Merge branch 'RED-10752-main' into 'main'

RED-10752: Enabled prometheus

See merge request fforesight/layout-parser!267

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/docstrum/model/Line.java

@@ -24,7 +24,7 @@ import lombok.EqualsAndHashCode;
 @EqualsAndHashCode(onlyExplicitlyIncluded = true, callSuper = false)
 public class Line extends TextBoundingBox {
 
-    private static final double WORD_DISTANCE_MULTIPLIER = 0.18;
+    private static final double WORD_DISTANCE_MULTIPLIER = 0.17;
 
     @EqualsAndHashCode.Include
     private final double x0;
@@ -157,6 +157,9 @@ public class Line extends TextBoundingBox {
 
     private void computeWords(List<Character> characters, double wordSpacing) {
 
+        // Imo, the width of space should be scaled with the font size, but it only depends on the median distance between horizontal neighbours.
+        // If there are large differences in fontsize on a page, this might lead to missing spaces for the smaller fonts and too many for larger fonts.
+        // I've just now changed the scaling factor. If you come across this comment with missing whitespaces again, try scaling the fontsize instead of simply changing the factor again.
         Word word = new Word();
         Character previous = null;
         for (Character current : characters) {

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/classification/TableOfContentsClassificationService.java

@@ -71,6 +71,9 @@ public class TableOfContentsClassificationService {
 
     private int identifyTOCItems(int start, List<TextBlockOnPage> textBlocks, ClassificationDocument document) {
 
+        if (start >= textBlocks.size()) {
+            return start;
+        }
         ClassificationPage startPage = textBlocks.get(start).page();
         List<TextBlockOnPage> initialLookAhead = textBlocks.subList(start, Math.min(start + SURROUNDING_BLOCKS_RADIUS, textBlocks.size()));
         HashMap<NumberWord, TextBlockOnPage> numberToBlockLookup = new HashMap<>();

layoutparser-service/layoutparser-service-server/build.gradle.kts

@@ -39,6 +39,7 @@ dependencies {
     implementation("org.springframework.cloud:spring-cloud-starter-openfeign:4.0.4")
     implementation("net.logstash.logback:logstash-logback-encoder:7.4")
     implementation("ch.qos.logback:logback-classic")
+    api("com.iqser.red.commons:metric-commons:2.3.0")
 
     implementation("com.pdftron:PDFNet:10.11.0")