The Preferences dialog gives control of the colors of some elements of
the gitk user interface, but many are hard-coded in the gitk script.
In order to allow these to be customized through the gitk config
file, these other colors are stored in variables which can be set
in the config file, thus providing a way for color schemes to be stored
and shared.

For win32, this makes the default foreground color that of window text
rather than button text.
Signed-off-by: NGauthier Östervall <gauthier@ostervall.se>
[paulus@samba.org: Reworded commit message to be clearer,
 changed filesepfgcolor to black]
Signed-off-by: NPaul Mackerras <paulus@samba.org>

gitk, when bound into the git.git project tree, used to live at the
root level, but in 62ba5143 (Move gitk to its own subdirectory,
2007-11-17) it was moved to a subdirectory.  The code used to track
changes to TCLTK_PATH (which should cause gitk to be rebuilt to
point at the new interpreter) was left in the main Makefile instead
of being moved to the new Makefile that was created for the gitk
project.

Also add .gitignore file to list build artifacts for the gitk
project.
Signed-off-by: NChristian Couder <chriscool@tuxfamily.org>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

When there are more than $maxrefs descendant heads to display in the
Branches field of the commit display, we currently just display "many",
which is not very informative.  To make the display more informative,
we now look for "master" and whichever head is currently checked out,
and display them even if there are too many heads to display them all.
The display then looks like "Branches: master and many more (33)" for
instance.
Signed-off-by: NPaul Mackerras <paulus@samba.org>

This provides a control in the preferences pane for the limit on
how many tags or heads are displayed with the commit details under
the Branch(es), Precedes and Follows headings.  This limit is now
saved in ~/.gitk so that changes are persistent.

This also applies word-wrapping to the list of tags or heads under
the Branch, Precedes and Follows headings, so that long lists are
more readable.
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Sometimes the code that divides commits up into arcs creates two
successive arcs, but the commit between them (the commit at the end
of the first arc and the beginning of the second arc) has only one
parent and one child.  If that commit is also the head of one or more
branches, those branches get omitted from the "Branches" field in the
commit display.

The omission occurs because the commit gets erroneously identified as
a commit which is part-way along an arc in [descheads].  This fixes it
by changing the test to look at the arcouts array, which only contains
elements for the commits at the start or end of an arc.
Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Signed-off-by: NPeter Krefting <peter@softwolves.pp.se>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

This fixes another regression that was introduced in b967135d ("gitk:
Synchronize highlighting in file view when scrolling diff"): when
searching for a string in tree mode, jumping to the next search hit
would highlight the "Comments" entry in the file list.
Signed-off-by: NStefan Haller <stefan@haller-berlin.de>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

When clicking on the line that connects two commit nodes, gitk
would bring up an error dialog saying "can't read "cflist_top":
no such variable".

This fixes a regression that was introduced with b967135d ("gitk:
Synchronize highlighting in file view when scrolling diff").
Signed-off-by: NStefan Haller <stefan@haller-berlin.de>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

When configured not to use themed widgets gitk may crash on launch with
a message that says that the image "bm-left disabled bm-left-gray"
doesn't exist. This happens when the left and right arrow buttons are
created.

The crash can be avoided by configuring the buttons differently
depending on whether or not themed widgets are used. If themed widgets
are not used then only set the images to bm-left and bm-right
respectively, and keep the old behavior when themed widgets are used.

The previous behaviour was added in f062e50f to work around a bug in Tk
on OS X where the disabled state did not display properly. The buttons
may still not display correctly, however the workaround added in
f062e50f will still apply if gitk is used with themed widgets.

Make gitk not crash on launch when not using themed widgets.
Signed-off-by: NMarcus Karlsson <mk@acc.umu.se>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Signed-off-by: NAndrew Wong <andrew.kw.w@gmail.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

The function includes a workaround for systems where F* keys are mapped
to XF86_Switch_VT_* when modifiers are used.
Signed-off-by: NAndrew Wong <andrew.kw.w@gmail.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

On Mac, the back and forward buttons show an empty rectange instead of
a grayed-out arrow when they are disabled. The reason is a Tk bug on Mac
that causes disabled images not to draw correctly (not to draw at all,
that is); see
<https://groups.google.com/forum/?fromgroups=#!topic/comp.lang.tcl/V-nW1JBq0eU>.

