRCLE(90f),
|
||||
HALF_CIRCLE(180f),
|
||||
THREE_QUARTER_CIRCLE(270f);
|
||||
|
||||
public static final String VALUE_STRING_SUFFIX = "°";
|
||||
|
||||
@JsonValue
|
||||
private final float degrees;
|
||||
private final float radians;
|
||||
|
||||
|
||||
TextDirection(float degreeValue) {
|
||||
|
||||
degrees = degreeValue;
|
||||
radians = (float) Math.toRadians(degreeValue);
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public String toString() {
|
||||
|
||||
return degrees + VALUE_STRING_SUFFIX;
|
||||
}
|
||||
|
||||
|
||||
@JsonCreator(mode = JsonCreator.Mode.DELEGATING)
|
||||
public static TextDirection fromDegrees(float degrees) {
|
||||
|
||||
for (var dir : TextDirection.values()) {
|
||||
if (degrees == dir.degrees) {
|
||||
return dir;
|
||||
}
|
||||
}
|
||||
|
||||
throw new IllegalArgumentException(String.format("A value of %f is not supported by TextDirection", degrees));
|
||||
}
|
||||
}
|
||||
-302
@@ -1,302 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.List;
|
||||
|
||||
import com.fasterxml.jackson.annotation.JsonIgnore;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.PageBlockType;
|
||||
import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizationUtilities;
|
||||
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Builder;
|
||||
import lombok.Data;
|
||||
import lombok.NoArgsConstructor;
|
||||
|
||||
@AllArgsConstructor
|
||||
@Builder
|
||||
@Data
|
||||
@NoArgsConstructor
|
||||
public class TextPageBlock extends AbstractPageBlock {
|
||||
|
||||
@Builder.Default
|
||||
private List<TextPositionSequence> sequences = new ArrayList<>();
|
||||
|
||||
@JsonIgnore
|
||||
private int rotation;
|
||||
|
||||
@JsonIgnore
|
||||
private String mostPopularWordFont;
|
||||
|
||||
@JsonIgnore
|
||||
private String mostPopularWordStyle;
|
||||
|
||||
@JsonIgnore
|
||||
private float mostPopularWordFontSize;
|
||||
|
||||
@JsonIgnore
|
||||
private float mostPopularWordHeight;
|
||||
|
||||
@JsonIgnore
|
||||
private float mostPopularWordSpaceWidth;
|
||||
|
||||
@JsonIgnore
|
||||
private float highestFontSize;
|
||||
|
||||
@JsonIgnore
|
||||
private PageBlockType classification;
|
||||
|
||||
|
||||
@JsonIgnore
|
||||
public TextDirection getDir() {
|
||||
|
||||
return sequences.get(0).getDir();
|
||||
}
|
||||
|
||||
|
||||
@JsonIgnore
|
||||
private float getPageHeight() {
|
||||
|
||||
return sequences.get(0).getPageHeight();
|
||||
}
|
||||
|
||||
|
||||
@JsonIgnore
|
||||
private float getPageWidth() {
|
||||
|
||||
return sequences.get(0).getPageWidth();
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* Returns the minX value in pdf coordinate system.
|
||||
* Note: This needs to use Pdf Coordinate System where {0,0} rotated with the page rotation.
|
||||
* 0 -> LowerLeft
|
||||
* 90 -> UpperLeft
|
||||
* 180 -> UpperRight
|
||||
* 270 -> LowerRight
|
||||
*
|
||||
* @return the minX value in pdf coordinate system
|
||||
*/
|
||||
@JsonIgnore
|
||||
public float getPdfMinX() {
|
||||
|
||||
if (getDir().getDegrees() == 90) {
|
||||
return minY;
|
||||
} else if (getDir().getDegrees() == 180) {
|
||||
return getPageWidth() - maxX;
|
||||
|
||||
} else if (getDir().getDegrees() == 270) {
|
||||
|
||||
return getPageWidth() - maxY;
|
||||
} else {
|
||||
return minX;
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* Returns the maxX value in pdf coordinate system.
|
||||
* Note: This needs to use Pdf Coordinate System where {0,0} rotated with the page rotation.
|
||||
* 0 -> LowerLeft
|
||||
* 90 -> UpperLeft
|
||||
* 180 -> UpperRight
|
||||
* 270 -> LowerRight
|
||||
*
|
||||
* @return the maxX value in pdf coordinate system
|
||||
*/
|
||||
@JsonIgnore
|
||||
public float getPdfMaxX() {
|
||||
|
||||
if (getDir().getDegrees() == 90) {
|
||||
return maxY;
|
||||
} else if (getDir().getDegrees() == 180) {
|
||||
return getPageWidth() - minX;
|
||||
} else if (getDir().getDegrees() == 270) {
|
||||
return getPageWidth() - minY;
|
||||
|
||||
} else {
|
||||
return maxX;
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* Returns the minY value in pdf coordinate system.
|
||||
* Note: This needs to use Pdf Coordinate System where {0,0} rotated with the page rotation.
|
||||
* 0 -> LowerLeft
|
||||
* 90 -> UpperLeft
|
||||
* 180 -> UpperRight
|
||||
* 270 -> LowerRight
|
||||
*
|
||||
* @return the minY value in pdf coordinate system
|
||||
*/
|
||||
@JsonIgnore
|
||||
public float getPdfMinY() {
|
||||
|
||||
if (getDir().getDegrees() == 90) {
|
||||
return minX;
|
||||
} else if (getDir().getDegrees() == 180) {
|
||||
return maxY;
|
||||
|
||||
} else if (getDir().getDegrees() == 270) {
|
||||
return getPageHeight() - maxX;
|
||||
|
||||
} else {
|
||||
return getPageHeight() - maxY;
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* Returns the maxY value in pdf coordinate system.
|
||||
* Note: This needs to use Pdf Coordinate System where {0,0} rotated with the page rotation.
|
||||
* 0 -> LowerLeft
|
||||
* 90 -> UpperLeft
|
||||
* 180 -> UpperRight
|
||||
* 270 -> LowerRight
|
||||
*
|
||||
* @return the maxY value in pdf coordinate system
|
||||
*/
|
||||
@JsonIgnore
|
||||
public float getPdfMaxY() {
|
||||
|
||||
if (getDir().getDegrees() == 90) {
|
||||
return maxX;
|
||||
} else if (getDir().getDegrees() == 180) {
|
||||
|
||||
return minY;
|
||||
} else if (getDir().getDegrees() == 270) {
|
||||
return getPageHeight() - minX;
|
||||
} else {
|
||||
return getPageHeight() - minY;
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
public TextPageBlock(float minX, float maxX, float minY, float maxY, List<TextPositionSequence> sequences, int rotation) {
|
||||
|
||||
this.minX = minX;
|
||||
this.maxX = maxX;
|
||||
this.minY = minY;
|
||||
this.maxY = maxY;
|
||||
this.sequences = sequences;
|
||||
this.rotation = rotation;
|
||||
}
|
||||
|
||||
|
||||
public TextPageBlock union(TextPositionSequence r) {
|
||||
|
||||
TextPageBlock union = this.copy();
|
||||
union.add(r);
|
||||
return union;
|
||||
}
|
||||
|
||||
|
||||
public TextPageBlock union(TextPageBlock r) {
|
||||
|
||||
TextPageBlock union = this.copy();
|
||||
union.add(r);
|
||||
return union;
|
||||
}
|
||||
|
||||
|
||||
public void add(TextPageBlock r) {
|
||||
|
||||
if (r.getMinX() < minX) {
|
||||
minX = r.getMinX();
|
||||
}
|
||||
if (r.getMaxX() > maxX) {
|
||||
maxX = r.getMaxX();
|
||||
}
|
||||
if (r.getMinY() < minY) {
|
||||
minY = r.getMinY();
|
||||
}
|
||||
if (r.getMaxY() > maxY) {
|
||||
maxY = r.getMaxY();
|
||||
}
|
||||
sequences.addAll(r.getSequences());
|
||||
}
|
||||
|
||||
|
||||
public void add(TextPositionSequence r) {
|
||||
|
||||
if (r.getMinXDirAdj() < minX) {
|
||||
minX = r.getMinXDirAdj();
|
||||
}
|
||||
if (r.getMaxXDirAdj() > maxX) {
|
||||
maxX = r.getMaxXDirAdj();
|
||||
}
|
||||
if (r.getMinYDirAdj() < minY) {
|
||||
minY = r.getMinYDirAdj();
|
||||
}
|
||||
if (r.getMaxYDirAdj() > maxY) {
|
||||
maxY = r.getMaxYDirAdj();
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
public TextPageBlock copy() {
|
||||
|
||||
return new TextPageBlock(minX, maxX, minY, maxY, sequences, rotation);
|
||||
}
|
||||
|
||||
|
||||
public void resize(float x1, float y1, float width, float height) {
|
||||
|
||||
set(x1, y1, x1 + width, y1 + height);
|
||||
}
|
||||
|
||||
|
||||
public void set(float x1, float y1, float x2, float y2) {
|
||||
|
||||
this.minX = Math.min(x1, x2);
|
||||
this.maxX = Math.max(x1, x2);
|
||||
this.minY = Math.min(y1, y2);
|
||||
this.maxY = Math.max(y1, y2);
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public String toString() {
|
||||
|
||||
StringBuilder builder = new StringBuilder();
|
||||
|
||||
for (int i = 0; i < sequences.size(); i++) {
|
||||
String sequenceAsString = sequences.get(i).toString();
|
||||
// Fix for missing Whitespace. This is recognized in getSequences method. See PDFTextStripper Line 1730.
|
||||
if (i != 0 && sequences.get(i - 1).charAt(sequences.get(i - 1).length() - 1) != ' ' && sequenceAsString.charAt(0) != ' ') {
|
||||
builder.append(' ');
|
||||
}
|
||||
builder.append(sequenceAsString);
|
||||
}
|
||||
|
||||
return builder.toString();
|
||||
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
@JsonIgnore
|
||||
public String getText() {
|
||||
|
||||
StringBuilder sb = new StringBuilder();
|
||||
|
||||
TextPositionSequence previous = null;
|
||||
for (TextPositionSequence word : sequences) {
|
||||
if (previous != null) {
|
||||
if (Math.abs(previous.getMaxYDirAdj() - word.getMaxYDirAdj()) > word.getTextHeight()) {
|
||||
sb.append('\n');
|
||||
} else {
|
||||
sb.append(' ');
|
||||
}
|
||||
}
|
||||
sb.append(word.toString());
|
||||
previous = word;
|
||||
}
|
||||
|
||||
return TextNormalizationUtilities.removeHyphenLineBreaks(sb.toString());
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
-301
@@ -1,301 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
|
||||
|
||||
import java.awt.geom.AffineTransform;
|
||||
import java.awt.geom.Point2D;
|
||||
import java.util.ArrayList;
|
||||
import java.util.List;
|
||||
import java.util.stream.Collectors;
|
||||
|
||||
import org.apache.pdfbox.text.TextPosition;
|
||||
|
||||
import com.fasterxml.jackson.annotation.JsonIgnore;
|
||||
import com.fasterxml.jackson.annotation.JsonIgnoreProperties;
|
||||
import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Point;
|
||||
import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Rectangle;
|
||||
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Builder;
|
||||
import lombok.Data;
|
||||
import lombok.NoArgsConstructor;
|
||||
import lombok.SneakyThrows;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
|
||||
@Slf4j
|
||||
@Data
|
||||
@Builder
|
||||
@NoArgsConstructor
|
||||
@AllArgsConstructor
|
||||
@JsonIgnoreProperties({"empty"})
|
||||
public class TextPositionSequence implements CharSequence {
|
||||
|
||||
public static final int HEIGHT_PADDING = 2;
|
||||
private int page;
|
||||
private List<RedTextPosition> textPositions = new ArrayList<>();
|
||||
|
||||
private TextDirection dir;
|
||||
private int rotation;
|
||||
private float pageHeight;
|
||||
private float pageWidth;
|
||||
|
||||
|
||||
public TextPositionSequence(int page) {
|
||||
|
||||
this.page = page;
|
||||
}
|
||||
|
||||
|
||||
public TextPositionSequence(List<TextPosition> textPositions, int page) {
|
||||
|
||||
this.textPositions = textPositions.stream().map(RedTextPosition::fromTextPosition).collect(Collectors.toList());
|
||||
this.page = page;
|
||||
this.dir = TextDirection.fromDegrees(textPositions.get(0).getDir());
|
||||
this.rotation = textPositions.get(0).getRotation();
|
||||
this.pageHeight = textPositions.get(0).getPageHeight();
|
||||
this.pageWidth = textPositions.get(0).getPageWidth();
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public int length() {
|
||||
|
||||
return textPositions.size();
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public char charAt(int index) {
|
||||
|
||||
RedTextPosition textPosition = textPositionAt(index);
|
||||
String text = textPosition.getUnicode();
|
||||
return text.charAt(0);
|
||||
}
|
||||
|
||||
|
||||
public char charAt(int index, boolean caseInSensitive) {
|
||||
|
||||
RedTextPosition textPosition = textPositionAt(index);
|
||||
String text = textPosition.getUnicode();
|
||||
return caseInSensitive ? text.toLowerCase().charAt(0) : text.charAt(0);
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public TextPositionSequence subSequence(int start, int end) {
|
||||
|
||||
var textPositionSequence = new TextPositionSequence();
|
||||
textPositionSequence.textPositions = textPositions.subList(start, end);
|
||||
textPositionSequence.page = page;
|
||||
textPositionSequence.dir = dir;
|
||||
textPositionSequence.rotation = rotation;
|
||||
textPositionSequence.pageHeight = pageHeight;
|
||||
textPositionSequence.pageWidth = pageWidth;
|
||||
|
||||
return textPositionSequence;
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public String toString() {
|
||||
|
||||
StringBuilder builder = new StringBuilder(length());
|
||||
for (int i = 0; i < length(); i++) {
|
||||
builder.append(charAt(i));
|
||||
}
|
||||
return builder.toString();
|
||||
}
|
||||
|
||||
|
||||
public RedTextPosition textPositionAt(int index) {
|
||||
|
||||
return textPositions.get(index);
|
||||
}
|
||||
|
||||
|
||||
public void add(TextPositionSequence textPositionSequence, RedTextPosition textPosition) {
|
||||
|
||||
this.textPositions.add(textPosition);
|
||||
this.page = textPositionSequence.getPage();
|
||||
this.dir = textPositionSequence.getDir();
|
||||
this.rotation = textPositionSequence.getRotation();
|
||||
this.pageHeight = textPositionSequence.getPageHeight();
|
||||
this.pageWidth = textPositionSequence.getPageWidth();
|
||||
}
|
||||
|
||||
|
||||
public void add(TextPosition textPosition) {
|
||||
|
||||
this.textPositions.add(RedTextPosition.fromTextPosition(textPosition));
|
||||
|
||||
this.dir = TextDirection.fromDegrees(textPositions.get(0).getDir());
|
||||
this.rotation = textPositions.get(0).getRotation();
|
||||
this.pageHeight = textPositions.get(0).getPageHeight();
|
||||
this.pageWidth = textPositions.get(0).getPageWidth();
|
||||
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* This value is adjusted so that 0,0 is upper left and it is adjusted based on the text direction.
|
||||
* This method ignores the page rotation but takes the text rotation and adjusts the coordinates to awt.
|
||||
*
|
||||
* @return the text direction adjusted minX value
|
||||
*/
|
||||
@JsonIgnore
|
||||
public float getMinXDirAdj() {
|
||||
|
||||
return textPositions.get(0).getXDirAdj();
|
||||
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* This value is adjusted so that 0,0 is upper left and it is adjusted based on the text direction.
|
||||
* This method ignores the page rotation but takes the text rotation and adjusts the coordinates to awt.
|
||||
*
|
||||
* @return the text direction adjusted maxX value
|
||||
*/
|
||||
@JsonIgnore
|
||||
public float getMaxXDirAdj() {
|
||||
|
||||
return textPositions.get(textPositions.size() - 1).getXDirAdj() + textPositions.get(textPositions.size() - 1).getWidthDirAdj() + HEIGHT_PADDING;
|
||||
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* This value is adjusted so that 0,0 is upper left and it is adjusted based on the text direction.
|
||||
* This method ignores the page rotation but takes the text rotation and adjusts the coordinates to awt.
|
||||
*
|
||||
* @return the text direction adjusted minY value. The upper border of the bounding box of the word.
|
||||
*/
|
||||
@JsonIgnore
|
||||
public float getMinYDirAdj() {
|
||||
|
||||
return textPositions.get(0).getYDirAdj() - getTextHeight();
|
||||
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* This value is adjusted so that 0,0 is upper left and it is adjusted based on the text direction.
|
||||
* This method ignores the page rotation but takes the text rotation and adjusts the coordinates to awt.
|
||||
*
|
||||
* @return the text direction adjusted maxY value. The lower border of the bounding box of the word.
|
||||
*/
|
||||
@JsonIgnore
|
||||
public float getMaxYDirAdj() {
|
||||
|
||||
return textPositions.get(0).getYDirAdj();
|
||||
|
||||
}
|
||||
|
||||
|
||||
@JsonIgnore
|
||||
public float getTextHeight() {
|
||||
|
||||
return textPositions.get(0).getHeightDir() + HEIGHT_PADDING;
|
||||
}
|
||||
|
||||
|
||||
@JsonIgnore
|
||||
public float getHeight() {
|
||||
|
||||
return getMaxYDirAdj() - getMinYDirAdj();
|
||||
}
|
||||
|
||||
|
||||
@JsonIgnore
|
||||
public float getWidth() {
|
||||
|
||||
return getMaxXDirAdj() - getMinXDirAdj();
|
||||
}
|
||||
|
||||
|
||||
@JsonIgnore
|
||||
public String getFont() {
|
||||
|
||||
return textPositions.get(0).getFontName().toLowerCase().replaceAll(",bold", "").replaceAll(",italic", "");
|
||||
}
|
||||
|
||||
|
||||
@JsonIgnore
|
||||
public String getFontStyle() {
|
||||
|
||||
String lowercaseFontName = textPositions.get(0).getFontName().toLowerCase();
|
||||
|
||||
if (lowercaseFontName.contains("bold") && lowercaseFontName.contains("italic")) {
|
||||
return "bold, italic";
|
||||
} else if (lowercaseFontName.contains("bold")) {
|
||||
return "bold";
|
||||
} else if (lowercaseFontName.contains("italic")) {
|
||||
return "italic";
|
||||
} else {
|
||||
return "standard";
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
|
||||
@JsonIgnore
|
||||
public float getFontSize() {
|
||||
|
||||
return textPositions.get(0).getFontSizeInPt();
|
||||
}
|
||||
|
||||
|
||||
@JsonIgnore
|
||||
public float getSpaceWidth() {
|
||||
|
||||
return textPositions.get(0).getWidthOfSpace();
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* This returns the bounding box of the word in Pdf Coordinate System where {0,0} rotated with the page rotation.
|
||||
* 0 -> LowerLeft
|
||||
* 90 -> UpperLeft
|
||||
* 180 -> UpperRight
|
||||
* 270 -> LowerRight
|
||||
*
|
||||
* @return bounding box of the word in Pdf Coordinate System
|
||||
*/
|
||||
@JsonIgnore
|
||||
@SneakyThrows
|
||||
public Rectangle getRectangle() {
|
||||
|
||||
log.debug("Page: '{}', Word: '{}', Rotation: '{}', textRotation {}", page, this, rotation, dir);
|
||||
|
||||
float textHeight = getTextHeight();
|
||||
|
||||
RedTextPosition firstTextPos = textPositions.get(0);
|
||||
RedTextPosition lastTextPos = textPositions.get(textPositions.size() - 1);
|
||||
|
||||
Point2D bottomLeft = new Point2D.Double(firstTextPos.getXDirAdj(), firstTextPos.getYDirAdj() - HEIGHT_PADDING);
|
||||
Point2D topRight = new Point2D.Double(lastTextPos.getXDirAdj() + lastTextPos.getWidthDirAdj(), lastTextPos.getYDirAdj() + textHeight + HEIGHT_PADDING);
|
||||
|
||||
AffineTransform transform = new AffineTransform();
|
||||
if (dir == TextDirection.ZERO || dir == TextDirection.HALF_CIRCLE) {
|
||||
transform.rotate(dir.getRadians(), pageWidth / 2f, pageHeight / 2f);
|
||||
transform.translate(0f, pageHeight + textHeight);
|
||||
transform.scale(1., -1.);
|
||||
} else if (dir == TextDirection.QUARTER_CIRCLE) {
|
||||
transform.rotate(dir.getRadians(), pageWidth / 2f, pageWidth / 2f);
|
||||
transform.translate(0f, pageWidth + textHeight);
|
||||
transform.scale(1., -1.);
|
||||
} else {
|
||||
transform.rotate(dir.getRadians(), pageHeight / 2f, pageHeight / 2f);
|
||||
transform.translate(0f, pageWidth + textHeight);
|
||||
transform.scale(1., -1.);
|
||||
}
|
||||
|
||||
bottomLeft = transform.transform(bottomLeft, null);
|
||||
topRight = transform.transform(topRight, null);
|
||||
|
||||
return new Rectangle( //
|
||||
new Point((float) bottomLeft.getX(), (float) bottomLeft.getY()),
|
||||
(float) (topRight.getX() - bottomLeft.getX()),
|
||||
(float) (topRight.getY() - bottomLeft.getY()),
|
||||
page);
|
||||
}
|
||||
|
||||
}
|
||||
-14
@@ -1,14 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
|
||||
|
||||
import java.util.List;
|
||||
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.Data;
|
||||
|
||||
@Data
|
||||
@AllArgsConstructor
|
||||
public class UnclassifiedText {
|
||||
|
||||
private List<TextPageBlock> textBlocks;
|
||||
|
||||
}
|
||||
-384
@@ -1,384 +0,0 @@
|
||||
/*
|
||||
* Licensed to the Apache Software Foundation (ASF) under one or more
|
||||
* contributor license agreements. See the NOTICE file distributed with
|
||||
* this work for additional information regarding copyright ownership.
|
||||
* The ASF licenses this file to You under the Apache License, Version 2.0
|
||||
* (the "License"); you may not use this file except in compliance with
|
||||
* the License. You may obtain a copy of the License at
|
||||
*
|
||||
* http://www.apache.org/licenses/LICENSE-2.0
|
||||
*
|
||||
* Unless required by applicable law or agreed to in writing, software
|
||||
* distributed under the License is distributed on an "AS IS" BASIS,
|
||||
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
* See the License for the specific language governing permissions and
|
||||
* limitations under the License.
