Permalink
Commits on Apr 16, 2014
Commits on Apr 15, 2014
  1. Merge pull request #1341 from mgalgs/no-merges

    magit-key-mode-popup-logging: support --no-merges option
    tarsius committed Apr 15, 2014
Commits on Apr 14, 2014
Commits on Apr 12, 2014
  1. magit-show: bugfix

    `magit-diff-range' returns a cons cell of atoms.
    tarsius committed Apr 12, 2014
  2. Merge pull request #1340 from dachary/wip-test

    stub test for magit-get-next-tag
    tarsius committed Apr 12, 2014
Commits on Apr 11, 2014
  1. stub test for magit-get-next-tag

    Signed-off-by: Loic Dachary <loic@dachary.org>
    dachary committed Apr 11, 2014
Commits on Apr 9, 2014
  1. Merge pull request #1331 from mgalgs/rewrite-diff-pending-stuff

    Don't try to run magit-rewrite-diff-pending from magit-status-sections-hook
    tarsius committed Apr 9, 2014
  2. Don't try to run magit-rewrite-diff-pending from magit-status-section…

    …s-hook
    
    In [9a4e236: "Transition machinery for magit-insert-pending-changes"] an
    obsolete function alias and "transition" function were introduced to
    help users that may have customized `magit-status-sections-hook' to know
    that `magit-rewrite-diff-pending' no longer exists. However, it really
    never makes sense to run `magit-rewrite-diff-pending' from
    `magit-status-sections-hook' and in some configurations the actual
    "help" message is hidden by the error from
    magit-rewrite-diff-pending. Fix this unnecessary confusion by *not*
    running `magit-rewrite-diff-pending' from the shim function.
    mgalgs committed Apr 9, 2014
Commits on Apr 7, 2014
  1. Merge pull request #1328 from mgalgs/obsolete-alias

    Transition machinery for magit-insert-pending-changes
    tarsius committed Apr 7, 2014
  2. Transition machinery for magit-insert-pending-changes

    If users have customized their `magit-status-sections-hook' then it can
    be confusing when they update magit and suddenly start getting error
    messages due to `magit-insert-pending-changes' not being defined
    anymore. Define a an obsolete function alias to a shim function that
    displays a message to the user about what's going on.
    
    See also: 3823518
    mgalgs committed Apr 7, 2014
Commits on Apr 6, 2014
  1. Merge pull request #1326 from jwiegley/master

    Fix magit-key-mode.el to refer to magit-diff-toggle-refine-hunk
    tarsius committed Apr 6, 2014
  2. add support for using unicode characters when drawing log graphs

    Add new function `magit-log-format-unicode-graph' and add it as a
    choice in option `magit-log-format-graph-function'.  Also add new
    option `magit-log-format-unicode-graph-alist' allowing users to
    choose the characters (unicode or not, actually) to be used.
    tarsius committed Apr 6, 2014
Commits on Apr 3, 2014
  1. sanitize some faces

    Remove the face `magit-header' instead make `magit-section-title'
    inherit from `header-line' directly, without the redirection through
    `magit-header'.
    
    The removed face was described as "Many Magit faces inherit from this
    one by default", but for those faces that are *not* used by "headers"
    aka "titles", doing so was wrong.  So do *not* change those faces to
    inherit from `magit-section-title' and instead use the same colors as
    for the respective `magit-log-head-label-*' faces, but don't put a
    `:box' around them.
    
    There is more wrong with these and other faces, but that will be
    fixed later.
    tarsius committed Apr 3, 2014
Commits on Apr 2, 2014
  1. Merge pull request #1316 from lunaryorn/patch-1

    Fix quoting for default-value
    tarsius committed Apr 2, 2014
  2. Merge pull request #1315 from treese/fix-init

    Make interactive form return a list for magit-init.
    tarsius committed Apr 2, 2014
  3. magit-rewrite-pending-diff: new function

    This is a substitute for `magit-insert-pending-changes'.  The latter was
    problematic because its implementation has not kept up with improvements
    elsewhere.  I.e. the section it inserted was initially expanded and
    worse when the user did collapse it, it kept being re-expanded on every
    refresh.
    
    Unfortunately there is no easy fix.  The only place where we can
    currently prevent a diff from being expanded is in the "raw diff
    washing" code.  The pending changes section, unlike staged and unstaged,
    does not use raw washing.  It would be a major undertaking to teach the
    raw washing how to deal with pending changes.  I don't want to do that
    especially because having an additional raw washing cycle is a mistake
    itself, which I have already fixed on the `next' branch.
    tarsius committed Apr 2, 2014
  4. Fix quoting for default-value

    The argument to default-value needs to be quoted
    lunaryorn committed Apr 2, 2014
Commits on Mar 31, 2014
  1. Merge pull request #1310 from dachary/wip-submodules

    proofread the submodule section
    tarsius committed Mar 31, 2014
  2. magit-commit-add-log: bugfix

    As suggested by Daniel Dehennin in #1309.
    tarsius committed Mar 31, 2014
  3. proofread the submodule section

    Signed-off-by: Loic Dachary <loic@dachary.org>
    dachary committed Mar 31, 2014
  4. Merge pull request #1308 from dachary/wip-c-c-c-a

    The C-c C-a binding in the change description is for Ack
    tarsius committed Mar 31, 2014
  5. The C-c C-a binding in the change description is for Ack

    and not for toggling amend.
    dachary committed Mar 31, 2014
Commits on Mar 28, 2014
  1. magit-insert-status-{local,remote}-line: use `branch' as section type

    Adjust `magit-copy-item-as-kill' accordingly.
    tarsius committed Mar 28, 2014
  2. Make possible to copy more stuff

    - local branch name
    - remote branch name
    - untracked files
    - (un)staged files
    Marian Schubert committed Mar 28, 2014
Commits on Mar 27, 2014
  1. magit-{insert,put-face-property}: don't specify overlay priority

    We cannot specify the default priority because the overlays created to
    refine hunks (by code in `smerge-mode.el') also are created without the
    priority being explicitly specified.  When the priority is undefined,
    then it defaults to 1.  In our case there are several overlays with that
    priority at the same buffer position, but it seems the correct on wins.
    
    Apparently if there are multiple overlays with the same priority then
    the one which was created later is prioritized.  It so happens that this
    is what we need here, but we just got lucky and this should be fixed in
    Emacs.
    tarsius committed Mar 26, 2014
Commits on Mar 25, 2014
  1. Merge branch 'mushroom'

    tarsius committed Mar 25, 2014