From 0c3194276a0184060dfc01abaff293a6fc070e54 Mon Sep 17 00:00:00 2001
From: yhampe <yannik.hampe@knecon.com>
Date: Mon, 8 Apr 2024 11:59:35 +0200
Subject: [PATCH] RED-8402: Header and footer are not indexed / searched

added headers and footers to simplifiedtext
---
 .../processor/model/graph/nodes/Document.java | 13 +++++++
 .../SimplifiedSectionTextService.java         | 37 ++++++++++++++++---
 2 files changed, 44 insertions(+), 6 deletions(-)

diff --git a/layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/Document.java b/layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/Document.java
index e31a27c..3c71fed 100644
--- a/layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/Document.java
+++ b/layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/Document.java
@@ -63,6 +63,19 @@ public class Document implements GenericSemanticNode {
                 .collect(Collectors.toList());
     }
 
+    public List<Header> getHeaderSections() {
+
+        return streamChildrenOfType(NodeType.HEADER).map(node -> (Header) node)
+                .collect(Collectors.toList());
+    }
+
+    public List<Footer> getFooterSections() {
+
+        return streamChildrenOfType(NodeType.FOOTER).map(node -> (Footer) node)
+                .collect(Collectors.toList());
+    }
+
+
 
     public Stream<TextBlock> streamTerminalTextBlocksInOrder() {
 
diff --git a/layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/SimplifiedSectionTextService.java b/layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/SimplifiedSectionTextService.java
index eb1e824..763d20a 100644
--- a/layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/SimplifiedSectionTextService.java
+++ b/layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/SimplifiedSectionTextService.java
@@ -1,26 +1,51 @@
 package com.knecon.fforesight.service.layoutparser.processor.services;
 
 import java.util.List;
+import java.util.stream.Collectors;
+import java.util.stream.Stream;
 
 import org.springframework.stereotype.Service;
 
 import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.SimplifiedSectionText;
-import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
-import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Section;
 import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.SimplifiedText;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.GenericSemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Section;
+
+import lombok.extern.slf4j.Slf4j;
 
 @Service
+@Slf4j
 public class SimplifiedSectionTextService {
 
     public SimplifiedText toSimplifiedText(Document document) {
 
-        List<SimplifiedSectionText> simplifiedSectionTexts = document.getMainSections().stream().map(this::toSimplifiedSectionText).toList();
-        return SimplifiedText.builder().numberOfPages(document.getNumberOfPages()).sectionTexts(simplifiedSectionTexts).build();
+        List<SimplifiedSectionText> simplifiedMainSectionsList = document.getMainSections()
+                .stream()
+                .map(this::toSimplifiedSectionText)
+                .toList();
+        List<SimplifiedSectionText> simplifiedHeadersList = document.getHeaderSections()
+                .stream()
+                .map(this::toSimplifiedSectionText)
+                .toList();
+        List<SimplifiedSectionText> simplifiedFootersList = document.getFooterSections()
+                .stream()
+                .map(this::toSimplifiedSectionText)
+                .toList();
+        List<SimplifiedSectionText> simplifiedText = Stream.of(simplifiedMainSectionsList, simplifiedHeadersList, simplifiedFootersList)
+                .flatMap(List::stream)
+                .collect(Collectors.toList());
+        return SimplifiedText.builder().numberOfPages(document.getNumberOfPages()).sectionTexts(simplifiedText).build();
     }
 
 
-    private SimplifiedSectionText toSimplifiedSectionText(Section section) {
+    private SimplifiedSectionText toSimplifiedSectionText(GenericSemanticNode section) {
 
-        return SimplifiedSectionText.builder().sectionNumber(section.getTreeId().get(0)).text(section.getTextBlock().getSearchText()).build();
+        return SimplifiedSectionText.builder()
+                .sectionNumber(section.getTreeId()
+                                       .get(0))
+                .text(section.getTextBlock().getSearchText())
+                .build();
     }
+
 }