As the title says, it would be nice to have detection of moved files when synchronizing for the first time.
My use case could amazingly benefit from this, as probably many others', and I think it shouldn't be that hard to code. We only need to detect duplicates as many other apps do (perhaps with a quick hash) and assume that the user is going to check what is assumed to be equal and moved. I'm not that proficient in C++ but I can code a short Python module with a proof of concept.
Let me know if you need any clarification or help!
Detection of moved files even when synchronizing a folder pair for the first time
- Posts: 2
- Joined: 22 Mar 2017
-
- Site Admin
- Posts: 7212
- Joined: 9 Dec 2007
Unfortunately there is no such thing as a "quick" hash... ;)
- Posts: 2
- Joined: 22 Mar 2017
Ohh you know what I'm talking about :). It can be made as an optional feature is performance is an issue. I (and probably many others) are currently stuck using two programs for our sync needs (one duplicate finder first, and then FreeFileSync) so the performance is really not that important.
You can compare only the first and lasts 64 Kb and then perform a complete hash if the comparison returns True, for that matter. Is there any other good reason for this not being in the feature backlog?
You can compare only the first and lasts 64 Kb and then perform a complete hash if the comparison returns True, for that matter. Is there any other good reason for this not being in the feature backlog?
- Posts: 1
- Joined: 6 Sep 2017
A bit of an outdated topic but seeing that this is still not implemented, I'd like to push this up back into the view of the developers.
Not sure why this is not such a big deal for pretty much everyone. The biggest problem is the difference in folder structures for files that already exist.
Would be great if FFS could check for a duplicates first and intelligently detect "moved" or duplicate files between the two folders.
Alternatively, allow the user to choose a "reference" folder or the folder who's structure the user wants to keep. Then find duplicates through the directory tree.
(Can also use Date of Folder Creation + modification as to which folder structure is newer but let the user pick one as mentioned above)
Allow the user to mirror the folder tree to the target base folder, shift/move files (instead of copying them again) to the newer structure, and then go back to the usual operation of updates, new files etc as it does now.
Even better would be allow custom "workflows" for the operations which are a combination of de-dup, folder tree mirroring, file updates (where no data is copied but only files are moved, updated or if the file content is the same then simply update the "attributes") etc.
It's biggest drawback is the program's limitation of
a) lack of ability to detect folder structure differences and assuming that files need to be updated and copied when all that's different is a diff folder structure
b) inability to de-dup
c) simply being able to update file attributes and time stamps without copying data when the files are exactly the same except the timestamps and/or attributes have changed (even windows robocopy is capable of doing that)
a few more, but If you can put that functionality in it, this will be the ONLY file-sync program anyone ever needs
Lastly, I agree with the OP, for people who are power users and want complex functionality and multi-step workflows are willing to sacrifice performance for functionality. Besides once the above mentioned workflow is completed and the folders are in sync, we're back to high-performance snappy sync mode...it may only be the first time when the full scan with hashes and folder tree differences are scanned/detected
Cheers
Al
Not sure why this is not such a big deal for pretty much everyone. The biggest problem is the difference in folder structures for files that already exist.
Would be great if FFS could check for a duplicates first and intelligently detect "moved" or duplicate files between the two folders.
Alternatively, allow the user to choose a "reference" folder or the folder who's structure the user wants to keep. Then find duplicates through the directory tree.
(Can also use Date of Folder Creation + modification as to which folder structure is newer but let the user pick one as mentioned above)
Allow the user to mirror the folder tree to the target base folder, shift/move files (instead of copying them again) to the newer structure, and then go back to the usual operation of updates, new files etc as it does now.
Even better would be allow custom "workflows" for the operations which are a combination of de-dup, folder tree mirroring, file updates (where no data is copied but only files are moved, updated or if the file content is the same then simply update the "attributes") etc.
It's biggest drawback is the program's limitation of
a) lack of ability to detect folder structure differences and assuming that files need to be updated and copied when all that's different is a diff folder structure
b) inability to de-dup
c) simply being able to update file attributes and time stamps without copying data when the files are exactly the same except the timestamps and/or attributes have changed (even windows robocopy is capable of doing that)
a few more, but If you can put that functionality in it, this will be the ONLY file-sync program anyone ever needs
Lastly, I agree with the OP, for people who are power users and want complex functionality and multi-step workflows are willing to sacrifice performance for functionality. Besides once the above mentioned workflow is completed and the folders are in sync, we're back to high-performance snappy sync mode...it may only be the first time when the full scan with hashes and folder tree differences are scanned/detected
Cheers
Al
