CafeTran

Following the discussion on the Newbie forum (but not posting it there, because this forum should stay 'clean' in my opinion):

How are alternative translations handled in TMs for Terms?

I mean:

In a glossary we have:

Haus<TAB>woning;huis;behuizing;kast

How does this looking a TMX file?

Underlying reason to ask:

If someone starts with tab-delimited glossaries, how can the conversion of alternative target terms be handled?

Of course I can investigate this myself, but perhaps someone knows it off the top of his head.

>How does this looking a TMX file?

How does this look in a TMX file?

(I'm working too long today, getting tired.)

Hmm, just did a quick test. Added:

Haus<TAB>woning;huis;behuizing;kast to my Glossary

Then did Memory > Import > Import entries from glossary

But only "kast" made it into the TMX, represented thus: the

<tuv xml:lang="nl-NL"><seg>Haus</seg></tuv><tuv xml:lang="en-GB"><seg>kast</seg></tuv>

Alternatives in a TMX are handled like this:

<tuv xml:lang="nl-NL"><seg>Haus</seg></tuv><tuv xml:lang="en-GB"><seg>kast</seg></tuv>

<tuv xml:lang="nl-NL"><seg>Haus</seg></tuv><tuv xml:lang="en-GB"><seg>behuizing</seg></tuv>

<tuv xml:lang="nl-NL"><seg>Haus</seg></tuv><tuv xml:lang="en-GB"><seg>huis</seg></tuv>

<tuv xml:lang="nl-NL"><seg>Haus</seg></tuv><tuv xml:lang="en-GB"><seg>huis</seg></tuv>

That is, a line into 4 lines.

It should be relatively trivial for Igor to change Memory > Import > Import entries from glossary, so those alternatives get sent across too, and CT splits them into several lines in the target TMX.

Would the TMX specification also allow:

<tuv xml:lang="nl-NL"><seg>Haus</seg></tuv>

<tuv xml:lang="en-GB"><seg>kast</seg></tuv>

<tuv xml:lang="en-GB"><seg>behuizing</seg></tuv>

<tuv xml:lang="en-GB"><seg>huis</seg></tuv>

<tuv xml:lang="en-GB"><seg>woning</seg></tuv>

This is most likely invalid, but you get the point.

Anyway, one must make sure that the TMX isn't set to only store Newer segments, I guess?