Adapter la récupération de langues aux TMX multilingues

filtres_TMX_pour_Calc/LISEZMOI.txt

@@ -0,0 +1,16 @@
+Le logiciel TMX-filters fournit deux filtres XSLT qui permettent d’ouvrir des mémoires de traduction TMX dans LibreOffice (ou OpenOffice.org) Calc et surtout d’enregistrer des phrases alignées dans deux colonnes ou plus d’un tableau.
+
+Installation : Télécharger ici le paquet TMX-filters.jar. Il doit être installé dans LibreOffice par la boîte de dialogue ordinaire de la commande Macros > Paramétrage des filtres XML… du menu Outils. (Ou directement la commande Paramétrage des filtres XML… du menu Outils dans des versions plus anciennes.) Cliquer le bouton Ouvrir le package… et naviguer jusqu’au paquet. (In het nederlands, [Macros >, volgens versie] XMLfilterinstellingen… van het menu Extra. Klik Open pakket… en blader naar het pakket.) Le paquet doit impérativement avoir l’extension .jar pour être reconnu, mais c’est, entre autres, un fichier au format ZIP. On peut donc l’ouvrir (au besoin en en prenant une copie renommée en .zip au lieu de .jar) et, si l’on veut, y modifier les deux filtres écrits en XSL. (Merci de me communiquer des modifications intéressantes.) Il contient un fichier TMX-readme.txt en anglais. Ce logiciel, © 2013, 2020 Dominique Meeùs et 2022 Philippe Tourigny, est distribué sous licence LGPL version 3 (voir plus loin).
+
+Limitation : Pour le filtre d'exportation, il faut mettre les codes de langues dans la première rangée du tableau, avec la langue source dans la cellule A1 et les langues cibles dans chacune des colonnes suivantes. Le code fixe la première langue cible (dans la cellule B1) comme langue administrative du fichier TMX. Si vous désirez utiliser une langue administrative autre que l'une de vos langues cible, il vous faudra soit modifier le fichier TMX après l'exportation, soit le code du filtre.
+
+Pour le filtre d'importation, le code suppose que toutes les entrées contiennent le même nombre de langues, dans le même ordre. Il récupère les codes de langues du fichier TMX dans le premier élément <tu> du fichier, et les utilise comme entête de colonne dans la première rangée du fichier Calc. Par conséquent, si vous importez une TMX contenant un plus grand nombre de langues dans certain éléments <tu> (comme c'est le cas dans certaines mémoires offertes par la DGT), certaines rangées contiendront des segments dans des colonnes sans entête, et certains segments aboutiront dans la colonne d'une autre langue.
+
+La possibilité de récupérer toutes les langues (depuis l'élément <tu> contenant le plus de sous-éléments <tuv> par exemple) et de placer chaque segment dans la bonne colonne (laissant vides celles qui n'ont pas de traduction) est une amélioration à apporter, mais n'étant que bidouilleurs plutôt que programmeurs (XSLT ou autre), nous ne savons pas encore comment. Si vous y connaissez plus que nous, n’hésitez pas à nous communiquer vos idées.
+
+Utilisation : Pour lire une mémoire TMX dans LibreOffice Calc, il suffit de faire Fichier, Ouvrir… Aligner des phrases en deux colones ou plus dans LibreOffice Calc. Il faut saisir les codes de langue voulus dans chacune des colonnes de la première rangée, avec la langue source dans la première colonne et la langue qui servira de langue administrative pour le fichier TMX dans la deuxième. On obtient ces phrases en mémoire de traduction TMX en faisant simplement Enregistrer sous… et en choisissant le type de fichier TMX (qui a toutes les chances de se retrouver tout au bas d’une longue liste).
+
+Licence : Ce logiciel peut être utilisé sous licence LGPL version 3 ou ultérieure. On trouve sur le site mentionné ci-dessous des explications dans de nombreuses langues dont le français.
+
+Copyright 2013, 2020, Dominique Meeùs, et 2022 Philippe Tourigny.
+This program is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details. You should have received a copy of the GNU Lesser General Public License along with this program. If not, see www.gnu.org/licenses/.

