Merge branch 'RED-7461' into 'main'
Red 7461 See merge request fforesight/layout-parser!51
This commit is contained in:
commit
3722fff476
@ -2,15 +2,19 @@ package com.knecon.fforesight.service.layoutparser.processor;
|
||||
|
||||
import static java.lang.String.format;
|
||||
|
||||
import java.awt.geom.Rectangle2D;
|
||||
import java.io.ByteArrayOutputStream;
|
||||
import java.io.IOException;
|
||||
import java.util.ArrayList;
|
||||
import java.util.HashMap;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
|
||||
import com.knecon.fforesight.service.layoutparser.processor.utils.MarkedContentUtils;
|
||||
import org.apache.pdfbox.pdmodel.PDDocument;
|
||||
import org.apache.pdfbox.pdmodel.PDPage;
|
||||
import org.apache.pdfbox.pdmodel.common.PDRectangle;
|
||||
import org.apache.pdfbox.pdmodel.documentinterchange.markedcontent.PDMarkedContent;
|
||||
import org.springframework.stereotype.Service;
|
||||
|
||||
import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingFinishedEvent;
|
||||
@ -162,9 +166,12 @@ public class LayoutParsingPipeline {
|
||||
stripper.getMaxCharHeight());
|
||||
|
||||
ClassificationPage classificationPage = switch (layoutParsingType) {
|
||||
case REDACT_MANAGER -> redactManagerBlockificationService.blockify(stripper.getTextPositionSequences(), cleanRulings.getHorizontal(), cleanRulings.getVertical());
|
||||
case TAAS -> taasBlockificationService.blockify(stripper.getTextPositionSequences(), cleanRulings.getHorizontal(), cleanRulings.getVertical());
|
||||
case DOCUMINE -> docuMineBlockificationService.blockify(stripper.getTextPositionSequences(), cleanRulings.getHorizontal(), cleanRulings.getVertical());
|
||||
case REDACT_MANAGER ->
|
||||
redactManagerBlockificationService.blockify(stripper.getTextPositionSequences(), cleanRulings.getHorizontal(), cleanRulings.getVertical());
|
||||
case TAAS ->
|
||||
taasBlockificationService.blockify(stripper.getTextPositionSequences(), cleanRulings.getHorizontal(), cleanRulings.getVertical());
|
||||
case DOCUMINE ->
|
||||
docuMineBlockificationService.blockify(stripper.getTextPositionSequences(), cleanRulings.getHorizontal(), cleanRulings.getVertical());
|
||||
};
|
||||
classificationPage.setCleanRulings(cleanRulings);
|
||||
classificationPage.setRotation(rotation);
|
||||
@ -173,6 +180,9 @@ public class LayoutParsingPipeline {
|
||||
classificationPage.setPageWidth(cropbox.getWidth());
|
||||
classificationPage.setPageHeight(cropbox.getHeight());
|
||||
|
||||
// MarkedContent needs to be converted at this point, otherwise it leads to GC Problems in Pdfbox.
|
||||
classificationPage.setMarkedContentBboxPerType(convertMarkedContents(stripper.getMarkedContents()));
|
||||
|
||||
// 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)) {
|
||||
classificationPage.setImages(pdfImages.get(pageNumber));
|
||||
@ -201,6 +211,13 @@ public class LayoutParsingPipeline {
|
||||
}
|
||||
|
||||
|
||||
private Map<String, Rectangle2D> convertMarkedContents(List<PDMarkedContent> pdMarkedContents) {
|
||||
Map<String, Rectangle2D> markedContentBboxes = new HashMap<>();
|
||||
markedContentBboxes.put(MarkedContentUtils.HEADER, MarkedContentUtils.getMarkedContentBboxPerLine(pdMarkedContents, MarkedContentUtils.HEADER));
|
||||
markedContentBboxes.put(MarkedContentUtils.FOOTER, MarkedContentUtils.getMarkedContentBboxPerLine(pdMarkedContents, MarkedContentUtils.FOOTER));
|
||||
