Why did I get so many errors when importing SDLXLIFF files into a Studio TM?
Thread poster: Fredrik Pettersson

Fredrik Pettersson  Identity Verified
Hong Kong
Member (2009)
English to Swedish
+ ...
Aug 1, 2015

Why did I get so many errors when importing SDLXLIFF files into a Studio TM?

The SDLXLIFF files originally come from memoQ, and my customer created them from there. Then I created an empty Studio TM and imported these SDLXLIFF files to that empty TM, because earlier my customer had tried to create a TMX export from memoQ which in the end turned out to be empty on translation units (nothing in it) according to SDL Support.

So my last resort now to get this memoQ project to work in my SDL Trados Studio 2014 is to import all the 23 SDLXLIFF files from memoQ into my Studio TM and have a working TM that way instead of using TMX import.

But it doesn't look well now, so I don't know what else remains to try. The analysis after having imported these SDLXLIFF files shows on 15,600 new words (should be around 12,000 new words).


Direct link Reply with quote
 

John Fossey  Identity Verified
Canada
Local time: 03:51
Member (2008)
French to English
Differences in segmentation Aug 1, 2015

AS I mentioned at http://www.proz.com/forum/sdl_trados_support/288973-why_does_the_tms_exported_from_memoq_get_disabled_and_cant_be_activated_in_studio_2014.html, memoQ and SDL Studio segment SDLXLIFF files slightly differently, so there will be a certain number of errors. Ask your client to send a TMX file as well - you might be able to import some of the missing segments with it.

Direct link Reply with quote
 

SDL Community  Identity Verified
United Kingdom
Local time: 09:51
English
The main reason for this... Aug 1, 2015

Fredrik Pettersson wrote:

Why did I get so many errors when importing SDLXLIFF files into a Studio TM?



... and they may not really be true errors, is that if the sdlxliff contain segments with no target translation, and duplicates, then Studio will automatically exclude them and report this as an information error.

I also don't see the point of using the sdlxliff files you have for tranlsation as the basis of a TM. You already have them populated so what's the point of creating a TM from them?

The analysis difference is most likely because you are now analysing against an empty TM in Studio, whereas in memoQ there is a TM, so of course the new words list goes up. At least from your description this is what it all sounds like.

I'd recommend you speak to your client again and try to get them to have another go at creating the TM from memoQ.

Regards

Paul
SDL Community Support


Direct link Reply with quote
 

Fredrik Pettersson  Identity Verified
Hong Kong
Member (2009)
English to Swedish
+ ...
TOPIC STARTER
No need for TM now Aug 1, 2015

You are right Paul, thanks. The SDLXLIFF files are already populated as you said. So I just create a new TM to get hits from the translations I do from now on myself.

Direct link Reply with quote
 

Fredrik Pettersson  Identity Verified
Hong Kong
Member (2009)
English to Swedish
+ ...
TOPIC STARTER
But why are there 15,600 new words now? Aug 1, 2015

Now when these SDLXLIFF files are populated, why do I still get 15,600 new words when there should be only around 12,000? Shouldn't these be reported as 100 % match, these segments that have been populated from the memoQ TM?

Direct link Reply with quote
 

Fredrik Pettersson  Identity Verified
Hong Kong
Member (2009)
English to Swedish
+ ...
TOPIC STARTER
I definitely need the TMX file Aug 1, 2015

I still would need the TMX file even though I have these pre-translated SDLXLIFF files because otherwise I get no fuzzy match hits and I can't make Concordance searches for terms in the TM.

Direct link Reply with quote
 

Stepan Konev  Identity Verified
Russian Federation
Local time: 11:51
English to Russian
why do I still get 15,600 new words when there should be only around 12,000? Aug 1, 2015

When you open in Studio an sdlxliff file previously translated in memoQ (I guess, this is how you got those populated segments - by translating them in memoQ, right?), the status of 100% translated segments from memoQ appears as 'confirmed' (a pencil with a green tick) in Studio.
However the translation does not go to TM. As a result, you see some (or most) segments confirmed, but if you click on one of them, you see 'No matches found' in the Translation Results section. So, they are confirmed but not in TM. That is why your file statistics does not change.
Possible soultion:
1. Filter segments using the 'New translated content' option.
2. Look though the result to make sure that only 100% confirmed segments returned (to avoid adding fuzzy matches into TM). This step is just to be on the safe side. AFAIK, the 'New translated content' filtering will not return fuzzy matches.
3. Using the left mouse button, select all segments in the Target column. If there are too many segments to scroll through manually, then click at the beginning of the first segment and then press Ctrl+Shit+End.
4. Now you have all segments (100% translated in memoQ and appearing confirmed in Studio) selected.
5. Press Ctrl+Enter. This will make the confirmed segments go to TM.
*If you have thousands of segments, this may take up to several minutes. So be patient.
Hope this helps.

