Application: gvSIG desktop

/*

 * To change this license header, choose License Headers in Project Properties.

 * To change this template file, choose Tools | Templates

 * and open the template in the editor.

 */

package org.gvsig.fmap.dal.store.gml.virtualrows;

import java.io.File;

import java.io.FileNotFoundException;

import java.io.IOException;

import java.io.InputStream;

import java.io.Reader;

import java.nio.charset.Charset;

import java.util.ArrayList;

import java.util.List;

import java.util.Locale;

import javax.xml.parsers.SAXParser;

import javax.xml.parsers.SAXParserFactory;

import org.apache.commons.lang3.StringUtils;

import org.apache.tika.utils.CharsetUtils;

import org.gvsig.fmap.dal.DALLocator;

import org.gvsig.fmap.dal.DataManager;

import org.gvsig.fmap.dal.store.gml.virtualrows.xmlinfo.XMLAttributeInfoImpl;

import org.gvsig.fmap.dal.store.gml.virtualrows.xmlinfo.XMLInfoImpl;

import org.gvsig.tools.ToolsLocator;

import org.gvsig.tools.i18n.I18nManager;

import org.gvsig.tools.task.SimpleTaskStatus;

import org.slf4j.Logger;

import org.slf4j.LoggerFactory;

import org.xml.sax.Attributes;

import org.xml.sax.InputSource;

import org.xml.sax.Locator;

import org.xml.sax.SAXException;

import org.xml.sax.helpers.DefaultHandler;

/**

 *

 * @author jjdelcerro

 */

public class StructureExtractorImpl { //implements StructureExtractor {

    private static final Logger LOGGER = LoggerFactory.getLogger(StructureExtractorImpl.class);

    @SuppressWarnings("UseSpecificCatch")

    private void extractTags(XMLInfoImpl xmlinfo, Reader reader, SimpleTaskStatus status) {

        if( reader == null ) {

            throw new IllegalArgumentException("reader is null");

        }

        try {

            final DataManager dataManager = DALLocator.getDataManager();

            I18nManager i18n = ToolsLocator.getI18nManager();

            status.message(i18n.getTranslation("_Reading_xml")+" 1/5");

            status.setRangeOfValues(0, xmlinfo.getCountLines());

            SAXParserFactory spf = SAXParserFactory.newInstance();

            spf.setNamespaceAware(true);

            SAXParser saxParser = spf.newSAXParser();

            InputSource is = new InputSource(reader);

            List<String> path = new ArrayList<>();

            saxParser.parse(is, new DefaultHandler() {

                private Locator locator;

                int size;

                int refreshInterval = 1;

                StringBuilder chars = new StringBuilder();

                @Override

                public void setDocumentLocator(Locator locator) {

                    this.locator = locator;

                }

                @Override

                public void startElement(String uri, String localName, String qName, Attributes attributes) throws SAXException {

                    int line = this.locator.getLineNumber();

                    int column = this.locator.getColumnNumber()-2-localName.length();

                    if(line % refreshInterval == 0) {

                        status.setCurValue(line);

                    }

                    if(line > 100000){

                        refreshInterval = 10000;

                    } else if(line > 10000){

                        refreshInterval = 1000;

                    } else if(line > 1000){

                        refreshInterval = 100;

                    } else if(line > 100){

                        refreshInterval = 10;

                    }

                    String idvalue = dataManager.createUniqueID();

                    path.add(localName);

                    String path_s = StringUtils.join(path, "/");

                    XMLAttributeInfoImpl info = xmlinfo.getTag(path_s);

                    if( info == null ) {

                        info = new XMLAttributeInfoImpl(xmlinfo.getLocale(), path_s, getNs(qName));

                        xmlinfo.addTag(info);

                    }

                    if( path.size()>1 ) {

                        List<String> parentpath = path.subList(0, path.size()-1);

                        String parentpath_s = StringUtils.join(parentpath, "/");

                        XMLAttributeInfoImpl parentinfo = xmlinfo.getTag(parentpath_s);

                        parentinfo.incrChildCount(localName);

                        parentinfo.setLastChildID(localName, idvalue);

                    }

                    for (int i = 0; i < attributes.getLength(); i++) {

                        String name = attributes.getLocalName(i);                        

                        String value = attributes.getValue(i);

                        String idvalueChild = dataManager.createUniqueID();

                        XMLAttributeInfoImpl infoChild = xmlinfo.getTag(path_s+"/"+name);