return markedContentBboxes;
|
||||
}
|
||||
|
||||
private void increaseDocumentStatistics(ClassificationPage classificationPage, ClassificationDocument document) {
|
||||
|
||||
if (!classificationPage.isLandscape()) {
|
||||
|
||||
@ -1,7 +1,10 @@
|
||||
package com.knecon.fforesight.service.layoutparser.processor.model;
|
||||
|
||||
import java.awt.geom.Rectangle2D;
|
||||
import java.util.ArrayList;
|
||||
import java.util.HashMap;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
|
||||
import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Rectangle;
|
||||
import com.knecon.fforesight.service.layoutparser.processor.model.image.ClassifiedImage;
|
||||
@ -11,6 +14,7 @@ import com.knecon.fforesight.service.layoutparser.processor.model.text.StringFre
|
||||
import lombok.Data;
|
||||
import lombok.NonNull;
|
||||
import lombok.RequiredArgsConstructor;
|
||||
import org.apache.pdfbox.pdmodel.documentinterchange.markedcontent.PDMarkedContent;
|
||||
|
||||
@Data
|
||||
@RequiredArgsConstructor
|
||||
@ -38,4 +42,6 @@ public class ClassificationPage {
|
||||
|
||||
CleanRulings cleanRulings;
|
||||
|
||||
private Map<String, Rectangle2D> markedContentBboxPerType = new HashMap<>();
|
||||
|
||||
}
|
||||
|
||||
@ -73,6 +73,7 @@ public class TextPageBlock extends AbstractPageBlock {
|
||||
|
||||
return sequences.get(0).getPageWidth();
|
||||
}
|
||||
|
||||
|
||||
public static TextPageBlock merge(List<TextPageBlock> textBlocksToMerge) {
|
||||
|
||||
|
||||
@ -2,6 +2,7 @@ package com.knecon.fforesight.service.layoutparser.processor.services;
|
||||
|
||||
import java.util.List;
|
||||
|
||||
import com.knecon.fforesight.service.layoutparser.processor.utils.MarkedContentUtils;
|
||||
import org.springframework.stereotype.Service;
|
||||
|
||||
import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Point;
|
||||
@ -20,7 +21,6 @@ import com.knecon.fforesight.service.layoutparser.processor.utils.PositionUtils;
|
||||
public class BodyTextFrameService {
|
||||
|
||||
|
||||
|
||||
public void setBodyTextFrames(ClassificationDocument classificationDocument, LayoutParsingType layoutParsingType) {
|
||||
|
||||
Rectangle bodyTextFrame = calculateBodyTextFrame(classificationDocument.getPages(), classificationDocument.getFontSizeCounter(), false, layoutParsingType);
|
||||
@ -157,6 +157,11 @@ public class BodyTextFrameService {
|
||||
continue;
|
||||
}
|
||||
|
||||
if (MarkedContentUtils.intersects(textBlock, page.getMarkedContentBboxPerType(), MarkedContentUtils.HEADER)
|
||||
|| MarkedContentUtils.intersects(textBlock, page.getMarkedContentBboxPerType(), MarkedContentUtils.FOOTER)) {
|
||||
continue;
|
||||
}
|
||||
|
||||
float approxLineCount = PositionUtils.getApproxLineCount(textBlock);
|
||||
if (layoutParsingType.equals(LayoutParsingType.DOCUMINE) && approxLineCount < approximateHeaderLineCount && textBlock.getMaxY() >= page.getPageHeight() - (page.getPageHeight() / 10) || !layoutParsingType.equals(
|
||||
LayoutParsingType.DOCUMINE) && approxLineCount < approximateHeaderLineCount) {
|
||||
@ -233,4 +238,4 @@ public class BodyTextFrameService {
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
}
|
||||
@ -5,6 +5,7 @@ import java.util.Locale;
|
||||
import java.util.regex.Matcher;
|
||||
import java.util.regex.Pattern;
|
||||
|
||||
import com.knecon.fforesight.service.layoutparser.processor.utils.MarkedContentUtils;
|
||||
import org.springframework.stereotype.Service;
