Index: pdfbox/src/main/java/org/apache/pdfbox/ExtractText.java =================================================================== --- pdfbox/src/main/java/org/apache/pdfbox/ExtractText.java (revision 1027422) +++ pdfbox/src/main/java/org/apache/pdfbox/ExtractText.java (working copy) @@ -28,6 +28,7 @@ import org.apache.pdfbox.pdmodel.encryption.AccessPermission; import org.apache.pdfbox.pdmodel.encryption.StandardDecryptionMaterial; import org.apache.pdfbox.util.PDFText2HTML; +import org.apache.pdfbox.util.PDFText2XML; import org.apache.pdfbox.util.PDFTextStripper; /** @@ -47,6 +48,7 @@ private static final String SORT = "-sort"; private static final String IGNORE_BEADS = "-ignoreBeads"; private static final String HTML = "-html"; // jjb - added simple HTML output + private static final String XML = "-xml"; // drodrigguez - added XML output private static final String FORCE = "-force"; //enables pdfbox to skip corrupt objects /** @@ -68,6 +70,7 @@ { boolean toConsole = false; boolean toHTML = false; + boolean toXML = false; boolean force = false; boolean sort = false; boolean separateBeads = true; @@ -113,6 +116,11 @@ toHTML = true; ext = ".html"; } + else if(args[i].equals(XML)) { + toXML = true; + ext = ".xml"; + } + else if( args[i].equals( SORT ) ) { sort = true; @@ -227,7 +235,11 @@ } else { - stripper = new PDFTextStripper(encoding); + if(toXML) { + stripper = new PDFText2XML(encoding); + } else { + stripper = new PDFTextStripper(encoding); + } } stripper.setForceParsing( force ); stripper.setSortByPosition( sort ); @@ -260,6 +272,7 @@ " -encoding (ISO-8859-1,UTF-16BE,UTF-16LE,...)\n" + " -console Send text to console instead of file\n" + " -html Output in HTML format instead of raw text\n" + + " -xml Output in XML format instead of raw text\n" + " -sort Sort the text before writing\n" + " -ignoreBeads Disables the separation by beads\n" + " -force Enables pdfbox to ignore corrupt objects\n" + Index: pdfbox/src/main/java/org/apache/pdfbox/util/Matrix.java =================================================================== --- pdfbox/src/main/java/org/apache/pdfbox/util/Matrix.java (revision 1027422) +++ pdfbox/src/main/java/org/apache/pdfbox/util/Matrix.java (working copy) @@ -295,11 +295,26 @@ * sqrt(x2) = * abs(x) */ + + /** + * DR: The previous din't take int an account that + * a rotation the θ angle could be 180º so + * single[1] and single[3] would be zero. + * + * If single[1] and single[3] are equal to zero we have to check + * against single[0] sign. + */ if( !(single[1]==0.0f && single[3]==0.0f) ) { xScale = (float)Math.sqrt(Math.pow(single[0], 2)+ Math.pow(single[1], 2)); } + + if( single[1]==0.0f && single[3]==0.0f && single[0] < 0.0f ) + { + xScale = -xScale; + } + return xScale; } @@ -315,10 +330,55 @@ yScale = (float)Math.sqrt(Math.pow(single[3], 2)+ Math.pow(single[4], 2)); } + + if( single[1]==0.0f && single[3]==0.0f && single[0] < 0.0f) + { + yScale = -yScale; + } return yScale; } /** + * Get the rotation angle of this matrix + * @return The rotation angle of this matrix + */ + public float getRotation() + { + + // We need to know if there could be a rotation wether not. + boolean rotation = !(single[1]==0.0f && single[3]==0.0f); + rotation = rotation | (!rotation && single[0] < 0.0f); + + if(rotation) + { + // Since there is rotation, we need to compute the angle + if(single[0] < 0.0f) return (float)(Math.PI / 2); + + // We obtain xScale and yScale. + float xScale = (float)Math.sqrt(Math.pow(single[0], 2)+ + Math.pow(single[1], 2)); + + float yScale = (float)Math.sqrt(Math.pow(single[3], 2)+ + Math.pow(single[4], 2)); + + if(xScale != 0) { + if(single[0] != 0.0f) + return (float)Math.acos(single[0]/xScale); + return (float)Math.asin(single[1]/xScale); + + } + + if(yScale != 0) { + if(single[0] != 0.0f) + return (float)Math.asin(single[3]/yScale); + return (float)Math.asin(single[4]/yScale); + + } + } + return 0.0f; + } + + /** * Get the x position in the matrix. * @return The x-position. */ Index: pdfbox/src/main/java/org/apache/pdfbox/util/PDFMarkedContentExtractor.java =================================================================== --- pdfbox/src/main/java/org/apache/pdfbox/util/PDFMarkedContentExtractor.java (revision 1027422) +++ pdfbox/src/main/java/org/apache/pdfbox/util/PDFMarkedContentExtractor.java (working copy) @@ -168,8 +168,8 @@ { showCharacter = false; String textCharacter = text.getCharacter(); - float textX = text.getX(); - float textY = text.getY(); + float textX = text.getXDirAdj(); + float textY = text.getYDirAdj(); List sameTextCharacters = this.characterListMapping.get( textCharacter ); if( sameTextCharacters == null ) { @@ -194,12 +194,12 @@ { TextPosition character = (TextPosition)sameTextCharacters.get( i ); String charCharacter = character.getCharacter(); - float charX = character.getX(); - float charY = character.getY(); + float charX = character.getXDirAdj(); + float charY = character.getYDirAdj(); //only want to suppress if( charCharacter != null && - //charCharacter.equals( textCharacter ) && + Math.abs(text.getDir()-character.getDir()) < 0.3/Math.PI && within( charX, textX, tolerance ) && within( charY, textY, Index: pdfbox/src/main/java/org/apache/pdfbox/util/PDFText2XML.java =================================================================== --- pdfbox/src/main/java/org/apache/pdfbox/util/PDFText2XML.java (revision 0) +++ pdfbox/src/main/java/org/apache/pdfbox/util/PDFText2XML.java (revision 0) @@ -0,0 +1,163 @@ +/* + * 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 org.apache.pdfbox.util; + +import java.io.IOException; + +import org.apache.pdfbox.pdmodel.PDDocument; + + +/** + * Wrap stripped text in a XML way, it just treat it words in an indepedent + * way in order to show word value, word position, line and page. It also + * incudes PDF metadata information. + * + * @author David Rodriguez - http://www.divisait.com - + * @version $Revision$ + * @since PDFBOX 1.4 + */ +public class PDFText2XML extends PDFTextStripper +{ + /** + * Initial builder size + */ + private static final int INITIAL_PDF_TO_XML_BYTES = 8192; + + /** + * Constructor. + * @throws IOException If there is an error during initialization. + */ + public PDFText2XML(String encoding) throws IOException + { + super((encoding==null)?"UTF-8":encoding); + setLineSeparator(systemLineSeparator); + setParagraphStart(""+systemLineSeparator); + setParagraphEnd(""+systemLineSeparator); + setPageStart("

"); + setPageEnd("

"+systemLineSeparator); + setArticleStart(""+systemLineSeparator); + setArticleEnd(""+systemLineSeparator); + setArticleStart(systemLineSeparator); + setArticleEnd(systemLineSeparator); + } + + /** + * {@inheritDoc} + */ + public void endDocument(PDDocument pdf) throws IOException + { + output.write(""); + } + + @Override + protected void writeWord(Word word, + String normalizedWord) + throws IOException { + StringBuilder sb = new StringBuilder(); + boolean rtl = false; + rtl = !normalizedWord.equals(word.getWord()); + sb.append(""); + + sb.append(escapeXML(normalizedWord)); + sb.append("").append(systemLineSeparator); + output.write(sb.toString()); + } + + @Override + protected void startDocument(PDDocument pdf) throws IOException { + super.startDocument(pdf); + StringBuilder buf = new StringBuilder(INITIAL_PDF_TO_XML_BYTES); + buf.append("").append(systemLineSeparator); + buf.append("").append(systemLineSeparator); + String title = document.getDocumentInformation().getTitle(); + String author = document.getDocumentInformation().getAuthor(); + String creator = document.getDocumentInformation().getCreator(); + String keywords = document.getDocumentInformation().getKeywords(); + String subject = document.getDocumentInformation().getSubject(); + + if(title!