| Age | Commit message (Collapse) | Author |
|---|
|
I'm not incredibly confident that I performed a faithful conversion,
but the things I've tested seem to be working. Keep the popup
configuration around since I still use it for guix.el and
snakemake.el.
|
|
I haven't used this since I added it.
|
|
This is useful information to have.
magit-branch-shelve already does this. (It's one of the callers of
the magit--rename-reflog-file helper.) At some point I should
probably just switch to using magit-branch-shelve, which didn't exist
when I added km/magit-branch-archive. The main differences are that
they use a different ref namespace, magit-branch-shelve has a nice
counterpart command for restoring the branch, and
km/magit-branch-archive supports multiple ref selections. I do use
the multi-ref feature a good amount from the refs buffer, but that
might be pretty particular to my workflow and not something that
should be added to magit-branch-shelve.
|
|
... so that the reflog is copied as well.
|
|
This should have been done in 6ea76aa (magit-branch-backup-current:
Don't use "b/" prefix, 2017-03-19).
|
|
|
|
|
|
|
|
|
|
The main motivation is to make this work with hexshas surrounded by `.
|
|
|
|
|
|
|
|
Magit will warn about an inappropriate value when the hook has a
non-function value. Instead, use the same approach used by
km/magit-cherry-toggle-upstream-section.
|
|
|
|
I used many of these heavily at some point, but due to some
combination of changes in my preferences/workflows and changes in
Magit, I rarely use any of these now.
|
|
That predated Magit's 24c96608 (git-commit-turn-on-flyspell: avoid
unnecessary work, 2017-12-03) and shouldn't be necessary now.
|
|
|
|
|
|
|
|
Don't assume that the question of interest is usually whether the
revision at point is the descendant.
|
|
These have been obsolete for a while and are removed in Magit's master
branch.
|
|
The latter has been obsolete for a while and is removed in Magit's
master branch.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
When magit-log-show-refname-after-summary was non-nil, the previous
logic failed. Use magit-rev-format instead of regular expressions to
extract the subjects. As a result, we don't make any assumptions
about where the subject is positioned in a magit-log-select-mode line.
|
|
|
|
|
|
Quite a while ago Magit stopped displaying commit messages with the
leading spaces, so there is no need anymore to strip them out.
|
|
|
|
|
|
This should have been updated with f0280cf9 (Update for
magit-copy-as-kill rename, 2015-10-04).
|
|
|
|
|
|
Appending "-tmp" to the current branch name isn't a great default
value, but it allows me to quickly pull it in with M-n and base the
new branch name off of the current one, which I often do, especially
for trying out a different approach on a topic branch.
|
|
|
|
|
|
|
|
|
|
This is no longer useful because this functionality was added to Magit
in 3584aaba (magit-stage: support --intent-to-add, 2016-05-04).
|
|
If only an integer is used, the end value used for the search is no
longer accurate because more characters are inserted for each line
where "fixup" or "squash" is substituted for "pick". This discrepancy
is unlikely to have an observable effect when git-rebase-show-instructions
is non-nil because the end value includes many characters beyond the
last commit line. But when git-rebase-show-instructions is nil, this
can easily result in the last commit line not matching the regexp
because the end value has shifted to the middle of the line.
|
|
I'm not sure why I preferred this to using --sort=-committerdate in
the magit-show-refs-popup.
|