|
||||
|
||||
import com.knecon.fforesight.service.layoutparser.processor.model.AbstractPageBlock;
|
||||
@ -62,12 +63,16 @@ public class DocuMineClassificationService {
|
||||
textBlock.setClassification(PageBlockType.OTHER);
|
||||
return;
|
||||
}
|
||||
if (PositionUtils.isOverBodyTextFrame(bodyTextFrame, textBlock, page.getRotation()) && (document.getFontSizeCounter()
|
||||
.getMostPopular() == null || textBlock.getHighestFontSize() <= document.getFontSizeCounter().getMostPopular())) {
|
||||
if (MarkedContentUtils.intersects(textBlock, page.getMarkedContentBboxPerType(), MarkedContentUtils.HEADER)
|
||||
|| 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())) {
|
||||
} else if (MarkedContentUtils.intersects(textBlock, page.getMarkedContentBboxPerType(), MarkedContentUtils.FOOTER)
|
||||
|| 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()
|
||||
@ -107,4 +112,4 @@ public class DocuMineClassificationService {
|
||||
}
|
||||
}
|
||||
|
||||
}
|
||||
}
|
||||
@ -3,6 +3,7 @@ package com.knecon.fforesight.service.layoutparser.processor.services.classifica
|
||||
import java.util.List;
|
||||
import java.util.regex.Pattern;
|
||||
|
||||
import com.knecon.fforesight.service.layoutparser.processor.utils.MarkedContentUtils;
|
||||
import org.springframework.stereotype.Service;
|
||||
|
||||
import com.knecon.fforesight.service.layoutparser.processor.model.AbstractPageBlock;
|
||||
@ -51,11 +52,13 @@ public class RedactManagerClassificationService {
|
||||
textBlock.setClassification(PageBlockType.OTHER);
|
||||
return;
|
||||
}
|
||||
if (PositionUtils.isOverBodyTextFrame(bodyTextFrame, textBlock, page.getRotation()) && (document.getFontSizeCounter()
|
||||
if (MarkedContentUtils.intersects(textBlock, page.getMarkedContentBboxPerType(), MarkedContentUtils.HEADER)
|
||||
|| 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()
|
||||
} else if (MarkedContentUtils.intersects(textBlock, page.getMarkedContentBboxPerType(), MarkedContentUtils.FOOTER)
|
||||
|| 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,
|
||||
|
||||
@ -3,6 +3,7 @@ package com.knecon.fforesight.service.layoutparser.processor.services.classifica
|
||||
import java.util.List;
|
||||
import java.util.regex.Pattern;
|
||||
|
||||
import com.knecon.fforesight.service.layoutparser.processor.utils.MarkedContentUtils;
|
||||
import org.springframework.stereotype.Service;
|
||||
|
||||
import com.knecon.fforesight.service.layoutparser.processor.model.AbstractPageBlock;
|
||||
@ -56,9 +57,11 @@ public class TaasClassificationService {
|
||||
textBlock.setClassification(PageBlockType.OTHER);
|
||||
return;
|
||||
}
|
||||
if (PositionUtils.isOverBodyTextFrame(bodyTextFrame, textBlock, page.getRotation())) {
|
||||
if (MarkedContentUtils.intersects(textBlock, page.getMarkedContentBboxPerType(), MarkedContentUtils.HEADER)
|
||||
|| PositionUtils.isOverBodyTextFrame(bodyTextFrame, textBlock, page.getRotation())) {
|
||||
textBlock.setClassification(PageBlockType.HEADER);
|
||||
} else if (PositionUtils.isUnderBodyTextFrame(bodyTextFrame, textBlock, page.getRotation())) {
|
||||
} else if (MarkedContentUtils.intersects(textBlock, page.getMarkedContentBboxPerType(), MarkedContentUtils.FOOTER)
|
||||
|| PositionUtils.isUnderBodyTextFrame(bodyTextFrame, textBlock, page.getRotation())) {
|
||||
textBlock.setClassification(PageBlockType.FOOTER);
|
||||
