Christoph Anton Mitterer

documentation: use actual path for ssh_filter_btrbk

2

In it’s manpage, use the actual installation pathname of `ssh_filter_btrbk.sh`. In `README.md`, which should be readable without building, use the usual pathname, which is `/usr/share/btrbk/scripts/ssh_filter_btrbk.sh`. Signed-off-by: Christoph Anton Mitterer

Hey. Some ideas… Not so sure about the last commit, since I haven’t found why you had put `AccuracySec` in at the first place. So if there was a good...

Hey. Just an absolutely low priority idea for the records: It would perhaps be nice if the user can be specified as part of the `ssh://` URL like in `ssh://user@host/…`,...

Hey. I think it would be nice if at least here: https://github.com/digint/btrbk/blob/af2d7b2c99bac7c5af2f9b2c5f0b1947be525597/doc/ssh_filter_btrbk.1.asciidoc?plain=1#L48 the pathname would match that where it's actually installed. Does `asciidoc` provide any replacement mechanism or should I...

feature-request: format-string-like fields like %h for host or %u for username

2

Hey. For pathnames, and there especially for the target pathnames it would be nice if one had format-string-like fields. Consider e.g. on has multiple hosts that run backups all to...

Hey Andrew. I found a case where I think `crecord` causes corruption (of the merged data), unfortunately I couldn't find a simple reproducer, so I'll mail you my test files...

using `crecord` on a repo with already staged changes still shows those within `crecord`

7

Hey. When using `crecord` on a repo with already staged changes it still shows those within `crecord`, which I guess is unnecessary. Especially, it does not seem to have any...

add a mode that allows `crecord` to operate on the staging area

2

Hey. #52 brought me to the idea that it would indeed be nice, if `crecord` had a mode where it operates on the stating area. In that mode it would...

Hey. Would be nice if there was bash/command completion for `crecord`. Right now, it seems that not even `git crecord` alone is completed, though I'd have expected that this would...

cosmetic: still says "commit" when doing staging

1

Hey. When using `s` to stage, the program still asks whether one is sure to commit, despite that it actually does stage. :-) Cheers, Chris.