                        if( infoChild == null ) {

                            infoChild = new XMLAttributeInfoImpl(xmlinfo.getLocale(), path_s+"/"+name, getNs(qName));

                            infoChild.setIsAttr(true);

                            xmlinfo.addTag(infoChild);

                        }

                        info.incrChildCount(infoChild.getName());

                        info.setLastChildID(infoChild.getName(), idvalueChild);

                        infoChild.addValue(value);

                    }

                    chars.setLength(0);

                }

                @Override

                public void endElement(String uri, String localName, String qName) throws SAXException {

                    int line = this.locator.getLineNumber();

//                    status.setCurValue(line);

                    String path_s = StringUtils.join(path, "/");

                    XMLAttributeInfoImpl info = xmlinfo.getTag(path_s);

                    if( info == null ) {

                        info = new XMLAttributeInfoImpl(xmlinfo.getLocale(), path_s, getNs(qName));

                        xmlinfo.addTag(info);

                    }

                    XMLAttributeInfoImpl parentinfo = null;

                    if( path.size()>1 ) {

                        List<String> parentpath = path.subList(0, path.size()-1);

                        String parentpath_s = StringUtils.join(parentpath, "/");

                        parentinfo = xmlinfo.getTag(parentpath_s);

                    }

                    if( info.hasChilds() || (parentinfo != null && (parentinfo.getChildsCount(localName)>1))) {

                        String value = this.chars.toString();

                        if( StringUtils.isNotBlank(value) ) {

                            String name = info.getName()+"$v";                        

                            String idvalueChild = dataManager.createUniqueID();

                            XMLAttributeInfoImpl infoChild = xmlinfo.getTag(path_s+"/"+name);

                            if( infoChild == null ) {

                                infoChild = new XMLAttributeInfoImpl(xmlinfo.getLocale(), path_s+"/"+name, getNs(qName));

                                xmlinfo.addTag(infoChild);

                            }

                            info.incrChildCount(infoChild.getName());

                            info.setLastChildID(infoChild.getName(), idvalueChild);

                            infoChild.addValue(value);

                        }

                    } else {

                        String value = this.chars.toString();

                        info.addValue(value);

                    }

                    info.consolidateChildCounters();

                    path.remove(path.size()-1);

                    chars.setLength(0);

                }

                @Override

                public void characters(char[] ch, int start, int length) throws SAXException {

                    int line = this.locator.getLineNumber();

//                    status.setCurValue(line);

                    this.chars.append(ch, start, length);

                }

            });

        } catch (Exception ex) {

            throw new RuntimeException("Can't extract tags.", ex);

        }

    }

    public XMLInfoImpl extractStructure(File xml, Charset charset, Locale locale, SimpleTaskStatus status) throws FileNotFoundException, IOException {

        XMLInfoImpl xmlinfo = new XMLInfoImpl();

        xmlinfo.setLocale(locale);

        long count = XmlCommons.countLines(xml, charset, status);

        xmlinfo.setCountLines(count);

        InputSource is = XmlCommons.openReader(xml,charset);

        return extractStructure(is, xmlinfo, status);

    }

    public XMLInfoImpl extractStructure(InputStream xml, Charset charset, Locale locale, SimpleTaskStatus status) throws IOException  {

        XMLInfoImpl xmlinfo = new XMLInfoImpl();

        xmlinfo.setLocale(locale);

        xmlinfo.setCountLines(-1);

        InputSource is = XmlCommons.openReader(xml, charset);

        return extractStructure(is, xmlinfo, status);

    }

    public XMLInfoImpl extractStructure(Reader reader, Locale locale, SimpleTaskStatus status) {

        XMLInfoImpl xmlinfo = new XMLInfoImpl();

        xmlinfo.setLocale(locale);

        InputSource is = new InputSource(reader);

        return extractStructure(is, xmlinfo, status);

    }

    public XMLInfoImpl extractStructure(InputSource is, XMLInfoImpl xmlinfo, SimpleTaskStatus status) {

        if( xmlinfo.getCharset()==null ) {

            xmlinfo.setCharset(CharsetUtils.forName(is.getEncoding()));

        }

        extractTags(xmlinfo, is.getCharacterStream(), status);

        return xmlinfo;

    }

    private String getNs(String s){

        String[] ss = s.split(":");

        if(ss.length == 1){

            return null;

        }

        return ss[0];

    }

}