|
||||
*/
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.parsing;
|
||||
|
||||
import java.io.IOException;
|
||||
import java.io.InputStream;
|
||||
import java.util.Map;
|
||||
import java.util.WeakHashMap;
|
||||
|
||||
import org.apache.commons.logging.Log;
|
||||
import org.apache.commons.logging.LogFactory;
|
||||
import org.apache.fontbox.ttf.TrueTypeFont;
|
||||
import org.apache.fontbox.util.BoundingBox;
|
||||
import org.apache.pdfbox.contentstream.PDFStreamEngine;
|
||||
import org.apache.pdfbox.contentstream.operator.DrawObject;
|
||||
import org.apache.pdfbox.contentstream.operator.state.Concatenate;
|
||||
import org.apache.pdfbox.contentstream.operator.state.Restore;
|
||||
import org.apache.pdfbox.contentstream.operator.state.Save;
|
||||
import org.apache.pdfbox.contentstream.operator.state.SetGraphicsStateParameters;
|
||||
import org.apache.pdfbox.contentstream.operator.state.SetMatrix;
|
||||
import org.apache.pdfbox.contentstream.operator.text.BeginText;
|
||||
import org.apache.pdfbox.contentstream.operator.text.EndText;
|
||||
import org.apache.pdfbox.contentstream.operator.text.MoveText;
|
||||
import org.apache.pdfbox.contentstream.operator.text.MoveTextSetLeading;
|
||||
import org.apache.pdfbox.contentstream.operator.text.NextLine;
|
||||
import org.apache.pdfbox.contentstream.operator.text.SetCharSpacing;
|
||||
import org.apache.pdfbox.contentstream.operator.text.SetFontAndSize;
|
||||
import org.apache.pdfbox.contentstream.operator.text.SetTextHorizontalScaling;
|
||||
import org.apache.pdfbox.contentstream.operator.text.SetTextLeading;
|
||||
import org.apache.pdfbox.contentstream.operator.text.SetTextRenderingMode;
|
||||
import org.apache.pdfbox.contentstream.operator.text.SetTextRise;
|
||||
import org.apache.pdfbox.contentstream.operator.text.SetWordSpacing;
|
||||
import org.apache.pdfbox.contentstream.operator.text.ShowText;
|
||||
import org.apache.pdfbox.contentstream.operator.text.ShowTextAdjusted;
|
||||
import org.apache.pdfbox.contentstream.operator.text.ShowTextLine;
|
||||
import org.apache.pdfbox.contentstream.operator.text.ShowTextLineAndSpace;
|
||||
import org.apache.pdfbox.cos.COSDictionary;
|
||||
import org.apache.pdfbox.pdmodel.PDPage;
|
||||
import org.apache.pdfbox.pdmodel.common.PDRectangle;
|
||||
import org.apache.pdfbox.pdmodel.font.PDCIDFont;
|
||||
import org.apache.pdfbox.pdmodel.font.PDCIDFontType2;
|
||||
import org.apache.pdfbox.pdmodel.font.PDFont;
|
||||
import org.apache.pdfbox.pdmodel.font.PDFontDescriptor;
|
||||
import org.apache.pdfbox.pdmodel.font.PDSimpleFont;
|
||||
import org.apache.pdfbox.pdmodel.font.PDTrueTypeFont;
|
||||
import org.apache.pdfbox.pdmodel.font.PDType0Font;
|
||||
import org.apache.pdfbox.pdmodel.font.PDType3Font;
|
||||
import org.apache.pdfbox.pdmodel.font.encoding.GlyphList;
|
||||
import org.apache.pdfbox.pdmodel.graphics.state.PDGraphicsState;
|
||||
import org.apache.pdfbox.text.TextPosition;
|
||||
import org.apache.pdfbox.util.Matrix;
|
||||
import org.apache.pdfbox.util.Vector;
|
||||
|
||||
/**
|
||||
* LEGACY text calculations which are known to be incorrect but are depended on by PDFTextStripper.
|
||||
* <p>
|
||||
* This class exists only so that we don't break the code of users who have their own subclasses of
|
||||
* PDFTextStripper. It replaces the mostly empty implementation of showGlyph() in PDFStreamEngine
|
||||
* with a heuristic implementation which is backwards compatible.
|
||||
* <p>
|
||||
* DO NOT USE THIS CODE UNLESS YOU ARE WORKING WITH PDFTextStripper.
|
||||
* THIS CODE IS DELIBERATELY INCORRECT, USE PDFStreamEngine INSTEAD.
|
||||
*/
|
||||
@SuppressWarnings({"PMD", "checkstyle:all"})
|
||||
class LegacyPDFStreamEngine extends PDFStreamEngine {
|
||||
|
||||
private static final Log LOG = LogFactory.getLog(LegacyPDFStreamEngine.class);
|
||||
|
||||
private int pageRotation;
|
||||
private PDRectangle pageSize;
|
||||
private Matrix translateMatrix;
|
||||
private final GlyphList glyphList;
|
||||
private final Map<COSDictionary, Float> fontHeightMap = new WeakHashMap<COSDictionary, Float>();
|
||||
|
||||
|
||||
/**
|
||||
* Constructor.
|
||||
*/
|
||||
LegacyPDFStreamEngine() throws IOException {
|
||||
|
||||
addOperator(new BeginText());
|
||||
addOperator(new Concatenate());
|
||||
addOperator(new DrawObject()); // special text version
|
||||
addOperator(new EndText());
|
||||
addOperator(new SetGraphicsStateParameters());
|
||||
addOperator(new Save());
|
||||
addOperator(new Restore());
|
||||
addOperator(new NextLine());
|
||||
addOperator(new SetCharSpacing());
|
||||
addOperator(new MoveText());
|
||||
addOperator(new MoveTextSetLeading());
|
||||
addOperator(new SetFontAndSize());
|
||||
addOperator(new ShowText());
|
||||
addOperator(new ShowTextAdjusted());
|
||||
addOperator(new SetTextLeading());
|
||||
addOperator(new SetMatrix());
|
||||
addOperator(new SetTextRenderingMode());
|
||||
addOperator(new SetTextRise());
|
||||
addOperator(new SetWordSpacing());
|
||||
addOperator(new SetTextHorizontalScaling());
|
||||
addOperator(new ShowTextLine());
|
||||
addOperator(new ShowTextLineAndSpace());
|
||||
|
||||
// load additional glyph list for Unicode mapping
|
||||
String path = "/org/apache/pdfbox/resources/glyphlist/additional.txt";
|
||||
InputStream input = GlyphList.class.getResourceAsStream(path);
|
||||
glyphList = new GlyphList(GlyphList.getAdobeGlyphList(), input);
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* This will initialize and process the contents of the stream.
|
||||
*
|
||||
* @param page the page to process
|
||||
* @throws java.io.IOException if there is an error accessing the stream.
|
||||
*/
|
||||
@Override
|
||||
public void processPage(PDPage page) throws IOException {
|
||||
|
||||
this.pageRotation = page.getRotation();
|
||||
this.pageSize = page.getCropBox();
|
||||
|
||||
if (pageSize.getLowerLeftX() == 0 && pageSize.getLowerLeftY() == 0) {
|
||||
translateMatrix = null;
|
||||
} else {
|
||||
// translation matrix for cropbox
|
||||
translateMatrix = Matrix.getTranslateInstance(-pageSize.getLowerLeftX(), -pageSize.getLowerLeftY());
|
||||
}
|
||||
super.processPage(page);
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* Called when a glyph is to be processed. The heuristic calculations here were originally
|
||||
* written by Ben Litchfield for PDFStreamEngine.
|
||||
*/
|
||||
@Override
|
||||
protected void showGlyph(Matrix textRenderingMatrix, PDFont font, int code, String unicode, Vector displacement) throws IOException {
|
||||
//
|
||||
// legacy calculations which were previously in PDFStreamEngine
|
||||
//
|
||||
// DO NOT USE THIS CODE UNLESS YOU ARE WORKING WITH PDFTextStripper.
|
||||
// THIS CODE IS DELIBERATELY INCORRECT
|
||||
//
|
||||
|
||||
PDGraphicsState state = getGraphicsState();
|
||||
Matrix ctm = state.getCurrentTransformationMatrix();
|
||||
float fontSize = state.getTextState().getFontSize();
|
||||
float horizontalScaling = state.getTextState().getHorizontalScaling() / 100f;
|
||||
Matrix textMatrix = getTextMatrix();
|
||||
|
||||
float displacementX = displacement.getX();
|
||||
// the sorting algorithm is based on the width of the character. As the displacement
|
||||
// for vertical characters doesn't provide any suitable value for it, we have to
|
||||
// calculate our own
|
||||
if (font.isVertical()) {
|
||||
displacementX = font.getWidth(code) / 1000;
|
||||
// there may be an additional scaling factor for true type fonts
|
||||
TrueTypeFont ttf = null;
|
||||
if (font instanceof PDTrueTypeFont) {
|
||||
ttf = ((PDTrueTypeFont) font).getTrueTypeFont();
|
||||
} else if (font instanceof PDType0Font) {
|
||||
PDCIDFont cidFont = ((PDType0Font) font).getDescendantFont();
|
||||
if (cidFont instanceof PDCIDFontType2) {
|
||||
ttf = ((PDCIDFontType2) cidFont).getTrueTypeFont();
|
||||
}
|
||||
}
|
||||
if (ttf != null && ttf.getUnitsPerEm() != 1000) {
|
||||
displacementX *= 1000f / ttf.getUnitsPerEm();
|
||||
}
|
||||
}
|
||||
|
||||
//
|
||||
// legacy calculations which were previously in PDFStreamEngine
|
||||
//
|
||||
// DO NOT USE THIS CODE UNLESS YOU ARE WORKING WITH PDFTextStripper.
|
||||
// THIS CODE IS DELIBERATELY INCORRECT
|
||||
//
|
||||
|
||||
// (modified) combined displacement, this is calculated *without* taking the character
|
||||
// spacing and word spacing into account, due to legacy code in TextStripper
|
||||
float tx = displacementX * fontSize * horizontalScaling;
|
||||
float ty = displacement.getY() * fontSize;
|
||||
|
||||
// (modified) combined displacement matrix
|
||||
Matrix td = Matrix.getTranslateInstance(tx, ty);
|
||||
|
||||
// (modified) text rendering matrix
|
||||
Matrix nextTextRenderingMatrix = td.multiply(textMatrix).multiply(ctm); // text space -> device space
|
||||
float nextX = nextTextRenderingMatrix.getTranslateX();
|
||||
float nextY = nextTextRenderingMatrix.getTranslateY();
|
||||
|
||||
// (modified) width and height calculations
|
||||
float dxDisplay = nextX - textRenderingMatrix.getTranslateX();
|
||||
Float fontHeight = fontHeightMap.get(font.getCOSObject());
|
||||
if (fontHeight == null) {
|
||||
fontHeight = computeFontHeight(font);
|
||||
fontHeightMap.put(font.getCOSObject(), fontHeight);
|
||||
}
|
||||
float dyDisplay = fontHeight * textRenderingMatrix.getScalingFactorY();
|
||||
|
||||
//
|
||||
// start of the original method
|
||||
//
|
||||
|
||||
// Note on variable names. There are three different units being used in this code.
|
||||
// Character sizes are given in glyph units, text locations are initially given in text
|
||||
// units, and we want to save the data in display units. The variable names should end with
|
||||
// Text or Disp to represent if the values are in text or disp units (no glyph units are
|
||||
// saved).
|
||||
|
||||
float glyphSpaceToTextSpaceFactor = 1 / 1000f;
|
||||
if (font instanceof PDType3Font) {
|
||||
glyphSpaceToTextSpaceFactor = font.getFontMatrix().getScaleX();
|
||||
}
|
||||
|
||||
float spaceWidthText = 0;
|
||||
try {
|
||||
// to avoid crash as described in PDFBOX-614, see what the space displacement should be
|
||||
spaceWidthText = font.getSpaceWidth() * glyphSpaceToTextSpaceFactor;
|
||||
} catch (Throwable exception) {
|
||||
LOG.warn(exception, exception);
|
||||
}
|
||||
|
||||
if (spaceWidthText == 0) {
|
||||
spaceWidthText = font.getAverageFontWidth() * glyphSpaceToTextSpaceFactor;
|
||||
// the average space width appears to be higher than necessary so make it smaller
|
||||
spaceWidthText *= .80f;
|
||||
}
|
||||
if (spaceWidthText == 0) {
|
||||
spaceWidthText = 1.0f; // if could not find font, use a generic value
|
||||
}
|
||||
|
||||
// the space width has to be transformed into display units
|
||||
float spaceWidthDisplay = spaceWidthText * textRenderingMatrix.getScalingFactorX();
|
||||
|
||||
// use our additional glyph list for Unicode mapping
|
||||
String unicodeMapping = font.toUnicode(code, glyphList);
|
||||
|
||||
// when there is no Unicode mapping available, Acrobat simply coerces the character code
|
||||
// into Unicode, so we do the same. Subclasses of PDFStreamEngine don't necessarily want
|
||||
// this, which is why we leave it until this point in PDFTextStreamEngine.
|
||||
if (unicodeMapping == null) {
|
||||
if (font instanceof PDSimpleFont) {
|
||||
char c = (char) code;
|
||||
unicodeMapping = new String(new char[]{c});
|
||||
} else {
|
||||
// Acrobat doesn't seem to coerce composite font's character codes, instead it
|
||||
// skips them. See the "allah2.pdf" TestTextStripper file.
|
||||
return;
|
||||
}
|
||||
}
|
||||
|
||||
// adjust for cropbox if needed
|
||||
Matrix translatedTextRenderingMatrix;
|
||||
if (translateMatrix == null) {
|
||||
translatedTextRenderingMatrix = textRenderingMatrix;
|
||||
} else {
|
||||
translatedTextRenderingMatrix = Matrix.concatenate(translateMatrix, textRenderingMatrix);
|
||||
nextX -= pageSize.getLowerLeftX();
|
||||
nextY -= pageSize.getLowerLeftY();
|
||||
}
|
||||
|
||||
// This is a hack for unicode letter with 2 chars e.g. RA see unicodeProblem.pdf
|
||||
if (unicodeMapping.length() == 2) {
|
||||
processTextPosition(new TextPosition(pageRotation,
|
||||
pageSize.getWidth(),
|
||||
pageSize.getHeight(),
|
||||
translatedTextRenderingMatrix,
|
||||
nextX,
|
||||
nextY,
|
||||
Math.abs(dyDisplay),
|
||||
dxDisplay,
|
||||
Math.abs(spaceWidthDisplay),
|
||||
Character.toString(unicodeMapping.charAt(0)),
|
||||
new int[]{code},
|
||||
font,
|
||||
fontSize,
|
||||
(int) (fontSize * textMatrix.getScalingFactorX())));
|
||||
processTextPosition(new TextPosition(pageRotation,
|
||||
pageSize.getWidth(),
|
||||
pageSize.getHeight(),
|
||||
translatedTextRenderingMatrix,
|
||||
nextX,
|
||||
nextY,
|
||||
Math.abs(dyDisplay),
|
||||
dxDisplay,
|
||||
Math.abs(spaceWidthDisplay),
|
||||
Character.toString(unicodeMapping.charAt(1)),
|
||||
new int[]{code},
|
||||
font,
|
||||
fontSize,
|
||||
(int) (fontSize * textMatrix.getScalingFactorX())));
|
||||
} else {
|
||||
|
||||
processTextPosition(new TextPosition(pageRotation,
|
||||
pageSize.getWidth(),
|
||||
pageSize.getHeight(),
|
||||
translatedTextRenderingMatrix,
|
||||
nextX,
|
||||
nextY,
|
||||
Math.abs(dyDisplay),
|
||||
dxDisplay,
|
||||
Math.abs(spaceWidthDisplay),
|
||||
unicodeMapping,
|
||||
new int[]{code},
|
||||
font,
|
||||
fontSize,
|
||||
(int) (fontSize * textMatrix.getScalingFactorX())));
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* Compute the font height. Override this if you want to use own calculations.
|
||||
*
|
||||
* @param font the font.
|
||||
* @return the font height.
|
||||
* @throws IOException if there is an error while getting the font bounding box.
|
||||
*/
|
||||
protected float computeFontHeight(PDFont font) throws IOException {
|
||||
|
||||
BoundingBox bbox = font.getBoundingBox();
|
||||
if (bbox.getLowerLeftY() < Short.MIN_VALUE) {
|
||||
// PDFBOX-2158 and PDFBOX-3130
|
||||
// files by Salmat eSolutions / ClibPDF Library
|
||||
bbox.setLowerLeftY(-(bbox.getLowerLeftY() + 65536));
|
||||
}
|
||||
// 1/2 the bbox is used as the height todo: why?
|
||||
float glyphHeight = bbox.getHeight() / 2;
|
||||
|
||||
// sometimes the bbox has very high values, but CapHeight is OK
|
||||
PDFontDescriptor fontDescriptor = font.getFontDescriptor();
|
||||
if (fontDescriptor != null) {
|
||||
float capHeight = fontDescriptor.getCapHeight();
|
||||
if (Float.compare(capHeight, 0) != 0 && (capHeight < glyphHeight || Float.compare(glyphHeight, 0) == 0)) {
|
||||
glyphHeight = capHeight;
|
||||
}
|
||||
// PDFBOX-3464, PDFBOX-4480, PDFBOX-4553:
|
||||
// sometimes even CapHeight has very high value, but Ascent and Descent are ok
|
||||
float ascent = fontDescriptor.getAscent();
|
||||
float descent = fontDescriptor.getDescent();
|
||||
if (capHeight > ascent && ascent > 0 && descent < 0 && ((ascent - descent) / 2 < glyphHeight || Float.compare(glyphHeight, 0) == 0)) {
|
||||
glyphHeight = (ascent - descent) / 2;
|
||||
}
|
||||
}
|
||||
|
||||
// transformPoint from glyph space -> text space
|
||||
float height;
|
||||
if (font instanceof PDType3Font) {
|
||||
height = font.getFontMatrix().transformPoint(0, glyphHeight).y;
|
||||
} else {
|
||||
height = glyphHeight / 1000;
|
||||
}
|
||||
|
||||
return height;
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* A method provided as an event interface to allow a subclass to perform some specific
|
||||
* functionality when text needs to be processed.
|
||||
*
|
||||
* @param text The text to be processed.
|
||||
*/
|
||||
protected void processTextPosition(TextPosition text) {
|
||||
// subclasses can override to provide specific functionality
|
||||
}
|
||||
|
||||
}
|
||||
-335
@@ -1,335 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.parsing;
|
||||
|
||||
import java.awt.geom.Point2D;
|
||||
import java.io.IOException;
|
||||
import java.util.ArrayList;
|
||||
import java.util.Comparator;
|
||||
import java.util.List;
|
||||
|
||||
import org.apache.pdfbox.contentstream.operator.Operator;
|
||||
import org.apache.pdfbox.contentstream.operator.OperatorName;
|
||||
import org.apache.pdfbox.contentstream.operator.color.SetNonStrokingColor;
|
||||
import org.apache.pdfbox.contentstream.operator.color.SetNonStrokingColorN;
|
||||
import org.apache.pdfbox.contentstream.operator.color.SetNonStrokingColorSpace;
|
||||
import org.apache.pdfbox.contentstream.operator.color.SetNonStrokingDeviceCMYKColor;
|
||||
import org.apache.pdfbox.contentstream.operator.color.SetNonStrokingDeviceGrayColor;
|
||||
import org.apache.pdfbox.contentstream.operator.color.SetNonStrokingDeviceRGBColor;
|
||||
import org.apache.pdfbox.contentstream.operator.color.SetStrokingColor;
|
||||
import org.apache.pdfbox.contentstream.operator.color.SetStrokingColorN;
|
||||
import org.apache.pdfbox.contentstream.operator.color.SetStrokingColorSpace;
|
||||
import org.apache.pdfbox.contentstream.operator.color.SetStrokingDeviceCMYKColor;
|
||||
import org.apache.pdfbox.contentstream.operator.color.SetStrokingDeviceGrayColor;
|
||||
import org.apache.pdfbox.contentstream.operator.color.SetStrokingDeviceRGBColor;
|
||||
import org.apache.pdfbox.contentstream.operator.state.SetFlatness;
|
||||
import org.apache.pdfbox.contentstream.operator.state.SetLineCapStyle;
|
||||
import org.apache.pdfbox.contentstream.operator.state.SetLineDashPattern;
|
||||
import org.apache.pdfbox.contentstream.operator.state.SetLineJoinStyle;
|
||||
import org.apache.pdfbox.contentstream.operator.state.SetLineMiterLimit;
|
||||
import org.apache.pdfbox.contentstream.operator.state.SetLineWidth;
|
||||
import org.apache.pdfbox.contentstream.operator.state.SetRenderingIntent;
|
||||
import org.apache.pdfbox.contentstream.operator.text.SetFontAndSize;
|
||||
import org.apache.pdfbox.cos.COSBase;
|
||||
import org.apache.pdfbox.cos.COSNumber;
|
||||
import org.apache.pdfbox.pdmodel.PDDocument;
|
||||
import org.apache.pdfbox.pdmodel.PDPage;
|
||||
import org.apache.pdfbox.text.TextPosition;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Ruling;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.RedTextPosition;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
|
||||
|
||||
import lombok.Getter;
|
||||
import lombok.Setter;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
|
||||
@Slf4j
|
||||
public class PDFLinesTextStripper extends PDFTextStripper {
|
||||
|
||||
@Getter
|
||||
private final List<TextPositionSequence> textPositionSequences = new ArrayList<>();
|
||||
@Getter
|
||||
private final List<Ruling> rulings = new ArrayList<>();
|
||||
private final List<Ruling> graphicsPath = new ArrayList<>();
|
||||
@Setter
|
||||
protected PDPage pdpage;
|
||||
@Getter
|
||||
private int minCharWidth;
|
||||
@Getter
|
||||
private int maxCharWidth;
|
||||
@Getter
|
||||
private int minCharHeight;
|
||||
@Getter
|
||||
private int maxCharHeight;
|
||||
|
||||
private float path_x;
|
||||
private float path_y;
|
||||
|
||||
@Setter
|
||||
private int pageNumber;
|
||||
|
||||
|
||||
public PDFLinesTextStripper() throws IOException {
|
||||
|
||||
super();
|
||||
this.addOperator(new SetStrokingColorSpace());
|
||||
this.addOperator(new SetNonStrokingColorSpace());
|
||||
this.addOperator(new SetLineDashPattern());
|
||||
this.addOperator(new SetStrokingDeviceGrayColor());
|
||||
this.addOperator(new SetNonStrokingDeviceGrayColor());
|
||||
this.addOperator(new SetFlatness());
|
||||
this.addOperator(new SetLineJoinStyle());
|
||||
this.addOperator(new SetLineCapStyle());
|
||||
this.addOperator(new SetStrokingDeviceCMYKColor());
|
||||
this.addOperator(new SetNonStrokingDeviceCMYKColor());
|
||||
this.addOperator(new SetLineMiterLimit());
|
||||
this.addOperator(new SetStrokingDeviceRGBColor());
|
||||
this.addOperator(new SetNonStrokingDeviceRGBColor());
|
||||
this.addOperator(new SetRenderingIntent());
|
||||
this.addOperator(new SetStrokingColor());
|
||||
this.addOperator(new SetNonStrokingColor());
|
||||
this.addOperator(new SetStrokingColorN());
|
||||
this.addOperator(new SetNonStrokingColorN());
|
||||
this.addOperator(new SetFontAndSize());
|
||||
this.addOperator(new SetLineWidth());
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
protected void processOperator(Operator operator, List<COSBase> arguments) throws IOException {
|
||||
|
||||
String operation = operator.getName();
|
||||
|
||||
//move
|
||||
switch (operation) {
|
||||
case OperatorName.MOVE_TO:
|
||||
if (arguments.size() == 2) {
|
||||
Point2D.Float pos = transformPosition(floatValue(arguments.get(0)), floatValue(arguments.get(1)));
|
||||
path_x = (float) pos.getX();
|
||||
path_y = (float) pos.getY();
|
||||
}
|
||||
break;
|
||||
|
||||
//line
|
||||
case OperatorName.LINE_TO:
|
||||
if (arguments.size() == 2) {
|
||||
Point2D.Float pos = transformPosition(floatValue(arguments.get(0)), floatValue(arguments.get(1)));
|
||||
|
||||
// The direction of vertical lines must always be from bottom to top for the table extraction algorithm.