To work around this, we explicitly provide gray images for the disabled
state; I think this looks better than the default stipple effect that you
get on Windows as well, but that may be a matter of taste.
Signed-off-by: NStefan Haller <stefan@haller-berlin.de>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

When typing in the "Search" field, select the current search result (so
that it gets highlighted in orange). This makes it easier to understand
what will happen if you then type Ctrl-S.
Signed-off-by: NStefan Haller <stefan@haller-berlin.de>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

When searching for text in the diff, and there are multiple occurrences
of the search string, the current one is highlighted in orange, and the
other ones in yellow. This makes it much easier to understand what happens
when you then click the Search button or hit Ctrl-S repeatedly.
Signed-off-by: NStefan Haller <stefan@haller-berlin.de>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Whenever the diff pane scrolls, highlight the corresponding file in the
file list on the right.  For a large commit with many files and long
per-file diffs, this makes it easier to keep track of what you're looking
at.

This allows simplifying the prevfile and nextfile functions, because
all they have to do is scroll the diff pane.

In some situations we want to suppress this mechanism, for example when
clicking on a file in the file list to select it, or when searching in the
diff, in which case we want to highlight based on the current search hit
and not the top line visible. In these cases it's not sufficient to set
a "suppress" flag before scrolling and reset it afterwards, because the
scrolltext notification is sent deferred from a timer or some such; so we
need to remember the scroll position for which we want to suppress the
auto-highlighting until the next call to scrolltext; a bit ugly, but does
the job.
Signed-off-by: NStefan Haller <stefan@haller-berlin.de>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

to get two commits from David Aguilar.

Name the 'tagcontents' variable similarly to the rest of the
variables cleared in the changedrefs() function.

This makes the naming consistent and provides a hint that it
should be cleared when reloading gitk's cache.
Suggested-by: NJunio C Hamano <gitster@pobox.com>
Signed-off-by: NDavid Aguilar <davvid@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Tag contents, once read, are forever cached in memory.
This makes gitk unable to notice when tag contents change.

Allow users to cause a reload of the tag contents by using
the "File->Reread references" action.
Reported-by: NTim McCormack <cortex@brainonfire.net>
Suggested-by: NJunio C Hamano <gitster@pobox.com>
Signed-off-by: NDavid Aguilar <davvid@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Check if GIT_TRACE env var is set and unset it if it is.

If the environment var GIT_TRACE=1 exists gitk will fail when trying
to find gitdir:
$ git rev-parse --git-dir
trace: built-in: git 'rev-parse' '--git-dir'
.git

Other git commands will also show debug output hence not work as
intended.
Signed-off-by: NAske Olsson <askeolsson@gmail.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Meta1-F5 is commonly mapped by window managers and what not.
Use Shift-F5 instead.
Signed-off-by: NFelipe Contreras <felipe.contreras@gmail.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

The changes to the dialog window tree broke the preview of the selected
font on the button. This corrects that issue.
Signed-off-by: NPat Thoyts <patthoyts@users.sourceforge.net>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

In 8.5 the incr command creates the target variable if it does not exist
but in 8.4 using incr on a non-existing variable raises an error. Ensure
we have created our counter variable when creating the tabbed dialog for
non-themed preferences.
Reported-by: NRamsay Jones <ramsay@ramsay1.demon.co.uk>
Signed-off-by: NPat Thoyts <patthoyts@users.sourceforge.net>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

In early days, all projects managed by git (except for git itself) had the
product of a fairly mature development history in their first commit, and
it was deemed unnecessary clutter to show additions of these thousands of
paths as a patch.

"git log" learned to show the patch for the initial commit without requiring
--root command line option at 0f03ca94 (config option log.showroot to show
the diff of root commits, 2006-11-23).

Teach gitk to respect log.showroot.

[paulus@samba.org: Cleaned up the Tcl a bit, use --bool on the
 git config call]
Signed-off-by: NMarcus Karlsson <mk@acc.umu.se>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Sometimes one wants to see the different between two commits that are
a long distance apart in the graph display.  This is difficult to do
with the "Diff this -> selected" and "Diff selected -> this" menu
items because the need to maintain the selection means that one can't
use the find facilities or the reference list window to navigate from
one to the other.

