![]() ![]() # = exclude temp/cache files/folders everywhere # but will not match "many/other/folders/ProgramData/Micro/" (the name of the folder is "Micro" and not "Microsoft") # "many/other/folders/ProgramData/Microsoft/" (the name of the folder is "Microsoft" so it acts also as a "is exactly") # "many/other/folders/ProgramData/Microsoft OneDrive/" (the name of the folder is "Microsoft OneDrive" so it acts as a "starts with") # Example: "(?i)^+/ProgramData/Microsoft*/" matches both # The difference between this and the above is the "*" instead of "+", which means 0 or more. */: has the purpose of matching a folder which "starts with" or "contains" or "is exactly this" (depending on where it is placed). # TODO TODO TODO TODO TODO TODO TODO TODO TODO When the repository is used in the normal mode: # folders and files which are in HDD root (c:, d:, etc.), when that root is symlinked from the repository. # In other words, the purpose of that regex is to handle excluding, at least in Windows, # "some_sym_link/Important App/Recovery/" nor # with the difference that it assures that any path only has 1 folder before (1 "/"). # "C_abcd_link/leAdministrateur/AppData/Local/someotherfolder/Steam/", as there has to be ONLY 1 FOLDER in between "AppData" and "Steam" folders. # "C_abcd_link/leAdministrateur/AppData/Local/Steam/", but it doesnt match: # Example: "(?i)/AppData/+/Steam/" matches: "+/": has the purpose of assuring that there is only 1 folder in the matched path (only 1 "/") # it seems the regex still works only with "/", so it's very easy to test. # Secondly, on, if the selected language is golang, # In my opinion using only "/" instead of "/" looks cleaner. Reasoning of "/" vs "/" for folder delimitation: # This file can be modified as needed, eventually also propose said modifications, if they are generic enough, to the main duplicacy repository on github. ![]() # This file is a generic file which adds some folders and files to the duplicacy backup ignore list ![]()
0 Comments
Leave a Reply. |