|
||||
if (pos.getY() > path_y) {
|
||||
graphicsPath.add(new Ruling(new Point2D.Float(path_x, path_y), new Point2D.Float((float) pos.getX(), (float) pos.getY())));
|
||||
} else {
|
||||
graphicsPath.add(new Ruling(new Point2D.Float(path_x, (float) pos.getY()), new Point2D.Float((float) pos.getX(), path_y)));
|
||||
}
|
||||
|
||||
path_x = (float) pos.getX();
|
||||
path_y = (float) pos.getY();
|
||||
}
|
||||
break;
|
||||
|
||||
//rectangle
|
||||
case OperatorName.APPEND_RECT:
|
||||
|
||||
if (arguments.size() == 4) {
|
||||
float x = floatValue(arguments.get(0));
|
||||
float y = floatValue(arguments.get(1));
|
||||
float width = floatValue(arguments.get(2));
|
||||
float height = floatValue(arguments.get(3));
|
||||
|
||||
Point2D p1 = transformPosition(x, y);
|
||||
Point2D p2 = transformPosition(x + width, y + height);
|
||||
|
||||
// Horizontal lines
|
||||
graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p1.getY()), new Point2D.Float((float) p2.getX(), (float) p1.getY())));
|
||||
graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p2.getY()), new Point2D.Float((float) p2.getX(), (float) p2.getY())));
|
||||
|
||||
// Vertical lines, direction must always be from bottom to top for the table extraction algorithm.
|
||||
if (p2.getY() > p1.getY()) {
|
||||
graphicsPath.add(new Ruling(new Point2D.Float((float) p2.getX(), (float) p1.getY()), new Point2D.Float((float) p2.getX(), (float) p2.getY())));
|
||||
} else {
|
||||
graphicsPath.add(new Ruling(new Point2D.Float((float) p2.getX(), (float) p2.getY()), new Point2D.Float((float) p2.getX(), (float) p1.getY())));
|
||||
}
|
||||
if (p2.getY() > p1.getY()) {
|
||||
graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p1.getY()), new Point2D.Float((float) p1.getX(), (float) p2.getY())));
|
||||
} else {
|
||||
graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p2.getY()), new Point2D.Float((float) p1.getX(), (float) p1.getY())));
|
||||
}
|
||||
}
|
||||
break;
|
||||
|
||||
//fill
|
||||
case OperatorName.FILL_NON_ZERO:
|
||||
case OperatorName.LEGACY_FILL_NON_ZERO:
|
||||
case OperatorName.FILL_EVEN_ODD:
|
||||
addVisibleRulings(graphicsPath, false);
|
||||
graphicsPath.clear();
|
||||
break;
|
||||
|
||||
//stroke
|
||||
case OperatorName.STROKE_PATH:
|
||||
addVisibleRulings(graphicsPath, true);
|
||||
graphicsPath.clear();
|
||||
break;
|
||||
|
||||
//cancel path
|
||||
case OperatorName.ENDPATH:
|
||||
graphicsPath.clear();
|
||||
break;
|
||||
|
||||
}
|
||||
|
||||
super.processOperator(operator, arguments);
|
||||
}
|
||||
|
||||
|
||||
private float floatValue(COSBase value) {
|
||||
|
||||
if (value instanceof COSNumber) {
|
||||
return ((COSNumber) value).floatValue();
|
||||
} else {
|
||||
return 0;
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
private Point2D.Float transformPosition(float x, float y) {
|
||||
|
||||
return super.transformedPoint(x, y);
|
||||
}
|
||||
|
||||
|
||||
private void addVisibleRulings(List<Ruling> path, boolean stroke) throws IOException {
|
||||
|
||||
try {
|
||||
if (stroke && !getGraphicsState().getStrokingColor().isPattern() && getGraphicsState().getStrokingColor()
|
||||
.toRGB() == 0 || !stroke && !getGraphicsState().getNonStrokingColor().isPattern() && getGraphicsState().getNonStrokingColor().toRGB() == 0) {
|
||||
rulings.addAll(path);
|
||||
}
|
||||
} catch (UnsupportedOperationException e) {
|
||||
log.debug("UnsupportedOperationException: " + getGraphicsState().getStrokingColor().getColorSpace().getName() + " or " + getGraphicsState().getNonStrokingColor()
|
||||
.getColorSpace()
|
||||
.getName() + " does not support toRGB");
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public void writeString(String text, List<TextPosition> textPositions) throws IOException {
|
||||
|
||||
int startIndex = 0;
|
||||
RedTextPosition previous = null;
|
||||
|
||||
textPositions.sort(Comparator.comparing(TextPosition::getXDirAdj));
|
||||
|
||||
for (int i = 0; i <= textPositions.size() - 1; i++) {
|
||||
|
||||
if (!textPositionSequences.isEmpty()) {
|
||||
previous = textPositionSequences.get(textPositionSequences.size() - 1)
|
||||
.getTextPositions()
|
||||
.get(textPositionSequences.get(textPositionSequences.size() - 1).getTextPositions().size() - 1);
|
||||
}
|
||||
|
||||
int charWidth = (int) textPositions.get(i).getWidthDirAdj();
|
||||
if (charWidth < minCharWidth) {
|
||||
minCharWidth = charWidth;
|
||||
}
|
||||
if (charWidth > maxCharWidth) {
|
||||
maxCharWidth = charWidth;
|
||||
}
|
||||
|
||||
int charHeight = (int) textPositions.get(i).getHeightDir();
|
||||
if (charHeight < minCharHeight) {
|
||||
minCharHeight = charHeight;
|
||||
}
|
||||
if (charWidth > maxCharHeight) {
|
||||
maxCharHeight = charHeight;
|
||||
}
|
||||
|
||||
if (i == 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i).getUnicode().equals("\u00A0") || textPositions.get(i).getUnicode().equals("\t"))) {
|
||||
startIndex++;
|
||||
continue;
|
||||
}
|
||||
|
||||
// Strange but sometimes this is happening, for example: Metolachlor2.pdf
|
||||
if (i > 0 && textPositions.get(i).getXDirAdj() < textPositions.get(i - 1).getXDirAdj()) {
|
||||
List<TextPosition> sublist = textPositions.subList(startIndex, i);
|
||||
if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0).getUnicode().equals("\u00A0") || sublist.get(0)
|
||||
.getUnicode()
|
||||
.equals("\t")))) {
|
||||
textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
|
||||
}
|
||||
startIndex = i;
|
||||
}
|
||||
|
||||
if (textPositions.get(i).getRotation() == 0 && i > 0 && textPositions.get(i).getX() > textPositions.get(i - 1).getEndX() + 1) {
|
||||
List<TextPosition> sublist = textPositions.subList(startIndex, i);
|
||||
if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0).getUnicode().equals("\u00A0") || sublist.get(0)
|
||||
.getUnicode()
|
||||
.equals("\t")))) {
|
||||
textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
|
||||
}
|
||||
startIndex = i;
|
||||
}
|
||||
|
||||
if (i > 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i).getUnicode().equals("\u00A0") || textPositions.get(i)
|
||||
.getUnicode()
|
||||
.equals("\t")) && i <= textPositions.size() - 2) {
|
||||
List<TextPosition> sublist = textPositions.subList(startIndex, i);
|
||||
if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0).getUnicode().equals("\u00A0") || sublist.get(0)
|
||||
.getUnicode()
|
||||
.equals("\t")))) {
|
||||
|
||||
// Remove false sequence ends (whitespaces)
|
||||
if (previous != null && sublist.get(0).getYDirAdj() == previous.getYDirAdj() && sublist.get(0)
|
||||
.getXDirAdj() - (previous.getXDirAdj() + previous.getWidthDirAdj()) < 0.01) {
|
||||
for (TextPosition textPosition : sublist) {
|
||||
textPositionSequences.get(textPositionSequences.size() - 1).add(textPosition);
|
||||
}
|
||||
} else {
|
||||
textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
|
||||
}
|
||||
}
|
||||
startIndex = i + 1;
|
||||
}
|
||||
}
|
||||
|
||||
List<TextPosition> sublist = textPositions.subList(startIndex, textPositions.size());
|
||||
if (!sublist.isEmpty() && (sublist.get(sublist.size() - 1).getUnicode().equals(" ") || sublist.get(sublist.size() - 1)
|
||||
.getUnicode()
|
||||
.equals("\u00A0") || sublist.get(sublist.size() - 1).getUnicode().equals("\t"))) {
|
||||
sublist = sublist.subList(0, sublist.size() - 1);
|
||||
}
|
||||
|
||||
if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0).getUnicode().equals("\u00A0") || sublist.get(0)
|
||||
.getUnicode()
|
||||
.equals("\t")))) {
|
||||
if (previous != null && sublist.get(0).getYDirAdj() == previous.getYDirAdj() && sublist.get(0)
|
||||
.getXDirAdj() - (previous.getXDirAdj() + previous.getWidthDirAdj()) < 0.01) {
|
||||
for (TextPosition t : sublist) {
|
||||
textPositionSequences.get(textPositionSequences.size() - 1).add(t);
|
||||
}
|
||||
} else {
|
||||
textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
|
||||
}
|
||||
}
|
||||
super.writeString(text);
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public String getText(PDDocument doc) throws IOException {
|
||||
|
||||
minCharWidth = Integer.MAX_VALUE;
|
||||
maxCharWidth = 0;
|
||||
minCharHeight = Integer.MAX_VALUE;
|
||||
maxCharHeight = 0;
|
||||
textPositionSequences.clear();
|
||||
rulings.clear();
|
||||
graphicsPath.clear();
|
||||
path_x = 0.0f;
|
||||
path_y = 0.0f;
|
||||
|
||||
return super.getText(doc);
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
-2070
File diff suppressed because it is too large
Load Diff
-25
@@ -1,25 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
|
||||
|
||||
import static java.util.stream.Collectors.toSet;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.Comparator;
|
||||
import java.util.Iterator;
|
||||
import java.util.List;
|
||||
|
||||
import org.springframework.stereotype.Service;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.FloatFrequencyCounter;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.Orientation;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Ruling;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.StringFrequencyCounter;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.RulingTextDirAdjustUtil;
|
||||
|
||||
public interface BlockificationService {
|
||||
|
||||
ClassificationPage blockify(List<TextPositionSequence> textPositions, List<Ruling> horizontalRulingLines, List<Ruling> verticalRulingLines);
|
||||
}
|
||||
-169
@@ -1,169 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
|
||||
|
||||
import java.util.List;
|
||||
|
||||
import org.springframework.beans.factory.annotation.Value;
|
||||
import org.springframework.stereotype.Service;
|
||||
|
||||
import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Point;
|
||||
import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Rectangle;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.FloatFrequencyCounter;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Cell;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.TablePageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.PositionUtils;
|
||||
|
||||
@Service
|
||||
public class BodyTextFrameService {
|
||||
|
||||
@Value("${application.type:RedactManager}")
|
||||
private String applicationType;
|
||||
|
||||
private static final float APPROXIMATE_HEADER_LINE_COUNT = 2.9f;
|
||||
|
||||
|
||||
/**
|
||||
* Adjusts and sets the body text frame to a page.
|
||||
* Note: This needs to use Pdf Coordinate System where {0,0} rotated with the page rotation.
|
||||
* 0 -> LowerLeft
|
||||
* 90 -> UpperLeft
|
||||
* 180 -> UpperRight
|
||||
* 270 -> LowerRight
|
||||
* The aspect ratio of the page is also regarded.
|
||||
*
|
||||
* @param page The page
|
||||
* @param bodyTextFrame frame that contains the main text on portrait pages
|
||||
* @param landscapeBodyTextFrame frame that contains the main text on landscape pages
|
||||
*/
|
||||
public void setBodyTextFrameAdjustedToPage(ClassificationPage page, Rectangle bodyTextFrame, Rectangle landscapeBodyTextFrame) {
|
||||
|
||||
Rectangle textFrame = page.isLandscape() ? landscapeBodyTextFrame : bodyTextFrame;
|
||||
|
||||
if (page.getPageWidth() > page.getPageHeight() && page.getRotation() == 270) {
|
||||
textFrame = new Rectangle(new Point(textFrame.getTopLeft().getY(), page.getPageHeight() - textFrame.getTopLeft().getX() - textFrame.getWidth()),
|
||||
textFrame.getHeight(),
|
||||
textFrame.getWidth(),
|
||||
0);
|
||||
} else if (page.getPageWidth() > page.getPageHeight() && page.getRotation() != 0) {
|
||||
textFrame = new Rectangle(new Point(textFrame.getTopLeft().getY(), textFrame.getTopLeft().getX()), textFrame.getHeight(), textFrame.getWidth(), page.getPageNumber());
|
||||
} else if (page.getRotation() == 180) {
|
||||
textFrame = new Rectangle(new Point(textFrame.getTopLeft().getX(), page.getPageHeight() - textFrame.getTopLeft().getY() - textFrame.getHeight()),
|
||||
textFrame.getWidth(),
|
||||
textFrame.getHeight(),
|
||||
0);
|
||||
}
|
||||
page.setBodyTextFrame(textFrame);
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* Calculates the frame that contains the main text, text outside the frame will be e.g. headers or footers.
|
||||
* Note: This needs to use Pdf Coordinate System where {0,0} rotated with the page rotation.
|
||||
* 0 -> LowerLeft
|
||||
* 90 -> UpperLeft
|
||||
* 180 -> UpperRight
|
||||
* 270 -> LowerRight
|
||||
* The aspect ratio of the page is also regarded.
|
||||
*
|
||||
* @param pages List of all pages
|
||||
* @param documentFontSizeCounter Statistics of the document
|
||||
* @param landscape Calculate for landscape or portrait
|
||||
* @return Rectangle of the text frame
|
||||
*/
|
||||
public Rectangle calculateBodyTextFrame(List<ClassificationPage> pages, FloatFrequencyCounter documentFontSizeCounter, boolean landscape) {
|
||||
|
||||
BodyTextFrameExpansionsRectangle expansionsRectangle = new BodyTextFrameExpansionsRectangle();
|
||||
|
||||
for (ClassificationPage page : pages) {
|
||||
|
||||
if (page.getTextBlocks().isEmpty() || landscape != page.isLandscape()) {
|
||||
continue;
|
||||
}
|
||||
|
||||
for (AbstractPageBlock container : page.getTextBlocks()) {
|
||||
|
||||
if (container instanceof TextPageBlock) {
|
||||
TextPageBlock textBlock = (TextPageBlock) container;
|
||||
if (textBlock.getMostPopularWordFont() == null || textBlock.getMostPopularWordStyle() == null) {
|
||||
continue;
|
||||
}
|
||||
|
||||
float approxLineCount = PositionUtils.getApproxLineCount(textBlock);
|
||||
if (applicationType.equals("DocuMine") && approxLineCount < APPROXIMATE_HEADER_LINE_COUNT && textBlock.getMaxY() >= page.getPageHeight() - (page.getPageHeight() / 10)
|
||||
|| applicationType.equals("RedactManager") && approxLineCount < APPROXIMATE_HEADER_LINE_COUNT){
|
||||
continue;
|
||||
}
|
||||
|
||||
if (documentFontSizeCounter.getMostPopular() != null && textBlock.getMostPopularWordFontSize() >= documentFontSizeCounter.getMostPopular()) {
|
||||
|
||||
expandRectangle(textBlock, page, expansionsRectangle);
|
||||
}
|
||||
}
|
||||
|
||||
if (container instanceof TablePageBlock) {
|
||||
TablePageBlock table = (TablePageBlock) container;
|
||||
for (List<Cell> row : table.getRows()) {
|
||||
for (Cell cell : row) {
|
||||
|
||||
if (cell == null || cell.getTextBlocks() == null) {
|
||||
continue;
|
||||
}
|
||||
for (TextPageBlock textBlock : cell.getTextBlocks()) {
|
||||
expandRectangle(textBlock, page, expansionsRectangle);
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
return new Rectangle(new Point(expansionsRectangle.minX, expansionsRectangle.minY),
|
||||
expansionsRectangle.maxX - expansionsRectangle.minX,
|
||||
expansionsRectangle.maxY - expansionsRectangle.minY,
|
||||
0);
|
||||
}
|
||||
|
||||
|
||||
private void expandRectangle(TextPageBlock textBlock, ClassificationPage page, BodyTextFrameExpansionsRectangle expansionsRectangle) {
|
||||
|
||||
if (page.getPageWidth() > page.getPageHeight() && page.getRotation() != 0) {
|
||||
if (textBlock.getPdfMinY() < expansionsRectangle.minX) {
|
||||
expansionsRectangle.minX = textBlock.getPdfMinY();
|
||||
}
|
||||
if (textBlock.getPdfMaxY() > expansionsRectangle.maxX) {
|
||||
expansionsRectangle.maxX = textBlock.getPdfMaxY();
|
||||
}
|
||||
if (textBlock.getPdfMinX() < expansionsRectangle.minY) {
|
||||
expansionsRectangle.minY = textBlock.getPdfMinX();
|
||||
}
|
||||
if (textBlock.getPdfMaxX() > expansionsRectangle.maxY) {
|
||||
expansionsRectangle.maxY = textBlock.getPdfMaxX();
|
||||
}
|
||||
} else {
|
||||
if (textBlock.getPdfMinX() < expansionsRectangle.minX) {
|
||||
expansionsRectangle.minX = textBlock.getPdfMinX();
|
||||
}
|
||||
if (textBlock.getPdfMaxX() > expansionsRectangle.maxX) {
|
||||
expansionsRectangle.maxX = textBlock.getPdfMaxX();
|
||||
}
|
||||
if (textBlock.getPdfMinY() < expansionsRectangle.minY) {
|
||||
expansionsRectangle.minY = textBlock.getPdfMinY();
|
||||
}
|
||||
if (textBlock.getPdfMaxY() > expansionsRectangle.maxY) {
|
||||
expansionsRectangle.maxY = textBlock.getPdfMaxY();
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
private class BodyTextFrameExpansionsRectangle {
|
||||
|
||||
float minX = 10000;
|
||||
float maxX = -100;
|
||||
float minY = 10000;
|
||||
float maxY = -100;
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
-7
@@ -1,7 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationDocument;
|
||||
|
||||
public interface ClassificationService {
|
||||
void classifyDocument(ClassificationDocument document);
|
||||
}
|
||||
-149
@@ -1,149 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
|
||||
|
||||
import java.awt.geom.Rectangle2D;
|
||||
import java.util.Collection;
|
||||
import java.util.Comparator;
|
||||
import java.util.HashMap;
|
||||
import java.util.LinkedList;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
import java.util.stream.IntStream;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
|
||||
|
||||
import lombok.experimental.UtilityClass;
|
||||
|
||||
@UtilityClass
|
||||
public class ColumnDetectionService {
|
||||
|
||||
private static final double SPLITTABLE_LINE_PERCENTAGE_THRESHOLD = 0.6;
|
||||
private static final int MAX_NUMBER_OF_COLUMNS = 4;
|
||||
|
||||
|
||||
public List<Rectangle2D> detectColumns(List<TextPositionSequence> textPositionSequences, Rectangle2D mainBodyTextFrame) {
|
||||
|
||||
if (textPositionSequences.size() < 2) {
|
||||
return List.of(mainBodyTextFrame);
|
||||
}
|
||||
|
||||
List<List<Rectangle2D>> linesWithGaps = LineDetectionService.findLinesWithGaps(textPositionSequences);
|
||||
|
||||
Map<Integer, List<Integer>> linesWithMatchingGapIndices = new HashMap<>();
|
||||
for (int numberOfColumns = 2; numberOfColumns <= MAX_NUMBER_OF_COLUMNS; numberOfColumns++) {
|
||||
linesWithMatchingGapIndices.put(numberOfColumns, findConsecutiveLinesWithMatchingGaps(linesWithGaps, mainBodyTextFrame.getWidth(), numberOfColumns));
|
||||
}
|
||||
|
||||
int optimalNumberOfColumns = findOptimalNumberOfColumns(linesWithMatchingGapIndices, linesWithGaps.size());
|
||||
if (optimalNumberOfColumns == 1) {
|
||||
return List.of(mainBodyTextFrame);
|
||||
}
|
||||
return buildColumns(mainBodyTextFrame, getLinesWithMatchingGaps(linesWithMatchingGapIndices.get(optimalNumberOfColumns), linesWithGaps), optimalNumberOfColumns);
|
||||
}
|
||||
|
||||
|
||||
private static List<Integer> findConsecutiveLinesWithMatchingGaps(List<List<Rectangle2D>> linesWithGaps, double width, int numberOfColumns) {
|
||||
|
||||
List<Boolean> booleans = lineHasMatchingGap(linesWithGaps, width, numberOfColumns);
|
||||
return findConsecutiveTrueIndicesWithMaxLengthRun(booleans);
|
||||
}
|
||||
|
||||
|
||||
private List<Boolean> lineHasMatchingGap(List<List<Rectangle2D>> linesWithGaps, double width, int numberOfColumns) {
|
||||
|
||||
return linesWithGaps.stream()
|
||||
.map(blocksWithGaps -> IntStream.range(1, numberOfColumns)
|
||||
.allMatch(columnIndex -> noBlocksIntersectX(blocksWithGaps, calculateGapLocation(width, numberOfColumns, columnIndex))))
|
||||
.toList();
|
||||
}
|
||||
|
||||
|
||||
private List<Integer> findConsecutiveTrueIndicesWithMaxLengthRun(List<Boolean> booleans) {
|
||||
|
||||
List<Integer> maxConsecutiveTrueIndices = new LinkedList<>();
|
||||
List<Integer> currentConsecutiveTrueIndices = new LinkedList<>();
|
||||
for (int i = 0; i < booleans.size(); i++) {
|
||||
if (!booleans.get(i)) {
|
||||
if (currentConsecutiveTrueIndices.isEmpty()) {
|
||||
continue;
|
||||
}
|
||||
if (currentConsecutiveTrueIndices.size() > maxConsecutiveTrueIndices.size()) {
|
||||
maxConsecutiveTrueIndices = currentConsecutiveTrueIndices;
|
||||
}
|
||||
currentConsecutiveTrueIndices = new LinkedList<>();
|
||||
continue;
|
||||
}
|
||||
currentConsecutiveTrueIndices.add(i);
|
||||
}
|
||||
if (currentConsecutiveTrueIndices.size() > maxConsecutiveTrueIndices.size()) {
|
||||
return currentConsecutiveTrueIndices;
|
||||
}
|
||||
return maxConsecutiveTrueIndices;
|
||||
}
|
||||
|
||||
|
||||
private static int findOptimalNumberOfColumns(Map<Integer, List<Integer>> linesWithMatchingGapIndices, Integer numberOfLines) {
|
||||
|
||||
return linesWithMatchingGapIndices.entrySet()
|
||||
.stream()
|
||||
.max(comparePercentages(numberOfLines))
|
||||
.filter(entry -> percentageIsAboveThreshold(entry, numberOfLines))
|
||||
.map(Map.Entry::getKey)
|
||||
.orElse(1);
|
||||
}
|
||||
|
||||
|
||||
private List<Rectangle2D> buildColumns(Rectangle2D mainBodyTextFrame, List<Rectangle2D> rectanglesToMerge, int optimalColumnCount) {
|
||||
|
||||
if (optimalColumnCount == 1 || rectanglesToMerge.isEmpty()) {
|
||||
return List.of(mainBodyTextFrame);
|
||||
}
|
||||
|
||||
double maxY = rectanglesToMerge.get(0).getMaxY();
|
||||
double minY = rectanglesToMerge.get(rectanglesToMerge.size() - 1).getMinY();
|
||||
|
||||
List<Rectangle2D> columns = new LinkedList<>();
|
||||
double width = mainBodyTextFrame.getWidth() / optimalColumnCount;
|
||||
double height = maxY - minY;
|
||||
for (int i = 0; i < optimalColumnCount; i++) {
|
||||
columns.add(new Rectangle2D.Double(mainBodyTextFrame.getMinY() + i * width, minY, width, height));
|
||||
}
|
||||
return columns;
|
||||
}
|
||||
|
||||
|
||||
private Comparator<Map.Entry<Integer, List<Integer>>> comparePercentages(Integer numberOfLines) {
|
||||
|
||||
return Comparator.comparingDouble(entry -> calculatePercentage(entry.getValue().size(), numberOfLines));
|
||||
}
|
||||
|
||||
|
||||
private List<Rectangle2D> getLinesWithMatchingGaps(List<Integer> linesWithMatchingGapIndices, List<List<Rectangle2D>> linesWithGaps) {
|
||||
|
||||
return linesWithMatchingGapIndices.stream().map(linesWithGaps::get).flatMap(Collection::stream).toList();
|
||||
}
|
||||
|
||||
|
||||
private boolean percentageIsAboveThreshold(Map.Entry<Integer, List<Integer>> entry, Integer numberOfLines) {
|
||||
|
||||
return calculatePercentage(entry.getValue().size(), numberOfLines) > SPLITTABLE_LINE_PERCENTAGE_THRESHOLD;
|
||||
}
|
||||
|
||||
|
||||
private double calculatePercentage(Integer numberOfMatchingLines, Integer numberOfLines) {
|
||||
|
||||
return ((double) numberOfMatchingLines) / ((double) numberOfLines);
|
||||
}
|
||||
|
||||
|
||||
private double calculateGapLocation(double pageWidth, int numberOfColumns, int columnIndex) {
|
||||
|
||||
return (pageWidth / numberOfColumns) * columnIndex;
|
||||
}
|
||||
|
||||
|
||||
private Boolean noBlocksIntersectX(List<Rectangle2D> blocksWithGaps, double x) {
|
||||
|
||||
return blocksWithGaps.stream().noneMatch(rect -> rect.getMaxX() > x && rect.getMinX() < x);
|
||||
}
|
||||
|
||||
}
|
||||
-200
@@ -1,200 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
|
||||
|
||||
import static java.util.stream.Collectors.toSet;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.Comparator;
|
||||
import java.util.Iterator;
|
||||
import java.util.List;
|
||||
import java.util.regex.Matcher;
|
||||
import java.util.regex.Pattern;
|
||||
|
||||
import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
|
||||
import org.springframework.stereotype.Service;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.FloatFrequencyCounter;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.Orientation;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Ruling;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.StringFrequencyCounter;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.RulingTextDirAdjustUtil;
|
||||
|
||||
@Service
|
||||
@ConditionalOnProperty(prefix = "application", name = "type", havingValue = "DocuMine")
|
||||
@SuppressWarnings("all")
|
||||
public class DocuMineBlockificationService implements BlockificationService{
|
||||
|
||||
static final float THRESHOLD = 1f;
|
||||
|
||||
|
||||
/**
|
||||
* This method is building blocks by expanding the minX/maxX and minY/maxY value on each word that is not split by the conditions.
