File deduplication

File deduplication finds files with byte-identical content across your case, even when filenames differ. Open the dialog from the case menu, stay on the Files tab, check the count, and click Continue. Hintyr keeps one copy of each document, moves custodians and tags from the removed copies onto it, and deletes the rest in the background while the dialog shows live progress. Produced and redacted documents are always kept, and email families stay intact.

What counts as a duplicate

Hintyr compares file content, not filenames. Two files with different names but identical bytes are duplicates. Contract_MSA_2024.pdf and Contract_MSA_2024 (1).pdf? Same content, same group. The detection is exact. Near-duplicates (one extra comma, one different scan resolution) are not grouped here. Use Emails mode for fuzzy thread-level overlap.

How to deduplicate files

1. Open the case menu from the top navigation and click Deduplicate.
2. Confirm the dialog opens on the Filestab. If it doesn't, switch to it.
3. Hintyr scans the case and tells you how many duplicate files it will delete. If some duplicates are produced, redacted, or still processing, a short note under the count says how many are kept or skipped.
4. Click Continue. Deletion starts in the background and the dialog shows a live progress line.
5. Wait for the result, or click Close. Closing is safe: deletion keeps going after the dialog is gone.

Options and fields

Duplicate count

The headline asks you to confirm, for example "Delete 7 duplicate files?". The number is exactly how many files the run will delete. If nothing qualifies, only Closeis shown and the dialog says "No duplicate files found." When duplicates exist but every one of them is kept or skipped (see the notes below), it says "No duplicate files to delete." instead.

Kept and skipped notes

These lines appear only when they apply. "2 duplicates are kept because they were produced or redacted" counts copies Hintyr never deletes. "1 duplicate is still processing and will be skipped" counts copies whose upload or processing hasn't finished yet. Run deduplication again once processing completes to catch them.

Continue

Starts the deletion. It stays unavailable while the count is loading and while the case is on legal hold. Once a run starts, the button goes away so the same run can't be started twice.

Cancel and Close

Cancel closes the dialog without deleting anything. After you click Continue, the button reads Close. Closing never stops a deletion that is already running.

Progress line

"Deleting duplicates... 4 of 7" with a progress bar. The second number is how many files are being deleted; the first is how many have been handled so far. It updates every couple of seconds. If the dialog briefly can't reach Hintyr, a warning says progress couldn't be checked. Deletion continues either way.

Result message

• Success: every duplicate was deleted, for example "Deleted 7 duplicate files."
• Finished with leftovers: the deleted count plus how many could not be deleted and how many were already removed (for example, a colleague deleted a copy during the run). Run deduplication again to retry anything that couldn't be deleted.
• Stopped by a legal hold: a hold was placed on the case during the run, so deletion stopped. The message says how many files were deleted before it stopped.
• Stopped early: the run stopped before it finished for another reason. The message says how many files were deleted. Run deduplication again to remove the rest.
• Nothing left to delete: by the time you clicked Continue, the duplicates were already gone or could no longer be deleted.

Which copy is kept

Hintyr picks the copy to keep for you. A produced or redacted copy always wins. Otherwise it keeps the original: the copy that wasn't flagged as a duplicate when it was uploaded. If that doesn't settle it, Hintyr prefers a copy that is already searchable, then the oldest upload. Custodian assignments and tags from every removed copy move onto the kept copy, so you don't lose track of who held the document or how it was coded.

Emails and attachments

File deduplication keeps email families intact. Every email that stays keeps a complete set of attachments.

• When a duplicate email is removed, any attachment that exists only under that email moves to the kept copy of the email. Nothing is left without its parent.
• If the kept email then holds two identical attachments, only one stays.
• Attachments of different emails are never deleted, even when the files are identical. Each email keeps its own attachment.
• A loose copy of an attachment (the same file uploaded on its own, outside any email) is removed. Its custodians and tags move to the attachment.
• The same rules apply at every level, for example a zip file attached to an email and the files inside it.

What is never deleted

• Produced documents: anything Bates-stamped or included in a production or export is always kept, even when an identical copy exists.
• Redacted documents: a document with redactions is always kept.
• Files still processing: files that are still uploading or processing are skipped. If any part of an email family is still processing, the whole family is skipped. Run deduplication again later to catch them.

When to run file dedup

Common scenarios:

• After importing a large batch where the same document was collected from multiple sources.
• When team members upload independently and some overlap.
• Before production, so a single document doesn't produce twice.
• When consolidating collections from custodians who shared the same attachments.

Edge cases and limits

• Removal is permanent. There's no trash. Re-upload if you need a removed copy back.
• Cases on legal hold can't run file dedup. If a hold is placed while a run is going, deletion stops. Release the hold, then run deduplication again.
• Only one file dedup runs per case at a time. If you click Continue while another run is going (for example, one a colleague started), the dialog says a deduplication is already running.
• If you close the dialog during a run and reopen it, you see a fresh count instead of the earlier run's progress. That count still includes files the earlier run is about to delete, so it can look higher than what is actually left, and Continue tells you a deduplication is already running. The earlier run still finishes on its own. Reopen the dialog after it finishes for an accurate count.
• You can run file dedup repeatedly. If no duplicates remain, the dialog reports none found.

Frequently asked questions