Merging into a csv will not surround strings with a text qualifier

Issue #187 resolved
Former user created an issue

Original [issue 187](https://code.google.com/p/okapi/issues/detail?id=187) created by kucharczyk.lu... on 2011-09-19T13:59:18.000Z:

What steps will reproduce the problem? 1. Generate OmegaT package from a csv file with comma as a delimiter 2. Put a comma in a text segment in OmegaT 3. Merge it back

What is the expected output? What do you see instead? The resulting csv should have the value surrounded by a text qualifier as not to divide the string into two values. However, the resulting csv will not have the value surrounded by a text qualifier, thus resulting in separating the value into multiple values.

What version of the product are you using? On what operating system? Rainbow 6.0.1.3 on Windows 7 64/bit.

Please provide any additional information below. This can be amended by putting the surrounding quotes around the text manually in OmegaT, but I \_think\_ that this should be accounted for during the merging process.

Comments (12)

  1. Former user Account Deleted

    Comment [4.](https://code.google.com/p/okapi/issues/detail?id=187#c4) originally posted by @ysavourel on 2011-09-19T15:52:33.000Z:

    I created a test file test1.txt (attached), filtered it with okf\_table@ copy-of-csv.fprm parameters, and created a OmegaT package in Rainbow.

    After "translation" in OmegaT (target test1.txt.xlf) and merging in Rainbow, the final test1.out.txt looks good.

  2. Former user Account Deleted

    Comment [6.](https://code.google.com/p/okapi/issues/detail?id=187#c6) originally posted by @ysavourel on 2011-09-20T10:32:08.000Z:

    Just to make sure: the table filter is not auto-adding qualifiers that are not present in the source text. Only if they are in the source, they will be present in the target.

    The two-way filters of Okapi only replace translated content, and are not allowed to add elements like qualifiers.

  3. Former user Account Deleted

    Comment [7.](https://code.google.com/p/okapi/issues/detail?id=187#c7) originally posted by kucharczyk.lu... on 2011-09-20T11:33:18.000Z:

    Aha! As I said in the initial post, I though that this should be handled by the filters, because usage of commas (which are common in written text, as you'd surely aqree) will break the resulting document. Do you think it could be added as a checkbox option? My translation package has about 5000 segments and checking manually whether the text qualifiers are present whenever there is a comma in the target text is a chore, to say the least.

  4. Log in to comment