|
||||
* This method must use text direction adjusted postions (DirAdj). Where {0,0} is on the upper left. Never try to change this!
|
||||
* Rulings (Table lines) must be adjusted to the text directions as well, when checking if a block is split by a ruling.
|
||||
* @param textPositions The words of a page.
|
||||
* @param horizontalRulingLines Horizontal table lines.
|
||||
* @param verticalRulingLines Vertical table lines.
|
||||
* @return Page object that contains the Textblock and text statistics.
|
||||
*/
|
||||
public ClassificationPage blockify(List<TextPositionSequence> textPositions, List<Ruling> horizontalRulingLines, List<Ruling> verticalRulingLines) {
|
||||
|
||||
List<TextPositionSequence> chunkWords = new ArrayList<>();
|
||||
List<AbstractPageBlock> chunkBlockList1 = new ArrayList<>();
|
||||
|
||||
float minX = 1000, maxX = 0, minY = 1000, maxY = 0;
|
||||
TextPositionSequence prev = null;
|
||||
|
||||
boolean wasSplitted = false;
|
||||
Float splitX1 = null;
|
||||
for (TextPositionSequence word : textPositions) {
|
||||
|
||||
boolean lineSeparation = prev != null && word.getMinYDirAdj() - maxY > Math.min(word.getHeight(), prev.getHeight()) * 1.25;
|
||||
boolean startFromTop = prev != null && word.getMinYDirAdj() < prev.getMinYDirAdj() - prev.getTextHeight();
|
||||
boolean splitByX = prev != null && maxX + 50 < word.getMinXDirAdj() && prev.getMinYDirAdj() == word.getMinYDirAdj();
|
||||
boolean negativeXGap = prev != null && word.getMinXDirAdj() - minX < -5;
|
||||
boolean newLineAfterSplit = prev != null && word.getMinYDirAdj() != prev.getMinYDirAdj() && wasSplitted && splitX1 != word.getMinXDirAdj();
|
||||
boolean isSplitByRuling = isSplitByRuling(minX, minY, maxX, maxY, word, horizontalRulingLines, verticalRulingLines);
|
||||
boolean splitByDir = prev != null && !prev.getDir().equals(word.getDir());
|
||||
boolean splitByOtherFontAndOtherY = prev != null && prev.getMaxYDirAdj() != word.getMaxYDirAdj() && (word.getFontStyle().contains("bold") && !prev.getFontStyle().contains("bold") || prev.getFontStyle().contains("bold") && !word.getFontStyle().contains("bold"));
|
||||
|
||||
if (prev != null && (lineSeparation || startFromTop || splitByDir || isSplitByRuling || splitByOtherFontAndOtherY || negativeXGap)) {
|
||||
|
||||
Orientation prevOrientation = null;
|
||||
if (!chunkBlockList1.isEmpty()) {
|
||||
prevOrientation = chunkBlockList1.get(chunkBlockList1.size() - 1).getOrientation();
|
||||
}
|
||||
|
||||
TextPageBlock cb1 = buildTextBlock(chunkWords);
|
||||
chunkBlockList1.add(cb1);
|
||||
chunkWords = new ArrayList<>();
|
||||
|
||||
if (splitByX && !isSplitByRuling) {
|
||||
wasSplitted = true;
|
||||
cb1.setOrientation(Orientation.LEFT);
|
||||
splitX1 = word.getMinXDirAdj();
|
||||
} else if (newLineAfterSplit && !isSplitByRuling) {
|
||||
wasSplitted = false;
|
||||
cb1.setOrientation(Orientation.RIGHT);
|
||||
splitX1 = null;
|
||||
} else if (prevOrientation != null && prevOrientation.equals(Orientation.RIGHT) && (lineSeparation || !startFromTop || !splitByX || !newLineAfterSplit || !isSplitByRuling)) {
|
||||
cb1.setOrientation(Orientation.LEFT);
|
||||
}
|
||||
|
||||
minX = 1000;
|
||||
maxX = 0;
|
||||
minY = 1000;
|
||||
maxY = 0;
|
||||
prev = null;
|
||||
}
|
||||
|
||||
chunkWords.add(word);
|
||||
|
||||
prev = word;
|
||||
if (word.getMinXDirAdj() < minX) {
|
||||
minX = word.getMinXDirAdj();
|
||||
}
|
||||
if (word.getMaxXDirAdj() > maxX) {
|
||||
maxX = word.getMaxXDirAdj();
|
||||
}
|
||||
if (word.getMinYDirAdj() < minY) {
|
||||
minY = word.getMinYDirAdj();
|
||||
}
|
||||
if (word.getMaxYDirAdj() > maxY) {
|
||||
maxY = word.getMaxYDirAdj();
|
||||
}
|
||||
}
|
||||
|
||||
TextPageBlock cb1 = buildTextBlock(chunkWords);
|
||||
if (cb1 != null) {
|
||||
chunkBlockList1.add(cb1);
|
||||
}
|
||||
|
||||
return new ClassificationPage(chunkBlockList1);
|
||||
}
|
||||
|
||||
|
||||
private boolean equalsWithThreshold(float f1, float f2) {
|
||||
|
||||
return Math.abs(f1 - f2) < THRESHOLD;
|
||||
}
|
||||
|
||||
|
||||
private TextPageBlock buildTextBlock(List<TextPositionSequence> wordBlockList) {
|
||||
|
||||
TextPageBlock textBlock = null;
|
||||
|
||||
FloatFrequencyCounter lineHeightFrequencyCounter = new FloatFrequencyCounter();
|
||||
FloatFrequencyCounter fontSizeFrequencyCounter = new FloatFrequencyCounter();
|
||||
FloatFrequencyCounter spaceFrequencyCounter = new FloatFrequencyCounter();
|
||||
StringFrequencyCounter fontFrequencyCounter = new StringFrequencyCounter();
|
||||
StringFrequencyCounter styleFrequencyCounter = new StringFrequencyCounter();
|
||||
|
||||
for (TextPositionSequence wordBlock : wordBlockList) {
|
||||
|
||||
lineHeightFrequencyCounter.add(wordBlock.getTextHeight());
|
||||
fontSizeFrequencyCounter.add(wordBlock.getFontSize());
|
||||
spaceFrequencyCounter.add(wordBlock.getSpaceWidth());
|
||||
fontFrequencyCounter.add(wordBlock.getFont());
|
||||
styleFrequencyCounter.add(wordBlock.getFontStyle());
|
||||
|
||||
if (textBlock == null) {
|
||||
textBlock = new TextPageBlock(wordBlock.getMinXDirAdj(), wordBlock.getMaxXDirAdj(), wordBlock.getMinYDirAdj(), wordBlock.getMaxYDirAdj(), wordBlockList, wordBlock.getRotation());
|
||||
} else {
|
||||
TextPageBlock spatialEntity = textBlock.union(wordBlock);
|
||||
textBlock.resize(spatialEntity.getMinX(), spatialEntity.getMinY(), spatialEntity.getWidth(), spatialEntity.getHeight());
|
||||
}
|
||||
}
|
||||
|
||||
if (textBlock != null) {
|
||||
textBlock.setMostPopularWordFont(fontFrequencyCounter.getMostPopular());
|
||||
textBlock.setMostPopularWordStyle(styleFrequencyCounter.getMostPopular());
|
||||
textBlock.setMostPopularWordFontSize(fontSizeFrequencyCounter.getMostPopular());
|
||||
textBlock.setMostPopularWordHeight(lineHeightFrequencyCounter.getMostPopular());
|
||||
textBlock.setMostPopularWordSpaceWidth(spaceFrequencyCounter.getMostPopular());
|
||||
textBlock.setHighestFontSize(fontSizeFrequencyCounter.getHighest());
|
||||
}
|
||||
|
||||
if (textBlock != null && textBlock.getSequences() != null && textBlock.getSequences().stream().map(t -> round(t.getMinYDirAdj(), 3)).collect(toSet()).size() == 1) {
|
||||
textBlock.getSequences().sort(Comparator.comparing(TextPositionSequence::getMinXDirAdj));
|
||||
}
|
||||
return textBlock;
|
||||
}
|
||||
|
||||
|
||||
private boolean isSplitByRuling(float minX,
|
||||
float minY,
|
||||
float maxX,
|
||||
float maxY,
|
||||
TextPositionSequence word,
|
||||
List<Ruling> horizontalRulingLines,
|
||||
List<Ruling> verticalRulingLines) {
|
||||
|
||||
return isSplitByRuling(maxX, minY, word.getMinXDirAdj(), word.getMinYDirAdj(), verticalRulingLines, word.getDir().getDegrees(), word.getPageWidth(), word.getPageHeight()) //
|
||||
|| isSplitByRuling(minX, minY, word.getMinXDirAdj(), word.getMaxYDirAdj(), horizontalRulingLines, word.getDir().getDegrees(), word.getPageWidth(), word.getPageHeight()) //
|
||||
|| isSplitByRuling(maxX, minY, word.getMinXDirAdj(), word.getMinYDirAdj(), horizontalRulingLines, word.getDir().getDegrees(), word.getPageWidth(), word.getPageHeight()) //
|
||||
|| isSplitByRuling(minX, minY, word.getMinXDirAdj(), word.getMaxYDirAdj(), verticalRulingLines, word.getDir().getDegrees(), word.getPageWidth(), word.getPageHeight()); //
|
||||
}
|
||||
|
||||
|
||||
private boolean isSplitByRuling(float previousX2, float previousY1, float currentX1, float currentY1, List<Ruling> rulingLines, float dir, float pageWidth, float pageHeight) {
|
||||
|
||||
for (Ruling ruling : rulingLines) {
|
||||
var line = RulingTextDirAdjustUtil.convertToDirAdj(ruling, dir, pageWidth, pageHeight);
|
||||
if (line.intersectsLine(previousX2, previousY1, currentX1, currentY1)) {
|
||||
return true;
|
||||
}
|
||||
}
|
||||
return false;
|
||||
}
|
||||
|
||||
|
||||
private double round(float value, int decimalPoints) {
|
||||
|
||||
var d = Math.pow(10, decimalPoints);
|
||||
return Math.round(value * d) / d;
|
||||
}
|
||||
|
||||
}
|
||||
-118
@@ -1,118 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
|
||||
|
||||
import java.util.List;
|
||||
import java.util.Locale;
|
||||
import java.util.regex.Matcher;
|
||||
import java.util.regex.Pattern;
|
||||
|
||||
import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
|
||||
import org.springframework.stereotype.Service;
|
||||
|
||||
import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Rectangle;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationDocument;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.PageBlockType;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.PositionUtils;
|
||||
import com.iqser.red.service.redaction.v1.server.redaction.utils.Patterns;
|
||||
|
||||
import lombok.RequiredArgsConstructor;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
|
||||
@Slf4j
|
||||
@Service
|
||||
@RequiredArgsConstructor
|
||||
@ConditionalOnProperty(prefix = "application", name = "type", havingValue = "DocuMine")
|
||||
public class DocuMineClassificationService implements ClassificationService {
|
||||
|
||||
private final BodyTextFrameService bodyTextFrameService;
|
||||
|
||||
|
||||
public void classifyDocument(ClassificationDocument document) {
|
||||
|
||||
Rectangle bodyTextFrame = bodyTextFrameService.calculateBodyTextFrame(document.getPages(), document.getFontSizeCounter(), false);
|
||||
Rectangle landscapeBodyTextFrame = bodyTextFrameService.calculateBodyTextFrame(document.getPages(), document.getFontSizeCounter(), true);
|
||||
List<Float> headlineFontSizes = document.getFontSizeCounter().getHighterThanMostPopular();
|
||||
|
||||
log.debug("Document FontSize counters are: {}", document.getFontSizeCounter().getCountPerValue());
|
||||
|
||||
for (ClassificationPage page : document.getPages()) {
|
||||
bodyTextFrameService.setBodyTextFrameAdjustedToPage(page, bodyTextFrame, landscapeBodyTextFrame);
|
||||
classifyPage(page, document, headlineFontSizes);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
private void classifyPage(ClassificationPage page, ClassificationDocument document, List<Float> headlineFontSizes) {
|
||||
|
||||
for (AbstractPageBlock textBlock : page.getTextBlocks()) {
|
||||
if (textBlock instanceof TextPageBlock) {
|
||||
classifyBlock((TextPageBlock) textBlock, page, document, headlineFontSizes);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
private void classifyBlock(TextPageBlock textBlock, ClassificationPage page, ClassificationDocument document, List<Float> headlineFontSizes) {
|
||||
|
||||
log.debug("headlineFontSizes: {}", headlineFontSizes);
|
||||
var bodyTextFrame = page.getBodyTextFrame();
|
||||
|
||||
var pattern = Patterns.getCompiledPattern("^(\\d{1,1}\\.?){1,3}\\d{1,2}\\.?\\s[0-9A-Za-z\\[\\]\\-]{2,50}", true);
|
||||
var pattern2 = Patterns.getCompiledPattern(".*\\d{4}$", true);
|
||||
var pattern3 = Patterns.getCompiledPattern("^(\\d{1,1}\\.){1,3}\\d{1,2}\\.?\\s[a-z]{1,2}\\/[a-z]{1,2}.*", false);
|
||||
|
||||
Matcher matcher = pattern.matcher(textBlock.toString());
|
||||
Matcher matcher2 = pattern2.matcher(textBlock.toString());
|
||||
Matcher matcher3 = pattern3.matcher(textBlock.toString());
|
||||
|
||||
if (document.getFontSizeCounter().getMostPopular() == null) {
|
||||
textBlock.setClassification(PageBlockType.OTHER);
|
||||
return;
|
||||
}
|
||||
if (textBlock.getText().length() > 6 && (textBlock.getMostPopularWordHeight() > document.getTextHeightCounter()
|
||||
.getMostPopular() || textBlock.getMostPopularWordFontSize() > document.getFontSizeCounter().getMostPopular()) && PositionUtils.getApproxLineCount(textBlock) < 5.9
|
||||
|
||||
&& (textBlock.getMostPopularWordStyle().contains("bold") && Character.isDigit(textBlock.toString().charAt(0)) && !matcher2.matches() && !textBlock.toString()
|
||||
.contains(":") || textBlock.toString().equals(textBlock.toString().toUpperCase(Locale.ROOT)) && !matcher2.matches() && !textBlock.toString()
|
||||
.contains(":") || textBlock.toString().startsWith("APPENDIX") || textBlock.toString().startsWith("FIGURE") || textBlock.toString().startsWith("TABLE")) && !textBlock.toString().endsWith(":")) {
|
||||
textBlock.setClassification(PageBlockType.getHeadlineType(1));
|
||||
document.setHeadlines(true);
|
||||
|
||||
} else if (matcher.find() && PositionUtils.getApproxLineCount(textBlock) < 2.9 && !matcher3.matches() && !matcher2.matches()) {
|
||||
textBlock.setClassification(PageBlockType.getHeadlineType(2));
|
||||
document.setHeadlines(true);
|
||||
} else if (PositionUtils.isOverBodyTextFrame(bodyTextFrame, textBlock, page.getRotation()) && (document.getFontSizeCounter()
|
||||
.getMostPopular() == null || textBlock.getHighestFontSize() <= document.getFontSizeCounter().getMostPopular())) {
|
||||
textBlock.setClassification(PageBlockType.HEADER);
|
||||
|
||||
} else if (PositionUtils.isUnderBodyTextFrame(bodyTextFrame, textBlock, page.getRotation()) && (document.getFontSizeCounter()
|
||||
.getMostPopular() == null || textBlock.getHighestFontSize() <= document.getFontSizeCounter().getMostPopular())) {
|
||||
textBlock.setClassification(PageBlockType.FOOTER);
|
||||
} else if (page.getPageNumber() == 1 && (PositionUtils.getHeightDifferenceBetweenChunkWordAndDocumentWord(textBlock,
|
||||
document.getTextHeightCounter().getMostPopular()) > 2.5 && textBlock.getHighestFontSize() > document.getFontSizeCounter().getMostPopular() || page.getTextBlocks()
|
||||
.size() == 1)) {
|
||||
if (!Pattern.matches("[0-9]+", textBlock.toString())) {
|
||||
textBlock.setClassification(PageBlockType.TITLE);
|
||||
}
|
||||
} else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getMostPopularWordFontSize() == document.getFontSizeCounter()
|
||||
.getMostPopular() && textBlock.getMostPopularWordStyle().equals("bold") && !document.getFontStyleCounter().getMostPopular().equals("bold")) {
|
||||
textBlock.setClassification(PageBlockType.PARAGRAPH_BOLD);
|
||||
} else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getMostPopularWordFont()
|
||||
.equals(document.getFontCounter().getMostPopular()) && textBlock.getMostPopularWordStyle()
|
||||
.equals(document.getFontStyleCounter().getMostPopular()) && textBlock.getMostPopularWordFontSize() == document.getFontSizeCounter().getMostPopular()) {
|
||||
textBlock.setClassification(PageBlockType.PARAGRAPH);
|
||||
} else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getMostPopularWordFontSize() == document.getFontSizeCounter()
|
||||
.getMostPopular() && textBlock.getMostPopularWordStyle().equals("italic") && !document.getFontStyleCounter()
|
||||
.getMostPopular()
|
||||
.equals("italic") && PositionUtils.getApproxLineCount(textBlock) < 2.9) {
|
||||
textBlock.setClassification(PageBlockType.PARAGRAPH_ITALIC);
|
||||
} else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock)) {
|
||||
textBlock.setClassification(PageBlockType.PARAGRAPH_UNKNOWN);
|
||||
} else {
|
||||
textBlock.setClassification(PageBlockType.OTHER);
|
||||
}
|
||||
}
|
||||
|
||||
}
|
||||
-115
@@ -1,115 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
|
||||
|
||||
import java.awt.geom.Rectangle2D;
|
||||
import java.util.Collections;
|
||||
import java.util.LinkedList;
|
||||
import java.util.List;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.document.utils.RectangleTransformations;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.document.utils.TextPositionSequenceComparator;
|
||||
|
||||
import lombok.AllArgsConstructor;
|
||||
import lombok.experimental.UtilityClass;
|
||||
|
||||
@UtilityClass
|
||||
public class LineDetectionService {
|
||||
|
||||
private static final double X_GAP_FACTOR = 1; // multiplied with average text height, determines the minimum distance of gaps in lines
|
||||
|
||||
|
||||
public static List<List<Rectangle2D>> findLinesWithGaps(List<TextPositionSequence> textPositionSequences) {
|
||||
|
||||
if (textPositionSequences.isEmpty()) {
|
||||
return Collections.emptyList();
|
||||
}
|
||||
|
||||
final double avgTextPositionHeight = textPositionSequences.stream().mapToDouble(TextPositionSequence::getHeight).average().orElseThrow();
|
||||
|
||||
Context context = Context.init();
|
||||
|
||||
List<TextPositionSequence> sortedTextPositionSequence = textPositionSequences.stream().sorted(new TextPositionSequenceComparator()).toList();
|
||||
|
||||
var previousTextPosition = sortedTextPositionSequence.get(0);
|
||||
context.textPositionsToMerge.add(previousTextPosition);
|
||||
for (TextPositionSequence currentTextPosition : sortedTextPositionSequence.subList(1, sortedTextPositionSequence.size())) {
|
||||
if (isNewLine(currentTextPosition, previousTextPosition, avgTextPositionHeight) || isSplitByOrientation(currentTextPosition, previousTextPosition)) {
|
||||
addBlockToLine(context);
|
||||
startNewLine(currentTextPosition, context);
|
||||
} else if (isXGap(currentTextPosition, previousTextPosition, avgTextPositionHeight)) {
|
||||
addBlockToLine(context);
|
||||
startNewBlock(currentTextPosition, context);
|
||||
} else {
|
||||
context.textPositionsToMerge.add(currentTextPosition);
|
||||
}
|
||||
previousTextPosition = currentTextPosition;
|
||||
}
|
||||
addBlockToLine(context);
|
||||
return context.linesWithGaps;
|
||||
}
|
||||
|
||||
|
||||
private static boolean isXGap(TextPositionSequence currentTextPosition, TextPositionSequence previousTextPosition, double avgTextPositionHeight) {
|
||||
|
||||
return Math.abs(previousTextPosition.getMaxXDirAdj() - currentTextPosition.getMinXDirAdj()) > (avgTextPositionHeight * X_GAP_FACTOR);
|
||||
}
|
||||
|
||||
|
||||
private static boolean isSplitByOrientation(TextPositionSequence currentTextPosition, TextPositionSequence previousTextPosition) {
|
||||
|
||||
return !previousTextPosition.getDir().equals(currentTextPosition.getDir());
|
||||
}
|
||||
|