filtres_TMX_pour_Calc/TMX-export.xsl

@@ -1,9 +1,15 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <!-- Dominique Meeùs, created 1-10-2013, version 0.9.
-     XSLT transformation of an Open Document Format spreadsheet in two columns
+     XSLT transformation of an Open Document Format spreadsheet in five columns
      into a TMX translation memory exchange file.
      Filter to install as an export filter for LibreOffice Calc. -->
-<!-- Copyright 2013 Dominique Meeùs.
+<!-- Dominique Meeùs, modified 21-5-2020, version 0.92.
+     Hardcoded languages of the columns are : nl-BE, fr-BE, de-DE, en-GB, es-ES. -->
+<!-- Philippe Tourigny, modified 12-9-2022, version 0.94
+     Enable the filter to read the language code from the first column
+     in each row. Also, add a SYSTEM DOCTYPE declaration to the output
+     XML file.-->
+<!-- Copyright 2013, 2020 Dominique Meeùs, and 2022 Philippe Tourigny@.
      This program is free software: you can redistribute it and/or modify it
      under the terms of the GNU Lesser General Public License
      as published by the Free Software Foundation,
@@ -20,50 +26,88 @@
                 xmlns:text="urn:oasis:names:tc:opendocument:xmlns:text:1.0"
                 exclude-result-prefixes="office table text">
            <!-- Namespaces needed to access parts of the document -->
-  <xsl:output method = "xml" indent = "yes" encoding = "UTF-8" omit-xml-declaration = "no"/>
 
-  <xsl:variable name="headerCell" select="//table:table/table:table-row[1]/table:table-cell"></xsl:variable>
+   <!-- Version 0.94: Add a SYSTEM DOCTYPE to output -->
+   <xsl:output method = "xml" indent = "yes" encoding = "UTF-8"
+    doctype-system="tmx14.dtd" omit-xml-declaration = "no"/>
 
