Laboratoire ICAR » Plateforme TXM

package org.txm.scripts.importer.xtz

import org.txm.scripts.importer.*

import org.xml.sax.Attributes

import org.txm.importer.scripts.filters.*

import java.util.ArrayList

import javax.xml.parsers.*

import javax.xml.stream.*

import java.net.URL

import org.xml.sax.InputSource

import org.xml.sax.helpers.DefaultHandler

import java.io.FileFilter

import org.txm.importer.xtz.*

class XTZFacsPagerStep extends Step {

        private def url

        private def inputData

        private def factory

        private XMLStreamReader parser

        OutputStreamWriter writer

        StaxStackWriter pagedWriter = null

        File editionDir

        File imageDirectory // contains all the images sub-diretories

        int iImages

        File textImagesDirectory // contains the images for the txtname

        File xmlFile

        File htmlFile

        def pages = []

        def imageAttributes = [:];

        def tag, attribute, txtname, corpusname

        boolean firstWord

        boolean cutBefore = true;

        String wtag;

        boolean debug = false;

        public XTZFacsPagerStep(File xmlFile, File editionDir, File imageDirectory, String txtname, String corpusname, String tag, String attribute, String wtag, boolean debug) {

                inputData = xmlFile.toURI().toURL().openStream()

                factory = XMLInputFactory.newInstance()

                parser = factory.createXMLStreamReader(inputData)

                this.xmlFile = xmlFile

                this.editionDir = editionDir

                editionDir.mkdirs()

                this.imageDirectory = imageDirectory

                this.tag = tag

                this.attribute = attribute

                this.txtname = txtname

                this.corpusname= corpusname

                this.wtag= wtag;

                this.debug = debug

                imageAttributes = ["src":"",

                        "id":"txm_image",

                        "alt":"Le fac-similé de cette page n'est pas accessible",

                        //"style":"display:block;width:100%;",

                        "onLoad":"viewer.toolbarImages='images/icons';viewer.onload=viewer.toolbar;new viewer({image: this, frame: ['100%','100%']});"

                        ]

                if (imageDirectory != null)

                        textImagesDirectory = new File(imageDirectory, txtname)

                iImages = 0;

        }

        int n = 1;

        private boolean createNextOutput() {

                try {

                        def tags = closeMultiWriter();

                        for (int i = 0 ; i < tags.size() ; i++) {

                                String tag = tags[i]

                                if ("div" != tag) {

                                        tags.remove(i--)

                                } else {

                                        tags.remove(i--) // remove first "div"

                                        break; // remove elements until first "div" tag

                                }

                        }

                        if (wordid != null) {//wordid = "w_0";

                                //println " add page $n $wordid, page=$pages"

                                pages << [htmlFile, wordid] // store the previous page

                                n++

                        }

                        // Page suivante

                        htmlFile = new File(editionDir, "${txtname}_${n}.html")

                        firstWord = true

                        //println "SET FIRST WORD=true"

                        pagedWriter = new StaxStackWriter(htmlFile, "UTF-8");

                        if (debug) println "Create file $htmlFile"

                        pagedWriter.writeStartDocument("UTF-8", "1.0")

                        pagedWriter.writeStartElement("html");

                        pagedWriter.writeStartElement("head");

                        pagedWriter.writeStartElement("title")

                        pagedWriter.writeCharacters(corpusname+" Edition - Page "+n)

                        pagedWriter.writeEndElement(); // </title>

                        pagedWriter.writeStartElement("meta", ["http-equiv":"Content-Type", "content":"text/html","charset":"UTF-8"]);

                        pagedWriter.writeEndElement()

                        pagedWriter.writeStartElement("link", ["rel":"stylesheet", "media":"all", "type":"text/css","href":"txm.css"]);

                        pagedWriter.writeEndElement()

                        pagedWriter.writeStartElement("link", ["rel":"stylesheet", "media":"all", "type":"text/css","href":"${corpusname}.css"]);

                        pagedWriter.writeEndElement()

                        pagedWriter.writeStartElement("script", ["type":"text/javascript", "src":"js/viewer/Simple_Viewer_beta_1.1-min.js"]);

