Marmelator/redaction-service
1
0
Fork 0
Code Issues Pull Requests 14 Actions Packages Projects Releases 622 Wiki Activity

Merge branch 'DM-307' into 'master'

Browse Source 
Resolve DM-307

Closes DM-307

See merge request redactmanager/redaction-service!43
This commit is contained in:
Dominique Eifländer 2023-07-11 09:35:18 +02:00
commit ff749ab88d
2 changed files with 16 additions and 21 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/service/PdfSegmentationService.java
View File

@ -14,6 +14,7 @@ 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;
@ -47,6 +48,9 @@ public class PdfSegmentationService {
private final ImageServiceResponseAdapter imageServiceResponseAdapter;
private final TableServiceResponseAdapter tableServiceResponseAdapter;
@Value("${application.type:RedactManager}")
private String applicationType;
public ClassificationDocument parseDocument(String dossierId,
String fileId,
@ -108,6 +112,11 @@ public class PdfSegmentationService {
stripper.setStartPage(pageNumber);
stripper.setEndPage(pageNumber);
stripper.setPdpage(pdPage);
if(applicationType.equals("DocuMine")){
stripper.setSortByPosition(true);
}
stripper.getText(pdDocument);
PDRectangle pdr = pdPage.getMediaBox();

28
redaction-service-v1/redaction-service-server-v1/src/test/resources/drools/documine_flora.drl
View File

@ -568,15 +568,9 @@ rule "DOC.12.1: Guideline Deviation in text"
rule "DOC.13.0: Clinical Signs"
when
FileAttribute(label == "OECD Number", value == "425")
$section: Section(
(
getHeadline().containsString("Clinical Signs")
|| getHeadline().containsString("Macroscopic Findings")
)
&& !getHeadline().containsString("TABLE")
)
$headline: Headline(containsAnyStringIgnoreCase("Clinical Signs", "Macroscopic Findings") && !containsString("TABLE"))
then
entityCreationService.bySemanticNodeParagraphsOnly($section, "clinical_signs", EntityType.ENTITY)
entityCreationService.bySemanticNodeParagraphsOnly($headline.getParent(), "clinical_signs", EntityType.ENTITY)
.forEach(entity -> entity.apply("DOC.13.0", "Clinical Signs found", "n-a"));
end
@ -718,18 +712,9 @@ rule "DOC.23.2: Bodyweight changes"
rule "DOC.23.0: Bodyweight changes"
when
FileAttribute(label == "OECD Number", value == "403")
$section: Section(
(
getHeadline().containsString("Bodyweight")
|| getHeadline().containsString("Bodyweights")
|| getHeadline().containsString("Body Weights")
|| getHeadline().containsString("Body Weight")
)
&& !getHeadline().containsStringIgnoreCase("Appendix")
&& !getHeadline().containsStringIgnoreCase("TABLE")
)
$headline: Headline(containsAnyStringIgnoreCase("Bodyweight", "Bodyweights", "Body Weights", "Body Weight"), !containsAnyStringIgnoreCase("Appendix", "TABLE"))
then
entityCreationService.bySemanticNodeParagraphsOnly($section, "bodyweight_changes", EntityType.ENTITY)
entityCreationService.bySemanticNodeParagraphsOnly($headline.getParent(), "bodyweight_changes", EntityType.ENTITY)
.forEach(entity -> entity.apply("DOC.23.0", "Bodyweight section found", "n-a"));
end
@ -968,7 +953,7 @@ rule "DOC.43.0: Dose Mortality"
when
FileAttribute(label == "OECD Number", value == "425")
$table: Table(
(hasHeader("Mortality") || hasHeader("Long Term Results") || hasHeader("Long Term Outcome") || hasHeader("Comments") || hasHeader("Viability / Mortality"))
(hasHeader("Mortality") || hasHeader("Long Term Results") || hasHeader("Long Term Outcome") || hasHeader("Comments") || hasHeader("Viability / Mortality") || hasHeader("Viability/Mortality"))
&&
(hasHeader("Dose [mg/kg bodyweight]") || hasHeader("Dose [mg/kg body weight]") ||hasHeader("Dose (mg/kg)") || hasHeader("Dose levei (mg/kg)") || hasHeader("Dose Level (mg/kg)") || hasHeader("Dose level (mg/kg)") || hasHeader("Dosage [mg/kg body weight]"))
)
@ -977,7 +962,8 @@ rule "DOC.43.0: Dose Mortality"
$table.streamTableCellsWithHeader("Comments"),
$table.streamTableCellsWithHeader("Long Term Results"),
$table.streamTableCellsWithHeader("Long Term Outcome"),
$table.streamTableCellsWithHeader("Viability / Mortality")
$table.streamTableCellsWithHeader("Viability / Mortality"),
$table.streamTableCellsWithHeader("Viability/Mortality")
).flatMap(a -> a)
.map(tableCell -> entityCreationService.bySemanticNode(tableCell, "dose_mortality", EntityType.ENTITY))
.filter(Optional::isPresent)