Hi,
we plan to translate the Localization export files with CAT tools, which is usually not a problem with XLIFF files.
However, if we export formatted text with the Gainsight Localization feature, the XLIFF files contain undesired segmentations that make working with a Translation Memory almost impossible.
Example: The engagement text “This is a sentence with a bold word” is segmented as follows in the XLIFF export:
<?xml version="1.0"?>
<xliff xmlns="urn:oasis:names:tc:xliff:document:2.0" version="2.0" srcLang="en" trgLang="de-DE">
<file id="1">
<unit id="thisisasentencewitha">
<segment>
<source>This is a sentence with a </source>
</segment>
</unit>
<unit id="bold">
<segment>
<source>bold</source>
</segment>
</unit>
<unit id="word">
<segment>
<source>word. </source>
</segment>
</unit>
</file>
</xliff>
This kind of segmentation causes a lot of difficulties and manual effort during translation: Either we merge the individual segments manually or we get wrong bilingual sentence pairs in our translation memories, depending on the syntax of the target language.
Why is the inline bold formatting, which is correctly represented by a <strong> </strong> tag pair in the source code in the Gainsight Editor, converted into a ‘unit id’ followed by a separate segment?
I would expect this sentence to appear like that:
<unit id="thisisasentencewithaboldword">
<segment>
<source>This is a sentence with a <strong>bold</strong> word.</source>
</segment>
</unit>
A structure like that could be easily processed by our CAT tool.
Is there a way to change these ‘segmentation rules’?
Maybe an option we have overlooked during configuration?
