Recently I’m getting a lot of translations in WordFast Pro files. The first translation I did using WF demo and I didn’t liked it, so when I got the next ones, I looked for a way to process them in a much more comfortable environment, that is memoQ. As it came out, it isn’t so hard.
What do we need:
- WordFast Pro demo (not necessary, but recommended)
- Excel/Calc
- memoQ
- Word/Writer (optional)
- Open the file to translate (.txml) in the WordFast.
- Use Ctrl-Alt-Ins to copy all source segments contents into target segments.
- Save the file, close WordFast (repeat the procedure for additional files if necessary). If, for some reason, you don’t want or can’t use the WordFast demo, you can use the search/replace procedure described here.
- Download this filter file and save it on your disc.
- Start the memoQ.
- From the Tools menu choose Resource console, then Filter configurations and Import new from the left pane. Select the file saved in step 4.
- Create new project, in the Add document window select Add document as. You’ll have to switch file type filter in the lower part of the window from All supported files to All files (*.*). Select the txml file(s) saved in step 3.
- In the Document import settings click the yellow folder icon (Load filter configuration) and select WordFast filter imported in step 6. The filter works by hiding the original source segments (<source>), displaying for edition target segments (<target>) and defining handling of WordFast tags. See picture below. Tip: WordFast tags, displayed natively as consecutive numbers {1}, {2}, {3}, in memoQ are displayed as “inline” tags. By default you can insert them using F9 key, but in my there are more convenient key combinations, for example Ctrl-Alt-Down, like in Trados. To modify keyboard settings choose Tools > Resource console > Keyboard shortcuts > Clone (for the Default) > Edit.
- After translation export the finished file using Export command. Excellent QA features of memoQ won’t allow you to export files with tags mismatch and will show you the affected segments (if any).
- Just to be extra sure you can open the translated file in WordFast, to check if everything is OK.
- Open the TM file in Word (or OpenOffice Writer) (option). In the WF TM tags are encoded as &tA;, &tB;, &tC;, etc. It is possible to remove them all at once using Word. Open Search & Replace window (Ctrl-H), click More >> button and select Use wildcards. Enter \&t[A-Z]\; in the Find field. Leave the Replace field empty. Click Replace all button. Now deselect Use wildcards checkbox and replace &’A9; with ©, and &’AE; with ® (of course, if there are such strings in your TM. Save as text file using UTF-8 encoding.
- Start Excel (or OpenOffice Calc), import the TM file by selecting Data > External data > From text. Choose Tab as a column separator and import the first row checkbox.
- Remove all columns except the ones with source and target texts. In the example above we have to remove columns A, B, C, D, F, H and I, leaving only E and G. As an alternative we can leave also column B, with the translators ID.
- Enter the language identifiers into first row of source and target columns. If necessary, we can add columns with additional information, like the translation author, domain, client, etc.
- If step 1 was skipped, you have to remove tags – unfortunately, Excel requires removing them one by one, i.e. you have to enter manually all possible letter combinations (see step 1). Save the file: select round office button > Save as > Other formats > Unicode text. Close Excel.
- In memoQ select an existing memory or create a new one. Choose the Import from TMX/CSV command. Select the file saved in step 5.
- In the Translation memory CSV import settings window select proper settings:
- File encoding – UTF-8
- Delimiter – Tab
- First row is header
- Voila.