mistake in regex?

[tophee]

Are you sure that this will work:

duplicacy-utils/filters

Line 169 in 834473e

e:/AppData/Local/Packages/

Shouldn't it be

e:AppData/Local/Packages/.*

(and likewise all the other exclusions in the same format)?

[NiJoao]

Regarding the first "/":
If you want to support selecting a users home folder it can't be there.
Otherwise it's irrelevant.

Regarding the last ".*":
(See my very recent post here: gilbertchen/duplicacy#355 (comment) )
As per the Documentation:

When a directory is excluded, all files and subdirectories under it will also be excluded. Therefore, to include a subdirectory, all parent directories must be explicitly included.

The folders are tested first, and their children are only listed and tested if the parent wasn't excluded, which implies that all of the following exclusions do the same, since nothing inside folder/ will be tested.

(^|/)folder/
(^|/)folder/$
(^|/)folder/.*
(^|/)folder/.*$

And keep in mind that Duplicacy uses the MatchString method, that searches for a sub-string in any position. https://golang.org/pkg/regexp/#example_MatchString

[TheBestPessimist]

Regarding the first "/":
I don't think there's any reason. I just like the / so that is "yells" it's only these folders!!!!!.

Regarding the last ".*"
I agree with @NiJoao: there shouldn't be a need for that (at least in this case i think).

What I want to do is ignore the content of <anyUsernameHere>/AppData/Local/Packages/.* but anything else in AppData can be backed up as needed.

I wrote (i think) in the file a comment for my reason to ignore that (and opened a bug for duplicacy here: gilbertchen/duplicacy#309): there's some weird skype file which can't be backed up for some reason and duplicacy just stops the backup (so that no data loss occurs).

[tophee]

Thanks for both of your answers. Yes, I suspected thar the trailing .* is unnecessary but since the filter didn't work for me out of the box (because I was using it on a user's home directory), I was thinking about possible sources of error and those were the teo that came to mind. But I was unsure, since I know @TheBestPessimist has invested quite some thought into this.

In any case, I now think the first slash shouldn't be there.

As regards the MatchString method: I'm not a programmer, so if someone could explain if/how needs to be considered when composing filter rules?

[TheBestPessimist]

There's a much more detailed discussion in gilbertchen/duplicacy#355, you could want t look there.

From that discussion i might get to change a lot of the regexes (again :D).

[TheBestPessimist]

todo: check where it would make sense to remove the first slash as per @tophee's problem