=null || author!=null || creator!=null || keywords!=null || subject!=null) { + buf.append("").append(systemLineSeparator); + + if(title!=null) { + buf.append("").append(escapeXML(title)).append("").append(systemLineSeparator); + } + if(author!=null) { + buf.append("").append(escapeXML(author)).append("").append(systemLineSeparator); + } + if(creator!=null) { + buf.append("").append(escapeXML(creator)).append("").append(systemLineSeparator); + } + if(keywords!=null) { + buf.append("").append(escapeXML(keywords)).append("").append(systemLineSeparator); + } + if(subject!=null) { + buf.append("").append(escapeXML(subject)).append("").append(systemLineSeparator); + } + buf.append("").append(systemLineSeparator); + } + + output.write(buf.toString()); + } + + /** + * Escape special XML Characters + * + * @param chars String to be written to the stream + * @throws IOException + * If there is an error writing to the stream. + */ + protected String escapeXML(String chars) + { + StringBuilder output = new StringBuilder(chars.length()); + + for (int i = 0; i < chars.length(); i++) + { + char c = chars.charAt(i); + // write non-ASCII as named entities + if (c < 32) + { + int charAsInt = c; + output.append("&#" + charAsInt + ";"); + } + else + { + switch (c) + { + case 34: + output.append("""); + break; + case 38: + output.append("&"); + break; + case 60: + output.append("<"); + break; + case 62: + output.append(">"); + break; + default: + output.append(c); + } + } + } + return output.toString(); + } +} Property changes on: pdfbox\src\main\java\org\apache\pdfbox\util\PDFText2XML.java ___________________________________________________________________ Added: svn:eol-style + native Index: pdfbox/src/main/java/org/apache/pdfbox/util/PDFTextStripper.java =================================================================== --- pdfbox/src/main/java/org/apache/pdfbox/util/PDFTextStripper.java (revision 1027422) +++ pdfbox/src/main/java/org/apache/pdfbox/util/PDFTextStripper.java (working copy) @@ -631,28 +631,16 @@ previousAveCharWidth = -1; } - float positionX; - float positionY; - float positionWidth; - float positionHeight; + /** + * DR: We always use adjusted coordinates, we don't really mind + * page orientation, but text orientation. + */ + float positionX = position.getXDirAdj(); + float positionY = position.getYDirAdj();; + float positionWidth = position.getWidthDirAdj(); + float positionHeight = position.getHeightDir(); + float positionTheta = position.getDir(); - /* If we are sorting, then we need to use the text direction - * adjusted coordinates, because they were used in the sorting. */ - if (getSortByPosition()) - { - positionX = position.getXDirAdj(); - positionY = position.getYDirAdj(); - positionWidth = position.getWidthDirAdj(); - positionHeight = position.getHeightDir(); - } - else - { - positionX = position.getX(); - positionY = position.getY(); - positionWidth = position.getWidth(); - positionHeight = position.getHeight(); - } - //The current amount of characters in a word int wordCharCount = position.getIndividualWidths().length; @@ -722,9 +710,15 @@ /* XXX BC: In theory, this check should really check if the next char is in full range * seen in this line. This is what I tried to do with minYTopForLine, but this caused a lot * of regression test failures. So, I'm leaving it be for now. */ - if(!overlap(positionY, positionHeight, maxYForLine, maxHeightForLine)) + + /* DR: We check if theta is the same, since if the rotation angle is + * different, we have a different line, we allow a little angle + * deviation (0.3 degree) + */ + if(!overlap(positionY, positionHeight, maxYForLine, maxHeightForLine) || + Math.abs(positionTheta-lastPosition.getTextPosition().getDir()) > 0.3/Math.PI) { - writeLine(normalize(line,isRtlDominant,hasRtl),isRtlDominant); + writeLine(normalize(line),isRtlDominant,hasRtl); line.clear(); lastLineStartPosition = handleLineSeparation(current, lastPosition, lastLineStartPosition); @@ -779,7 +773,7 @@ // print the final line if (line.size() > 0) { - writeLine(normalize(line,isRtlDominant,hasRtl),isRtlDominant); + writeLine(normalize(line),isRtlDominant,hasRtl); writeParagraphEnd(); } @@ -828,6 +822,21 @@ { output.write(getWordSeparator()); } + + /** + * Write the word to the output stream + * + * @param word The original word to write (with its position, etc.) + * @param normalizedWord Normalized word - string - + * + * @throws IOException + * If there is a problem writing out the wordseparator to the document. + */ + protected void writeWord(Word word, String normalizedWord) + throws IOException + { + output.write(normalizedWord); + } /** * Write the string in TextPosition to the output stream. @@ -877,8 +886,8 @@ { showCharacter = false; String textCharacter = text.getCharacter(); - float textX = text.getX(); - float textY = text.getY(); + float textX = text.getXDirAdj(); + float textY = text.getYDirAdj(); List sameTextCharacters = (List)characterListMapping.get( textCharacter ); if( sameTextCharacters == null ) { @@ -903,11 +912,12 @@ { TextPosition character = sameTextCharacters.get( i ); String charCharacter = character.getCharacter(); - float charX = character.getX(); - float charY = character.getY(); + float charX = character.getXDirAdj(); + float charY = character.getYDirAdj(); //only want to suppress if( charCharacter != null && + Math.abs(text.getDir()-character.getDir()) < 0.3/Math.PI && //charCharacter.equals( textCharacter ) && within( charX, textX, tolerance ) && within( charY, @@ -1803,21 +1813,38 @@ * Write a list of string containing a whole line of a document. * @param line a list with the words of the given line * @param isRtlDominant determines if rtl or ltl is dominant + * @param hasRtl determines if lines contains rtl formatted text(parts) * @throws IOException if something went wrong */ - private void writeLine(List line, boolean isRtlDominant)throws IOException{ + private void writeLine(List line, boolean isRtlDominant, boolean hasRtl) + throws IOException{ + int numberOfStrings = line.size(); if (isRtlDominant) { for(int i=numberOfStrings-1; i>=0; i--){ if (i > 1) writeWordSeparator(); - writeString(line.get(i)); + + String word = line.get(i).getWord(); + if (hasRtl) { + word = normalize.makeLineLogicalOrder(word,isRtlDominant); + } + word = normalize.normalizePres(word); + + writeWord(line.get(i),word); } } else { for(int i=0; i normalize(List line, boolean isRtlDominant, boolean hasRtl){ - LinkedList normalized = new LinkedList(); - StringBuilder lineBuilder = new StringBuilder(); - for(TextPosition text : line){ - if (text instanceof WordSeparator) { - String lineStr = lineBuilder.toString(); - if (hasRtl) { - lineStr = normalize.makeLineLogicalOrder(lineStr,isRtlDominant); + private List normalize(List line){ + /* DR: We have to reorder the tokens in this line in a set of words */ + Word currentWord = null; + + ArrayList words = new ArrayList(); + + for(TextPosition text : line) { + if(text instanceof WordSeparator) { + if(!currentWord.isEmpty()) { + words.add(currentWord); } - lineStr = normalize.normalizePres(lineStr); - normalized.add(lineStr); - lineBuilder = new StringBuilder(); + currentWord = null; + + } else { + // DR: We create a new word to store data, firstly we have + // to check if currentPostion stores only one char, or severals, + // in last case we have to look for spaces in order to split + // current token into severals. + if(currentWord == null) { + currentWord = new Word(); + } + int _result = currentWord.addTextPosition(text); + if(_result == -1 ) { + // If word is not empty we add to the list of words + if(!currentWord.isEmpty()) { + words.add(currentWord); + } + currentWord = null; + } + } - else { - lineBuilder.append(text.getCharacter()); - } } - if (lineBuilder.length() > 0) { - String lineStr = lineBuilder.toString(); - if (hasRtl) { - lineStr = normalize.makeLineLogicalOrder(lineStr,isRtlDominant); - } - lineStr = normalize.normalizePres(lineStr); - normalized.add(lineStr); - } - return normalized; + // Si la palabra no esta vacía, la añadimos. + if(currentWord!