Laboratoire ICAR » Plateforme TXM

package org.txm.macro.pdf

// STANDARD DECLARATIONS

import org.kohsuke.args4j.*

import groovy.transform.Field

import org.txm.rcp.swt.widget.parameters.*

import org.apache.pdfbox.Loader

import org.apache.pdfbox.text.PDFTextStripperByArea

import org.apache.pdfbox.cos.COSName

import java.awt.geom.Rectangle2D

import java.awt.geom.Rectangle2D.Float

import org.apache.pdfbox.cos.COSArray

import org.apache.pdfbox.cos.COSFloat

import org.apache.pdfbox.pdmodel.common.PDRectangle

// BEGINNING OF PARAMETERS

@Field @Option(name="input_file", usage=".pdf input file", widget="File", required=false, def="")

def input_file

@Field @Option(name="input_dir", usage="The directory containing the .pdf files to read", widget="Folder", required=false, def="")

def input_dir

// Open the parameters input dialog box

if (!ParametersDialog.open(this)) return

// END OF PARAMETERS

ArrayList<String> highlightedTexts = new ArrayList<>()

if (input_dir != null &&  input_dir.exists()) {

	nFiles = 0

	input_dir.eachFileMatch(~/.*.pdf/) { f ->

		if (f.getAbsolutePath().toUpperCase().endsWith(".PDF")) {

	 		processFile(f)

	 		nFiles++

		}

	}

	println "Processed "+nFiles+" files."

} else {

	if (input_file.getAbsolutePath().toUpperCase().endsWith(".PDF")) {

	 processFile(input_file)

	}

}

def processFile(input_file) {

		name = input_file.getName()

		dir = input_file.getParentFile()

		idx = name.lastIndexOf(".")

		if (idx > 0) name = name.substring(0, idx)

		outputFile = new File(input_file.getParentFile(), name + "-annotations.tsv")

		println "Processing "+name+"..."

		try {

			outputFile.withWriter("UTF-8") { writer ->

			doc = Loader.loadPDF(input_file)

			writer.println "page\ttype\thighlight\tcomment"

			for (npage = 0; npage < doc.getNumberOfPages(); ++npage) {

				page = doc.getPage(npage)

				al = page.getAnnotations()

				stripper = new PDFTextStripperByArea()

				stripper.setSortByPosition(true)

				al.each {

					if (it.getSubtype().equals("Highlight")) {

						s = (npage+1)+"\t"+it.getSubtype()

						writer.print s

						// adapted from https://www.anycodings.com/1questions/5047580/java-apache-pdfbox-extract-highlighted-text

						quadsArray = it.getCOSObject().getDictionaryObject(COSName.getPDFName("QuadPoints"))

            					str = null

            					for(int j=1, k=0; j<=(quadsArray.size()/8); j++) {

                					ULX = quadsArray.get(0+k)

                					ULY = quadsArray.get(1+k)

                					URX = quadsArray.get(2+k)

                					URY = quadsArray.get(3+k)

                					LLX = quadsArray.get(4+k)

                					LLY = quadsArray.get(5+k)

                					LRX = quadsArray.get(6+k)

                					LRY = quadsArray.get(7+k)

                					k+=8

                					ulx = ULX.floatValue() - 1			// upper left x.

                					uly = ULY.floatValue()				// upper left y.

                					width = URX.floatValue() - LLX.floatValue()	// calculated by upperRightX - lowerLeftX.

                					height = URY.floatValue() - LLY.floatValue()	// calculated by upperRightY - lowerLeftY.

                					PDRectangle pageSize = page.getMediaBox()

                					uly = pageSize.getHeight() - uly

                					rectangle_2 = new Rectangle2D.Double(ulx, uly, width, height)

                					stripper.addRegion("highlightedRegion", rectangle_2)

                					stripper.extractRegions(page)

                					highlightedText = stripper.getTextForRegion("highlightedRegion").trim()

                					if(j > 1) {

                					    str = str+" "+highlightedText

                					} else {

                					    str = highlightedText

                					}

            					}

            					if (str.length() > 0) {

            						writer.print '\t"'+str.replaceAll('"', '""')+'"'

            					} else {

            						writer.print '\t""'

            					}

            					if (it.getContents() != null) {

							writer.println '\t"'+it.getContents().replaceAll('"', '""')+'"'

						} else {

            						writer.println '\t""'

            					}

					} else {

						if (it.getContents() != null) {

							s = (npage+1)+"\t"+it.getSubtype()

							writer.print s

							s = '\t""\t"'+it.getContents().replaceAll('"', '""')+'"'

							writer.println s

            					}

					}

			    	}

			}

			doc.close()

			writer.close()

			} // writer

		} catch (Exception e) {

			e.printStackTrace()

		}

}

@Field @Option(name="set_sort_by_position", usage="""The order of the text tokens in a PDF file may not be in the same as they appear visually on the screen. For example, a PDF writer may write out all text by font, so all bold or larger text, then make a second pass and write out the normal text.

The default is to not sort by position.

A PDF writer could choose to write each character in a different order. By default PDFBox does not sort the text tokens before processing them due to performance reasons.""", widget="Boolean", required=true, def="false")

def set_sort_by_position

		if (f.getAbsolutePath().toUpperCase().endsWith(".PDF")) {

	 		processFile(f)

	 		nFiles++

	if (input_file.getAbsolutePath().toUpperCase().endsWith(".PDF")) {

	 processFile(input_file)

	}

}

def processFile(input_file) {

		println "Processing "+name+"..."

/*

*/

/*

			println "SeparateByBeads = "+strip.getSeparateByBeads()

			// this will tell if the text stripper should separate by beads

			println "ArticleStart = "+strip.getArticleStart()

			// the string which will be used at the beginning of an article

			println "ArticleEnd = "+strip.getArticleEnd()

			// the string which will be used at the end of an article

			println "AddMoreFormatting = "+strip.getAddMoreFormatting()

			// this will tell if the text stripper should add some more text formatting

			println "LineSeparator = "+strip.getLineSeparator()

			// the desired line separator for output text. The line.separator system property is used if the line separator preference is not set explicitly

			println "WordSeparator = "+strip.getWordSeparator()

			// set the desired word separator for output text. The PDFBox text extraction algorithm will output a space character if there is enough space between two words. By default a space character is used. If you need and accurate count of characters that are found in a PDF document then you might want to set the word separator to the empty string

*/

			strip.setSortByPosition(set_sort_by_position)

		if (f.getAbsolutePath().toUpperCase().endsWith(".PDF")) {

	 		processFile(f)

	 		nFiles++

	if (input_file.getAbsolutePath().toUpperCase().endsWith(".PDF")) {

	 processFile(input_file)

	}

}

def processFile(input_file) {

		println "Processing "+name+"..."