Under Community Review

Smart TM Cleanup: A Lifesaver for Accidental XLIFF Imports!

Dear Trados Team,

I hope there is already such a feature, that I am not aware of, so please pardon me if there is one!

I’d like to suggest a feature that could be a real game-changer for both novice and professional translators using Trados Studio.

Imagine this: a translator — even a seasoned one — mistakenly imports poorly translated XLIFF files into a high-quality Translation Memory (TM). The result? A compromised TM filled with inaccurate segments that are hard to undo.

Currently, there’s no quick or elegant way to reverse such an action. That’s why I propose a new "Segment Cleanup" feature:
 Arrow right️ The ability to remove TM segments that match those found in one or more XLIFF files.
This would allow users to “roll back” accidental imports by identifying and deleting all segments that came from flawed XLIFF files.

Of course, best practice dictates using a temporary or “dummy” TM first and importing final results afterward — but mistakes happen, especially under deadlines. A feature like this would provide an extra layer of safety and flexibility.

I believe this would be relatively straightforward to implement and would significantly boost user confidence in managing their TMs with precision and peace of mind.

Looking forward to your thoughts!

Best regards,
Sameh

Cartoon of a worried TM file with bandages on a hospital bed, surrounded by medical staff holding a clipboard labeled 'XLIFF Segment Match Removal Protocol'. Text reads 'We need to remove the contaminated segments, stat!' and 'When bad translations slip into your TM, Trados performs emergency cleanup surgery!'.

Sameh Elsharkawy

  • Filter the TM by (adding date)

    This will sort all segments you want to clean/delete

    Then batch delete

  • Hi Paul,

    Thanks for your reply!

    The issue is a bit more complex than what batch delete currently handles. In many real-world scenarios — especially when managing contributions from multiple translators — it’s not always possible to isolate the affected segments by date, username, or even import session.

    Often, the damage is accumulative and scattered, and we may not have a reliable "hook" like a consistent timestamp or author. That's why a more targeted and safe solution would be a feature that allows Trados Studio to compare a batch of XLIFF files directly with the TM and delete any matching segments.

    This way, users can surgically remove the problematic entries based on the exact source of the error (i.e., the XLIFF files), without impacting any of the good translations that may share similar metadata.

    Such a feature would be a lifesaver in large projects and collaborative environments where things don’t always follow a clean, traceable path. Let me know if my argument makes sense at all.

    Best regards,
    Sameh

  •  doesn't batch delete solve this?  You need to have something to hook onto... a date the import occurred, username maybe... but a batch delete would do it I think.  If not, why not?