This is a bug in Passolo Add-in Text Parser?

I have an error while parsing the following text file and can not parsing all its contents. This is a BUG? Please help me.

Files:
http://jp.wanfutrade.com/download/ErrorInTextParser.rar

Parents Reply Children
  • I've tried a lot of rules, but parsing error occurred.
    Now I put the project and rule files and error screen shot file uploaded to you, please help test it out.
    If possible, please help edit a rule.
    Thank you!

    Files:
    http://jp.wanfutrade.com/download/ErrorInTextParser.rar



  • I analyzed the file and the rule and I can reproduce the problem. Please note that some of your translatable text entries that are stored in one single text file line are complete XML files with > 10 kb in size. A single regular expression in the text file parser cannot evaluate more than 16 kb, but your expression is evaluating 2 lines and the line above the translatable content also contains the translatable string which doubles the size. Here are my recommendations that may help to solve the problem.

    • If you are able to influence the text file structure, do not export the translatable string in the first line. This will increase the number of entries that can be imported but there are still some entries that are too long.
    • Many of the XML files stored in the text file do not contain any translatable data. Simple do not export data that must not be translated. This will again increase the number of entries that can be imported but there may be still some entries that are too long.
    • Do not export translatable XML files into text files. Export this data into real XML files. Then the files will be processed by the XML parser which separates the XML structure from the translatable data. It protects the XML structure from accidental changes by translators and will reduce the amount of data the must be read before one can find out what is the translatable part of the XML file.

    Hope this helps.

  • Thank you for your detailed reply.

    Need further explanation is that this HCS file is hard-coded strings extracted from a PE file by my own macros.
    When these strings are translated, also need to write in the PE file to complete the localization.

    This HCS file contains the complete XML text, and some elements of these XML text also need to be translated.
    Of course, can export the XML text as XML files and to parses by the XML parser.
    But write the translated XML text is a very troublesome thing because with third-party software is required to complete.
    If can use Passolo text parser to parses the XML text directly, will save a lot of trouble and improve work efficiency.

    So, I would like to ask:
    1, Why to limit length of a single string for the text parser? As far as I know, the regular expression do not have this limitation.
    2, Can lift this limitation in future versions?

    Thank you!

  • This is a very special requirement that didn’t come up in all the years the text file parser exists. SDL Passolo users usually are working with properly internationalized software, standard workflows and standard resource files types where shorter software strings (<16 kb) are stored instead of complete files that must be handled in one single text entry.

    I can insert this requirement into the product backlog for future releases but can promise when it will be implemented.

    For the time being I would suggest to use one of my proposed work-around and delete untranslatable entries from the file and stored longer entries into separate files.