||||
|
||||
private static boolean isNewLine(TextPositionSequence currentTextPosition, TextPositionSequence previousTextPosition, double avgTextPositionHeight) {
|
||||
|
||||
return Math.abs(previousTextPosition.getMinYDirAdj() - currentTextPosition.getMinYDirAdj()) > avgTextPositionHeight;
|
||||
}
|
||||
|
||||
|
||||
private static void startNewBlock(TextPositionSequence currentTextPosition, Context context) {
|
||||
|
||||
context.textPositionsToMerge = new LinkedList<>();
|
||||
context.textPositionsToMerge.add(currentTextPosition);
|
||||
}
|
||||
|
||||
|
||||
private static void addBlockToLine(Context context) {
|
||||
|
||||
context.blocksInLine.add(textPositionBBox(context.textPositionsToMerge));
|
||||
}
|
||||
|
||||
|
||||
private static void startNewLine(TextPositionSequence current, Context context) {
|
||||
|
||||
context.blocksInLine = new LinkedList<>();
|
||||
startNewBlock(current, context);
|
||||
context.linesWithGaps.add(context.blocksInLine);
|
||||
}
|
||||
|
||||
|
||||
private Rectangle2D textPositionBBox(List<TextPositionSequence> textPositionSequences) {
|
||||
|
||||
return RectangleTransformations.rectangleBBox(textPositionSequences.stream().map(TextPositionSequence::getRectangle).toList());
|
||||
}
|
||||
|
||||
|
||||
@AllArgsConstructor
|
||||
private class Context {
|
||||
|
||||
List<List<Rectangle2D>> linesWithGaps;
|
||||
List<Rectangle2D> blocksInLine;
|
||||
List<TextPositionSequence> textPositionsToMerge;
|
||||
|
||||
|
||||
public static Context init() {
|
||||
|
||||
List<List<Rectangle2D>> initialLinesWithGaps = new LinkedList<>();
|
||||
List<Rectangle2D> initialBlocksInLine = new LinkedList<>();
|
||||
initialLinesWithGaps.add(initialBlocksInLine);
|
||||
return new Context(initialLinesWithGaps, initialBlocksInLine, new LinkedList<>());
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
-186
@@ -1,186 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
|
||||
|
||||
import java.io.File;
|
||||
import java.io.FileOutputStream;
|
||||
import java.io.IOException;
|
||||
import java.io.InputStream;
|
||||
import java.util.ArrayList;
|
||||
import java.util.HashMap;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
|
||||
import org.apache.commons.io.IOUtils;
|
||||
import org.apache.pdfbox.io.MemoryUsageSetting;
|
||||
import org.apache.pdfbox.pdmodel.PDDocument;
|
||||
import org.apache.pdfbox.pdmodel.PDPage;
|
||||
import org.apache.pdfbox.pdmodel.common.PDRectangle;
|
||||
import org.springframework.beans.factory.annotation.Value;
|
||||
import org.springframework.stereotype.Service;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.ImageServiceResponseAdapter;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.TableServiceResponseAdapter;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table.PdfTableCell;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationDocument;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.image.ClassifiedImage;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.CleanRulings;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.parsing.PDFLinesTextStripper;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.FileUtils;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.document.utils.RectangleTransformations;
|
||||
import com.iqser.red.service.redaction.v1.server.settings.RedactionServiceSettings;
|
||||
|
||||
import lombok.RequiredArgsConstructor;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
|
||||
@Slf4j
|
||||
@Service
|
||||
@RequiredArgsConstructor
|
||||
public class PdfSegmentationService {
|
||||
|
||||
private final RedactionServiceSettings redactionServiceSettings;
|
||||
private final RulingCleaningService rulingCleaningService;
|
||||
private final TableExtractionService tableExtractionService;
|
||||
private final BlockificationService blockificationService;
|
||||
private final ClassificationService classificationService;
|
||||
private final SectionsBuilderService sectionsBuilderService;
|
||||
private final ImageServiceResponseAdapter imageServiceResponseAdapter;
|
||||
private final TableServiceResponseAdapter tableServiceResponseAdapter;
|
||||
|
||||
@Value("${application.type:RedactManager}")
|
||||
private String applicationType;
|
||||
|
||||
|
||||
public ClassificationDocument parseDocument(String dossierId,
|
||||
String fileId,
|
||||
InputStream documentInputStream,
|
||||
Map<Integer, List<ClassifiedImage>> pdfImages) throws IOException {
|
||||
|
||||
PDDocument pdDocument = null;
|
||||
File tempFile = null;
|
||||
try {
|
||||
Map<Integer, List<PdfTableCell>> pdfTableCells = new HashMap<>();
|
||||
if (redactionServiceSettings.isCvTableParsingEnabled()) {
|
||||
pdfTableCells = tableServiceResponseAdapter.convertTables(dossierId, fileId);
|
||||
}
|
||||
|
||||
tempFile = FileUtils.createTempFile("document", ".pdf");
|
||||
try (var fos = new FileOutputStream(tempFile)) {
|
||||
IOUtils.copy(documentInputStream, fos);
|
||||
|
||||
// initialize required variables
|
||||
ClassificationDocument document = new ClassificationDocument();
|
||||
List<ClassificationPage> pages = new ArrayList<>();
|
||||
|
||||
pdDocument = PDDocument.load(tempFile, MemoryUsageSetting.setupMixed(67108864L));
|
||||
pdDocument.setAllSecurityToBeRemoved(true);
|
||||
long pageCount = pdDocument.getNumberOfPages();
|
||||
|
||||
for (int pageNumber = 1; pageNumber <= pageCount; pageNumber++) {
|
||||
processPage(pdfImages, pdDocument, pdfTableCells, document, pages, pageNumber);
|
||||
}
|
||||
|
||||
document.setPages(pages);
|
||||
|
||||
classificationService.classifyDocument(document);
|
||||
sectionsBuilderService.buildSections(document);
|
||||
sectionsBuilderService.addImagesToSections(document);
|
||||
|
||||
return document;
|
||||
}
|
||||
} finally {
|
||||
if (pdDocument != null) {
|
||||
pdDocument.close();
|
||||
}
|
||||
|
||||
FileUtils.deleteFile(tempFile);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
private void processPage(Map<Integer, List<ClassifiedImage>> pdfImages,
|
||||
PDDocument pdDocument,
|
||||
Map<Integer, List<PdfTableCell>> pdfTableCells,
|
||||
ClassificationDocument document,
|
||||
List<ClassificationPage> pages,
|
||||
int pageNumber) throws IOException {
|
||||
|
||||
PDFLinesTextStripper stripper = new PDFLinesTextStripper();
|
||||
PDPage pdPage = pdDocument.getPage(pageNumber - 1);
|
||||
stripper.setPageNumber(pageNumber);
|
||||
stripper.setStartPage(pageNumber);
|
||||
stripper.setEndPage(pageNumber);
|
||||
stripper.setPdpage(pdPage);
|
||||
|
||||
if(applicationType.equals("DocuMine")){
|
||||
stripper.setSortByPosition(true);
|
||||
}
|
||||
|
||||
stripper.getText(pdDocument);
|
||||
|
||||
PDRectangle pdr = pdPage.getMediaBox();
|
||||
|
||||
int rotation = pdPage.getRotation();
|
||||
boolean isLandscape = pdr.getWidth() > pdr.getHeight() && (rotation == 0 || rotation == 180) || pdr.getHeight() > pdr.getWidth() && (rotation == 90 || rotation == 270);
|
||||
|
||||
PDRectangle cropbox = pdPage.getCropBox();
|
||||
CleanRulings cleanRulings = rulingCleaningService.getCleanRulings(pdfTableCells.get(pageNumber),
|
||||
stripper.getRulings(),
|
||||
stripper.getMinCharWidth(),
|
||||
stripper.getMaxCharHeight());
|
||||
// var columns = ColumnDetectionService.detectColumns(stripper.getTextPositionSequences(), RectangleTransformations.toRectangle2D(pdPage.getCropBox()));
|
||||
ClassificationPage page = blockificationService.blockify(stripper.getTextPositionSequences(), cleanRulings.getHorizontal(), cleanRulings.getVertical());
|
||||
|
||||
page.setRotation(rotation);
|
||||
page.setLandscape(isLandscape);
|
||||
page.setPageNumber(pageNumber);
|
||||
page.setPageWidth(cropbox.getWidth());
|
||||
page.setPageHeight(cropbox.getHeight());
|
||||
|
||||
// If images is ocr needs to be calculated before textBlocks are moved into tables, otherwise findOcr algorithm needs to be adopted.
|
||||
if (pdfImages != null && pdfImages.containsKey(pageNumber)) {
|
||||
page.setImages(pdfImages.get(pageNumber));
|
||||
imageServiceResponseAdapter.findOcr(page);
|
||||
}
|
||||
|
||||
tableExtractionService.extractTables(cleanRulings, page);
|
||||
buildPageStatistics(page);
|
||||
increaseDocumentStatistics(page, document);
|
||||
|
||||
pages.add(page);
|
||||
}
|
||||
|
||||
|
||||
private void increaseDocumentStatistics(ClassificationPage page, ClassificationDocument document) {
|
||||
|
||||
if (!page.isLandscape()) {
|
||||
document.getFontSizeCounter().addAll(page.getFontSizeCounter().getCountPerValue());
|
||||
}
|
||||
document.getFontCounter().addAll(page.getFontCounter().getCountPerValue());
|
||||
document.getTextHeightCounter().addAll(page.getTextHeightCounter().getCountPerValue());
|
||||
document.getFontStyleCounter().addAll(page.getFontStyleCounter().getCountPerValue());
|
||||
}
|
||||
|
||||
|
||||
private void buildPageStatistics(ClassificationPage page) {
|
||||
|
||||
// Collect all statistics for the page, except from blocks inside tables, as tables will always be added to BodyTextFrame.
|
||||
for (AbstractPageBlock textBlock : page.getTextBlocks()) {
|
||||
if (textBlock instanceof TextPageBlock) {
|
||||
if (((TextPageBlock) textBlock).getSequences() == null) {
|
||||
continue;
|
||||
}
|
||||
for (TextPositionSequence word : ((TextPageBlock) textBlock).getSequences()) {
|
||||
page.getTextHeightCounter().add(word.getTextHeight());
|
||||
page.getFontCounter().add(word.getFont());
|
||||
page.getFontSizeCounter().add(word.getFontSize());
|
||||
page.getFontStyleCounter().add(word.getFontStyle());
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
-280
@@ -1,280 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
|
||||
|
||||
import static java.util.stream.Collectors.toSet;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.Comparator;
|
||||
import java.util.Iterator;
|
||||
import java.util.List;
|
||||
|
||||
import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
|
||||
import org.springframework.stereotype.Service;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.FloatFrequencyCounter;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.Orientation;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Ruling;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.StringFrequencyCounter;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.RulingTextDirAdjustUtil;
|
||||
|
||||
@Service
|
||||
@SuppressWarnings("all")
|
||||
@ConditionalOnProperty(prefix = "application", name = "type", havingValue = "RedactManager")
|
||||
public class RedactManagerBlockificationService implements BlockificationService {
|
||||
|
||||
static final float THRESHOLD = 1f;
|
||||
|
||||
|
||||
/**
|
||||
* This method is building blocks by expanding the minX/maxX and minY/maxY value on each word that is not split by the conditions.
|
||||
* This method must use text direction adjusted postions (DirAdj). Where {0,0} is on the upper left. Never try to change this!
|
||||
* Rulings (TablePageBlock lines) must be adjusted to the text directions as well, when checking if a block is split by a ruling.
|
||||
*
|
||||
* @param textPositions The words of a page.
|
||||
* @param horizontalRulingLines Horizontal table lines.
|
||||
* @param verticalRulingLines Vertical table lines.
|
||||
* @return Page object that contains the Textblock and text statistics.
|
||||
*/
|
||||
public ClassificationPage blockify(List<TextPositionSequence> textPositions, List<Ruling> horizontalRulingLines, List<Ruling> verticalRulingLines) {
|
||||
|
||||
int indexOnPage = 0;
|
||||
List<TextPositionSequence> chunkWords = new ArrayList<>();
|
||||
List<AbstractPageBlock> chunkBlockList = new ArrayList<>();
|
||||
|
||||
float minX = 1000, maxX = 0, minY = 1000, maxY = 0;
|
||||
TextPositionSequence prev = null;
|
||||
|
||||
boolean wasSplitted = false;
|
||||
Float splitX1 = null;
|
||||
for (TextPositionSequence word : textPositions) {
|
||||
|
||||
boolean lineSeparation = word.getMinYDirAdj() - maxY > word.getHeight() * 1.25;
|
||||
boolean startFromTop = prev != null && word.getMinYDirAdj() < prev.getMinYDirAdj() - prev.getTextHeight();
|
||||
boolean splitByX = prev != null && maxX + 50 < word.getMinXDirAdj() && prev.getMinYDirAdj() == word.getMinYDirAdj();
|
||||
boolean xIsBeforeFirstX = prev != null && word.getMinXDirAdj() < minX;
|
||||
boolean newLineAfterSplit = prev != null && word.getMinYDirAdj() != prev.getMinYDirAdj() && wasSplitted && splitX1 != word.getMinXDirAdj();
|
||||
boolean isSplitByRuling = isSplitByRuling(minX, minY, maxX, maxY, word, horizontalRulingLines, verticalRulingLines);
|
||||
boolean splitByDir = prev != null && !prev.getDir().equals(word.getDir());
|
||||
|
||||
if (prev != null && (lineSeparation || startFromTop || splitByX || splitByDir || isSplitByRuling)) {
|
||||
|
||||
Orientation prevOrientation = null;
|
||||
if (!chunkBlockList.isEmpty()) {
|
||||
prevOrientation = chunkBlockList.get(chunkBlockList.size() - 1).getOrientation();
|
||||
}
|
||||
|
||||
TextPageBlock cb1 = buildTextBlock(chunkWords, indexOnPage);
|
||||
indexOnPage++;
|
||||
|
||||
chunkBlockList.add(cb1);
|
||||
chunkWords = new ArrayList<>();
|
||||
|
||||
if (splitByX && !isSplitByRuling) {
|
||||
wasSplitted = true;
|
||||
cb1.setOrientation(Orientation.LEFT);
|
||||
splitX1 = word.getMinXDirAdj();
|
||||
} else if (newLineAfterSplit && !isSplitByRuling) {
|
||||
wasSplitted = false;
|
||||
cb1.setOrientation(Orientation.RIGHT);
|
||||
splitX1 = null;
|
||||
} else if (prevOrientation != null && prevOrientation.equals(Orientation.RIGHT) && (lineSeparation || !startFromTop || !splitByX || !newLineAfterSplit || !isSplitByRuling)) {
|
||||
cb1.setOrientation(Orientation.LEFT);
|
||||
}
|
||||
|
||||
minX = 1000;
|
||||
maxX = 0;
|
||||
minY = 1000;
|
||||
maxY = 0;
|
||||
prev = null;
|
||||
}
|
||||
|
||||
chunkWords.add(word);
|
||||
|
||||
prev = word;
|
||||
if (word.getMinXDirAdj() < minX) {
|
||||
minX = word.getMinXDirAdj();
|
||||
}
|
||||
if (word.getMaxXDirAdj() > maxX) {
|
||||
maxX = word.getMaxXDirAdj();
|
||||
}
|
||||
if (word.getMinYDirAdj() < minY) {
|
||||
minY = word.getMinYDirAdj();
|
||||
}
|
||||
if (word.getMaxYDirAdj() > maxY) {
|
||||
maxY = word.getMaxYDirAdj();
|
||||
}
|
||||
}
|
||||
|
||||
TextPageBlock cb1 = buildTextBlock(chunkWords, indexOnPage);
|
||||
if (cb1 != null) {
|
||||
chunkBlockList.add(cb1);
|
||||
}
|
||||
|
||||
Iterator<AbstractPageBlock> itty = chunkBlockList.iterator();
|
||||
|
||||
TextPageBlock previousLeft = null;
|
||||
TextPageBlock previousRight = null;
|
||||
while (itty.hasNext()) {
|
||||
TextPageBlock block = (TextPageBlock) itty.next();
|
||||
|
||||
if (previousLeft != null && block.getOrientation().equals(Orientation.LEFT)) {
|
||||
if (previousLeft.getMinY() > block.getMinY() && block.getMaxY() + block.getMostPopularWordHeight() > previousLeft.getMinY()) {
|
||||
previousLeft.add(block);
|
||||
itty.remove();
|
||||
continue;
|
||||
}
|
||||
}
|
||||
|
||||
if (previousRight != null && block.getOrientation().equals(Orientation.RIGHT)) {
|
||||
if (previousRight.getMinY() > block.getMinY() && block.getMaxY() + block.getMostPopularWordHeight() > previousRight.getMinY()) {
|
||||
previousRight.add(block);
|
||||
itty.remove();
|
||||
continue;
|
||||
}
|
||||
}
|
||||
|
||||
if (block.getOrientation().equals(Orientation.LEFT)) {
|
||||
previousLeft = block;
|
||||
} else if (block.getOrientation().equals(Orientation.RIGHT)) {
|
||||
previousRight = block;
|
||||
}
|
||||
}
|
||||
|
||||
itty = chunkBlockList.iterator();
|
||||
TextPageBlock previous = null;
|
||||
while (itty.hasNext()) {
|
||||
TextPageBlock block = (TextPageBlock) itty.next();
|
||||
|
||||
if (previous != null && previous.getOrientation().equals(Orientation.LEFT) && block.getOrientation().equals(Orientation.LEFT) && equalsWithThreshold(block.getMaxY(),
|
||||
previous.getMaxY()) || previous != null && previous.getOrientation().equals(Orientation.LEFT) && block.getOrientation()
|
||||
.equals(Orientation.RIGHT) && equalsWithThreshold(block.getMaxY(), previous.getMaxY())) {
|
||||
previous.add(block);
|
||||
itty.remove();
|
||||
continue;
|
||||
}
|
||||
|
||||
previous = block;
|
||||
}
|
||||
|
||||
return new ClassificationPage(chunkBlockList);
|
||||
}
|
||||
|
||||
|
||||
private boolean equalsWithThreshold(float f1, float f2) {
|
||||
|
||||
return Math.abs(f1 - f2) < THRESHOLD;
|
||||
}
|
||||
|
||||
|
||||
private TextPageBlock buildTextBlock(List<TextPositionSequence> wordBlockList, int indexOnPage) {
|
||||
|
||||
TextPageBlock textBlock = null;
|
||||
|
||||
FloatFrequencyCounter lineHeightFrequencyCounter = new FloatFrequencyCounter();
|
||||
FloatFrequencyCounter fontSizeFrequencyCounter = new FloatFrequencyCounter();
|
||||
FloatFrequencyCounter spaceFrequencyCounter = new FloatFrequencyCounter();
|
||||
StringFrequencyCounter fontFrequencyCounter = new StringFrequencyCounter();
|
||||
StringFrequencyCounter styleFrequencyCounter = new StringFrequencyCounter();
|
||||
|
||||
for (TextPositionSequence wordBlock : wordBlockList) {
|
||||
|
||||
lineHeightFrequencyCounter.add(wordBlock.getTextHeight());
|
||||
fontSizeFrequencyCounter.add(wordBlock.getFontSize());
|
||||
spaceFrequencyCounter.add(wordBlock.getSpaceWidth());
|
||||
fontFrequencyCounter.add(wordBlock.getFont());
|
||||
styleFrequencyCounter.add(wordBlock.getFontStyle());
|
||||
|
||||
if (textBlock == null) {
|
||||
textBlock = new TextPageBlock(wordBlock.getMinXDirAdj(),
|
||||
wordBlock.getMaxXDirAdj(),
|
||||
wordBlock.getMinYDirAdj(),
|
||||
wordBlock.getMaxYDirAdj(),
|
||||
wordBlockList,
|
||||
wordBlock.getRotation());
|
||||
} else {
|
||||
TextPageBlock spatialEntity = textBlock.union(wordBlock);
|
||||