=null && !currentWord.isEmpty()) { + words.add(currentWord); + } + return words; } /** Index: pdfbox/src/main/java/org/apache/pdfbox/util/TextPosition.java =================================================================== --- pdfbox/src/main/java/org/apache/pdfbox/util/TextPosition.java (revision 1027422) +++ pdfbox/src/main/java/org/apache/pdfbox/util/TextPosition.java (working copy) @@ -36,7 +36,7 @@ private float endY; private float maxTextHeight; // maximum height of text, in display units - private int rot; // 0, 90, 180, 270 degrees of page rotation + private float rot; // 0, 90, 180, 270 degrees of page rotation private float x = Float.NEGATIVE_INFINITY; private float y = Float.NEGATIVE_INFINITY; private float pageHeight; @@ -92,12 +92,11 @@ this.endX = textPositionEnd.getXPosition(); this.endY = textPositionEnd.getYPosition(); - this.rot = page.findRotation(); - // make sure it is 0 to 270 and no negative numbers - if(this.rot < 0) - { - rot += 360; + int _rot = page.findRotation(); + if(_rot < 0 ) { + _rot += 360; } + this.rot = ((float)_rot)/(float)Math.PI; this.maxTextHeight = maxFontH; this.pageHeight = page.findMediaBox().getHeight(); @@ -148,7 +147,12 @@ this.endX = textPositionEnd.getXPosition(); this.endY = textPositionEnd.getYPosition(); - this.rot = pageRotation; + int _rot = pageRotation; + if(_rot < 0 ) { + _rot += 360; + } + this.rot = ((float)_rot)/(float)Math.PI; + // make sure it is 0 to 270 and no negative numbers if(this.rot < 0) { @@ -189,90 +193,54 @@ /** * Return the direction/orientation of the string in this object * based on its text matrix. - * @return The direction of the text (0, 90, 180, or 270) + * @return The direction of the text */ public float getDir() { - float a = textPos.getValue(0,0); - float b = textPos.getValue(0,1); - float c = textPos.getValue(1,0); - float d = textPos.getValue(1,1); - - // 12 0 left to right - // 0 12 - if ((a > 0) && (Math.abs(b) < d) && (Math.abs(c) < a) && (d > 0)) - { - return 0; - } - // -12 0 right to left (upside down) - // 0 -12 - else if ((a < 0) && (Math.abs(b) < Math.abs(d)) && (Math.abs(c) < Math.abs(a)) && (d < 0)) - { - return 180; - } - // 0 12 up - // -12 0 - else if ((Math.abs(a) < Math.abs(c)) && (b > 0) && (c < 0) && (Math.abs(d) < b)) - { - return 90; - } - // 0 -12 down - // 12 0 - else if ((Math.abs(a) < c) && (b < 0) && (c > 0) && (Math.abs(d) < Math.abs(b))) - { - return 270; - } - return 0; + float theta = textPos.getRotation(); + if(theta < 0) theta = theta+(float)Math.PI; + return theta; } /** * Return the X starting coordinate of the text, adjusted by - * the given rotation amount. The rotation adjusts where the 0,0 - * location is relative to the text. + * the given rotation amount. * - * @param rotation Rotation to apply (0, 90, 180, or 270). 0 will perform no adjustments. + * @param rotation Rotation to apply * @return X coordinate */ private float getXRot(float rotation) { - if (rotation == 0) - { - return textPos.getValue(2,0); - } - else if (rotation == 90) - { - return textPos.getValue(2,1); - } - else if (rotation == 180) - { - return pageWidth - textPos.getValue(2,0); - } - else if (rotation == 270) - { - return pageHeight - textPos.getValue(2,1); - } - return 0; + float [] rot = new float[]{(float)Math.cos(rotation),(float)Math.sin(rotation)}; + + // We obtain xPos and yPos, in order to compute + // coordinates transformation + float xPosition = textPos.getXPosition(); + float yPosition = textPos.getYPosition(); + + float pageWidthRotated = Math.abs(pageWidth*rot[0]+pageHeight*rot[1]); + + float xRotated = (xPosition*rot[0] + yPosition*rot[1]); + if(xRotated < 0 ) xRotated += pageWidthRotated; + return xRotated; } /** * This will get the page rotation adjusted x position of the character. - * This is adjusted based on page rotation so that the upper - * left is 0,0. * * @return The x coordinate of the character. */ public float getX() { + if(x==Float.NEGATIVE_INFINITY){ - x = getXRot(rot); + x = getXRot(rot); } return x; } /** - * This will get the text direction adjusted x position of the character. - * This is adjusted based on text direction so that the first character - * in that direction is in the upper left at 0,0. + * This will get the text direction adjusted x position of the character * * @return The x coordinate of the text. */ @@ -282,72 +250,48 @@ } /** - * This will get the y position of the character with 0,0 in lower left. - * This will be adjusted by the given rotation. - * @param rotation Rotation to apply to text to adjust the 0,0 location (0,90,180,270) + * This will get the y position this will be adjusted by the rotation * + * @param rotation Rotation to apply + * * @return The y coordinate of the text */ - private float getYLowerLeftRot(float rotation) + private float getYRot(float rotation) { - if (rotation == 0) - { - return textPos.getValue(2,1); - } - else if (rotation == 90) - { - return pageWidth - textPos.getValue(2,0); - } - else if (rotation == 180) - { - return pageHeight - textPos.getValue(2,1); - } - else if (rotation == 270) - { - return textPos.getValue(2,0); - } - return 0; + + float [] rot = new float[]{(float)Math.cos(rotation),(float)Math.sin(rotation)}; + + // We obtain xPos and yPos, in order to compute + // coordinates transformation + float xPosition = textPos.getXPosition(); + float yPosition = textPos.getYPosition(); + + + // We compute pageHeight rotation + return (-xPosition*rot[1] + yPosition*rot[0]); } /** - * This will get the y position of the text, adjusted so that 0,0 is upper left and - * it is adjusted based on the page rotation. + * This will get the y position of the text * * @return The adjusted y coordinate of the character. */ public float getY() { if(y==Float.NEGATIVE_INFINITY){ - if ((rot == 0) || (rot == 180)) - { - y = pageHeight - getYLowerLeftRot(rot); - } - else - { - y = pageWidth - getYLowerLeftRot(rot); - } + y = getYRot(rot); } return y; } /** - * 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. + * This will get the y position of the text * * @return The adjusted y coordinate of the character. */ public float getYDirAdj() { - float dir = getDir(); - // some PDFBox code assumes that the 0,0 point is in upper left, not lower left - if ((dir == 0) || (dir == 180)) - { - return pageHeight - getYLowerLeftRot(dir); - } - else - { - return pageWidth - getYLowerLeftRot(dir); - } + return getYRot(getDir()); } @@ -355,19 +299,17 @@ /** * Get the length or width of the text, based on a given rotation. * - * @param rotation Rotation that was used to determine coordinates (0,90,180,270) + * @param rotation Rotation that was used to determine coordinates * @return Width of text in display units */ private float getWidthRot(float rotation) { - if ((rotation == 90) || (rotation == 270)) - { - return Math.abs(endY - textPos.getYPosition()); - } - else - { - return Math.abs(endX - textPos.getXPosition()); - } + float [] rot = new float[]{(float)Math.cos(rotation),(float)Math.sin(rotation)}; + + float width = Math.abs(endX - textPos.getXPosition()); + float height = Math.abs(endY - textPos.getYPosition()); + return Math.abs((width*rot[0] + height*rot[1])); + } /** @@ -517,6 +459,13 @@ double tp2Xstart = tp2.getXDirAdj(); double tp2Xend = tp2.getXDirAdj() + tp2.getWidthDirAdj(); + + /* + * We check also the rotation angle. + */ + if(Math.abs(getDir()-tp2.getDir()) > 0.3/Math.PI) { + return false; + } /* * No