} else if (page.getPageNumber() == 1 && (PositionUtils.getHeightDifferenceBetweenChunkWordAndDocumentWord(textBlock,
|
||||
document.getTextHeightCounter().getMostPopular()) > 2.5 && textBlock.getHighestFontSize() > document.getFontSizeCounter().getMostPopular() || page.getTextBlocks()
|
||||
|
||||
@ -1,33 +1,18 @@
|
||||
package com.knecon.fforesight.service.layoutparser.processor.services.parsing;
|
||||
|
||||
import java.awt.color.CMMException;
|
||||
import java.awt.geom.Point2D;
|
||||
import java.io.IOException;
|
||||
import java.util.ArrayList;
|
||||
import java.util.Arrays;
|
||||
import java.util.List;
|
||||
|
||||
import com.knecon.fforesight.service.layoutparser.processor.model.table.Ruling;
|
||||
import com.knecon.fforesight.service.layoutparser.processor.model.text.RedTextPosition;
|
||||
import com.knecon.fforesight.service.layoutparser.processor.model.text.TextPositionSequence;
|
||||
import lombok.Getter;
|
||||
import lombok.Setter;
|
||||
import lombok.SneakyThrows;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
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.color.*;
|
||||
import org.apache.pdfbox.contentstream.operator.markedcontent.BeginMarkedContentSequenceWithProperties;
|
||||
import org.apache.pdfbox.contentstream.operator.markedcontent.EndMarkedContentSequence;
|
||||
import org.apache.pdfbox.contentstream.operator.state.*;
|
||||
import org.apache.pdfbox.contentstream.operator.text.SetFontAndSize;
|
||||
import org.apache.pdfbox.cos.COSBase;
|
||||
import org.apache.pdfbox.cos.COSNumber;
|
||||
@ -36,14 +21,11 @@ import org.apache.pdfbox.pdmodel.PDPage;
|
||||
import org.apache.pdfbox.pdmodel.graphics.color.PDColor;
|
||||
import org.apache.pdfbox.text.TextPosition;
|
||||
|
||||
import com.knecon.fforesight.service.layoutparser.processor.model.table.Ruling;
|
||||
import com.knecon.fforesight.service.layoutparser.processor.model.text.RedTextPosition;
|
||||
import com.knecon.fforesight.service.layoutparser.processor.model.text.TextPositionSequence;
|
||||
|
||||
import lombok.Getter;
|
||||
import lombok.Setter;
|
||||
import lombok.SneakyThrows;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
import java.awt.color.CMMException;
|
||||
import java.awt.geom.Point2D;
|
||||
import java.io.IOException;
|
||||
import java.util.ArrayList;
|
||||
import java.util.List;
|
||||
|
||||
@Getter
|
||||
@Slf4j
|
||||
@ -59,6 +41,7 @@ public class PDFLinesTextStripper extends PDFTextStripper {
|
||||
private int minCharHeight;
|
||||
private int maxCharHeight;
|
||||
|
||||
|
||||
private float path_x;
|
||||
private float path_y;
|
||||
|
||||
@ -89,6 +72,12 @@ public class PDFLinesTextStripper extends PDFTextStripper {
|
||||
this.addOperator(new SetNonStrokingColorN());
|
||||
this.addOperator(new SetFontAndSize());
|
||||
this.addOperator(new SetLineWidth());
|
||||
|
||||
|
||||
addOperator(new BeginMarkedContentSequenceWithProperties());
|
||||
// addOperator(new BeginMarkedContentSequence());
|
||||
addOperator(new EndMarkedContentSequence());
|
||||
|
||||
}
|
||||
|
||||
|
||||
@ -99,6 +88,7 @@ public class PDFLinesTextStripper extends PDFTextStripper {
|
||||
|
||||
//move
|
||||
switch (operation) {
|
||||
|
||||
case OperatorName.MOVE_TO:
|
||||
if (arguments.size() == 2) {
|
||||
Point2D.Float pos = transformPosition(floatValue(arguments.get(0)), floatValue(arguments.get(1)));
|
||||
