I am trying to view a TSV file(containing transcription data for code-mixed speech, primarily Hindi and English) that also has colons within the transcription. ELAN chooses the colons as the default delimiter for the file. When I change the delimiter to ‘tab’, ELAN does not split on tab uniformly. In a row with 8 tab characters, none of the tab characters are recognized as a delimiter, generating a single column with the entire row instead of the expected 9 columns.
Is this a known bug? Please let me know if sharing a sample file will help with this problem. I could email it to you separately since I am unable to attach it here.
Look forward to hearing back from you soon.
I don’t think this problem was reported before, but I happened to run into it recently when trying to improve the import of tab-delimited or csv files.
It should be fixed in a next release. If you want you can send a sample file to me so that I can check if the fix really works (han.sloetjes AT mpi.nl).
I’m afraid the sample file was not very useful; it contains only two rows (lines) and the second line does not contain any tabs or other suitable delimiter, only white spaces.
It is always difficult to estimate when a next release will be ready, but it might be a new (test) version will be available within one or two months and the fixes/changes that have been implemented until then, will be included.
But if you create a text file with tab-delimited lines, it can be imported in the current (and previous) version(s).
