Source segments that are in the target language - how to automatically put all in target
Thread poster: Scott Saylor

Scott Saylor  Identity Verified
Local time: 03:26
Japanese to English
Mar 21, 2011

icon_confused.gificon_confused.gificon_confused.gificon_confused.gif
Hello,

TRADOS 2009 question.icon_confused.gif

I have a Japanese source where some of the segments are consecutive numbers like

GFT100
GFT101
GFT102

and then sometimes the first numbers change (like from GFT in the example above to something like ABC):

ABC100
ABC101
ABC102

Number localization is not catching this. Everytime I paste one of these to the target (example GFT101), then everything thereafter just becomes GFT101.

I would like to know three things, listed below in order of priority for me.

1) How can I have English text that is in the source be copied to the target automatically for all instances of exclusively English text that is in the source segments.

2) How can I fix number localization so that it works in a manner that is more helpful. (like when I have 1月 2月 3月 4月 I want Trados to change those to January, February, March, April. or at least not having every month thereafter become the month I just typed (ex: I get to 2月 and type February, and then 3, 4, 5, 6月 -> 12 月 become February.

3) This one drives me crazy:icon_evil.gif when I verify numbers, literally almost every segment gets a big red X next to it even if its right. This makes verification very ineffective and time consuming for me.

Any help would be GREATLY appreciated. Thank you!icon_cool.gif

Scott

[Edited at 2011-03-21 21:51 GMT]

[Edited at 2011-03-21 21:52 GMT]

[Edited at 2011-03-21 21:52 GMT]


 

Emma Goldsmith  Identity Verified
Spain
Local time: 12:26
Member (2010)
Spanish to English
A couple of suggestions Mar 21, 2011

I think I can answer 2 questions:
1. Alt+shift+ins copies all source to target. Maybe that would help? Or Ctrl+alt+K to copy an individual segment to the target.
3. Tools>options>verification>QA checker>numbers. Change Check numbers from error to warning or note, or just uncheck the box if you don't want any warning at all.

Not sure about your 2nd question, so I hope someone else can help on that one.

[Edited at 2011-03-21 22:31 GMT]


 

Scott Saylor  Identity Verified
Local time: 03:26
Japanese to English
TOPIC STARTER
thank you for your answers. Mar 21, 2011

Hi Emma, thank you for your answers.

What I am trying to do is have a swooping, all-encompasing, all-at-once transfer of all source segments in the document that are in English only to go over to the target. However, I don't want all the Japanese to be copied over to the target, if possible.

As for the warning, error, note suggestion, thats a really good feature that I like, but it doesn't really accomplish what I am trying to accomplish: accurate verification of numbers in the document. I want the program to check all of the numbers in the document and come back with the correct segments that have errors. Not 365 units as wrong, when 363 are absolutely fine and 2 were missing a number.

I am curious about the Alt+shift+ins command, and I am going to explore that one.

I really appreciate your taking the time to help me!

Scott


 

FarkasAndras
Local time: 12:26
English to Hungarian
+ ...
Should be easy Mar 22, 2011

All you need to do is filter out the segments that you want to keep unchanged and then do a "copy source to target" on them.

Filtering:
- Enable the Display Filter toolbar in View/Toolbars
- Put your search term in the "Containing:" box. In your case it could be
[A-Z]{3}[0-9]{3}
to get segments made up of 3 letters plus 3 digits, or
^[^ ]$
to get segments that don't have a space character in them etc.

Then select all segments and copy source to target using the right-click context menu. Change the segment status as desired and remove the filter to get back to translating the rest of the segments.


 

Scott Saylor  Identity Verified
Local time: 03:26
Japanese to English
TOPIC STARTER
Wow! Mar 22, 2011

Hi Farkas,

Thank you very much for your response. I am very excited to learn how to use the filter function. This definitely seems like it will work. I will let you know when I try it for my next project (I already copied everything (including the Japanese) to the target for this one).

Thank you for your help!

Scott


 

Scott Saylor  Identity Verified
Local time: 03:26
Japanese to English
TOPIC STARTER
Wildcard expression for all English in the source? Mar 22, 2011

Hi Farkas,

Do you happen to know the wildcard expression for all English sentences in the source?

Scott


 

FarkasAndras
Local time: 12:26
English to Hungarian
+ ...
? Mar 22, 2011

Writing regex (regular expressions) is a bit of an art form.
You have to think about how you can best formulate what you want. You also need to think carefully about all the weird edge cases that could occur.
"All English sentences" is not a category that can be covered by a regex, you need to give a lot more information than that. What are the segments that you want to exclude?
If they are like "ABC100" and "GFT102" then one possible pattern is:
\b[a-z]+\b
which translates to [word boundary][any number of characters between a and z][word boundary]
or:
\b[^0-9]+\b
or a million other things depending on what your input data looks like.

Here's a regex tutorial.
Trados seems to mostly follow the perl regex syntax, which is brillinant, because it offers a widely used and immensely powerful toolset. To what extent Trados follows perl convention is anyone's guess, I don't think SDL ever published a full spec.

Some basics:
[] encloses a character class (group of characters)^. So [ert] matches e, r or t and [a-c] matches a, b or c [^a-c] means anything but a, b or c. [a-z] doesn't match á or ö.
* means "0 or more occurrences of"
+ means "1 or more occurrences of"
{5} means "5 occurrences of", {2,4} means "2-4 occurrences of", {4,} means "4 or more occurrences of"
^ anchors to the start of the segment, $ to the end
\b is a word boundary
\w is a "word-character", i.e. a letter, number or underscore
\W is a "non-word character"
\d is a digit, i.e. [0-9]
\D is [^0-9]
\s is whitespace

Based on the above, \d[as]{2,5}x matches 7ssx, 5asasax or 0aaaaax but not 6daax, 2sx or 8asb.
All of this is standard perl stuff and seems to work in Studio.

[Edited at 2011-03-22 23:20 GMT]


 

Scott Saylor  Identity Verified
Local time: 03:26
Japanese to English
TOPIC STARTER
Thank you Mar 24, 2011

Hi Farkas,

Thank you for your help on regular expressions / wildcards. You really know them well. You gave me a lot of really helpful hints and many new ideas to explore. Thanks for taking the time to write such a thoughtful and thorough response to my question. This is going to help me a lot, and I am sure it will help others, too. Thank you.

Scott


 

FarkasAndras
Local time: 12:26
English to Hungarian
+ ...
Welcome Mar 24, 2011

You're welcome. I think everyone should know a bit of regex syntax, so this is me teaching the world about regex one translator at a time. I guess I could proselytize for worse causes, right?

Incidentally, if anyone knows if and how Trados can paste matches back in with search and replace, do tell.
For instance, replacing the grouping separator in numbers should be something like:
replace (\d),(\d{3})
with \1 \2
or perhaps $1 $2

Neither of these works in Trados. Did they forget this basic feature?


 


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


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

Source segments that are in the target language - how to automatically put all in target

Advanced search







SDL Trados Studio 2019 Freelance
The leading translation software used by over 250,000 translators.

SDL Trados Studio 2019 has evolved to bring translators a brand new experience. Designed with user experience at its core, Studio 2019 transforms how new users get up and running and helps experienced users make the most of the powerful features.

More info »
TM-Town
Manage your TMs and Terms ... and boost your translation business

Are you ready for something fresh in the industry? TM-Town is a unique new site for you -- the freelance translator -- to store, manage and share translation memories (TMs) and glossaries...and potentially meet new clients on the basis of your prior work.

More info »



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