API tools faq
paste
Advertisement
KosIvantsov

Write XLIFF (new)

KosIvantsov
Apr 2nd, 2016
154
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
Groovy 7.89 KB | None | 0 0
raw download clone embed print report
  1. /* :name=Write XLIFF (new) :description=
  2.  * @Purpose:    Export the whole project to XLF file for other CAT tools
  3.  * @author:     Kos Ivantsov
  4.  * @date:       2016-04-02
  5.  * @version:    1.0
  6.  */
  7.  
  8.  
  9. /* set to true to write a settings file for Okapi Rainbow that can be
  10.  * used to convert the XLF file produced by this script, to TMX.
  11.  * Otherwise set to false    */
  12. rainbow = true
  13.  
  14. /* set to true to output only approved entries from XLF to TMX during
  15.  * conversion in Rainbow    */
  16. get_only_approved = true
  17.  
  18. /* add xml:space='preserve' to each TU in the resultant XLF */
  19. preserve_spaces = true
  20.  
  21. /* Changing this variable lets the user specify
  22.  * what to do with untranslated segments
  23.  * "copy" for copying source to target
  24.  * "empty" for leaving the target empty
  25.  * "ignore" for not including target element    */
  26. xliffUntranslated = "copy"
  27.  
  28. /* The value set for the following variable will
  29.  * be inserted in empty translations as target   */
  30. emptyTranslation = ''
  31.  
  32.  
  33. import static javax.swing.JOptionPane.*
  34. import static org.omegat.util.Platform.*
  35. import org.omegat.util.StringUtil
  36. import org.omegat.core.data.ProtectedPart
  37.  
  38. def prop = project.projectProperties
  39. if (!prop) {
  40.     final def title = 'Export project to XLIFF file(s)'
  41.     final def msg   = 'Please try again after you open a project.'
  42.     showMessageDialog null, msg, title, INFORMATION_MESSAGE
  43.     return
  44. }
  45. def folder = prop.projectRoot+'script_output'+File.separator
  46. projname = new File(prop.getProjectRoot()).getName()
  47. // create folder if it doesn't exist
  48. if (! (new File (folder)).exists()) {
  49.     (new File(folder)).mkdir()
  50.     }
  51. filename = folder + projname +'.xlf'
  52. xliff_file = new File(filename).newWriter('UTF-8')
  53.  
  54. count = transcount = untranscount = emptytranscount = ignorecount = writecount = 0
  55.  
  56. sourceLocale = prop.getSourceLanguage().toString().toLowerCase()
  57. targetLocale = prop.getTargetLanguage().toString().toLowerCase()
  58. xmlspace = preserve_spaces ? "xml:space=\'preserve\'" : ""
  59.  
  60. xliff_file.write("""<?xml version="1.0" encoding="UTF-8"?>
  61. <xliff xmlns="urn:oasis:names:tc:xliff:document:1.2" version="1.2">
  62. """)
  63.  
  64. files = project.projectFiles
  65. //files = project.projectFiles.subList(editor.@displayedFileIndex, editor.@displayedFileIndex + 1);
  66.     for (i in 0 ..< files.size())
  67.     {
  68.         fi = files[i]
  69.         xliff_file.append("""  <file original="$fi.filePath" source-language="$sourceLocale" target-language="$targetLocale" datatype="x-application/x-tmx+xml">
  70.    <body>
  71.      <trans-unit id="0" approved="yes">
  72.        <source xml:lang="$sourceLocale"><ph id="filename">==FILENAME: "$fi.filePath"==</ph>
  73.        </source>
  74.        <target xml:lang="$targetLocale" state="final"><ph id="filename">==FILENAME: "$fi.filePath"==</ph>
  75.        </target>
  76.      </trans-unit>\n""")
  77.         for (j in 0 ..< fi.entries.size()) {
  78.             def state
  79.             def approved = ''
  80.             def unitnote = ''
  81.             def ignore = ''
  82.             ste = fi.entries[j]
  83.             seg_num = ste.entryNum()
  84.             source = ste.getSrcText()
  85.             info = project.getTranslationInfo(ste)
  86.             target = info ? info.translation : null
  87.             def alltags = ''
  88.             List xmlTags = []
  89.             for (ProtectedPart pp : ste.getProtectedParts()) {
  90.                 alltags += pp.getTextInSourceSegment()
  91.                 xmlTags.add(StringUtil.makeValidXML(pp.getTextInSourceSegment()))
  92.             }
  93.             if (source == alltags) {
  94.                 count++
  95.                 ignorecount++
  96.                 continue
  97.             }
  98.             if (target != null) {
  99.                 approved = 'approved="yes"'
  100.                 state = 'state="final" state-qualifier="exact-match"'
  101.                 transcount++
  102.                 if (target.size() == 0 ) {
  103.                     target = emptyTranslation
  104.                     emptytranscount++
  105.                     }
  106.             }else{
  107.                 state = 'state="needs-translation"'
