Forum: SDL Trados support
Topic: Studio 2011: no blanket "copy source" for non-alpha segments?
Poster: Andrew052
Post title: the question to SDL support
[quote]PapagenoX wrote:
[quote]SDL Support wrote:
... the first is to use a regex in the display filter to select only the segments that contain number only segments and then copy source to target for everything displayed.
The second is to do something similar using the SDLXLIFF Toolkit from the openexchange, but this time you can do the entire Project in one go without having to merge the files together first.
SDLXLIFF Toolkit : [url removed]
Regards
Paul [/quote]
Hi Paul, thanks for the reply, but I'm unclear on what a "regex" is, although I have located the display filter, so I'll go from there.
[Edited at 2014-02-10 23:17 GMT] [/quote]
Hi,
1) I would be grateful, if you explain, how "regex" works, as I've never heard about it before.
2) You have said something about "mering the files together" -- what did you mean. I am really interested in trying to find new ways to tackle this issue and if this "merging process" does something to make it quicker to work with "number-only" segments or range the segments alphabetically in trados, I would be grateful to find you more about it.
Asker, Please find below the way I fist prepare the TM in memo and pre-translate the file in the Trados then on the basis of the memo TM:
I regularly have to deal with the same issue.
I receive a file that has a table (50-60 pages in the file). The table contains the names of the spare parts, the item numbers, the systems in which these spare parts are used and descriptive info.
the names of the spare parts, the systems are repeated through the file (like there can be 5 systems, repeated gazzilion times per 100-page file)
the item numbers are just either 1) numbers (356234) or 2) numbers + letters (53T35TSJ)
I have asked around when I started receiving such files and people say they just make do with what you have in Trados.
But, in Trados -- in terms of display filters -- you cannot do thing that memo does:
place the segment alphabetically.
Why is it important: 1) all the item numbers will be placed together and you will see, where the numbers segments finish and the text segments start. So you just mark the numbers segments and approve them all at once. 2) you filter out the number segments and have only text segments. Now, you can find out, for instance, that there are 50 segments like "bushing for system 1", 50 segments like "bushing for system 2" yada yada yada. (working in trados, you would, of course, have to translate only once "bush for system 1" to automatically translate all the same segments, but the benefit in memo is that you don't have to look up the name of the system in the termbase each time you have "bushing for system 2". Also, if you have even slight variations, for instance, "bushing for system 2""bushing for system No. 2", "bushing for system Number 2", you will have to look up the previous translation in the matches window, but in memo you will see the following list:
bushing for system 2
bushing for system 2
bushing for system 2
bushing for system No. 2
bushing for system No. 2
bushing for system No. 2
bushing for system Number 2
bushing for system Number 2
bushing for system Number 2
bushing for system Number 2
this is because the segments are ranged alphabetically, and the difference between the segments starts only after the word "system", while the previous words are the same. Using memo you can quickly scroll thru such lists, mark them and approve'em.
Thus, you have created the TM with the 1) number segments, and 2) text segments (the "bushing for system 2" type). You can export and import the TM in Trados and pre-translate the file using the TM.
Benefits and Unbenefits Analysis:
Benefits:
1. clearly great with "bushing for system 2" segments
Unbenefits:
1. (I have a problem with spell check iin memo -- it just doesn't work.
2. The number segments will not at all times be incoporated in your trados file ("345452" segments MIGHT NOT IN ALL CASES BE INCORPORATED into your trados file), but if you have like 20K of stuff like 134GFDF44 segments, it's great to get rid of them from the start, because it is a drag to try and handle them in trados.
3. If you have to send back the trados file, you have to go back to Trados anyhow...
But, I am sure that if you a file that has like 20K of the same, but slightly different segments memo can help a lot.
P.S.
Memo also enabled to range the segments by the number of symbols in a segment. This can also be extremely helpful in some cases.
I would like to ask others to contribute the ways they handle such translations.
