Richard Williams

Which version has your tmx file? Is it valid? Are the language codes set properly? And on top of this: what kind of error message do you get?

Sorry, I do not know what you mean by version. It comes from another translation programme - Linguatec Personal Translator. I always saved my archives in PT and tmx formats for exporting to other programmes. Do you mean by language codes Ausgangstext/Zieltext? If so, German/English. The message is: Import fehlgeschlagen. Die Datei konnte nicht verarbeitet werden. 

Where? If I choose Translation memory in SDL 2021 the only option is sdltm. 

That’s because you are confusing Open with Import!

I do not use open with import when creating a new project as this option is inactive and becomes available only after a sdltm file has been chosen. So this question is not solved. All I want is using a tmx file that has been exported from e.g. MemoQ or any other translation program, which is used as the only memory in SDL without any import. In other words SDL should convert it to a sdltm file so that it can be used during translation.    

Tjasa Kuerpick

Tjasa Kuerpick said:

So this question is not solved.

You're right.  It's not solved because the user has still not responded. The question was how he imports his TMX into Studio because he has a problem with the file.  You responded and gave him some reasonable advice on how to edit the TMX to achieve this, but then you made a comment that is completely incorrect.  You said:

Tjasa Kuerpick said:

Please note that the 2021 version of SDL does not support the import of a TMX file any more

This is incorrect.  It absolutely does does support the import of a TMX and always has done.

It's now clear that you are not importing at all.  You are simply using the TMX as your TM and converting it to an SDTM on the fly using the quick upgrade feature.  This capability was indeed removed in 2021.  I think it was a mistake to remove it, but nonetheless you can get this feature back by installing this app:

https://appstore.sdl.com/language/app/sdl-trados-compatibility-and-migration-power-pack/1101/

You can read why this was removed here:

https://multifarious.filkin.com/2020/08/15/some-you-win-some-you-lose/

So, if you have some other question you want answered I suggest you create a new thread so we can mark this onet as resolved if we manage to do so when the user responds.

Sorry for the delay. What I am now hoping to learn is how to use Olifant to enable me to export my Linguatec tmx file into a Trados tmx file. Do you know how to do this?

Richard Williams

Richard Williams said:

What I am now hoping to learn is how to use Olifant to enable me to export my Linguatec tmx file into a Trados tmx file. Do you know how to do this?

I wouldn't even entertain this route without seeing the file first.  Can you provide the details I asked for earlier?

The most common problem with importing TMX files, if the file is well formed and a valid TMX, is the information in the header.  Can you use a text editor to delete all the contents of your TMX apart from the header and the first translation unit, then share with us?  Like this for example:

<?xml version="1.0" encoding="utf-8"?>
<tmx version="1.4">
  <header creationtool="SDL Language Platform" creationtoolversion="8.1" o-tmf="SDL TM8 Format" datatype="xml" segtype="sentence" adminlang="en-US" srclang="en-US" creationdate="20170918T092759Z" creationid="GLOBAL\pfilkin">
    <prop type="x-Model:SinglePicklist">Aston Martin,Austro-Daimler,Bugatti,Duesenberg,Ferrari,Horse Carriage,Jaguar,McLaren,Mercedes,Shelby,Stanley</prop>
    <prop type="x-Year:Integer"></prop>
    <prop type="x-Quality:Integer"></prop>
    <prop type="x-SourceFile:SingleString"></prop>
    <prop type="x-TargetFile:SingleString"></prop>
    <prop type="x-Recognizers">RecognizeAll</prop>
    <prop type="x-IncludesContextContent">True</prop>
    <prop type="x-TMName">en(US) - de(DE) {Fastest Production Cars}</prop>
    <prop type="x-TokenizerFlags">DefaultFlags</prop>
    <prop type="x-WordCountFlags">DefaultFlags</prop>
  </header>
  <body>
    <tu creationdate="20170918T093522Z" creationid="GLOBAL\pfilkin" changedate="20170918T093844Z" changeid="GLOBAL\pfilkin" lastusagedate="20200817T084158Z" usagecount="1">
      <prop type="x-LastUsedBy">GLOBAL\pfilkin</prop>
      <prop type="x-Context">0, 0</prop>
      <prop type="x-Origin">TM</prop>
      <prop type="x-ConfirmationLevel">Translated</prop>
      <prop type="x-Year:Integer">1880</prop>
      <prop type="x-Model:SinglePicklist">Horse Carriage</prop>
      <tuv xml:lang="en-US">
        <seg>I know, I know, this is not a car, it’s a horse-drawn carriage — but we needed a reference point.</seg>
      </tuv>
      <tuv xml:lang="de-DE">
        <seg>Ich weiß, ich weiß, das ist kein Auto, sondern eine Pferdekutsche - aber wir brauchten einen Bezugspunkt.</seg>
      </tuv>
    </tu>
  </body>
</tmx>

I wanted to send a text file with an example, but I got the following message: Leider kann das Dokument x nicht geöffnet werden, da sein Inhalt Probleme verursacht.

Details: Ungültiges Unicode-Zeichen

Position: Zeile 566623, Spalte 112 

Which "Dokument"? File name extension? Where did you try to open the document? And the text says clearly, that your xml is invalid, bad character in line 566623 column 112. Open the file in Notepad++ and go to this place, remove the offending character and try again.

Richard Williams

Can you send it anyway and we can sort it out.

Sorry, I answered your email. That obviously doesn't work. Thank you for your clear advice. I found the line and column, but how can I remove the character, which is invisible. Should I just delete the whole TU? Is the TM then valid or whatever? It is a large archive and would presumably help me a lot, if I could transfer it into Trados.

My main problem is actually as follows: I have my own system, which has worked well for years. It is a little complicated to explain, but the problem boils down to this: I would like ideally to continue to use my method and just align the source and target files to produce an XLIFF file (in Trados or elsewhere) . This procedure seems to be cumbersome and time-consuming in Trados. Three-quarters of the segments don't match, so I would have to spend far too much time aligning them. Any suggestions? I would be most grateful, as I may lose my main client. If you can propose a solution, I would reward you with 12 bottles of the best Riesling in Germany.        

Sorry, I answered your email. That obviously doesn't work. Thank you for your clear advice. I found the line and column, but how can I remove the character, which is invisible. Should I just delete the whole TU? Is the TM then valid or whatever? It is a large archive and would presumably help me a lot, if I could transfer it into Trados.

My main problem is actually as follows: I have my own system, which has worked well for years. It is a little complicated to explain, but the problem boils down to this: I would like ideally to continue to use my method and just align the source and target files to produce an XLIFF file (in Trados or elsewhere) . This procedure seems to be cumbersome and time-consuming in Trados. Three-quarters of the segments don't match, so I would have to spend far too much time aligning them. Any suggestions? I would be most grateful, as I may lose my main client. If you can propose a solution, I would reward you with 12 bottles of the best Riesling in Germany.