Commits on Sep 26, 2016
  1. magit-branch-delete: fallback to delete tracking branch

    When the deletion of a remote branch fails, then that's probably the
    case because it no longer exists on the remote but the local tracking
    branch still exists.  Users should use pruning to deal with this, but
    if they don't, then this command now falls back to deleting the local
    tracking branch.
    Closes #2778.
    tarsius committed Sep 26, 2016
  2. git-rebase-comment-{hash,heading}: new faces

    As requested in #2769.
    tarsius committed Sep 26, 2016
  3. git-rebase-command-descriptions: use correct line movement commands

    We use `forward-line' instead of `next-line' because we do want
    point to go to the beginning of the line, likewise for going upward.
    Because `forward-line' and `git-rebase-backward-line' (obviously)
    aren't bound in the global keymap, this also fixes #2769 for most
    tarsius committed Sep 26, 2016
Commits on Sep 24, 2016
  1. magit-commit-squash-internal: drop --gpg-sign for instant variants

    I.e. when running `git commit' synchronously and therefor without
    being able to read the passphrase.  In this case it is pointless
    to sign because the commit will be discarded instantly anyway.
    Fixes #2799.
    tarsius committed Sep 24, 2016
  2. git-rebase-hash: inherit from magit-hash

    We can do that because `git-rebase' depends on
    `magit' now.  It used to be the other way around.
    tarsius committed Sep 24, 2016
Commits on Sep 23, 2016
  1. magit-worktree-{checkout,branch}: expand path

    Git does not understand the meaning of "~".
    Fixes #2796.
    tarsius committed Sep 23, 2016
Commits on Sep 21, 2016
  1. Add INSIDE_EMACS to magit-git-environment

    This follows core Emacs behavior. For example M-x shell, term and epa
    all export similar env vars. This will enable functionality such as the
    pinentry package distributed with Emacs 25.1 being used for the
    passphrase for git commit signing with gpg.
    gdobbins committed with tarsius Sep 21, 2016
Commits on Sep 20, 2016
  1. Update docs for magit-stash-drop keybinding

    The keybinding assigned to magit-stash-drop (in magit-stash.el) is z k,
    not z d.
    jschwab committed with tarsius Sep 20, 2016
  2. 2.8.1.txt: fix pasto

    tarsius committed Sep 20, 2016
Commits on Sep 19, 2016
Commits on Sep 18, 2016
  1. git-rebase: disable save check when showing commits

    The git-rebase-todo buffer is frequently in a modified state, but there
    is no need to save it before showing a commit.  It will be taken care of
    by with-editor-return when the user exits through with-editor-finish.
    The unsaved buffer should almost always be the git-rebase-todo because
    the initial 'git rebase' call will fail if there are unstaged changes.
    The user could unwisely go and modify some other tracked buffers after
    starting a rebase, and this could put things into a bad state [1], but
    this is unrelated to whether the user happened to call a command to show
    a commit.
    Closes #2770.
    [1] And, when the user has done this, I don't think there is really any
        good action to take to fix the situation.  It's bad whether or not
        the buffer is saved.  The user should just never start an
        interactive rebase and then change buffers of tracked files before
        triggering the rebase with with-editor-finish.
    kyleam committed Sep 15, 2016
  2. git-commit: Fix highlighting of second line

    Add a function to font-lock-extend-region-functions to correctly
    identity the multiline git-commit-summary-regexp to font-lock, as
    suggested in `(elisp) Multiline Font Lock'.  This causes the
    git-commit-nonempty-second-line face to be applied immediately, instead
    of having to run font-lock-fontify-block manually to trigger it.
    npostavs committed Sep 18, 2016
Commits on Sep 17, 2016
  1. fix typo

    ChillarAnand committed with kyleam Sep 17, 2016
Commits on Sep 11, 2016
  1. Update

    Fix donations link
    vermiculus committed on GitHub Sep 11, 2016
Commits on Sep 7, 2016
  1. magit-gpg-secret-key-hist: strip everything but the actual key

    But keep displaying the name and email address anyway
    as that makes it easier to identifying the correct key.
    Fixes #2761.
    tarsius committed Sep 7, 2016
  2. magit-stash-drop: log revision hash to process buffer too

    `magit-stash-drop' process already logged the revision hash in the
    *Messages* buffer to allow users who accidentally deleted a stash to
    recover it.  Now we also call `git rev-parse STASH' before deleting
    STASH.  That causes an additional entry in Magit's process buffer.
    Deleting a stash then looks like this:
      0 git … rev-parse stash@{0}
      0 git … reflog delete --updateref --rewrite stash@{0}
    Or after expanding:
      0 git … rev-parse stash@{0}
      0 git … reflog delete --updateref --rewrite stash@{0}
    There is no guarantee that a user who just accidentally deleted a stash,
    will realize that s/he can get to the hash by expanding that section,
    but it increases the odds of at least some users finding the hash.
    Re #2746.
    tarsius committed Sep 7, 2016
  3. magit-mode-map: bind S-tab to magit-section-cycle-global

    It appears this was always the intention.  Keep the bindings
    for s-tab and backtab, in case someone actually uses those.
    Fixes #2759.
    tarsius committed Sep 7, 2016
Commits on Sep 2, 2016
Commits on Aug 30, 2016
  1. magit-blame: swap description of -M and -C options

    bb2020 committed Aug 30, 2016
Commits on Aug 27, 2016
Commits on Aug 24, 2016
Commits on Aug 23, 2016
  1. magit-repolist-column-version: ignore empty repositories

    All output will be nil in a freshly initialized repository.
    kyleam committed Aug 23, 2016
  2. add v2.8.1 release notes stub

    kyleam committed Aug 23, 2016
Commits on Aug 21, 2016