This provides an alternative using the mark.  Having found one commit,
one marks it with the "Mark this commit" menu item, then navigates to
the other commit and uses the new "Diff this -> marked commit" and/or
"Diff marked commit -> this" menu items.
Signed-off-by: NPaul Mackerras <paulus@samba.org>

On large repositories such as the Linux kernel, it can take quite a
noticeable time (several seconds) for gitk to resolve short SHA1 IDs
to their long form.  This speeds up the process by maintaining lists
of IDs indexed by the first 4 characters of the SHA1 ID, speeding up
the search by a factor of 65536 on large repositories.
Signed-off-by: NPaul Mackerras <paulus@samba.org>

The following only concerns systems using X and the client-side font
rendering framework from freedesktop.org.  Windows and Mac OS X are
not affected.

Starting with version 8.5, Tk uses freetype and fontconfig by default
to render fonts on platforms that support it.  Gitk currently defaults
to the font Helvetica for the interface and Courier for diffs, and
both unfortunately look rather bad on screen in the default
configuration on many Linux distros with anti-aliasing and poor
hinting.

It is better to default to "sans" and "monospace", which are mapped by
fontconfig to some appropriate font of the sysadmin and user's
choosing (typically Bitstream Vera Sans and Mono).  The result looks
more sensible and it makes gitk feel like a well-behaved software
citizen since its fonts match other native apps.

This patch does not change the appearance of gitk for users that have
already run it, since gitk uses the remembered UI and diff font names
from ~/.gitk.
Requested-by: NMichael Biebl <biebl@debian.org>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>
Acked-by: NMark Hills <mark@pogo.org.uk>
Signed-off-by: NJonathan Nieder <jrnieder@gmail.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

This prevents a search for a number like "105" on "All Fields" from
matching against the raw author and commit timestamps.  These
timestamps were already not searchable by themselves, and the
displayed format does not match the query string anyway.
Signed-off-by: NFrédéric Brière <fbriere@fbriere.net>
Signed-off-by: NJonathan Nieder <jrnieder@gmail.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Automake's contribution guidelines suggest using "git describe" output
in commit logs to reference previous commits.  By contrast, in
coreutils, I had acquired the habit of using a bare SHA1 prefix (8 hex
digits), since gitk creates clickable links for that, and not for "git
describe" output.

I prefer the readability of the full "git describe" output, yet want
to retain the gitk links, so this renders as clickable not just
SHA1-like strings, but also an SHA1-like string that is prefixed by
"-g".
Signed-off-by: NJim Meyering <meyering@redhat.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Launching 'gitk -- .' or 'gitk -- ..\t' restricts the display to files
under the given directory but the file list is left empty. This is because
the path_filter function fails to match the filenames which are relative
to the working tree to the filter which is filessytem relative.
This solves the problem by making both names fully qualified filesystem
paths before performing the comparison.
Tested-by: NDavid Aguilar <davvid@gmail.com>
Signed-off-by: NPat Thoyts <patthoyts@users.sourceforge.net>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

This commit converts the user preferences dialog into a tabbed property
sheet grouping general properties, colours and font selections onto
separate pages. The previous implementation was exceeding the screen
height on some systems and this avoids such problems and permits extension
using new pages in the future.

If themed Tk is unavailable or undesired a reasonable facsimile of the
tabbed notebook widget is used instead.
Signed-off-by: NPat Thoyts <patthoyts@users.sourceforge.net>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Previously, when run in a subdirectory, gitk would show the name
of this subdirectory as title, which was misleading. When run with
GIT_DIR set, it would show the cwd, which is even more misleading.

In case of non-bare repos, the .git suffix in the path is skipped.
Signed-off-by: NZbigniew Jędrzejewski-Szmek <zbyszek@in.waw.pl>

When commit 6e2dda35 (Add new keybindings, 2005-09-22) added vi-style
keybindings to gitk (an excellent idea!), instead of adopting the
usual "hjkl = left, down, up, right" bindings used by less, vi, rogue,
and many other programs, it used "ijkl = up, left, down, right" to
mimic the inverted-T formation of the arrow keys on a qwerty keyboard,
in the style of Lode runner.  So using 'j' and 'k' to scroll through
commits produces utterly confusing results to the vi user, as 'k'
moves down and 'j' moves to the previous commit.

