The support PCs are Windows: case is preserved but not significant. The staging area etc is all Linux.
Specific problems this causes:
dataset IDs are case sensitive, and would break if a user changed the case of a directory (think we should use case insensitive entryID lookups)
user names are case sensitive (but user folder names aren't): currently we lowercase the name of the user folder, hence create all lowercase user names. (Changing the case of a user folder would still cause all data to be duplicated)
changing the case of a datafile would probably cause a duplicate import
The support PCs are Windows: case is preserved but not significant. The staging area etc is all Linux.
Specific problems this causes:
Ugh.
original LH ticket
This ticket has 0 attachment(s).