                        pagedWriter.writeEndElement()

                        pagedWriter.writeStartElement("script", ["type":"text/javascript", "src":"js/viewer/toolbar-ext.js"]);

                        pagedWriter.writeEndElement()

                        pagedWriter.writeStartElement("link", ["rel":"stylesheet", "media":"all", "type":"text/css","href":"js/viewer/toolbar-ext.css"]);

                        pagedWriter.writeEndElement()

                        pagedWriter.writeEndElement() // </head>

                        pagedWriter.writeStartElement("body") //<body>

                        pagedWriter.writeStartElement("div", ["id":"txmeditionpage"]);

                        pagedWriter.writeStartElements(tags);

                        wordid = "${wtag}_0"; // default value if no word is found

                        return true;

                } catch (Exception e) {

                        System.out.println("Error while creating HTML file: "+e);

                        return false;

                }

        }

        private def closeMultiWriter()

        {

                if (pagedWriter != null) {

                        def tags = pagedWriter.getTagStack().clone();

                        if (firstWord) { // there was no words

                                pagedWriter.writeCharacters("");

                                pagedWriter.write("<span id=\"${wtag}_0\"/>");

                        }

                        pagedWriter.writeEndElements();

                        pagedWriter.close();

                        return tags;

                } else {

                        return [];

                }

        }

        private writeImg(String src) {

                if (src == null || src.length() == 0) {

                        pagedWriter.writeStartElement("p", ["class":"no-img"]);

                        pagedWriter.writeCharacters("Cette page n'est pas disponible en mode fac-similé.");

                        pagedWriter.writeEndElement(); // </div>

                } else {

                        imageAttributes["src"] = src

                        pagedWriter.writeEmptyElement("img", imageAttributes);

                }

        }

        /**

         * Process the XML-TXM file. if imageDirectory is not null, then it will be use to fetch the images, if not then the pb@facs attribute contains the image path

         */

        public boolean process() {

                if (textImagesDirectory != null) {

                        return processWithImages();

                } else {

                        return processWithoutImages();

                }

        }

        String wordid = null;

        def imagesFiles;

        public boolean processWithImages() {

                if (!textImagesDirectory.exists()) {

                        println ("No images directory found: "+textImagesDirectory)

                        return false;

                }

                // Scan textImagesDirectory for images

                imagesFiles = textImagesDirectory.listFiles(new FileFilter() {

                                        public boolean accept(File f) {

                                                return !f.isHidden() && !f.isDirectory()

                                        }

                                });

                if (imagesFiles == null || imagesFiles.size() == 0) {

                        println ("No facs images found in "+textImagesDirectory)

                        return false;

                }

                imagesFiles.sort(); // sort by name

                iImages = 0;

                //println "For text "+xmlFile.getName()

                //println " using images : $imagesFiles"

                // parse XML-TXM files

                boolean start = false

                String localname

                createNextOutput(); // empty page

                for (int event = parser.next(); event != XMLStreamConstants.END_DOCUMENT; event = parser.next()) {

                        switch (event) {

                                case XMLStreamConstants.START_ELEMENT:

                                        localname = parser.getLocalName();

                                        switch (localname) {

                                                case "text":

                                                        start = true

                                                        printStartPage();

                                                        break;

                                                        break;

                                                case wtag:

                                                        if (firstWord) {

                                                                wordid = parser.getAttributeValue(null, "id");

                                                                //println "found word: $wordid"

                                                                firstWord = false;

                                                        }

                                                        break;

                                                case tag:

                                                        if (debug) println "** TAG $tag $attribute : "+parser.getAttributeValue(null, attribute);

                                                        if (iImages >= imagesFiles.size() ) {

                                                                println " ERROR Cound not find the $iImages th image for file $xmlFile at location "+parser.getLocation().getLineNumber()

                                                        } else {

                                                                String imgPath = "res/images/$corpusname/${editionDir.getName()}/$txtname/"+imagesFiles[iImages].getName()

                                                                if (imgPath == null) {

                                                                        println " ERROR in $xmlFile no value found for tag=$tag attribute=@$attribute iImages=@iImages at location "+parser.getLocation().getLineNumber()

