[user]
	name = Karl Fogel
	email = kfogel@red-bean.com
[core]
	pager = /bin/cat
	excludesfile = /home/kfogel/.gitignore
[init]
	defaultBranch = main
[format]
	pretty = fuller
[status]
        short = true
[diff]
	noprefix = true
        # FWIW, this diff.renameLimit was motivated by a warning I got
        # when doing 'git log --name-status' on the 'feature-sharing'
        # branch of git@github.com:hmis-tools/hmis-api-server.git from
        # commit bc0f1e2ef, with git 2.9.3:
        #
        #   warning: inexact rename detection was skipped due to too many files.
        #   warning: you may want to set your diff.renameLimit variable to at least 2074 and retry the command.
	renameLimit = 2074
[branch]
	autosetupmerge = true
[push]
	default = matching
[pull]
        # On 2020-06-18 I suddenly started getting this new message
        # when I did 'git pull'.  That's why I added this setting.
        # 
        #   $ pwd
        #   $ ~/private/work/ots/clients/red-cross/dcsops/r/arcdata-wiki
        #   $ git pull
        #   
        #   warning: Pulling without specifying how to reconcile
        #   divergent branches is discouraged. You can squelch this
        #   message by running one of the following commands sometime
        #   before your next pull:
        #   
        #     git config pull.rebase false  # merge (the default strategy)
        #     git config pull.rebase true   # rebase
        #     git config pull.ff only       # fast-forward only
        #   
        #   You can replace "git config" with "git config --global" to
        #   set a default preference for all repositories. You can
        #   also pass --rebase, --no-rebase, or --ff-only on the
        #   command line to override the configured default per
        #   invocation.
        #   $ 
        #
        # For more context, see https://www.reddit.com/r/git/comments/\
        # h7kv0y/git_2270_starts_yelling_during_pull/ .
	ff = only
[rerere]
	enabled = 1
# Three 'fsckObject' settings are recommended in this thread...
#
#   https://groups.google.com/forum/#!topic/binary-transparency/f-BI4o8HZW0
#   From: Eric Myhre
#   Subject: git integrity
#   To: binary-transparency (Google Group)
#   Message-ID: <56ABBA5B.6020703@exultant.us>
#   Date: Fri, 29 Jan 2016 11:15:39 -0800
#
# ...however, it turns out that just setting transfer.fsckObjects is
# enough, as the other two default to that.  See Paul Eggert's post
# https://lists.gnu.org/archive/html/emacs-devel/2016-01/msg01807.html.
#
# See also bug reports mentioned by DKG in the "git integrity" thread:
#
#   https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=813157
#   https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=743227
#
# (See https://github.com/robbyrussell/oh-my-zsh/issues/4963 for
# discussion of an error that this setting can cause due to some
# gotchas in some repositories' git histories, though.)
[transfer]
        fsckObjects = true

# But, for some repositories, not mentioning any names (cough,
# kfogel@git.sv.gnu.org:/srv/git/emacs/elpa.git, cough), we then have
# to ignore one of the fsck messages.  See these threads for details:
#
#   https://github.com/robbyrussell/oh-my-zsh/issues/4963
#   https://github.com/holman/dotfiles/issues/231
#   http://thread.gmane.org/gmane.comp.version-control.git/288336
#
# Oh, but wait, it doesn't work anyway.  So we'll comment it out, and
# when checking out GNU ELPA, we'll comment out the above setting too.
#
# [fsck]
#         zeroPaddedFilemode = ignore

[alias]
# See https://news.ycombinator.com/item?id=4131460, which points to
# http://blog.kfish.org/2010/04/git-lola.html.
lol = log --graph --decorate --oneline
lola = log --graph --decorate --oneline --all

# The 'git edit' alias.
# ---------------------
#
# https://gist.github.com/slifty/68bd451896f6494b870280c00dc4fe9c
# https://chat.opentechstrategies.com/#narrow/stream/2-general/topic/some.20git.20magic/near/108695
#
# ## How it works
# 
# The alias does the following...
# 
# 1. Populate a critical status tracking variable (this is used to make
#    sure the stash is kept consistent)
# 
#      $ EDITED_STASH=0
# 
# 2. Start an interactive rebase to the specified commit, replacing
#    `pick` with `edit` for the first line (i.e. specifying `edit` for
#    the passed in commit hash).
# 
#      $ GIT_SEQUENCE_EDITOR=\"sed -i '1s/^pick/edit/'\" git rebase -i "$1"
# 
# 2. Remove the staged commit content
# 
#      $ git reset HEAD~1
# 
# 3. Mark all new files as being intended to commit (this allows you to
#    selectively include only a portion of the new file.)
# 
#      $ git add -AN
# 
# 4. Perform a patch add against the original commit content.  This is
#    where you can decide what NOT to include.
# 
#      $ git add -p
# 
# 5. Perform the commit with only the patch edited changes, using the
#    original commit message (and allowing you to edit that message).
# 
#      $ git commit -e -C HEAD@{1}
# 
# 6. Check to see if there is anything sitting around NOT staged for
#    commit (check to see if an edit actually happened)
# 
#      $ if [ -z \"$(git status --porcelain)\" ];
# 
# 7. If there were no changes (if git status --porcelain returned `""`),
#    then continue the rebase.
# 
#      $ then git rebase --continue;
# 
# 8. Otherwise there are changes left over which have been edited out of
#    the commit.  We need to:
# 
#    * Stage all of those changes.
#    
#         $ else git add -A
#    
#    * Stash those changes (and denote that we have stashed)
#    
#         $ git stash -q && EDITED_STASH=1
#    
#    * Invoke the rebase
#    
#         $ git rebase --continue
#    
#    * Pop the changes into our working directory 
#      (and register that the stash is clean)
#    
#         $ git stash pop --quiet && EDITED_STASH=0
#    
#    * Unstage the changes
#    
#         $ git reset;
# 
# 
# 9. And finally, if anything goes wrong at any step of the way, we want
#    to abort the rebase and clean up any lingering git cache and stash
#    changes, as if none of this ever happened.
# 
#      $ (echo 'Aborting the edit...' && if [ \"$EDITED_STASH\" = \"1\" ]; then git stash drop -q; fi && git rebase --abort && git add -A && git reset); }; f"
# 
# ## Some notes
# 
# * It isn't currently possible to edit away a deleted file, though it
#   *is* possible to edit away an added file.
# 
# * If you remove *all* changes from a commit the edit will be
#   aborted.
# 
# * If your edits cause a later merge conflict (for instance if you
#   remove a file from a commit that is edited later in your rebase)
#   then the edit will be aborted.
# 
# * If you ctrl-C out of the git add patch view, the alias will stop
#   short (and you will need to manually type `git rebase --abort`.
edit = !"f() { EDITED_STASH=0 && GIT_SEQUENCE_EDITOR=\"sed -i '1s/^pick/edit/'\" git rebase -i \"$1~\" && git reset HEAD~1 && git add -AN && git add -p && git commit -e -C HEAD@{1} && if [ -z \"$(git status --porcelain)\" ]; then git rebase --continue; else git add -A && git stash -q && EDITED_STASH=1 && git rebase --continue && git stash pop -q && EDITED_STASH=0 && git reset; fi || (echo 'Aborting the edit...' && if [ \"$EDITED_STASH\" = \"1\" ]; then git stash drop -q; fi && git rebase --abort && git add -A && git reset); }; f"