textBlock.resize(spatialEntity.getMinX(), spatialEntity.getMinY(), spatialEntity.getWidth(), spatialEntity.getHeight());
|
||||
}
|
||||
}
|
||||
|
||||
if (textBlock != null) {
|
||||
textBlock.setMostPopularWordFont(fontFrequencyCounter.getMostPopular());
|
||||
textBlock.setMostPopularWordStyle(styleFrequencyCounter.getMostPopular());
|
||||
textBlock.setMostPopularWordFontSize(fontSizeFrequencyCounter.getMostPopular());
|
||||
textBlock.setMostPopularWordHeight(lineHeightFrequencyCounter.getMostPopular());
|
||||
textBlock.setMostPopularWordSpaceWidth(spaceFrequencyCounter.getMostPopular());
|
||||
textBlock.setHighestFontSize(fontSizeFrequencyCounter.getHighest());
|
||||
}
|
||||
|
||||
if (textBlock != null && textBlock.getSequences() != null && textBlock.getSequences().stream().map(t -> round(t.getMinYDirAdj(), 3)).collect(toSet()).size() == 1) {
|
||||
textBlock.getSequences().sort(Comparator.comparing(TextPositionSequence::getMinXDirAdj));
|
||||
}
|
||||
return textBlock;
|
||||
}
|
||||
|
||||
|
||||
private boolean isSplitByRuling(float minX,
|
||||
float minY,
|
||||
float maxX,
|
||||
float maxY,
|
||||
TextPositionSequence word,
|
||||
List<Ruling> horizontalRulingLines,
|
||||
List<Ruling> verticalRulingLines) {
|
||||
|
||||
return isSplitByRuling(maxX,
|
||||
minY,
|
||||
word.getMinXDirAdj(),
|
||||
word.getMinYDirAdj(),
|
||||
verticalRulingLines,
|
||||
word.getDir().getDegrees(),
|
||||
word.getPageWidth(),
|
||||
word.getPageHeight()) //
|
||||
|| isSplitByRuling(minX,
|
||||
minY,
|
||||
word.getMinXDirAdj(),
|
||||
word.getMaxYDirAdj(),
|
||||
horizontalRulingLines,
|
||||
word.getDir().getDegrees(),
|
||||
word.getPageWidth(),
|
||||
word.getPageHeight()) //
|
||||
|| isSplitByRuling(maxX,
|
||||
minY,
|
||||
word.getMinXDirAdj(),
|
||||
word.getMinYDirAdj(),
|
||||
horizontalRulingLines,
|
||||
word.getDir().getDegrees(),
|
||||
word.getPageWidth(),
|
||||
word.getPageHeight()) //
|
||||
|| isSplitByRuling(minX,
|
||||
minY,
|
||||
word.getMinXDirAdj(),
|
||||
word.getMaxYDirAdj(),
|
||||
verticalRulingLines,
|
||||
word.getDir().getDegrees(),
|
||||
word.getPageWidth(),
|
||||
word.getPageHeight());
|
||||
}
|
||||
|
||||
|
||||
private boolean isSplitByRuling(float previousX2, float previousY1, float currentX1, float currentY1, List<Ruling> rulingLines, float dir, float pageWidth, float pageHeight) {
|
||||
|
||||
for (Ruling ruling : rulingLines) {
|
||||
var line = RulingTextDirAdjustUtil.convertToDirAdj(ruling, dir, pageWidth, pageHeight);
|
||||
if (line.intersectsLine(previousX2, previousY1, currentX1, currentY1)) {
|
||||
return true;
|
||||
}
|
||||
}
|
||||
return false;
|
||||
}
|
||||
|
||||
|
||||
private double round(float value, int decimalPoints) {
|
||||
|
||||
var d = Math.pow(10, decimalPoints);
|
||||
return Math.round(value * d) / d;
|
||||
}
|
||||
|
||||
}
|
||||
-117
@@ -1,117 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
|
||||
|
||||
import java.util.List;
|
||||
import java.util.regex.Pattern;
|
||||
|
||||
import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
|
||||
import org.springframework.stereotype.Service;
|
||||
|
||||
import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Rectangle;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationDocument;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.PageBlockType;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.PositionUtils;
|
||||
|
||||
import lombok.RequiredArgsConstructor;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
|
||||
@Slf4j
|
||||
@Service
|
||||
@RequiredArgsConstructor
|
||||
@ConditionalOnProperty(prefix = "application", name = "type", havingValue = "RedactManager")
|
||||
public class RedactManagerClassificationService implements ClassificationService {
|
||||
|
||||
private final BodyTextFrameService bodyTextFrameService;
|
||||
|
||||
|
||||
public void classifyDocument(ClassificationDocument document) {
|
||||
|
||||
Rectangle bodyTextFrame = bodyTextFrameService.calculateBodyTextFrame(document.getPages(), document.getFontSizeCounter(), false);
|
||||
Rectangle landscapeBodyTextFrame = bodyTextFrameService.calculateBodyTextFrame(document.getPages(), document.getFontSizeCounter(), true);
|
||||
List<Float> headlineFontSizes = document.getFontSizeCounter().getHighterThanMostPopular();
|
||||
|
||||
log.debug("Document FontSize counters are: {}", document.getFontSizeCounter().getCountPerValue());
|
||||
|
||||
for (ClassificationPage page : document.getPages()) {
|
||||
bodyTextFrameService.setBodyTextFrameAdjustedToPage(page, bodyTextFrame, landscapeBodyTextFrame);
|
||||
classifyPage(page, document, headlineFontSizes);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
private void classifyPage(ClassificationPage page, ClassificationDocument document, List<Float> headlineFontSizes) {
|
||||
|
||||
for (AbstractPageBlock textBlock : page.getTextBlocks()) {
|
||||
if (textBlock instanceof TextPageBlock) {
|
||||
classifyBlock((TextPageBlock) textBlock, page, document, headlineFontSizes);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
private void classifyBlock(TextPageBlock textBlock, ClassificationPage page, ClassificationDocument document, List<Float> headlineFontSizes) {
|
||||
|
||||
var bodyTextFrame = page.getBodyTextFrame();
|
||||
|
||||
if (document.getFontSizeCounter().getMostPopular() == null) {
|
||||
textBlock.setClassification(PageBlockType.OTHER);
|
||||
return;
|
||||
}
|
||||
if (PositionUtils.isOverBodyTextFrame(bodyTextFrame, textBlock, page.getRotation()) && (document.getFontSizeCounter()
|
||||
.getMostPopular() == null || textBlock.getHighestFontSize() <= document.getFontSizeCounter().getMostPopular())) {
|
||||
textBlock.setClassification(PageBlockType.HEADER);
|
||||
|
||||
} else if (PositionUtils.isUnderBodyTextFrame(bodyTextFrame, textBlock, page.getRotation()) && (document.getFontSizeCounter()
|
||||
.getMostPopular() == null || textBlock.getHighestFontSize() <= document.getFontSizeCounter().getMostPopular())) {
|
||||
textBlock.setClassification(PageBlockType.FOOTER);
|
||||
} else if (page.getPageNumber() == 1 && (PositionUtils.getHeightDifferenceBetweenChunkWordAndDocumentWord(textBlock,
|
||||
document.getTextHeightCounter().getMostPopular()) > 2.5 && textBlock.getHighestFontSize() > document.getFontSizeCounter().getMostPopular() || page.getTextBlocks()
|
||||
.size() == 1)) {
|
||||
if (!Pattern.matches("[0-9]+", textBlock.toString())) {
|
||||
textBlock.setClassification(PageBlockType.TITLE);
|
||||
}
|
||||
} else if (textBlock.getMostPopularWordFontSize() > document.getFontSizeCounter()
|
||||
.getMostPopular() && PositionUtils.getApproxLineCount(textBlock) < 4.9 && (textBlock.getMostPopularWordStyle().equals("bold") || !document.getFontStyleCounter()
|
||||
.getCountPerValue()
|
||||
.containsKey("bold") && textBlock.getMostPopularWordFontSize() > document.getFontSizeCounter().getMostPopular() + 1) && textBlock.getSequences()
|
||||
.get(0)
|
||||
.getTextPositions()
|
||||
.get(0)
|
||||
.getFontSizeInPt() >= textBlock.getMostPopularWordFontSize()) {
|
||||
|
||||
for (int i = 1; i <= headlineFontSizes.size(); i++) {
|
||||
if (textBlock.getMostPopularWordFontSize() == headlineFontSizes.get(i - 1)) {
|
||||
textBlock.setClassification(PageBlockType.getHeadlineType(i));
|
||||
document.setHeadlines(true);
|
||||
}
|
||||
}
|
||||
} else if (!textBlock.getText().startsWith("Figure ") && PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getMostPopularWordStyle()
|
||||
.equals("bold") && !document.getFontStyleCounter().getMostPopular().equals("bold") && PositionUtils.getApproxLineCount(textBlock) < 2.9 && textBlock.getSequences()
|
||||
.get(0)
|
||||
.getTextPositions()
|
||||
.get(0)
|
||||
.getFontSizeInPt() >= textBlock.getMostPopularWordFontSize()) {
|
||||
textBlock.setClassification(PageBlockType.getHeadlineType(headlineFontSizes.size() + 1));
|
||||
document.setHeadlines(true);
|
||||
} else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getMostPopularWordFontSize() == document.getFontSizeCounter()
|
||||
.getMostPopular() && textBlock.getMostPopularWordStyle().equals("bold") && !document.getFontStyleCounter().getMostPopular().equals("bold")) {
|
||||
textBlock.setClassification(PageBlockType.PARAGRAPH_BOLD);
|
||||
} else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getMostPopularWordFont()
|
||||
.equals(document.getFontCounter().getMostPopular()) && textBlock.getMostPopularWordStyle()
|
||||
.equals(document.getFontStyleCounter().getMostPopular()) && textBlock.getMostPopularWordFontSize() == document.getFontSizeCounter().getMostPopular()) {
|
||||
textBlock.setClassification(PageBlockType.PARAGRAPH);
|
||||
} else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getMostPopularWordFontSize() == document.getFontSizeCounter()
|
||||
.getMostPopular() && textBlock.getMostPopularWordStyle().equals("italic") && !document.getFontStyleCounter()
|
||||
.getMostPopular()
|
||||
.equals("italic") && PositionUtils.getApproxLineCount(textBlock) < 2.9) {
|
||||
textBlock.setClassification(PageBlockType.PARAGRAPH_ITALIC);
|
||||
} else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock)) {
|
||||
textBlock.setClassification(PageBlockType.PARAGRAPH_UNKNOWN);
|
||||
} else {
|
||||
textBlock.setClassification(PageBlockType.OTHER);
|
||||
}
|
||||
}
|
||||
|
||||
}
|
||||
-235
@@ -1,235 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
|
||||
|
||||
import java.awt.geom.Line2D;
|
||||
import java.awt.geom.Point2D;
|
||||
import java.util.ArrayList;
|
||||
import java.util.Collection;
|
||||
import java.util.Collections;
|
||||
import java.util.Comparator;
|
||||
import java.util.HashMap;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
|
||||
import org.springframework.stereotype.Service;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table.PdfTableCell;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.CleanRulings;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Ruling;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.Utils;
|
||||
import com.iqser.red.service.redaction.v1.server.settings.RedactionServiceSettings;
|
||||
|
||||
import lombok.RequiredArgsConstructor;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
|
||||
@Slf4j
|
||||
@Service
|
||||
@RequiredArgsConstructor
|
||||
public class RulingCleaningService {
|
||||
|
||||
private final RedactionServiceSettings redactionServiceSettings;
|
||||
|
||||
|
||||
public CleanRulings getCleanRulings(List<PdfTableCell> pdfTableCells, List<Ruling> rulings, float minCharWidth, float maxCharHeight) {
|
||||
|
||||
if (!rulings.isEmpty()) {
|
||||
snapPoints(rulings, minCharWidth, maxCharHeight);
|
||||
}
|
||||
|
||||
List<Ruling> vrs = new ArrayList<>();
|
||||
for (Ruling vr : rulings) {
|
||||
if (vr.vertical()) {
|
||||
vrs.add(vr);
|
||||
}
|
||||
}
|
||||
if (vrs.isEmpty() && redactionServiceSettings.isCvTableParsingEnabled()) {
|
||||
vrs.addAll(extractVerticalRulings(pdfTableCells));
|
||||
}
|
||||
List<Ruling> verticalRulingLines = collapseOrientedRulings(vrs);
|
||||
|
||||
List<Ruling> hrs = new ArrayList<>();
|
||||
for (Ruling hr : rulings) {
|
||||
if (hr.horizontal()) {
|
||||
hrs.add(hr);
|
||||
}
|
||||
}
|
||||
if (hrs.isEmpty() && redactionServiceSettings.isCvTableParsingEnabled()) {
|
||||
hrs.addAll(extractHorizontalRulings(pdfTableCells));
|
||||
}
|
||||
List<Ruling> horizontalRulingLines = collapseOrientedRulings(hrs);
|
||||
|
||||
return CleanRulings.builder().vertical(verticalRulingLines).horizontal(horizontalRulingLines).build();
|
||||
}
|
||||
|
||||
|
||||
public void snapPoints(List<? extends Line2D.Float> rulings, float xThreshold, float yThreshold) {
|
||||
|
||||
// collect points and keep a Line -> p1,p2 map
|
||||
Map<Line2D.Float, Point2D[]> linesToPoints = new HashMap<>();
|
||||
List<Point2D> points = new ArrayList<>();
|
||||
for (Line2D.Float r : rulings) {
|
||||
Point2D p1 = r.getP1();
|
||||
Point2D p2 = r.getP2();
|
||||
linesToPoints.put(r, new Point2D[]{p1, p2});
|
||||
points.add(p1);
|
||||
points.add(p2);
|
||||
}
|
||||
|
||||
// snap by X
|
||||
points.sort(Comparator.comparingDouble(Point2D::getX));
|
||||
|
||||
List<List<Point2D>> groupedPoints = new ArrayList<>();
|
||||
groupedPoints.add(new ArrayList<>(Collections.singletonList(points.get(0))));
|
||||
|
||||
for (Point2D p : points.subList(1, points.size() - 1)) {
|
||||
List<Point2D> last = groupedPoints.get(groupedPoints.size() - 1);
|
||||
if (Math.abs(p.getX() - last.get(0).getX()) < xThreshold) {
|
||||
groupedPoints.get(groupedPoints.size() - 1).add(p);
|
||||
} else {
|
||||
groupedPoints.add(new ArrayList<>(Collections.singletonList(p)));
|
||||
}
|
||||
}
|
||||
|
||||
for (List<Point2D> group : groupedPoints) {
|
||||
float avgLoc = 0;
|
||||
for (Point2D p : group) {
|
||||
avgLoc += p.getX();
|
||||
}
|
||||
avgLoc /= group.size();
|
||||
for (Point2D p : group) {
|
||||
p.setLocation(avgLoc, p.getY());
|
||||
}
|
||||
}
|
||||
// ---
|
||||
|
||||
// snap by Y
|
||||
points.sort(Comparator.comparingDouble(Point2D::getY));
|
||||
|
||||
groupedPoints = new ArrayList<>();
|
||||
groupedPoints.add(new ArrayList<>(Collections.singletonList(points.get(0))));
|
||||
|
||||
for (Point2D p : points.subList(1, points.size() - 1)) {
|
||||
List<Point2D> last = groupedPoints.get(groupedPoints.size() - 1);
|
||||
if (Math.abs(p.getY() - last.get(0).getY()) < yThreshold) {
|
||||
groupedPoints.get(groupedPoints.size() - 1).add(p);
|
||||
} else {
|
||||
groupedPoints.add(new ArrayList<>(Collections.singletonList(p)));
|
||||
}
|
||||
}
|
||||
|
||||
for (List<Point2D> group : groupedPoints) {
|
||||
float avgLoc = 0;
|
||||
for (Point2D p : group) {
|
||||
avgLoc += p.getY();
|
||||
}
|
||||
avgLoc /= group.size();
|
||||
for (Point2D p : group) {
|
||||
p.setLocation(p.getX(), avgLoc);
|
||||
}
|
||||
}
|
||||
// ---
|
||||
|
||||
// finally, modify lines
|
||||
for (Map.Entry<Line2D.Float, Point2D[]> ltp : linesToPoints.entrySet()) {
|
||||
Point2D[] p = ltp.getValue();
|
||||
ltp.getKey().setLine(p[0], p[1]);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
private Collection<? extends Ruling> extractVerticalRulings(List<PdfTableCell> pdfTableCells) {
|
||||
|
||||
List<Ruling> vrs = new ArrayList<>();
|
||||
|
||||
if (pdfTableCells != null) {
|
||||
for (PdfTableCell pdfTableCell : pdfTableCells) {
|
||||
Ruling leftLine = createRuling(pdfTableCell.getX0(), pdfTableCell.getX0(), pdfTableCell.getY0(), pdfTableCell.getY1());
|
||||
Ruling rightLine = createRuling(pdfTableCell.getX1(), pdfTableCell.getX1(), pdfTableCell.getY0(), pdfTableCell.getY1());
|
||||
vrs.add(leftLine);
|
||||
vrs.add(rightLine);
|
||||
}
|
||||
}
|
||||
return vrs;
|
||||
}
|
||||
|
||||
|
||||
private Collection<? extends Ruling> extractHorizontalRulings(List<PdfTableCell> pdfTableCells) {
|
||||
|
||||
List<Ruling> hrs = new ArrayList<>();
|
||||
|
||||
if (pdfTableCells != null) {
|
||||
for (PdfTableCell pdfTableCell : pdfTableCells) {
|
||||
Ruling topLine = createRuling(pdfTableCell.getX0(), pdfTableCell.getX1(), pdfTableCell.getY1(), pdfTableCell.getY1());
|
||||
Ruling baseLine = createRuling(pdfTableCell.getX0(), pdfTableCell.getX1(), pdfTableCell.getY0(), pdfTableCell.getY0());
|
||||
hrs.add(topLine);
|
||||
hrs.add(baseLine);
|
||||
}
|
||||
}
|
||||
|
||||
return hrs;
|
||||
}
|
||||
|
||||
|
||||
private Ruling createRuling(float tableCellX0, float tableCellX1, float tableCellY0, float tableCellY1) {
|
||||
|
||||
float x0 = tableCellX0;
|
||||
float x1 = tableCellX1;
|
||||
float y0 = tableCellY0;
|
||||
float y1 = tableCellY1;
|
||||
|
||||
if (x1 < x0) {
|
||||
x0 = tableCellX1;
|
||||
x1 = tableCellX0;
|
||||
}
|
||||
|
||||
if (y1 < y0) {
|
||||
y0 = tableCellY1;
|
||||
y1 = tableCellY0;
|
||||
}
|
||||
|
||||
return new Ruling(new Point2D.Float(x0, y0), new Point2D.Float(x1, y1));
|
||||
}
|
||||
|
||||
|
||||
private List<Ruling> collapseOrientedRulings(List<Ruling> lines) {
|
||||
|
||||
int COLINEAR_OR_PARALLEL_PIXEL_EXPAND_AMOUNT = 1;
|
||||
return collapseOrientedRulings(lines, COLINEAR_OR_PARALLEL_PIXEL_EXPAND_AMOUNT);
|
||||
}
|
||||
|
||||
|
||||
private List<Ruling> collapseOrientedRulings(List<Ruling> lines, int expandAmount) {
|
||||
|
||||
ArrayList<Ruling> rv = new ArrayList<>();
|
||||
lines.sort((a, b) -> {
|
||||
final float diff = a.getPosition() - b.getPosition();
|
||||
return Float.compare(diff == 0 ? a.getStart() - b.getStart() : diff, 0f);
|
||||
});
|
||||
|
||||
for (Ruling next_line : lines) {
|
||||
Ruling last = rv.isEmpty() ? null : rv.get(rv.size() - 1);
|
||||
// if current line colinear with next, and are "close enough": expand current line
|
||||
if (last != null && Utils.feq(next_line.getPosition(), last.getPosition()) && last.nearlyIntersects(next_line, expandAmount)) {
|
||||
final float lastStart = last.getStart();
|
||||
final float lastEnd = last.getEnd();
|
||||
|
||||
final boolean lastFlipped = lastStart > lastEnd;
|
||||
final boolean nextFlipped = next_line.getStart() > next_line.getEnd();
|
||||
|
||||
boolean differentDirections = nextFlipped != lastFlipped;
|
||||
float nextS = differentDirections ? next_line.getEnd() : next_line.getStart();
|
||||
float nextE = differentDirections ? next_line.getStart() : next_line.getEnd();
|
||||
|
||||
final float newStart = lastFlipped ? Math.max(nextS, lastStart) : Math.min(nextS, lastStart);
|
||||
final float newEnd = lastFlipped ? Math.min(nextE, lastEnd) : Math.max(nextE, lastEnd);
|
||||
last.setStartEnd(newStart, newEnd);
|
||||
assert !last.oblique();
|
||||
} else if (next_line.length() == 0) {
|
||||
continue;
|
||||
} else {
|
||||
rv.add(next_line);
|
||||
}
|
||||
}
|
||||
return rv;
|
||||
}
|
||||
|
||||
}
|
||||
-304
@@ -1,304 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.Collections;