@ -349,3 +339,4 @@ public class PDFLinesTextStripper extends PDFTextStripper {
|
||||
|
||||
}
|
||||
|
||||
|
||||
|
||||
@ -25,27 +25,20 @@ import java.io.StringWriter;
|
||||
import java.io.Writer;
|
||||
import java.text.Bidi;
|
||||
import java.text.Normalizer;
|
||||
import java.util.ArrayList;
|
||||
import java.util.Collections;
|
||||
import java.util.Comparator;
|
||||
import java.util.HashMap;
|
||||
import java.util.Iterator;
|
||||
import java.util.LinkedList;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
import java.util.SortedMap;
|
||||
import java.util.SortedSet;
|
||||
import java.util.StringTokenizer;
|
||||
import java.util.TreeMap;
|
||||
import java.util.TreeSet;
|
||||
import java.util.*;
|
||||
import java.util.regex.Pattern;
|
||||
|
||||
import lombok.Getter;
|
||||
import org.apache.commons.logging.Log;
|
||||
import org.apache.commons.logging.LogFactory;
|
||||
import org.apache.pdfbox.cos.COSDictionary;
|
||||
import org.apache.pdfbox.cos.COSName;
|
||||
import org.apache.pdfbox.pdmodel.PDDocument;
|
||||
import org.apache.pdfbox.pdmodel.PDPage;
|
||||
import org.apache.pdfbox.pdmodel.PDPageTree;
|
||||
import org.apache.pdfbox.pdmodel.common.PDRectangle;
|
||||
import org.apache.pdfbox.pdmodel.documentinterchange.markedcontent.PDMarkedContent;
|
||||
import org.apache.pdfbox.pdmodel.graphics.PDXObject;
|
||||
import org.apache.pdfbox.pdmodel.interactive.documentnavigation.outline.PDOutlineItem;
|
||||
import org.apache.pdfbox.pdmodel.interactive.pagenavigation.PDThreadBead;
|
||||
import org.apache.pdfbox.text.TextPosition;
|
||||
@ -63,6 +56,10 @@ public class PDFTextStripper extends LegacyPDFStreamEngine {
|
||||
private static float defaultDropThreshold = 2.5f;
|
||||
private static final boolean useCustomQuickSort;
|
||||
|
||||
@Getter
|
||||
protected final List<PDMarkedContent> markedContents = new ArrayList<>();
|
||||
protected final Deque<PDMarkedContent> currentMarkedContents = new ArrayDeque<>();
|
||||
|
||||
private static final Log LOG = LogFactory.getLog(PDFTextStripper.class);
|
||||
|
||||
// enable the ability to set the default indent/drop thresholds
|
||||
@ -196,6 +193,45 @@ public class PDFTextStripper extends LegacyPDFStreamEngine {
|
||||
}
|
||||
|
||||
|
||||
|
||||
public void beginMarkedContentSequence(COSName tag, COSDictionary properties)
|
||||
{
|
||||
PDMarkedContent markedContent = PDMarkedContent.create(tag, properties);
|
||||
if (this.currentMarkedContents.isEmpty())
|
||||
{
|
||||
this.markedContents.add(markedContent);
|
||||
}
|
||||
else
|
||||
{
|
||||
PDMarkedContent currentMarkedContent =
|
||||
this.currentMarkedContents.peek();
|
||||
if (currentMarkedContent != null)
|
||||
{
|
||||
currentMarkedContent.addMarkedContent(markedContent);
|
||||
}
|
||||
}
|
||||
this.currentMarkedContents.push(markedContent);
|
||||
}
|
||||
|
||||
@Override
|
||||
public void endMarkedContentSequence()
|
||||
{
|
||||
if (!this.currentMarkedContents.isEmpty())
|
||||
{
|
||||
this.currentMarkedContents.pop();
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
public void xobject(PDXObject xobject)
|
||||
{
|
||||
if (!this.currentMarkedContents.isEmpty())
|
||||
{
|
||||
this.currentMarkedContents.peek().addXObject(xobject);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* This will return the text of a document. See writeText. <br>
|
||||
* NOTE: The document must not be encrypted when coming into this method.
