Start a new topic

Excel filter, only already translated rows imported

I'm trying to import an Excel document with the the following settings:

- Import source column C

- Import target column D (many, but not all rows, have translations)

- Export target column D

Every time, only the translated rows are imported. Untranslated rows are not.

I've tried copy-pasting the cells in a new document, but I get the same symptom. Same if I try with different segmentation settings. I've also tried setting another column for export.

Any ideas? Should I open a ticket?


CTE assumes that if you define two Excel columns for import separately (in two Import fields of the filter), the second column is to be imported for the review of the existing translation. So there is the limitation of such a separate import to only existing rows (excluding the empty ones). If you specify two import columns separately, leave the export column empty. 

Thank you for this information, Igor. When I find some time, I will test it and update the document on file formats for future reference. For now, I have exported a TMX of the existing translations, created a new project while importing only the source column, inserted all exact matches, filtered translated segments, marked them as checked and was able to focus on translating the rest.

Igor, I have tested leaving the export column empty, but the imported column that takes precedence seems to be the target text one. 

This seems to be the intended behavior for the current implementation, as per your above explanation.

When both source and target text columns are set, any blank cells in target column result in the respective source/target cells not being imported.

I understand this can be helpful when reviewing completed translations.

However, it is not uncommon to receive Excel files with partly translated content. It would be nice to have the source column take precedence over the target one: If the source cell has content, import the respective source/target text even if target cell is blank.

I think this could be set as the default filter behavior, but maybe I’m missing something in the usefulness of the current implementation.

For now, solutions include the following:

Solution A: Create a project while importing existing translations and export a TM. Then, create a new project with only source column imported (empty target segments), then apply exact TM matches, and go on with reviewing existing translations and translating the rest.

Solution B: Add some characters in the empty target cells and import both source and target.

Maybe a checkbox for the import target column(s) could do the trick?


Login to post a comment