The xliff I created from a date node opened in the Swordfish xliff editor, however the sources were all copied to targets and moving to another segment gave the following error:
Error saving current target: [Fatal Error] 1:95 The prefix "xmrk" for attribute "xmrk:ancs" associated with an element type "g" is not bound.

CommentFileSizeAuthor
node-58.xlf_.tar_.gz637 bytesmeatron

Comments

rmraya’s picture

rmraya commented

The "xmark" prefix indicates the use of a custom extension in the XLIFF file. You can add the XML schema where this extension is defined to Swordfish's XML catalog.

If the XML schema is available under a redistributable license (not GPL), send the schema to tech@maxprograms.com and it will be included in Swordfish's catalog in future releases.

meatron’s picture

meatron commented

Thanks Rodolfo, you really are under every stone! This module is early alpha as announced, I sincerely hope the development continues. For sure it would be a great enhancment to any multi-lingual site.

pasqualle’s picture

pasqualle
Location 🇭🇺 Budapest
commented
Version: 6.x-1.0-beta1 » 6.x-2.x-dev
Status: Active » Fixed

this might be fixed in 6.x-2.x

meatron’s picture

meatron commented

Testing 6.x-2.x. Nodes with photographs or flash objects don't convert. A simple node converted and Swordfish opened it. But it segmented by paragraphs, and sources copied to targets again.

pasqualle’s picture

pasqualle
Location 🇭🇺 Budapest
commented
Title: Swordfish error » Leave the translation (target) empty when exporting
Category: bug » feature
Status: Fixed » Active

- My guess would be that nodes which can not be exported are not valid HTML, as we worked with images without problem.
- The paragraph segmenting is the way how it works. I do not see what is the problem with that.
- renamed the issue to match the feature request (make target empty)

meatron’s picture

meatron commented

My guess would be that nodes which can not be exported are not valid HTML, as we worked with images without problem.

Yes, I think the problem is with image assist and swf code.

The paragraph segmenting is the way how it works. I do not see what is the problem with that.

Most translators are used to sentence segmenting

Great you are working on this module again, I appreciate it very much!

meatron’s picture

meatron commented

Image assist works, but the code is of course inside the xliff. Along with tags like for formatted text. It would be a nice feature if tags were in Swordfish format «1» etc. But I just tested it on several nodes, and it worked perfectly! Except I would still prefer sentence segmentation of course :P
Kudos and thanks for the work!

meatron’s picture

meatron commented

Along with tags like for formatted text

Oops the tag got eaten -
Along with tags like <g id="strong-11">for formatted text