                                                                } else {

                                                                        if (cutBefore) {

                                                                                if (debug) println " cut before"

                                                                                createNextOutput()

                                                                                if (debug) println " write img $imgPath"

                                                                                writeImg(imgPath)

                                                                        } else {

                                                                                if (debug) println " write img $imgPath"

                                                                                writeImg(imgPath)

                                                                                if (debug) println " cut after"

                                                                                createNextOutput()

                                                                        }

                                                                }

                                                        }

                                                        iImages++ // next image file

                                                        break;

                                        }

                                        break;

                        }

                }

                closeMultiWriter()

                if (parser != null) parser.close();

                if (inputData != null) inputData.close();

                pages << [htmlFile, wordid] // add the last page (no pb encountered

                return pages.size() > 1

        }

        public boolean processWithoutImages() {

                // parse XML-TXM files

                boolean start = false

                String localname

                createNextOutput(); // empty page

                for (int event = parser.next(); event != XMLStreamConstants.END_DOCUMENT; event = parser.next()) {

                        switch (event) {

                                case XMLStreamConstants.START_ELEMENT:

                                        localname = parser.getLocalName();

                                        switch (localname) {

                                                case "text":

                                                        start = true

                                                        printStartPage()

                                                        break;

                                                case wtag:

                                                        if (firstWord) {

                                                                wordid = parser.getAttributeValue(null, "id");

                                                                //println "found word: $wordid"

                                                                firstWord = false;

                                                        }

                                                        break;

                                                case tag:

                                                        String url = "" // get image URL

                                                        for (int i = 0 ; i < parser.getAttributeCount() ; i++) {

                                                                if (parser.getAttributeLocalName(i).equals(attribute)) {

                                                                        url = parser.getAttributeValue(i);

                                                                        continue;

                                                                }

                                                        }

                                                        if (debug) println "** TAG $tag $attribute : $url"

                                                        if (cutBefore) {

                                                                if (debug) println " cut before"

                                                                createNextOutput()

                                                                if (debug) println " write img $url"

                                                                writeImg(url)

                                                        } else {

                                                                if (debug) println " write img $url"

                                                                writeImg(url)

                                                                if (debug) println " cut after"

                                                                url()

                                                        }

                                                        break;

                                        }

                                        break;

                        }

                }

                closeMultiWriter()

                pages << [htmlFile, wordid] // add the last page (no pb encountered

                return pages.size() > 1

        }

        public void printStartPage() {

                LinkedHashMap attributes = new LinkedHashMap();

                for (int i = 0 ; i < parser.getAttributeCount() ; i++) {

                        attributes[parser.getAttributeLocalName(i)] = parser.getAttributeValue(i).toString()

                }

                pagedWriter.write("\n");

                pagedWriter.writeStartElement("p")

                if (attributes.containsKey("id")) {

                        pagedWriter.writeElement("h3", attributes["id"])

                }

                pagedWriter.writeStartElement("table");

                for (String k : attributes.keySet()) {

                        if (k == "id") continue;

                        pagedWriter.writeStartElement("tr");

                        pagedWriter.writeElement("td", k);

                        pagedWriter.writeElement("td", attributes[k]);

                        pagedWriter.writeEndElement();

                }

                pagedWriter.writeEndElement() // table

                pagedWriter.writeEndElement() // p

                pagedWriter.writeCharacters("");

        }

        public def getPageFiles() {

                return pages

        }

        public static void main(String[] args) {

                File txmFile = new File("/home/mdecorde/TXM/corpora/BVHEPISTEMON2016/txm/BVHEPISTEMON2016/1546_RabelaisTL.xml")

                String txtname = "1532_RabelaisPrnstctn"

                String corpusname = "BVH"

                File newEditionDirectory = new File("/tmp/xtzpagertest/")

                newEditionDirectory.deleteDir()

                newEditionDirectory.mkdir()

                File imageDirectory = null;

                def ed = new XTZFacsPagerStep(txmFile, newEditionDirectory, imageDirectory, txtname, corpusname, "pb", "facs", "w", true);

                if (!ed.process()) {

                        println "Fail to build edition for text: $txmFile"

                }

        }

}