|
||||
@ -877,7 +913,12 @@ public class PDFTextStripper extends LegacyPDFStreamEngine {
|
||||
textList.add(text);
|
||||
}
|
||||
}
|
||||
if (!this.currentMarkedContents.isEmpty())
|
||||
{
|
||||
this.currentMarkedContents.peek().addText(text);
|
||||
}
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
|
||||
@ -2103,6 +2144,7 @@ public class PDFTextStripper extends LegacyPDFStreamEngine {
|
||||
this.isHangingIndent = true;
|
||||
}
|
||||
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
@ -0,0 +1,55 @@
|
||||
package com.knecon.fforesight.service.layoutparser.processor.utils;
|
||||
|
||||
import com.knecon.fforesight.service.layoutparser.processor.model.text.TextPageBlock;
|
||||
import com.knecon.fforesight.service.layoutparser.processor.model.text.TextPositionSequence;
|
||||
import lombok.experimental.UtilityClass;
|
||||
import org.apache.pdfbox.cos.COSName;
|
||||
import org.apache.pdfbox.pdmodel.documentinterchange.markedcontent.PDMarkedContent;
|
||||
import org.apache.pdfbox.text.TextPosition;
|
||||
|
||||
import java.awt.geom.Rectangle2D;
|
||||
import java.util.Collection;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
import java.util.stream.Collectors;
|
||||
|
||||
@UtilityClass
|
||||
public class MarkedContentUtils {
|
||||
|
||||
public static final String HEADER = "Header";
|
||||
public static final String FOOTER = "Footer";
|
||||
|
||||
public Rectangle2D getMarkedContentBboxPerLine(List<PDMarkedContent> markedContents, String subtype) {
|
||||
|
||||
if (markedContents == null) {
|
||||
return null;
|
||||
}
|
||||
|
||||
var markedContentByYPosition = markedContents.stream()
|
||||
.filter(m -> m.getTag().equals("Artifact"))
|
||||
.filter(m -> m.getProperties() != null)
|
||||
.filter(m -> m.getProperties().getItem("Subtype") != null)
|
||||
.filter(m -> ((COSName) m.getProperties().getItem("Subtype")).getName().equals(subtype))
|
||||
.map(PDMarkedContent::getContents).flatMap(Collection::stream)
|
||||
.filter(t -> t instanceof TextPosition)
|
||||
.map(t -> (TextPosition) t)
|
||||
.filter(t -> !t.getUnicode().equals(" "))
|
||||
.collect(Collectors.groupingBy(TextPosition::getY));
|
||||
|
||||
if (markedContentByYPosition.isEmpty()) {
|
||||
return null;
|
||||
}
|
||||
|
||||
return RectangleTransformations.rectangle2DBBox(markedContentByYPosition.values().stream()
|
||||
.map(textPositions -> new TextPositionSequence(textPositions.stream()
|
||||
.toList(), 0, true)
|
||||
.getRectangle())
|
||||
.map(t -> new Rectangle2D.Float(t.getTopLeft().getX(), t.getTopLeft().getY() - Math.abs(t.getHeight()), t.getWidth(), Math.abs(t.getHeight()))).collect(Collectors.toList()));
|
||||
}
|
||||
|
||||
|
||||
public boolean intersects(TextPageBlock textBlock, Map<String, Rectangle2D> markedContentBboxPerType, String type) {
|
||||
return markedContentBboxPerType.get(type) != null && markedContentBboxPerType.get(type).intersects(textBlock.getPdfMinX(), textBlock.getPdfMinY(), textBlock.getWidth(), textBlock.getHeight());
|
||||
}
|
||||
|
||||
}
|
||||
@ -26,7 +26,7 @@ public class DocumentGraphVisualizationTest extends BuildDocumentGraphTest {
|
||||
// @Disabled
|
||||
public void visualizeMetolachlor() {
|
||||
|
||||
String filename = "files/S-Metolachlor_RAR_01_Volume_1_2018-09-06.pdf";
|
||||
String filename = "files/SOLICITA_VICTRATO-GOLD-II_Item 21_Mutacao_Genica (4).pdf";
|
||||
visualizePdf(filename);
|
||||
}
|
||||
|
||||
|
||||
@ -4,6 +4,7 @@ import java.io.FileOutputStream;
|
||||
import java.nio.file.Path;
|
||||
|
||||
import org.apache.pdfbox.Loader;
|
||||
import org.junit.jupiter.api.Disabled;
|
||||
import org.junit.jupiter.api.Test;
|
||||
import org.springframework.core.io.ClassPathResource;
|
||||
|
||||
@ -17,13 +18,14 @@ import lombok.SneakyThrows;
|
||||
public class ViewerDocumentTest extends BuildDocumentGraphTest {
|
||||
|
||||
@Test
|
||||
@Disabled
|
||||
@SneakyThrows
|
||||
public void testViewerDocument() {
|
||||
|
||||
LayoutGridService layoutGridService = new LayoutGridService();
|
