I am in the process of evaluating CafeTran (CT) to see whether I can move over to this CAT tool from the one I am currently using. I am using the trial version of CT and have been trying to import a TMX file into CT but I get a "prolog" error. I also tried to import a tab separated file as an alternative, but nothing gets imported. What shoul dbe the structure of the tab separated file? Should the first line of the tab-separated file contain the headings? which ones? should they have a specific name? should the source text and target text for each TU in tab file be surrounded by the " character?. Why am I getting the prolog error when I try to import the TMX?
The Knowledge Base has sections on Glossaries and Memories but it doesn't cover the file conversions between other CAT tools yet.
Okay, here are the easy steps to import the Metatexis tab delimited memory to CafeTran:
1. Open your tab delimited memory in MS Excel. Make sure you that the file UTF-8 encoding is properly displayed in cells. You may need to choose the UTF-8 encoding during the file import.
2. Remove the " text delimiters.
I used LibreOffice Calc spreadsheet application instead of Excel for the above. It removed the " text delimited automatically.
3. Save the file as the Ms Excel file (.xlsx).
4. In CafeTran, go the the menu Memory > Import Ms Excel memory... and make sure that your source and target languages are selected at the import.
5. After the memory is loaded, go the menu Memory > Save as... and save it as a .tmx file.
Can you submit a support ticket attaching a sample TMX file that you are trying to load? I can check if it is in a TMX format indeed.