   <xsl:template match="/">
-    <!-- Get the source and target languages from the first and second column of the first row. Code based on this answer https://forum.openoffice.org/en/forum/viewtopic.php?p=56627&sid=d4ebce191acc01d99d587fef28377db4#p56627 from the Apache Office forum -->
-
-     
-    <xsl:variable name="adminlang" select="$headerCell[2]/text:p"></xsl:variable>
-    <xsl:variable name="srclang" select="$headerCell[1]/text:p"></xsl:variable>
-    <xsl:variable name="tgtlang" select="$headerCell[2]/text:p"></xsl:variable>
-    
     <tmx version="1.4">
+      <!-- Define variables to make code easier to manage.
+           Cells in the first row are language code headings.
+           The first target language is selected as the
+           administrative language for the output TMX.
+           The first column identifies the source language. -->
+
+      <xsl:variable name="headingCell"
+        select="//table:table/table:table-row[1]/table:table-cell"/>
+      <xsl:variable name="adminlang"
+        select="$headingCell[2]/text:p"/> 
+      <xsl:variable name="srclang"
+         select="$headingCell[1]/text:p"/>
+     
+        <!-- Define the TMX header
+             The <xsl:attribute> element is used because variables
+             are not recognized if entered directly in attributes. -->
+    
       <header>
-        <xsl:attribute name="creationtool">ods2tmx filter</xsl:attribute>
-        <xsl:attribute name="creationtoolversion">0.93</xsl:attribute>
+        <xsl:attribute name="creationtool">TMX-export for LibreOffice</xsl:attribute>
+        <xsl:attribute name="creationtoolversion">0.94</xsl:attribute>
         <xsl:attribute name="segtype">sentence</xsl:attribute>
         <xsl:attribute name="o-tmf">application/vnd.oasis.opendocument.spreadsheet</xsl:attribute>
-        <xsl:attribute name="adminlang"><xsl:value-of select="$adminlang"></xsl:value-of></xsl:attribute>
-        <xsl:attribute name="srclang"><xsl:value-of select="$srclang"></xsl:value-of></xsl:attribute>
+        <xsl:attribute name="adminlang">
+          <xsl:value-of select="$adminlang"/>
+        </xsl:attribute>
+        <xsl:attribute name="srclang">
+          <xsl:value-of select="$srclang"/>
+        </xsl:attribute>
         <xsl:attribute name="datatype">plaintext</xsl:attribute>
       </header>
+
+      <!-- Define the TMX body
+           Additional templates are called to make the code
+           easier to modify and maintain. -->      
       <body>
-        <xsl:for-each select="//table:table-row[position()>1]">
-          <tu>
-            <xsl:for-each select="table:table-cell">
-              <xsl:choose>
-                <xsl:when test="position()=1">
-                  <tuv>
-                    <xsl:attribute name="xml:lang"><xsl:value-of select="$srclang"></xsl:value-of></xsl:attribute>                  
-                    <seg><xsl:value-of select="text:p"/></seg>
-                  </tuv>
-                </xsl:when>
-                <xsl:when test="position()=2">
-                  <tuv>
-                    <xsl:attribute name="xml:lang"><xsl:value-of select="$tgtlang"></xsl:value-of></xsl:attribute> 
-                    <seg><xsl:value-of select="text:p"/></seg>
-                  </tuv>
-                </xsl:when>
-              </xsl:choose>
-            </xsl:for-each>
-          </tu>
-        </xsl:for-each>
+        <!-- Call the template to insert <tu> elements -->
+        <xsl:apply-templates select="//table:table"/>
       </body>
     </tmx>
   </xsl:template>
+
+  <!-- Templates to populate the <tu> and <tuv> elements-->
+  
+  <!-- Template for <tu> elements:
+       Loop through each row, skipping the first one. -->
+  <xsl:template match="table:table">
+    <xsl:for-each select="table:table-row">
+          <tu>
+        <xsl:apply-templates select="//table:table-row[position()>1]"/>
+      </tu>
+    </xsl:for-each>
+  </xsl:template>
+
+  <!-- Template for the <tuv> elements:
+       Loop through each cell in the row, and assign the value
+       of the language for that column to the xml:lang attribute.
+       After that, add the <seg> element and populated it with the
+       text in the current cell. -->
+  <xsl:template match="table:table-row[position()>1]">
+            <xsl:for-each select="table:table-cell">
+      <xsl:variable name="currentLang"
+        select="//table:table/table:table-row[1]/table:table-cell"/>
+      <xsl:variable name="currentColumn"
+        select="position()"/>
+      <xsl:if test="normalize-space(text:p) != ''">
+                  <tuv>
+          <xsl:attribute name="xml:lang">
+            <xsl:value-of select="$currentLang[$currentColumn]/text:p"/>
+          </xsl:attribute>
+          <seg>
+            <xsl:value-of select="text:p"/>
+          </seg>
+                  </tuv>
+      </xsl:if>
+        </xsl:for-each>
+  </xsl:template>
+
 </xsl:stylesheet>

filtres_TMX_pour_Calc/TMX-import.xsl

@@ -4,7 +4,10 @@
      XSLT transformation of a TMX translation memory exchange file
      into an Open Document Format spreadsheet in two columns.
      Filter to install as an import filter for LibreOffice Calc. -->
-<!-- Copyright 2013 Dominique Meeùs.
+<!-- Philippe Tourigny, modified 12-9-2022, version 0.99
+     Allow the filter to retrieve the languages in the TMX from its
+     first <tu> element, and create a column for each language. -->
+<!-- Copyright 2013 Dominique Meeùs, and 2022 Philippe Tourigny.
      This program is free software: you can redistribute it and/or modify it
      under the terms of the GNU Lesser General Public License
      as published by the Free Software Foundation,
@@ -42,65 +45,76 @@
       office:version="1.0">
 