|
||||
import java.util.HashMap;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
import java.util.stream.Collectors;
|
||||
|
||||
import org.apache.commons.collections4.CollectionUtils;
|
||||
import org.springframework.stereotype.Service;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationDocument;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationFooter;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationHeader;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationSection;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.PageBlockType;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.image.ClassifiedImage;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Cell;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.TablePageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.UnclassifiedText;
|
||||
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
|
||||
@Slf4j
|
||||
@Service
|
||||
public class SectionsBuilderService {
|
||||
|
||||
public void buildSections(ClassificationDocument document) {
|
||||
|
||||
List<AbstractPageBlock> chunkWords = new ArrayList<>();
|
||||
List<ClassificationSection> chunkBlockList = new ArrayList<>();
|
||||
List<ClassificationHeader> headers = new ArrayList<>();
|
||||
List<ClassificationFooter> footers = new ArrayList<>();
|
||||
List<UnclassifiedText> unclassifiedTexts = new ArrayList<>();
|
||||
|
||||
AbstractPageBlock prev = null;
|
||||
|
||||
String lastHeadline = "";
|
||||
TablePageBlock previousTable = null;
|
||||
for (ClassificationPage page : document.getPages()) {
|
||||
List<TextPageBlock> header = new ArrayList<>();
|
||||
List<TextPageBlock> footer = new ArrayList<>();
|
||||
List<TextPageBlock> unclassifiedText = new ArrayList<>();
|
||||
for (AbstractPageBlock current : page.getTextBlocks()) {
|
||||
|
||||
if (current.getClassification() == null) {
|
||||
continue;
|
||||
}
|
||||
|
||||
current.setPage(page.getPageNumber());
|
||||
|
||||
if (current.getClassification().equals(PageBlockType.HEADER)) {
|
||||
header.add((TextPageBlock) current);
|
||||
continue;
|
||||
}
|
||||
|
||||
if (current.getClassification().equals(PageBlockType.FOOTER)) {
|
||||
footer.add((TextPageBlock) current);
|
||||
continue;
|
||||
}
|
||||
|
||||
if (current.getClassification().equals(PageBlockType.OTHER)) {
|
||||
unclassifiedText.add((TextPageBlock) current);
|
||||
continue;
|
||||
}
|
||||
|
||||
if (prev != null && current.getClassification().isHeadline() && !prev.getClassification().isHeadline() || !document.isHeadlines()) {
|
||||
ClassificationSection chunkBlock = buildTextBlock(chunkWords, lastHeadline);
|
||||
chunkBlock.setHeadline(lastHeadline);
|
||||
if (document.isHeadlines()) {
|
||||
lastHeadline = current.getText();
|
||||
}
|
||||
chunkBlockList.add(chunkBlock);
|
||||
chunkWords = new ArrayList<>();
|
||||
if (CollectionUtils.isNotEmpty(chunkBlock.getTables())) {
|
||||
previousTable = chunkBlock.getTables().get(chunkBlock.getTables().size() - 1);
|
||||
}
|
||||
}
|
||||
if (current instanceof TablePageBlock table) {
|
||||
// Distribute header information for subsequent tables
|
||||
mergeTableMetadata(table, previousTable);
|
||||
previousTable = table;
|
||||
}
|
||||
chunkWords.add(current);
|
||||
prev = current;
|
||||
}
|
||||
|
||||
if (!header.isEmpty()) {
|
||||
headers.add(new ClassificationHeader(header));
|
||||
}
|
||||
if (!footer.isEmpty()) {
|
||||
footers.add(new ClassificationFooter(footer));
|
||||
}
|
||||
if (!unclassifiedText.isEmpty()) {
|
||||
unclassifiedTexts.add(new UnclassifiedText(unclassifiedText));
|
||||
}
|
||||
}
|
||||
|
||||
ClassificationSection chunkBlock = buildTextBlock(chunkWords, lastHeadline);
|
||||
chunkBlock.setHeadline(lastHeadline);
|
||||
chunkBlockList.add(chunkBlock);
|
||||
|
||||
document.setSections(chunkBlockList);
|
||||
document.setHeaders(headers);
|
||||
document.setFooters(footers);
|
||||
document.setUnclassifiedTexts(unclassifiedTexts);
|
||||
}
|
||||
|
||||
|
||||
public void addImagesToSections(ClassificationDocument document) {
|
||||
|
||||
Map<Integer, List<ClassificationSection>> sectionMap = new HashMap<>();
|
||||
for (ClassificationSection section : document.getSections()) {
|
||||
for (AbstractPageBlock container : section.getPageBlocks()) {
|
||||
|
||||
List<ClassificationSection> sectionsOnPage = sectionMap.computeIfAbsent(container.getPage(), c -> new ArrayList<>());
|
||||
if (sectionsOnPage.contains(section)) {
|
||||
continue;
|
||||
}
|
||||
sectionsOnPage.add(section);
|
||||
}
|
||||
}
|
||||
|
||||
if (sectionMap.isEmpty()) {
|
||||
ClassificationSection section = new ClassificationSection();
|
||||
document.getSections().add(section);
|
||||
sectionMap.computeIfAbsent(1, x -> new ArrayList<>()).add(section);
|
||||
}
|
||||
|
||||
// first page is always a paragraph, else we can't process pages 1..N,
|
||||
// where N is the first found page with a paragraph
|
||||
if (sectionMap.get(1) == null) {
|
||||
ClassificationSection section = new ClassificationSection();
|
||||
document.getSections().add(section);
|
||||
sectionMap.computeIfAbsent(1, x -> new ArrayList<>()).add(section);
|
||||
}
|
||||
|
||||
for (ClassificationPage page : document.getPages()) {
|
||||
for (ClassifiedImage image : page.getImages()) {
|
||||
List<ClassificationSection> sectionsOnPage = sectionMap.get(page.getPageNumber());
|
||||
if (sectionsOnPage == null) {
|
||||
int i = page.getPageNumber();
|
||||
while (sectionsOnPage == null) {
|
||||
sectionsOnPage = sectionMap.get(i);
|
||||
i--;
|
||||
}
|
||||
}
|
||||
for (ClassificationSection section : sectionsOnPage) {
|
||||
Float xMin = null;
|
||||
Float yMin = null;
|
||||
Float xMax = null;
|
||||
Float yMax = null;
|
||||
|
||||
for (AbstractPageBlock abs : section.getPageBlocks()) {
|
||||
if (abs.getPage() != page.getPageNumber()) {
|
||||
continue;
|
||||
}
|
||||
|
||||
if (abs.getMinX() < abs.getMaxX()) {
|
||||
if (xMin == null || abs.getMinX() < xMin) {
|
||||
xMin = abs.getMinX();
|
||||
}
|
||||
if (xMax == null || abs.getMaxX() > xMax) {
|
||||
xMax = abs.getMaxX();
|
||||
}
|
||||
} else {
|
||||
if (xMin == null || abs.getMaxX() < xMin) {
|
||||
xMin = abs.getMaxX();
|
||||
}
|
||||
if (xMax == null || abs.getMinX() > xMax) {
|
||||
xMax = abs.getMinX();
|
||||
}
|
||||
}
|
||||
|
||||
if (abs.getMinY() < abs.getMaxY()) {
|
||||
if (yMin == null || abs.getMinY() < yMin) {
|
||||
yMin = abs.getMinY();
|
||||
}
|
||||
if (yMax == null || abs.getMaxY() > yMax) {
|
||||
yMax = abs.getMaxY();
|
||||
}
|
||||
} else {
|
||||
if (yMin == null || abs.getMaxY() < yMin) {
|
||||
yMin = abs.getMaxY();
|
||||
}
|
||||
if (yMax == null || abs.getMinY() > yMax) {
|
||||
yMax = abs.getMinY();
|
||||
}
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
log.debug("Image position x: {}, y: {}", image.getPosition().getX(), image.getPosition().getY());
|
||||
log.debug("Paragraph position xMin: {}, xMax: {}, yMin: {}, yMax: {}", xMin, xMax, yMin, yMax);
|
||||
|
||||
if (xMin != null && xMax != null && yMin != null && yMax != null && image.getPosition().getX() >= xMin && image.getPosition()
|
||||
.getX() <= xMax && image.getPosition().getY() >= yMin && image.getPosition().getY() <= yMax) {
|
||||
section.getImages().add(image);
|
||||
image.setAppendedToSection(true);
|
||||
}
|
||||
}
|
||||
if (!image.isAppendedToSection()) {
|
||||
log.debug("Image uses first paragraph");
|
||||
sectionsOnPage.get(0).getImages().add(image);
|
||||
image.setAppendedToSection(true);
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
private void mergeTableMetadata(TablePageBlock currentTable, TablePageBlock previousTable) {
|
||||
|
||||
// Distribute header information for subsequent tables
|
||||
if (previousTable != null && hasInvalidHeaderInformation(currentTable) && hasValidHeaderInformation(previousTable)) {
|
||||
List<Cell> previousTableNonHeaderRow = getRowWithNonHeaderCells(previousTable);
|
||||
List<Cell> tableNonHeaderRow = getRowWithNonHeaderCells(currentTable);
|
||||
// Allow merging of tables if header row is separated from first logical non-header row
|
||||
if (previousTableNonHeaderRow.isEmpty() && previousTable.getRowCount() == 1 && previousTable.getRows().get(0).size() == tableNonHeaderRow.size()) {
|
||||
previousTableNonHeaderRow = previousTable.getRows().get(0).stream().map(cell -> {
|
||||
Cell fakeCell = new Cell(cell.getPoints()[0], cell.getPoints()[2]);
|
||||
fakeCell.setHeaderCells(Collections.singletonList(cell));
|
||||
return fakeCell;
|
||||
}).collect(Collectors.toList());
|
||||
}
|
||||
if (previousTableNonHeaderRow.size() == tableNonHeaderRow.size()) {
|
||||
for (int i = currentTable.getRowCount() - 1; i >= 0; i--) { // Non header rows are most likely at bottom of table
|
||||
List<Cell> row = currentTable.getRows().get(i);
|
||||
if (row.size() == tableNonHeaderRow.size() && row.stream().allMatch(cell -> cell.getHeaderCells().isEmpty())) {
|
||||
for (int j = 0; j < row.size(); j++) {
|
||||
row.get(j).setHeaderCells(previousTableNonHeaderRow.get(j).getHeaderCells());
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
private ClassificationSection buildTextBlock(List<AbstractPageBlock> wordBlockList, String lastHeadline) {
|
||||
|
||||
ClassificationSection section = new ClassificationSection();
|
||||
|
||||
for (AbstractPageBlock container : wordBlockList) {
|
||||
if (container instanceof TablePageBlock table) {
|
||||
|
||||
if (lastHeadline == null || lastHeadline.isEmpty()) {
|
||||
table.setHeadline("Text in table");
|
||||
} else {
|
||||
table.setHeadline("TablePageBlock in: " + lastHeadline);
|
||||
}
|
||||
|
||||
section.getPageBlocks().add(table);
|
||||
continue;
|
||||
}
|
||||
|
||||
TextPageBlock wordBlock = (TextPageBlock) container;
|
||||
section.getPageBlocks().add(wordBlock);
|
||||
}
|
||||
return section;
|
||||
}
|
||||
|
||||
|
||||
private boolean hasValidHeaderInformation(TablePageBlock table) {
|
||||
|
||||
return !hasInvalidHeaderInformation(table);
|
||||
}
|
||||
|
||||
|
||||
private boolean hasInvalidHeaderInformation(TablePageBlock table) {
|
||||
|
||||
return table.getRows().stream().flatMap(row -> row.stream().filter(cell -> CollectionUtils.isNotEmpty(cell.getHeaderCells()))).findAny().isEmpty();
|
||||
|
||||
}
|
||||
|
||||
|
||||
private List<Cell> getRowWithNonHeaderCells(TablePageBlock table) {
|
||||
|
||||
for (int i = table.getRowCount() - 1; i >= 0; i--) { // Non header rows are most likely at bottom of table
|
||||
List<Cell> row = table.getRows().get(i);
|
||||
if (row.size() == 1) {
|
||||
continue;
|
||||
}
|
||||
boolean allNonHeader = true;
|
||||
for (Cell cell : row) {
|
||||
if (cell.isHeaderCell()) {
|
||||
allNonHeader = false;
|
||||
break;
|
||||
}
|
||||
}
|
||||
if (allNonHeader) {
|
||||
return row;
|
||||
}
|
||||
}
|
||||
|
||||
return Collections.emptyList();
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
-337
@@ -1,337 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
|
||||
|
||||
import java.awt.geom.Point2D;
|
||||
import java.util.ArrayList;
|
||||
import java.util.Comparator;
|
||||
import java.util.HashMap;
|
||||
import java.util.HashSet;
|
||||
import java.util.Iterator;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
import java.util.Set;
|
||||
import java.util.stream.Collectors;
|
||||
|
||||
import org.springframework.stereotype.Service;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Cell;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.CleanRulings;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Rectangle;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Ruling;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.TablePageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.Utils;
|
||||
|
||||
@Service
|
||||
public class TableExtractionService {
|
||||
|
||||
private static final Comparator<Point2D> X_FIRST_POINT_COMPARATOR = (arg0, arg1) -> {
|
||||
|
||||
int rv = 0;
|
||||
float arg0X = Utils.round(arg0.getX(), 2);
|
||||
float arg0Y = Utils.round(arg0.getY(), 2);
|
||||
float arg1X = Utils.round(arg1.getX(), 2);
|
||||
float arg1Y = Utils.round(arg1.getY(), 2);
|
||||
|
||||
if (arg0X > arg1X) {
|
||||
rv = 1;
|
||||
} else if (arg0X < arg1X) {
|
||||
rv = -1;
|
||||
} else if (arg0Y > arg1Y) {
|
||||
rv = 1;
|
||||
} else if (arg0Y < arg1Y) {
|
||||
rv = -1;
|
||||
}
|
||||
return rv;
|
||||
};
|
||||
private static final Comparator<Point2D> POINT_COMPARATOR = (arg0, arg1) -> {
|
||||
|
||||
int rv = 0;
|
||||
float arg0X = Utils.round(arg0.getX(), 2);
|
||||
float arg0Y = Utils.round(arg0.getY(), 2);
|
||||
float arg1X = Utils.round(arg1.getX(), 2);
|
||||
float arg1Y = Utils.round(arg1.getY(), 2);
|
||||
|
||||
if (arg0Y > arg1Y) {
|
||||
rv = 1;
|
||||
} else if (arg0Y < arg1Y) {
|
||||
rv = -1;
|
||||
} else if (arg0X > arg1X) {
|
||||
rv = 1;
|
||||
} else if (arg0X < arg1X) {
|
||||
rv = -1;
|
||||
}
|
||||
return rv;
|
||||
};
|
||||
|
||||
|
||||
/**
|
||||
* Finds tables on a page and moves textblocks into cells of the found tables.
|
||||
* Note: This algorithm uses Pdf Coordinate System where {0,0} rotated with the page rotation.
|
||||
* 0 -> LowerLeft
|
||||
* 90 -> UpperLeft
|
||||
* 180 -> UpperRight
|
||||
* 270 -> LowerRight
|
||||
* <p>
|
||||
* DirAdj (Text direction adjusted) values can not be used here.
|
||||
*
|
||||
* @param cleanRulings The lines used to build the table.
|
||||
* @param page Page object that contains textblocks and statistics.
|
||||
*/
|
||||
public void extractTables(CleanRulings cleanRulings, ClassificationPage page) {
|
||||
|
||||
List<Cell> cells = findCells(cleanRulings.getHorizontal(), cleanRulings.getVertical());
|
||||
|
||||
List<TextPageBlock> toBeRemoved = new ArrayList<>();
|
||||
|
||||
for (AbstractPageBlock abstractPageBlock : page.getTextBlocks()) {
|
||||
TextPageBlock textBlock = (TextPageBlock) abstractPageBlock;
|
||||
for (Cell cell : cells) {
|
||||
if (cell.hasMinimumSize() && cell.intersects(textBlock.getPdfMinX(),
|
||||
textBlock.getPdfMinY(),
|
||||
textBlock.getPdfMaxX() - textBlock.getPdfMinX(),
|
||||
textBlock.getPdfMaxY() - textBlock.getPdfMinY())) {
|
||||
cell.addTextBlock(textBlock);
|
||||
toBeRemoved.add(textBlock);
|
||||
break;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
cells = new ArrayList<>(new HashSet<>(cells));
|
||||
Utils.sort(cells, Rectangle.ILL_DEFINED_ORDER);
|
||||
|
||||
List<Rectangle> spreadsheetAreas = findSpreadsheetsFromCells(cells).stream().filter(r -> r.getWidth() > 0f && r.getHeight() > 0f).collect(Collectors.toList());
|
||||
|
||||
List<TablePageBlock> tables = new ArrayList<>();
|
||||
for (Rectangle area : spreadsheetAreas) {
|
||||
|
||||
List<Cell> overlappingCells = new ArrayList<>();
|
||||
for (Cell c : cells) {
|
||||
if (c.hasMinimumSize() && c.intersects(area)) {
|
||||
overlappingCells.add(c);
|
||||
}
|
||||
}
|
||||
tables.add(new TablePageBlock(overlappingCells, area, page.getRotation()));
|
||||
}
|
||||
|
||||
for (TablePageBlock table : tables) {
|
||||
int position = -1;
|
||||
|
||||
Iterator<AbstractPageBlock> itty = page.getTextBlocks().iterator();
|
||||
while (itty.hasNext()) {
|
||||
AbstractPageBlock textBlock = itty.next();
|
||||
if (textBlock instanceof TextPageBlock ? table.containsBlock((TextPageBlock) textBlock) : table.contains(textBlock) && position == -1) {
|
||||
position = page.getTextBlocks().indexOf(textBlock);
|
||||
}
|
||||
}
|
||||
if (position != -1) {
|
||||
page.getTextBlocks().add(position, table);
|
||||
}
|
||||
}
|
||||
|
||||
page.getTextBlocks().removeAll(toBeRemoved);
|
||||
}
|
||||
|
||||
|
||||
public List<Cell> findCells(List<Ruling> horizontalRulingLines, List<Ruling> verticalRulingLines) {
|
||||
|
||||
List<Cell> cellsFound = new ArrayList<>();
|
||||
Map<Point2D, Ruling[]> intersectionPoints = Ruling.findIntersections(horizontalRulingLines, verticalRulingLines);
|
||||
List<Point2D> intersectionPointsList = new ArrayList<>(intersectionPoints.keySet());
|
||||
intersectionPointsList.sort(POINT_COMPARATOR);
|
||||
|
||||
for (int i = 0; i < intersectionPointsList.size(); i++) {
|
||||
Point2D topLeft = intersectionPointsList.get(i);
|
||||
Ruling[] hv = intersectionPoints.get(topLeft);
|
||||
|
||||
// CrossingPointsDirectlyBelow( topLeft );
|
||||
List<Point2D> xPoints = new ArrayList<>();
|
||||
// CrossingPointsDirectlyToTheRight( topLeft );
|
||||
List<Point2D> yPoints = new ArrayList<>();
|
||||
|
||||
for (Point2D p : intersectionPointsList.subList(i, intersectionPointsList.size())) {
|
||||
if (p.getX() == topLeft.getX() && p.getY() > topLeft.getY()) {
|
||||
xPoints.add(p);
|
||||
}
|
||||
if (p.getY() == topLeft.getY() && p.getX() > topLeft.getX()) {
|
||||
yPoints.add(p);
|
||||
}
|
||||
}
|
||||
outer:
|
||||
for (Point2D xPoint : xPoints) {
|
||||
// is there a vertical edge b/w topLeft and xPoint?
|
||||
if (!hv[1].equals(intersectionPoints.get(xPoint)[1])) {
|
||||
continue;
|
||||
}
|
||||
for (Point2D yPoint : yPoints) {
|
||||
// is there an horizontal edge b/w topLeft and yPoint ?
|
||||
if (!hv[0].equals(intersectionPoints.get(yPoint)[0])) {
|
||||
continue;
|
||||
}
|
||||
Point2D btmRight = new Point2D.Float((float) yPoint.getX(), (float) xPoint.getY());
|
||||
if (intersectionPoints.containsKey(btmRight) && intersectionPoints.get(btmRight)[0].equals(intersectionPoints.get(xPoint)[0]) && intersectionPoints.get(btmRight)[1].equals(
|
||||
intersectionPoints.get(yPoint)[1])) {
|
||||
cellsFound.add(new Cell(topLeft, btmRight));
|
||||
break outer;
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// TODO create cells for vertical ruling lines with aligned endpoints at the top/bottom of a grid
|
||||
// that aren't connected with an horizontal ruler?
