@Ed > As previously posted, we will allow this to remain for the time being, but please don't hop-link (I've edited all posted locations).
rekordbox mass-relocate tool available for download
There are many posts on the forums, as well as feature requests, related to the Relocate tool and how it doesn't search subdirectories.
I've written a simple command line tool for MacOS and Windows which might be able to help some of you out. It will search for missing files recursively and create a rekordbox XML file containing any tracks it was able to repair, then you can drag-drop those from the "rekordbox xml" area into your main collection. Cue points and loops are preserved through the process. I've written detailed instructions which should make everything clearer.
Along the way, my tool will also flag up any files on your disk which exceed the rekordbox limit of having a path longer than 255 characters. It will also tell you if it finds any files on your disk which you haven't imported into rekordbox, in case that's useful info.
Main page with installation and usage instructions: https://github.com/edkennard/rekordbox-repair
Page to download packaged version of the tool for Mac or Windows: https://github.com/edkennard/rekordbox-repair/releases
I hope this helps some of you out there!
Please sign in to leave a comment.
@Pulse - Sincere apologies for the link issue, and the extra work I made for you. That was not deliberate at all, but rather a copy/paste mistake I made which I guess was because the forum editor is a WYSIWYG so it hid the underlying URL from me
@Ed many thanks for making this tool available. It automates Rekordbox's current relocation capabilities for processing moved files in the whole collection and all folders in a single go. It also adds cool "gotcha" features (>255 char limit, forgotten imported files, etc)
Also many thanks for taking up the time to reply on several threads asking for this, which can now benefit from this work.
However I expect that the tool will still miss renamed files, even with the possible Levenshtein distance feature is implemented - there will always be false positives.To fix this, an unique hash identifier per file is required. For example, Traktor has an AUDIO_ID field for this.
I guess that the cloud analysis feature from the rekordbox 5.4.4 beta depends on such an ID to enable "track analysis sharing" between the users. In this case, I hope this could be used as well to mass-relocate files as well.
@Pedro, it made me very happy to read your post :)
The issue of file renaming is a sticky one indeed - my proposal to use something like a Levenshtein distance would have come after developing a full-blown UI, so that the user could interactively inspect potential matches there, listen to each file, then choose whether any matches were the right one.
But your suggestion to use an AUDIO_ID-like field is of course the ultimate, although I'd need to know exactly how that ID is being generated in order to reproduce them then match. I will have a look into the "track analysis sharing" feature and see what I can figure out.