Topic: Studio 2011: no blanket "copy source" for non-alpha segments?
Poster: Andrew052
Post title: the question to SDL support
[quote]PapagenoX wrote:
[quote]SDL Support wrote:
... the first is to use a regex in the display filter to select only the segments that contain number only segments and then copy source to target for everything displayed.
The second is to do something similar using the SDLXLIFF Toolkit from the openexchange, but this time you can do the entire Project in one go without having to merge the files together first.
SDLXLIFF Toolkit : [url removed]
Regards
Paul [/quote]
Hi Paul, thanks for the reply, but I'm unclear on what a "regex" is, although I have located the display filter, so I'll go from there.
[Edited at 2014-02-10 23:17 GMT] [/quote]
Hi,
1) I would be grateful, if you explain, how "regex" works, as I've never heard about it before.
2) You have said something about "mering the files together" -- what did you mean. I am really interested in trying to find new ways to tackle this issue and if this "merging process" does something to make it quicker to work with "number-only" segments or range the segments alphabetically in trados, I would be grateful to find you more about it.
Asker, Please find below the way I fist prepare the TM in memo and pre-translate the file in the Trados then on the basis of the memo TM:
I regularly have to deal with the same issue.
I receive a file that has a table (50-60 pages in the file). The table contains the names of the spare parts, the item numbers, the systems in which these spare parts are used and descriptive info.
the names of the spare parts, the systems are repeated through the file (like there can be 5 systems, repeated gazzilion times per 100-page file)
the item numbers are just either 1) numbers (356234) or 2) numbers + letters (53T35TSJ)
I have asked around when I started receiving such files and people say they just make do with what you have in Trados.
But, in Trados -- in terms of display filters -- you cannot do thing that memo does:
place the segment alphabetically.
Why is it important: 1) all the item numbers will be placed together and you will see, where the numbers segments finish and the text segments start. So you just mark the numbers segments and approve them all at once. 2) you filter out the number segments and have only text segments. Now, you can find out, for instance, that there are 50 segments like "bushing for system 1", 50 segments like "bushing for system 2" yada yada yada. (working in trados, you would, of course, have to translate only once "bush for system 1" to automatically translate all the same segments, but the benefit in memo is that you don't have to look up the name of the system in the termbase each time you have "bushing for system 2". Also, if you have even slight variations, for instance, "bushing for system 2""bushing for system No. 2", "bushing for system Number 2", you will have to look up the previous translation in the matches window, but in memo you will see the following list:
bushing for system 2
bushing for system 2
bushing for system 2
bushing for system No. 2
bushing for system No. 2
bushing for system No. 2
bushing for system Number 2
bushing for system Number 2
bushing for system Number 2
bushing for system Number 2
this is because the segments are ranged alphabetically, and the difference between the segments starts only after the word "system", while the previous words are the same. Using memo you can quickly scroll thru such lists, mark them and approve'em.
Thus, you have created the TM with the 1) number segments, and 2) text segments (the "bushing for system 2" type). You can export and import the TM in Trados and pre-translate the file using the TM.
Benefits and Unbenefits Analysis:
Benefits:
1. clearly great with "bushing for system 2" segments
Unbenefits:
1. (I have a problem with spell check iin memo -- it just doesn't work.
2. The number segments will not at all times be incoporated in your trados file ("345452" segments MIGHT NOT IN ALL CASES BE INCORPORATED into your trados file), but if you have like 20K of stuff like 134GFDF44 segments, it's great to get rid of them from the start, because it is a drag to try and handle them in trados.
3. If you have to send back the trados file, you have to go back to Trados anyhow...
But, I am sure that if you a file that has like 20K of the same, but slightly different segments memo can help a lot.
P.S.
Memo also enabled to range the segments by the number of symbols in a segment. This can also be extremely helpful in some cases.
I would like to ask others to contribute the ways they handle such translations.