However, try to import a tmx from memoQ TM into Studio TM with 'Use different translation provider...' first (http://www.proz.com/forum/sdl_trados_support/288973-why_does_the_tms_exported_from_memoq_get_disabled_and_cant_be_activated_in_studio_2014.html). This is much easier solution.


Direct link Reply with quote
 

SDL Community  Identity Verified
United Kingdom
Local time: 09:51
English
because you have no TM... Aug 1, 2015

Fredrik Pettersson wrote:

Now when these SDLXLIFF files are populated, why do I still get 15,600 new words when there should be only around 12,000? Shouldn't these be reported as 100 % match, these segments that have been populated from the memoQ TM?


... you won't get an analysis without a TM!

Regards

Paul
SDL Community Support


Direct link Reply with quote
 

SDL Community  Identity Verified
United Kingdom
Local time: 09:51
English
If you really want this... Aug 1, 2015

Fredrik Pettersson wrote:

I still would need the TMX file even though I have these pre-translated SDLXLIFF files because otherwise I get no fuzzy match hits and I can't make Concordance searches for terms in the TM.


... and you have the sdlxliff files that are basically the same as the TM in memoQ then just create a new TM for your project and update the sdlxliff files into your TM.

Regards

Paul
SDL Community Support


Direct link Reply with quote
 

Fredrik Pettersson  Identity Verified
Hong Kong
Member (2009)
English to Swedish
+ ...
TOPIC STARTER
Filter by 100 % match instead Aug 3, 2015

Stepan,

I think your solution will work. But when I opened all the files in the whole project all at the same time (opened them virtually), and got all segments in the whole project displayed, and then filtered by New contents only a few segments were displayed (not all 100 % matches).

These SDLXLIFF files from memoQ, they already contained a lot of confirmed 100 % matches.

So I guess I should filter by 100 % match instead. Apart from that, I will try now and follow your other instructions.


Direct link Reply with quote
 

Fredrik Pettersson  Identity Verified
Hong Kong
Member (2009)
English to Swedish
+ ...
TOPIC STARTER
How to deal with the locked segments? Aug 3, 2015

There are more than 60 % locked 100 % match also, and I think these won't get into my newly created translation memory by selecting all segments and confirming. Do I have to unlock these segments first in order to be able to get them into the TM?

Direct link Reply with quote
 

Fredrik Pettersson  Identity Verified
Hong Kong
Member (2009)
English to Swedish
+ ...
TOPIC STARTER
I have a working TM now with all TM segments imported, but still get wrong analysis Aug 3, 2015

I received two TMX files that I imported into 2 separate TMs for this project, and they contained more than 100,000 segments. But still when I added these TMs in the project settings (not All language pairs but Use specific ... for this language pairs), I get wrong wordcount (14,965 instead of around 12,300 new words). Why?

In the provided 23 SDLXLIFF files, a lot of segments have been committed by the customer already (in memoQ), so these are 100 % match (most of them locked segments, which must remain locked, I asked the customer if I could open them so I can commit them, but I can't that).

Should I just ignore the wordcount and start translating now? Is the discrepancy due to the fact that the 23 SDLXLIFF files already have been filled with translated segments from the customer's TMs? So that even though the wordcount displayed 14,965 new words to translate remains, in reality there are only 12,300 new remaining words to translate?


Direct link Reply with quote
 


To report site rules violations or get help, contact a site moderator:


You can also contact site staff by submitting a support request »

Why did I get so many errors when importing SDLXLIFF files into a Studio TM?

Advanced search







SDL MultiTerm 2017
Guarantee a unified, consistent and high-quality translation with terminology software by the industry leaders.

SDL MultiTerm 2017 allows translators to create one central location to store and manage multilingual terminology, and with SDL MultiTerm Extract 2017 you can automatically create term lists from your existing documentation to save time.

More info »
LSP.expert
You’re a freelance translator? LSP.expert helps you manage your daily translation jobs. It’s easy, fast and secure.

How about you start tracking translation jobs and sending invoices in minutes? You can also manage your clients and generate reports about your business activities. So you always keep a clear view on your planning, AND you get a free 30 day trial period!

More info »



Forums
  • All of ProZ.com
  • Term search
  • Jobs
  • Forums
  • Multiple search