||||
ViewerDocumentService viewerDocumentService = new ViewerDocumentService(layoutGridService);
|
||||
String fileName = "files/S-Metolachlor_RAR_01_Volume_1_2018-09-06.pdf";
|
||||
Document document = buildGraph(fileName, LayoutParsingType.REDACT_MANAGER);
|
||||
String fileName = "files/new/VV-511309_OCR.pdf";
|
||||
Document document = buildGraph(fileName, LayoutParsingType.DOCUMINE);
|
||||
String tmpFileName = "/tmp/" + Path.of(fileName).getFileName() + "_VIEWER.pdf";
|
||||
try (var pdDocument = Loader.loadPDF(new ClassPathResource(fileName).getInputStream()); var out = new FileOutputStream(tmpFileName)) {
|
||||
viewerDocumentService.createViewerDocument(pdDocument, document, out);
|
||||
|
||||
@ -1,24 +1,5 @@
|
||||
package com.knecon.fforesight.service.layoutparser.server.segmentation;
|
||||
|
||||
import static org.assertj.core.api.Assertions.assertThat;
|
||||
|
||||
import java.awt.geom.Rectangle2D;
|
||||
import java.io.IOException;
|
||||
import java.util.ArrayList;
|
||||
import java.util.Arrays;
|
||||
import java.util.Collections;
|
||||
import java.util.HashMap;
|
||||
import java.util.List;
|
||||
import java.util.Locale;
|
||||
import java.util.Map;
|
||||
import java.util.stream.Collectors;
|
||||
|
||||
import org.apache.pdfbox.Loader;
|
||||
import org.apache.pdfbox.pdmodel.PDDocument;
|
||||
import org.junit.jupiter.api.Test;
|
||||
import org.springframework.beans.factory.annotation.Autowired;
|
||||
import org.springframework.core.io.ClassPathResource;
|
||||
|
||||
import com.fasterxml.jackson.databind.ObjectMapper;
|
||||
import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingType;
|
||||
import com.knecon.fforesight.service.layoutparser.processor.LayoutParsingPipeline;
|
||||
@ -34,8 +15,19 @@ import com.knecon.fforesight.service.layoutparser.processor.python_api.model.tab
|
||||
import com.knecon.fforesight.service.layoutparser.processor.services.SectionsBuilderService;
|
||||
import com.knecon.fforesight.service.layoutparser.processor.services.classification.RedactManagerClassificationService;
|
||||
import com.knecon.fforesight.service.layoutparser.server.utils.AbstractTest;
|
||||
|
||||
import lombok.SneakyThrows;
|
||||
import org.apache.pdfbox.Loader;
|
||||
import org.apache.pdfbox.pdmodel.PDDocument;
|
||||
import org.junit.jupiter.api.Test;
|
||||
import org.springframework.beans.factory.annotation.Autowired;
|
||||
import org.springframework.core.io.ClassPathResource;
|
||||
|
||||
import java.awt.geom.Rectangle2D;
|
||||
import java.io.IOException;
|
||||
import java.util.*;
|
||||
import java.util.stream.Collectors;
|
||||
|
||||
import static org.assertj.core.api.Assertions.assertThat;
|
||||
|
||||
public class PdfSegmentationServiceTest extends AbstractTest {
|
||||
|
||||
@ -60,7 +52,7 @@ public class PdfSegmentationServiceTest extends AbstractTest {
|
||||
|
||||
public ClassificationDocument buildClassificationDocument(PDDocument originDocument) {
|
||||
|
||||
ClassificationDocument classificationDocument = layoutParsingPipeline.parseLayout(LayoutParsingType.REDACT_MANAGER,
|
||||
ClassificationDocument classificationDocument = layoutParsingPipeline.parseLayout(LayoutParsingType.DOCUMINE,
|
||||
originDocument,
|
||||
new ImageServiceResponse(),
|
||||
new TableServiceResponse());
|
||||
|
||||
@ -17,11 +17,10 @@ import lombok.SneakyThrows;
|
||||
class PageContentExtractorTest {
|
||||
|
||||
@Test
|
||||
@Disabled
|
||||
@SneakyThrows
|
||||
public void testTextPositionSequenceExtraction() {
|
||||
|
||||
String fileName = "files/invisible_tables/test-two-pages_ocred.pdf";
|
||||
String fileName = "files/SOLICITA_VICTRATO-GOLD-II_Item 21_Mutacao_Genica (4).pdf";
|
||||
var tmpFileName = Path.of("/tmp/").resolve(Path.of(fileName).getFileName() + "_TEXT_POSITION_SEQUENCES.pdf").toString();
|
||||
|
||||
List<PageContents> textPositionPerPage = PageContentExtractor.getSortedPageContents(fileName);
|
||||
|
||||
Binary file not shown.
Binary file not shown.
Loading…
x
Reference in New Issue
Block a user