  108.                 switch (xliffUntranslated) {
  109.                     case "copy" :
  110.                         target = source
  111.                         break
  112.                     case ["empty", "ignore"] :
  113.                         target = ''
  114.                         break
  115.                     default :
  116.                         target =  source
  117.                         break
  118.                 }
  119.                 untranscount++
  120.             }
  121.             unitnote = (info.hasNote()) ? "\n        <note>${StringUtil.makeValidXML(info.note)}</note>" : ""
  122.             if (source != alltags){
  123.                 source = StringUtil.makeValidXML(source)
  124.                 target = StringUtil.makeValidXML(target)
  125.                 xmlTags = xmlTags.unique()
  126.                 for (x in 0 ..< xmlTags.size()){
  127.                     source =  source.replace(xmlTags[x], "<ph id=\"$x\">" + xmlTags[x].toString() + "</ph>")
  128.                     target =  target.replace(xmlTags[x], "<ph id=\"$x\">" + xmlTags[x].toString() + "</ph>")
  129.                 }
  130.                 targetLine = (target == '' && xliffUntranslated == "ignore") ? '' : "\n        <target $state xml:lang=\"$targetLocale\"><mrk mid=\"0\" mtype=\"seg\">$target</mrk></target>"
  131.                 xliff_file.append("""\
  132.      <trans-unit id="$seg_num" $xmlspace $approved>
  133.        <source xml:lang="$sourceLocale">$source</source>
  134.        <seg-source><mrk mid="0" mtype="seg">$source</mrk></seg-source>\
  135. $targetLine\
  136. $unitnote
  137.      </trans-unit>\n""")
  138.                 writecount++
  139.             }
  140.         count++
  141.         }
  142.         xliff_file.append("    </body>\n  </file>\n")
  143.     }
  144. xliff_file.append("</xliff>")
  145. xliff_file.close()
  146. console.clear()
  147. console.println """\
  148. ${'*'*(filename.toString().size()+15)}
  149. Output file:   $filename
  150. ${'*'*(filename.toString().size()+15)}
  151. Segments processed: $count
  152. Translated: $transcount
  153. Untranslated: $untranscount
  154. Empty translation: $emptytranscount
  155. Segments written: $writecount
  156. Segments ignored: $ignorecount
  157. """
  158. mainWindow.statusLabel.setText("XLIFF Export: $filename written")
  159. Timer timer = new Timer().schedule({mainWindow.statusLabel.setText(null); console.clear()} as TimerTask, 10000)
  160.  
  161. if (rainbow == true) {
  162.     def approved = get_only_approved ? 'true' : 'false'
  163.     rainbowfile = new File(folder + projname +'.xlf2tmx.rnb')
  164.     rainbowfile.write("""\
  165. <?xml version="1.0" encoding="UTF-8"?>
  166. <rainbowProject version="4">
  167.    <fileSet id="1">
  168.        <root useCustom="0"></root>
  169.    </fileSet>
  170.    <fileSet id="2">
  171.        <root useCustom="0"></root>
  172.    </fileSet>
  173.    <fileSet id="3">
  174.        <root useCustom="0"></root>
  175.    </fileSet>
  176.    <output>
  177.        <root use="0"></root>
  178.        <subFolder use="0"></subFolder>
  179.        <extension use="1" style="0">.out</extension>
  180.        <replace use="0" oldText="" newText=""></replace>
  181.        <prefix use="0"></prefix>
  182.        <suffix use="0"></suffix>
  183.    </output>
  184.    <options sourceLanguage="$sourceLocale" sourceEncoding="UTF-8" targetLanguage="$targetLocale" targetEncoding="UTF-8"></options>
  185.    <parametersFolder useCustom="0"></parametersFolder>
  186.    <utilities xml:spaces="preserve"><params id="currentProjectPipeline">&lt;?xml version="1.0" encoding="UTF-8"?>
  187. &lt;rainbowPipeline version="1">&lt;step class="net.sf.okapi.steps.common.RawDocumentToFilterEventsStep">&lt;/step>
  188. &lt;step class="net.sf.okapi.steps.codesremoval.CodesRemovalStep">#v1
  189. stripSource.b=true
  190. stripTarget.b=true
  191. mode.i=0
  192. includeNonTranslatable.b=true
  193. replaceWithSpace.b=false&lt;/step>
  194. &lt;step class="net.sf.okapi.steps.formatconversion.FormatConversionStep">#v1
  195. singleOutput.b=false
  196. autoExtensions.b=true
  197. targetStyle.i=0
  198. outputPath=
  199. outputFormat=tmx
  200. useGenericCodes.b=false
  201. skipEntriesWithoutText.b=true
  202. approvedEntriesOnly.b=$approved
  203. overwriteSameSource.b=false&lt;/step>
  204. &lt;step class="net.sf.okapi.steps.common.FilterEventsToRawDocumentStep">&lt;/step>
  205. &lt;/rainbowPipeline>
  206. </params></utilities>
  207. </rainbowProject>
  208. """, 'UTF-8')
  209.     }
  210. return
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement
Public Pastes
Advertisement
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!