       <office:automatic-styles>
-        <!-- Properties of the table -->
+        <!-- PTable properties -->
         <style:style style:name="ta1" style:family="table" style:master-page-name="Default">
           <style:table-properties table:display="true" style:writing-mode="lr-tb"/>
         </style:style>
 
-        <!-- Properties of the columns -->
-        <!-- I consider the case of a two-languages TMX -->
-        <style:style style:name="co1" style:family="table-column"><!-- source language -->
-          <style:table-column-properties fo:break-before="auto" style:column-width="20.000cm"/>
-        </style:style>
-        <style:style style:name="co2" style:family="table-column"><!-- target language -->
-          <style:table-column-properties fo:break-before="auto" style:column-width="20.000cm"/>
+        <!-- Column properties (for all languages) -->
+        <style:style style:name="co1" style:family="table-column">
+          <style:table-column-properties fo:break-before="auto" style:column-width="14.000cm"/>
         </style:style>
 
-        <!-- Properties of the rows -->
-        <!-- The rows are “optimal height” but do not expand for the “wrap option” of the cells -->
+        <!-- Row properties -->
+        <!-- All rows are set to “optimal height” -->
         <style:style style:name="ro1" style:family="table-row">
           <style:table-row-properties fo:break-before="auto" style:use-optimal-row-height="true"/>
         </style:style>
 
-        <!-- Properties of the cells -->
-        <style:style style:name="ce1" style:family="table-cell" style:parent-style-name="Default">
-          <style:table-cell-properties fo:wrap-option="wrap"/>
+        <!-- Cell properties -->
+        <!-- Language code heading cells
+             The language codes are centered and set in bold
+             in the first column. -->
+        <style:style style:name="heading" style:family="table-cell"
+          style:parent-style-name="Default">
+          <style:table-cell-properties style:text-align-source="fix"
+            style:repeat-content="false" fo:wrap-option="wrap"/>
+          <style:paragraph-properties fo:text-align="center"/>
+          <style:text-properties fo:font-weight="bold"/>
         </style:style>
+
+        <!-- Style for cells with the segment text -->
         <style:style style:name="ce2" style:family="table-cell" style:parent-style-name="Default">
           <style:table-cell-properties fo:wrap-option="wrap"/>
         </style:style>
       </office:automatic-styles>
 
-        <!-- Get the source and target languages from the TMX file -->
-      <xsl:variable name="srclang" select="tmx/body/tu/tuv[1]/@*[local-name() = 'lang']"></xsl:variable>
-      <xsl:variable name="tgtlang" select="/tmx/body/tu/tuv[2]/@*[local-name() = 'lang']"></xsl:variable>
+      <!-- Define variables used to identify the languages
+           In a TMX with three or more languages. All translation unit
+           (<tuv>) elements are assumed to contain the same number of
+           languages, and the first <tu> is used to identify them. -->
+      
+      <!-- Todo: Identify the <tu> with the largest highest number of <tuv> elements to identify all languages in a TMX file with more languages in some <tu> elements than others. -->
+      <xsl:variable name="firstTU" select="tmx/body/tu[1]"/>
+      <xsl:variable name="numLangs" select="count($firstTU/tuv)"/>
 
       <office:body>
         <office:spreadsheet>
-          <table:table>
-            <!-- Format of the columns -->
-            <!-- How about a free number of columns ? -->
-            <table:table-column table:style-name="co1" table:default-cell-style-name="ce1"/>
-            <table:table-column table:style-name="co2" table:default-cell-style-name="ce2"/>
+            <table:table table:style-name="ta1">
+            <!-- Set the format for a number of columns equal to
+                 the number of languages in the imported TMX file -->
+            <table:table-column table:style-name="co1" table:number-columns-repeated="{$numLangs}"
+                  table:default-cell-style-name="ce2"/>
 
