Vault 3.3.1.8
I have a directory hierarchy with directories in question:
software\ra\src\drivers\drvlib\pcd\arm
...\inc
...\mipsiv_fp
(There are also many other directories which are peers of src, drivers, drvlib, and pcd, but I'm focused on an anomaly with these).
I already had $/software in Vault. I added to Vault source control the ra subdirectory, recursively. Since I had several DLL+EXE in some subdirectories which also needed checked in, I had removed all file and directory exclusions, and selected everything under ra.
The resuling tree in Vault now has ra, and many nested subfolders under it. However, I just discovered one subtree (src\drivers\drvlib\pcd) has no further folders under it. The arm, inc, and mipsiv_fp subdirectories I have on my system contain various files such as .h and .cpp source, so should have been added but were not.
Now, if I choose 'detect new files to add' at any Vault folder level all the way down to pcd itself, it incorrectly shows no subdirectories to add. If I choose 'add files/folders' now, it does offer arm, inc, and mipsiv_fp (and their file contents) even though it didn't add these before.
There may be other directories behaving the same way, but this is the only instance I've hit so far. Are there any known issues with add or detect files/folders that could explain this? Any tips to avoid this so I can ensure that all nested subfolders/files really do get added? Any experiments to try if you need more info to debug?
Thanks-
Aaron
Add new files (recursive) didn't add several subdirectories
Moderator: SourceGear
Detect new files only does files, not folders, so that would explain that issue.
Is it possible that the subfolders that weren't added had names in the folder exclusion list? Note that folder and file exclusions names are separate.
If not, can you send us the folder structure you have (perhaps with dummy files in them of the same name), so we can reproduce it here? Or perhaps try it again in a test repository, to verify it works (or doesn't work) in a reproducible way?
Is it possible that the subfolders that weren't added had names in the folder exclusion list? Note that folder and file exclusions names are separate.
If not, can you send us the folder structure you have (perhaps with dummy files in them of the same name), so we can reproduce it here? Or perhaps try it again in a test repository, to verify it works (or doesn't work) in a reproducible way?
I've been removing all entries from file and directory exclusions.
Trying to repro on clean new trees with add vs. detect and nothing vs. some folders already present, I get consistent behavior where straight add can hit all files, but detect will ignore the ones in new directories It looks like this must be an effect of 'detect/add new' only applying to files in existing folders instead of new folders, and I must have mistakenly used it instead of straight add new.
Is detecting new folders already on the wishlist for new features? I didn't realize it wasn't part of detect new files already, and would be extremely useful for all the same reasons that detect new files is.
Trying to repro on clean new trees with add vs. detect and nothing vs. some folders already present, I get consistent behavior where straight add can hit all files, but detect will ignore the ones in new directories It looks like this must be an effect of 'detect/add new' only applying to files in existing folders instead of new folders, and I must have mistakenly used it instead of straight add new.
Is detecting new folders already on the wishlist for new features? I didn't realize it wasn't part of detect new files already, and would be extremely useful for all the same reasons that detect new files is.