|
||||
// see: https://github.com/jazzido/tabula-extractor/issues/78#issuecomment-41481207
|
||||
|
||||
return cellsFound;
|
||||
}
|
||||
|
||||
|
||||
private List<Rectangle> findSpreadsheetsFromCells(List<? extends Rectangle> cells) {
|
||||
// via: http://stackoverflow.com/questions/13746284/merging-multiple-adjacent-rectangles-into-one-polygon
|
||||
List<Rectangle> rectangles = new ArrayList<>();
|
||||
Set<Point2D> pointSet = new HashSet<>();
|
||||
Map<Point2D, Point2D> edgesH = new HashMap<>();
|
||||
Map<Point2D, Point2D> edgesV = new HashMap<>();
|
||||
int i = 0;
|
||||
|
||||
for (Rectangle cell : cells) {
|
||||
for (Point2D pt : cell.getPoints()) {
|
||||
if (pointSet.contains(pt)) { // shared vertex, remove it
|
||||
pointSet.remove(pt);
|
||||
} else {
|
||||
pointSet.add(pt);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// X first sort
|
||||
List<Point2D> pointsSortX = new ArrayList<>(pointSet);
|
||||
pointsSortX.sort(X_FIRST_POINT_COMPARATOR);
|
||||
// Y first sort
|
||||
List<Point2D> pointsSortY = new ArrayList<>(pointSet);
|
||||
pointsSortY.sort(POINT_COMPARATOR);
|
||||
|
||||
while (i < pointSet.size()) {
|
||||
float currY = (float) pointsSortY.get(i).getY();
|
||||
while (i < pointSet.size() && Utils.feq(pointsSortY.get(i).getY(), currY)) {
|
||||
edgesH.put(pointsSortY.get(i), pointsSortY.get(i + 1));
|
||||
edgesH.put(pointsSortY.get(i + 1), pointsSortY.get(i));
|
||||
i += 2;
|
||||
}
|
||||
}
|
||||
|
||||
i = 0;
|
||||
while (i < pointSet.size()) {
|
||||
float currX = (float) pointsSortX.get(i).getX();
|
||||
while (i < pointSet.size() && Utils.feq(pointsSortX.get(i).getX(), currX)) {
|
||||
edgesV.put(pointsSortX.get(i), pointsSortX.get(i + 1));
|
||||
edgesV.put(pointsSortX.get(i + 1), pointsSortX.get(i));
|
||||
i += 2;
|
||||
}
|
||||
}
|
||||
|
||||
// Get all the polygons
|
||||
List<List<PolygonVertex>> polygons = new ArrayList<>();
|
||||
Point2D nextVertex;
|
||||
while (!edgesH.isEmpty()) {
|
||||
ArrayList<PolygonVertex> polygon = new ArrayList<>();
|
||||
Point2D first = edgesH.keySet().iterator().next();
|
||||
polygon.add(new PolygonVertex(first, Direction.HORIZONTAL));
|
||||
edgesH.remove(first);
|
||||
|
||||
while (true) {
|
||||
PolygonVertex curr = polygon.get(polygon.size() - 1);
|
||||
PolygonVertex lastAddedVertex;
|
||||
if (curr.direction == Direction.HORIZONTAL) {
|
||||
nextVertex = edgesV.get(curr.point);
|
||||
edgesV.remove(curr.point);
|
||||
lastAddedVertex = new PolygonVertex(nextVertex, Direction.VERTICAL);
|
||||
} else {
|
||||
nextVertex = edgesH.get(curr.point);
|
||||
edgesH.remove(curr.point);
|
||||
lastAddedVertex = new PolygonVertex(nextVertex, Direction.HORIZONTAL);
|
||||
}
|
||||
polygon.add(lastAddedVertex);
|
||||
|
||||
if (lastAddedVertex.equals(polygon.get(0))) {
|
||||
// closed polygon
|
||||
polygon.remove(polygon.size() - 1);
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
for (PolygonVertex vertex : polygon) {
|
||||
edgesH.remove(vertex.point);
|
||||
edgesV.remove(vertex.point);
|
||||
}
|
||||
polygons.add(polygon);
|
||||
}
|
||||
|
||||
// calculate grid-aligned minimum area rectangles for each found polygon
|
||||
for (List<PolygonVertex> poly : polygons) {
|
||||
float top = Float.MAX_VALUE;
|
||||
float left = Float.MAX_VALUE;
|
||||
float bottom = Float.MIN_VALUE;
|
||||
float right = Float.MIN_VALUE;
|
||||
for (PolygonVertex pt : poly) {
|
||||
top = (float) Math.min(top, pt.point.getY());
|
||||
left = (float) Math.min(left, pt.point.getX());
|
||||
bottom = (float) Math.max(bottom, pt.point.getY());
|
||||
right = (float) Math.max(right, pt.point.getX());
|
||||
}
|
||||
rectangles.add(new Rectangle(top, left, right - left, bottom - top));
|
||||
}
|
||||
|
||||
return rectangles;
|
||||
}
|
||||
|
||||
|
||||
private enum Direction {
|
||||
HORIZONTAL,
|
||||
VERTICAL
|
||||
}
|
||||
|
||||
static class PolygonVertex {
|
||||
|
||||
Point2D point;
|
||||
Direction direction;
|
||||
|
||||
|
||||
PolygonVertex(Point2D point, Direction direction) {
|
||||
|
||||
this.direction = direction;
|
||||
this.point = point;
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public boolean equals(Object other) {
|
||||
|
||||
if (this == other) {
|
||||
return true;
|
||||
}
|
||||
if (!(other instanceof PolygonVertex)) {
|
||||
return false;
|
||||
}
|
||||
return this.point.equals(((PolygonVertex) other).point);
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public int hashCode() {
|
||||
|
||||
return this.point.hashCode();
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public String toString() {
|
||||
|
||||
return String.format("%s[point=%s,direction=%s]", this.getClass().getName(), this.point.toString(), this.direction.toString());
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
-142
@@ -1,142 +0,0 @@
|
||||
/*
|
||||
* CohenSutherland.java
|
||||
* --------------------
|
||||
* (c) 2007 by Intevation GmbH
|
||||
*
|
||||
* @author Sascha L. Teichmann (teichmann@intevation.de)
|
||||
* @author Ludwig Reiter (ludwig@intevation.de)
|
||||
*
|
||||
* This program is free software under the LGPL (>=v2.1)
|
||||
* Read the file LICENSE.txt coming with the sources for details.
|
||||
*/
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils;
|
||||
|
||||
import java.awt.geom.Line2D;
|
||||
import java.awt.geom.Rectangle2D;
|
||||
|
||||
/**
|
||||
* Implements the well known Cohen Sutherland line
|
||||
* clipping algorithm (line against clip rectangle).
|
||||
*/
|
||||
@SuppressWarnings("all")
|
||||
public final class CohenSutherlandClipping {
|
||||
|
||||
private static final int INSIDE = 0;
|
||||
private static final int LEFT = 1;
|
||||
private static final int RIGHT = 2;
|
||||
private static final int BOTTOM = 4;
|
||||
private static final int TOP = 8;
|
||||
private double xMin;
|
||||
private double yMin;
|
||||
private double xMax;
|
||||
private double yMax;
|
||||
|
||||
|
||||
/**
|
||||
* Creates a Cohen Sutherland clipper with clip rect (0, 0, 0, 0).
|
||||
*/
|
||||
public CohenSutherlandClipping() {
|
||||
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* Creates a Cohen Sutherland clipper with the given clip rectangle.
|
||||
*
|
||||
* @param clip the clip rectangle to use
|
||||
*/
|
||||
public CohenSutherlandClipping(Rectangle2D clip) {
|
||||
|
||||
setClip(clip);
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* Sets the clip rectangle.
|
||||
*
|
||||
* @param clip the clip rectangle
|
||||
*/
|
||||
public void setClip(Rectangle2D clip) {
|
||||
|
||||
xMin = clip.getX();
|
||||
xMax = xMin + clip.getWidth();
|
||||
yMin = clip.getY();
|
||||
yMax = yMin + clip.getHeight();
|
||||
}
|
||||
|
||||
|
||||
private final int regionCode(double x, double y) {
|
||||
|
||||
int code = x < xMin ? LEFT : x > xMax ? RIGHT : INSIDE;
|
||||
if (y < yMin) {
|
||||
code |= BOTTOM;
|
||||
} else if (y > yMax) {
|
||||
code |= TOP;
|
||||
}
|
||||
return code;
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* Clips a given line against the clip rectangle.
|
||||
* The modification (if needed) is done in place.
|
||||
*
|
||||
* @param line the line to clip
|
||||
* @return true if line is clipped, false if line is
|
||||
* totally outside the clip rect.
|
||||
*/
|
||||
public boolean clip(Line2D.Float line) {
|
||||
|
||||
double p1x = line.getX1();
|
||||
double p1y = line.getY1();
|
||||
double p2x = line.getX2();
|
||||
double p2y = line.getY2();
|
||||
|
||||
double qx = 0d;
|
||||
double qy = 0d;
|
||||
|
||||
boolean vertical = p1x == p2x;
|
||||
|
||||
double slope = vertical ? 0d : (p2y - p1y) / (p2x - p1x);
|
||||
|
||||
int c1 = regionCode(p1x, p1y);
|
||||
int c2 = regionCode(p2x, p2y);
|
||||
|
||||
while (c1 != INSIDE || c2 != INSIDE) {
|
||||
|
||||
if ((c1 & c2) != INSIDE) {
|
||||
return false;
|
||||
}
|
||||
|
||||
int c = c1 == INSIDE ? c2 : c1;
|
||||
|
||||
if ((c & LEFT) != INSIDE) {
|
||||
qx = xMin;
|
||||
qy = (Utils.feq(qx, p1x) ? 0 : qx - p1x) * slope + p1y;
|
||||
} else if ((c & RIGHT) != INSIDE) {
|
||||
qx = xMax;
|
||||
qy = (Utils.feq(qx, p1x) ? 0 : qx - p1x) * slope + p1y;
|
||||
} else if ((c & BOTTOM) != INSIDE) {
|
||||
qy = yMin;
|
||||
qx = vertical ? p1x : (Utils.feq(qy, p1y) ? 0 : qy - p1y) / slope + p1x;
|
||||
} else if ((c & TOP) != INSIDE) {
|
||||
qy = yMax;
|
||||
qx = vertical ? p1x : (Utils.feq(qy, p1y) ? 0 : qy - p1y) / slope + p1x;
|
||||
}
|
||||
|
||||
if (c == c1) {
|
||||
p1x = qx;
|
||||
p1y = qy;
|
||||
c1 = regionCode(p1x, p1y);
|
||||
} else {
|
||||
p2x = qx;
|
||||
p2y = qy;
|
||||
c2 = regionCode(p2x, p2y);
|
||||
}
|
||||
}
|
||||
line.setLine(p1x, p1y, p2x, p2y);
|
||||
return true;
|
||||
}
|
||||
|
||||
}
|
||||
// end of file
|
||||
-56
@@ -1,56 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils;
|
||||
|
||||
import java.io.File;
|
||||
import java.io.IOException;
|
||||
import java.nio.file.Files;
|
||||
|
||||
import lombok.experimental.UtilityClass;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
|
||||
@Slf4j
|
||||
@UtilityClass
|
||||
public class FileUtils {
|
||||
|
||||
public File createTempFile(String filenamePrefix, String filenameSuffix) throws IOException {
|
||||
|
||||
File tempFile = Files.createTempFile(filenamePrefix, filenameSuffix).toFile();
|
||||
setRWPermissionsOnlyForOwner(tempFile);
|
||||
|
||||
return tempFile;
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* Deletes a file; logs a message with the reason if the deletion fails.
|
||||
* This method is null-safe.
|
||||
*
|
||||
* @param file The file to delete. Can be null.
|
||||
*/
|
||||
public void deleteFile(File file) {
|
||||
|
||||
if (file != null) {
|
||||
try {
|
||||
Files.deleteIfExists(file.toPath());
|
||||
} catch (IOException ex) {
|
||||
log.warn("Could not delete file!", ex);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
// We don't need to check the results of the permission setters below,
|
||||
// since we're manipulating a file we created ourselves.
|
||||
@SuppressWarnings({"ResultOfMethodCallIgnored", "squid:S899"})
|
||||
private void setRWPermissionsOnlyForOwner(File tempFile) {
|
||||
|
||||
try {
|
||||
tempFile.setReadable(true, true);
|
||||
tempFile.setWritable(true, true);
|
||||
tempFile.setExecutable(false);
|
||||
} catch (SecurityException ex) {
|
||||
// This should never happen since we're creating a temp file ourselves.
|
||||
log.warn("Caught an exception during temp file creation. This should not happend. Check the code.", ex);
|
||||
}
|
||||
}
|
||||
|
||||
}
|
||||
-121
@@ -1,121 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils;
|
||||
|
||||
import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Rectangle;
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
|
||||
|
||||
import lombok.experimental.UtilityClass;
|
||||
|
||||
@UtilityClass
|
||||
@SuppressWarnings("all")
|
||||
public final class PositionUtils {
|
||||
|
||||
// TODO This currently uses pdf coord system. In the futher this should use java coord system.
|
||||
// Note: DirAdj (TextDirection Adjusted) can not be user for this.
|
||||
public boolean isWithinBodyTextFrame(Rectangle btf, TextPageBlock textBlock) {
|
||||
|
||||
if (btf == null || textBlock == null) {
|
||||
return false;
|
||||
}
|
||||
|
||||
double threshold = textBlock.getMostPopularWordHeight() * 3;
|
||||
|
||||
if (textBlock.getPdfMinX() + threshold > btf.getTopLeft().getX() && textBlock.getPdfMaxX() - threshold < btf.getTopLeft()
|
||||
.getX() + btf.getWidth() && textBlock.getPdfMinY() + threshold > btf.getTopLeft().getY() && textBlock.getPdfMaxY() - threshold < btf.getTopLeft()
|
||||
.getY() + btf.getHeight()) {
|
||||
return true;
|
||||
} else {
|
||||
return false;
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
|
||||
// TODO This currently uses pdf coord system. In the futher this should use java coord system.
|
||||
// Note: DirAdj (TextDirection Adjusted) can not be user for this.
|
||||
public boolean isOverBodyTextFrame(Rectangle btf, TextPageBlock textBlock, int rotation) {
|
||||
|
||||
if (btf == null || textBlock == null) {
|
||||
return false;
|
||||
}
|
||||
|
||||
if (rotation == 90 && textBlock.getPdfMaxX() < btf.getTopLeft().getX()) {
|
||||
return true;
|
||||
}
|
||||
|
||||
if (rotation == 180 && textBlock.getPdfMaxY() < btf.getTopLeft().getY()) {
|
||||
return true;
|
||||
}
|
||||
|
||||
if (rotation == 270 && textBlock.getPdfMinX() > btf.getTopLeft().getX() + btf.getWidth()) {
|
||||
return true;
|
||||
}
|
||||
|
||||
if (rotation == 0 && textBlock.getPdfMinY() > btf.getTopLeft().getY() + btf.getHeight()) {
|
||||
return true;
|
||||
} else {
|
||||
return false;
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
|
||||
// TODO This currently uses pdf coord system. In the futher this should use java coord system.
|
||||
// Note: DirAdj (TextDirection Adjusted) can not be user for this.
|
||||
public boolean isUnderBodyTextFrame(Rectangle btf, TextPageBlock textBlock, int rotation) {
|
||||
|
||||
if (btf == null || textBlock == null) {
|
||||
return false;
|
||||
}
|
||||
|
||||
if (rotation == 90 && textBlock.getPdfMinX() > btf.getTopLeft().getX() + btf.getWidth()) {
|
||||
return true;
|
||||
}
|
||||
|
||||
if (rotation == 180 && textBlock.getPdfMinY() > btf.getTopLeft().getY() + btf.getHeight()) {
|
||||
return true;
|
||||
}
|
||||
|
||||
if (rotation == 270 && textBlock.getPdfMaxX() < btf.getTopLeft().getX()) {
|
||||
return true;
|
||||
}
|
||||
|
||||
if (rotation == 0 && textBlock.getPdfMaxY() < btf.getTopLeft().getY()) {
|
||||
return true;
|
||||
} else {
|
||||
return false;
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
|
||||
// TODO This currently uses pdf coord system. In the futher this should use java coord system.
|
||||
// Note: DirAdj (TextDirection Adjusted) can not be user for this.
|
||||
public boolean isTouchingUnderBodyTextFrame(Rectangle btf, TextPageBlock textBlock) {
|
||||
|
||||
//TODO Currently this is not working for rotated pages.
|
||||
|
||||
if (btf == null || textBlock == null) {
|
||||
return false;
|
||||
}
|
||||
|
||||
if (textBlock.getMinY() < btf.getTopLeft().getY()) {
|
||||
return true;
|
||||
} else {
|
||||
return false;
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
|
||||
public float getHeightDifferenceBetweenChunkWordAndDocumentWord(TextPageBlock textBlock, Float documentMostPopularWordHeight) {
|
||||
|
||||
return textBlock.getMostPopularWordHeight() - documentMostPopularWordHeight;
|
||||
}
|
||||
|
||||
|
||||
public Float getApproxLineCount(TextPageBlock textBlock) {
|
||||
|
||||
return textBlock.getHeight() / textBlock.getMostPopularWordHeight();
|
||||
}
|
||||
|
||||
}
|
||||
-109
@@ -1,109 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils;
|
||||
|
||||
import java.util.ArrayDeque;
|
||||
import java.util.Comparator;
|
||||
import java.util.Deque;
|
||||
import java.util.List;
|
||||
|
||||
import lombok.experimental.UtilityClass;
|
||||
|
||||
/**
|
||||
* Copied and minimal modified from PDFBox.
|
||||
*/
|
||||
@UtilityClass
|
||||
public final class QuickSort {
|
||||
|
||||
private static final Comparator<? extends Comparable> OBJCOMP = new Comparator<Comparable>() {
|
||||
@Override
|
||||
public int compare(Comparable object1, Comparable object2) {
|
||||
|
||||
return object1.compareTo(object2);
|
||||
}
|
||||
};
|
||||
|
||||
|
||||
/**
|
||||
* Sorts the given list using the given comparator.
|
||||
*
|
||||
* @param <T> type of the objects to be sorted.
|
||||
* @param list list to be sorted
|
||||
* @param cmp comparator used to compare the objects within the list
|
||||
*/
|
||||
public static <T> void sort(List<T> list, Comparator<? super T> cmp) {
|
||||
|
||||
int size = list.size();
|
||||
if (size < 2) {
|
||||
return;
|
||||
}
|
||||
quicksort(list, cmp);
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* Sorts the given list using compareTo as comparator.
|
||||
*
|
||||
* @param <T> type of the objects to be sorted.
|
||||
* @param list list to be sorted
|
||||
*/
|
||||
public static <T extends Comparable> void sort(List<T> list) {
|
||||
|
||||
sort(list, (Comparator<T>) OBJCOMP);
|
||||
}
|
||||
|
||||
|
||||
private static <T> void quicksort(List<T> list, Comparator<? super T> cmp) {
|
||||
|
||||
Deque<Integer> stack = new ArrayDeque<Integer>();
|
||||
stack.push(0);
|
||||
stack.push(list.size());
|
||||
while (!stack.isEmpty()) {
|
||||
int right = stack.pop();
|
||||
int left = stack.pop();
|
||||
if (right - left < 2) {
|
||||
continue;
|
||||
}
|
||||
int p = left + ((right - left) / 2);
|
||||
p = partition(list, cmp, p, left, right);
|
||||
|
||||
stack.push(p + 1);
|
||||
stack.push(right);
|
||||
|
||||
stack.push(left);
|
||||
stack.push(p);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
private static <T> int partition(List<T> list, Comparator<? super T> cmp, int p, int start, int end) {
|
||||
|
||||
int l = start;
|
||||
int h = end - 2;
|
||||
T piv = list.get(p);
|
||||
swap(list, p, end - 1);
|
||||
|
||||
while (l < h) {
|
||||
if (cmp.compare(list.get(l), piv) <= 0) {
|
||||
l++;
|
||||
} else if (cmp.compare(piv, list.get(h)) <= 0) {
|
||||
h--;
|
||||
} else {
|
||||
swap(list, l, h);
|
||||
}
|
||||
}
|
||||
int idx = h;
|
||||
if (cmp.compare(list.get(h), piv) < 0) {
|
||||
idx++;
|
||||
}
|
||||
swap(list, end - 1, idx);
|
||||
return idx;
|
||||
}
|
||||
|
||||
|
||||
private static <T> void swap(List<T> list, int i, int j) {
|
||||
|
||||
T tmp = list.get(i);
|
||||
list.set(i, list.get(j));
|
||||
list.set(j, tmp);
|
||||
}
|
||||
|
||||
}
|
||||
-64
@@ -1,64 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils;
|
||||
|
||||
import java.awt.geom.Line2D;
|
||||
import java.awt.geom.Point2D;
|
||||
|
||||
import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Ruling;
|
||||
|
||||
import lombok.experimental.UtilityClass;
|
||||
|
||||
@UtilityClass
|
||||
public final class RulingTextDirAdjustUtil {
|
||||
|
||||
/**
|
||||
* Converts a ruling (line of a table) the same way TextPositions are converted in PDFBox.
|
||||
* This will get the y position of the text, adjusted so that 0,0 is upper left and it is adjusted based on the text direction.
|
||||
* <p>
|
||||
* See org.apache.pdfbox.text.TextPosition
|
||||
*/
|
||||
public Line2D.Float convertToDirAdj(Ruling ruling, float dir, float pageWidth, float pageHeight) {
|
||||
|
||||
return new Line2D.Float(convertPoint(ruling.x1, ruling.y1, dir, pageWidth, pageHeight), convertPoint(ruling.x2, ruling.y2, dir, pageWidth, pageHeight));
|
||||
}
|
||||
|
||||
|
||||
private Point2D convertPoint(float x, float y, float dir, float pageWidth, float pageHeight) {
|
||||
|
||||
var xAdj = getXRot(x, y, dir, pageWidth, pageHeight);
|
||||
var yLowerLeftRot = getYLowerLeftRot(x, y, dir, pageWidth, pageHeight);
|
||||
var yAdj = dir == 0 || dir == 180 ? pageHeight - yLowerLeftRot : pageWidth - yLowerLeftRot;
|
||||
return new Point2D.Float(xAdj, yAdj);
|
||||
}
|
||||
|
||||
|
||||
@SuppressWarnings("SuspiciousNameCombination")
|
||||
private float getXRot(float x, float y, float dir, float pageWidth, float pageHeight) {
|
||||
|
||||
if (dir == 0) {
|
||||
return x;
|
||||
} else if (dir == 90) {
|
||||
return y;
|
||||
} else if (dir == 180) {
|
||||
return pageWidth - x;
|
||||
} else if (dir == 270) {
|
||||
return pageHeight - y;
|
||||
}
|
||||
return 0;
|
||||
}
|
||||
|
||||
|
||||
private float getYLowerLeftRot(float x, float y, float dir, float pageWidth, float pageHeight) {
|
||||
|
||||
if (dir == 0) {
|
||||
return y;
|
||||
} else if (dir == 90) {
|
||||
return pageWidth - x;
|
||||
} else if (dir == 180) {
|
||||
return pageHeight - y;
|
||||
} else if (dir == 270) {
|
||||
return x;
|
||||
}
|
||||
return 0;
|
||||
}
|
||||
|
||||
}
|
||||
-43
@@ -1,43 +0,0 @@
|
||||
package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils;
|
||||
|
||||
import java.math.BigDecimal;
|
||||
import java.util.Comparator;
|
||||
import java.util.List;
|
||||
|
||||
import lombok.experimental.UtilityClass;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
|
||||
@Slf4j
|
||||
@UtilityClass
|
||||
@SuppressWarnings("all")
|
||||
public final class Utils {
|
||||
|
||||
private final static float EPSILON = 0.1f;
|
||||
|
||||
|
||||
public static boolean feq(double f1, double f2) {
|
||||
|
||||
return (Math.abs(f1 - f2) < EPSILON);
|
||||
}
|
||||
|
||||
|
||||
public static float round(double d, int decimalPlace) {
|
||||
|
||||
BigDecimal bd = BigDecimal.valueOf(d);
|
||||
bd = bd.setScale(decimalPlace, BigDecimal.ROUND_HALF_UP);
|
||||
return bd.floatValue();
|
||||
}
|
||||
|
||||
|
||||
public static <T> void sort(List<T> list, Comparator<? super T> comparator) {
|
||||
|
||||
try {
|
||||
QuickSort.sort(list, comparator);
|
||||
} catch (IllegalArgumentException e) {
|
||||
// This should not happen since we use QuickSort from PDFBox
|
||||
log.warn(e.getMessage());
|
||||
}
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user