X overlap at all so return as soon as possible. Index: pdfbox/src/main/java/org/apache/pdfbox/util/TextPositionComparator.java =================================================================== --- pdfbox/src/main/java/org/apache/pdfbox/util/TextPositionComparator.java (revision 1027422) +++ pdfbox/src/main/java/org/apache/pdfbox/util/TextPositionComparator.java (working copy) @@ -54,15 +54,15 @@ float pos1YBottom = pos1.getYDirAdj(); float pos2YBottom = pos2.getYDirAdj(); - // note that the coordinates have been adjusted so 0,0 is in upper left - float pos1YTop = pos1YBottom - pos1.getHeightDir(); - float pos2YTop = pos2YBottom - pos2.getHeightDir(); + // 0,0 coordinates are in upper left + float pos1YTop = pos1YBottom + pos1.getHeightDir(); + float pos2YTop = pos2YBottom + pos2.getHeightDir(); float yDifference = Math.abs( pos1YBottom-pos2YBottom); //we will do a simple tolerance comparison. if( yDifference < .1 || - (pos2YBottom >= pos1YTop && pos2YBottom <= pos1YBottom) || - (pos1YBottom >= pos2YTop && pos1YBottom <= pos2YBottom)) + (pos2YBottom <= pos1YTop && pos2YBottom >= pos1YBottom) || + (pos1YBottom <= pos2YTop && pos1YBottom >= pos2YBottom)) { if( x1 < x2 ) { @@ -77,7 +77,7 @@ retval = 0; } } - else if( pos1YBottom < pos2YBottom ) + else if( pos1YBottom > pos2YBottom ) { retval = -1; } Index: pdfbox/src/main/java/org/apache/pdfbox/util/Word.java =================================================================== --- pdfbox/src/main/java/org/apache/pdfbox/util/Word.java (revision 0) +++ pdfbox/src/main/java/org/apache/pdfbox/util/Word.java (revision 0) @@ -0,0 +1,120 @@ +/* + * 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 org.apache.pdfbox.util; + +import java.util.ArrayList; + +/** + * Class used to store several TextPositions that conforms a word, a word + * is a set of positions splited by an space. + * 
+ * $Log$
+ *
+ * + * @author David Rodriguez - http://www.divisait.com - + * @since PDFBOX 1.4 + */ +public class Word { + /** + * Word start x position; + */ + private float wordX; + + /** + * Word start y position + */ + private float wordY; + + /** + * Word rotation angle + */ + private float wordTheta; + + /** + * Word max height + */ + private float wordMaxHeight; + + /** + * Word max width + */ + private float wordMaxWidth; + + /** + * Word data + */ + private StringBuilder word; + + /** + * This methods adds a position to the current word, we are making the + * asumption that a TextPosition doesn't store several words o word fragments. + * + * @param text The new position to add + * @return It returs 0 if position is added to the current word, or -1 + * if it founds a word-separator char + */ + public int addTextPosition(TextPosition text) { + if(text.getCharacter()==null || text.getCharacter().equals(" ")) return -1; + + if(word == null) { + word = new StringBuilder(); + wordX = text.getX(); + wordY = text.getY(); + wordTheta = text.getTextPos().getRotation(); + word = new StringBuilder(); + } + + // We modifiy height word, taking into account max letter height + if(text.getHeight() > wordMaxHeight) wordMaxHeight = text.getHeight(); + wordMaxWidth += text.getWidth(); + + // We append text character to current word + word.append(text.getCharacter()); + return 0; + } + + + public float getWordX() { + return wordX; + } + + public float getWordY() { + return wordY; + } + + public float getWordTheta() { + return wordTheta; + } + + public float getWordMaxHeight() { + return wordMaxHeight; + } + + public float getWordMaxWidth() { + return wordMaxWidth; + } + + public String getWord() { + return word.toString(); + } + + public boolean isEmpty() { + return (word == null ||word.length()==0); + } + + +} Property changes on: pdfbox\src\main\java\org\apache\pdfbox\util\Word.java ___________________________________________________________________ Added: svn:eol-style + native