-            <!-- Put the source and target languages retrieved from
-                 the TMX file in the first two columns of the first row -->
-            <table:table-row>
-              <table:table-cell>
+                <!-- Fill in the language headers in the first row
+                     The use of the "local-name()" function enables
+                     the filter to handle older versions that use the
+                     "lang" attribute as well as recent versions that
+                     use the "xml:lang"` attributes -->
+                <table:table-row table:style-name="ro1">
+                  <xsl:for-each select="$firstTU/tuv">
+                    <table:table-cell table:style-name="heading">
                 <text:p>
-                  <xsl:value-of select="$srclang"></xsl:value-of>
-                </text:p>
-              </table:table-cell>
-              <table:table-cell>
-                <text:p>
-                  <xsl:value-of select="$tgtlang"></xsl:value-of>
+                        <xsl:value-of select="@*[local-name()='lang']"/>
                 </text:p>
               </table:table-cell>
+                  </xsl:for-each> 
             </table:table-row>
 
-            <!-- Process XML of the input TMX file: one row for each tu, one cell for the segment in each tuv -->
+                <!-- Process the <tu> and <tuv> elements in the TMX file:
+                     One row per tu, one column per segment in each <tuv>. -->
             <xsl:for-each select="tmx/body/tu">
-              <table:table-row>
+              <table:table-row  table:style-name="ro1">
                 <xsl:for-each select="tuv">
                   <table:table-cell>
                     <text:p><xsl:value-of select="seg"/></text:p>

filtres_TMX_pour_Calc/TMX-readme.txt

@@ -4,7 +4,8 @@ Modified 28-11-2019 as 0.91 for the position of the command
 to install in new versions of LibreOffice.
 (If my surname is not shown correctly, mind the fact that this file is encoded in UTF-8.)
 (On would have written Meeùs in old HTML entities for older charachters encodings.)
-Copyright 2013 Dominique Meeùs, see below.
+Updated to version 0.99 on 12-9-2022 by Philippe Tourigny (kazephil@gmail.com) to reflect the addition of new functionality and changes to the limitations.
+Copyright 2013 Dominique Meeùs, and 2022 Philppe Tourigny. See below.
 
 The TMX-filters software provides two XSLT filters
 to import or export TMX translation memories
@@ -14,27 +15,25 @@ Installation
 ============
 The package TMX-filters-vx_y.jar has to be installed in LibreOffice
 by the standard dialog from the menu Tools,
-  command Macros > XML Filter Settings…
-  [command directly XML Filter Settings… in older versions]
+  Macros > XML Filter Settings…
+  [or just XML Filter Settings… in older versions]
 Click the button Open Package… and browse to the package.
 
-Limitation
-==========
-Source language nl-BE (Dutch) and target fr-BE (French)
-are hardcoded in the export filter.
-For other languages you have to search and replace these in the TMX
-or edit the export filter.
-Feel free to e-mail me if you think of a way to improve this.
+Limitations
+===========
+For the export filter, enter the appropriate language code in each column of the first row. The source language must be entered in column A, and the target language in column B is used as the administrative language for the translation memory. If you wish to use an administrative language other than one of your target languages, you will have to edit either the resulting TMX file or the code of the filter.
+The import filter assumes that all <tu> elements in the file have the same number of languages, in the same order, and retrieves the language code from the <tuv> elements in the first <tu>. This means that importing a TMX containing <tu> elements with more <tuv> elements than others will result in some rows having additional entries in a column without a header, and with some languages entered in the wrong column.
+Feel free to e-mail us if you think of a way to improve this.
 
 Use
 ===
 — To read a TMX in LibreOffice Calc: File, Open…
-— To produce a TMX sentences aligned in two columns in LibreOffice Calc,
+— To produce a TMX sentences aligned in two or more columns in LibreOffice Calc,
 File, Save as… and choose type TMX (likely all the way down a long list).
 
 License
 =======
-Copyright 2013 Dominique Meeùs.
+Copyright 2013 Dominique Meeùs and 2022 Philippe Tourigny.
 This program is free software: you can redistribute it and/or modify it
 under the terms of the GNU Lesser General Public License
 as published by the Free Software Foundation,