Luckily most non-vi-users are probably using an alternate set of keys
(cursor keys or z/x + n/p) anyway.  Switch to the expected vi/nethack
convention.
Requested-by: NJosh Triplett <josh@joshtriplett.org>
Signed-off-by: NJonathan Nieder <jrnieder@gmail.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Gitk can search for commits touching a specified path. The search text is
always treated as a regular expression, regardless of the matching option
selected (Exact, IgnCase, or Regexp). In particular, backslashes escape
the next character. This is inconvenient on Windows systems, where backslashes
are the norm for path specifiers, for example when copy/pasting from
Windows Explorer or a cmd shell -- these copy-pasted paths must be manually
modified in the gitk search text edit box before they will work.

This change uses the match option "Exact" to mean that a slash is a slash,
not part of a regular expression. Backslashes are converted to frontslashes
before searching, thus allowing easy copy/pasting of paths on Windows
systems. If the previous behaviour of "touching paths" search is desired,
simply select the "Regexp" search mode.

One potential drawback is that the default setting for the match option
($findtype in the code) is "Exact", and so this change alters the default
behaviour, which may confuse users and lead to bug reports.
Signed-off-by: NYggy King <yggy@zeroandone.ca>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

"git rev-parse --is-inside-work-tree" is currently used to determine
whether to show modified files in gitk (the red and green fake
commits). This does not work if the current directory is not inside
the work tree, as can be the case e.g. if GIT_WORK_TREE is
set. Instead, check if the repository is not bare and that we are not
inside the .git directory.
Signed-off-by: NMartin von Zweigbergk <martin.von.zweigbergk@gmail.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Since 5024baa4 ([PATCH] Make gitk work when launched in a subdirectory,
2007-01-09), gitk has used 'git rev-parse --git-dir' to find the .git
directory. However, gitk still first checks for the $GIT_DIR
environment variable and that the value returned from git-rev-parse
does not point to a file. Since git-rev-parse does both of these
checks already, the checks can safely be removed from gitk. This makes
the gitdir procedure small enough to inline.

This cleanup introduces a UI regression in that the error message will
now be "Cannot find a git repository here." even in the case where
GIT_DIR points to a file, for which the error message was previously
"Cannot find the git directory \"%s\".". It should be noted, though,
that even before this patch, 'gitk --git-dir=path/to/some/file' would
give the former error message.
Signed-off-by: NMartin von Zweigbergk <martin.von.zweigbergk@gmail.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

It seems like gitk has been setting the global variable 'gitdir' at
startup since aa81d974 (gitk: Fix Update menu item, 2006-02-28).  It
should therefore no longer be necessary to call the procedure with the
same name (more than once to set the global variable).  Remove the
other call sites and use the global variable instead.
Signed-off-by: NMartin von Zweigbergk <martin.von.zweigbergk@gmail.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

When running "External diff" from gitk, the "from" and "to" files will
first be copied into a directory that is currently
".git/../.gitk-tmp.$pid".  When gitk is closed, the directory is
deleted. When the work tree is not at ".git/.." (which is supported
since the previous commit), that directory may not even be git-related
and it does not seem unlikely that permissions may not allow the
temporary directory to be created there.  Move the directory inside
.git instead.

This introduces a regression in the case that the .git directory
is readonly, but .git/.. is writeable.
Signed-off-by: NMartin von Zweigbergk <martin.von.zweigbergk@gmail.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Running "External diff" to compare the index and work tree currently
brings up an empty blame view when the work tree is not the parent of
the git directory.  This is because the file that is taken from the
work tree is assumed to be in
$GIT_DIR/../<repo-relative-file-name>.  Fix it by feeding the diff tool
a path under $GIT_WORK_TREE instead of "$GIT_DIR/..".
Signed-off-by: NMartin von Zweigbergk <martin.von.zweigbergk@gmail.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Running "blame parent commit" currently brings up an empty blame view
when the the work tree is not the parent of the git directory.  Fix it
by feeding git-blame paths relative to $GIT_WORK_TREE instead of
"$GIT_DIR/..".
Signed-off-by: NMartin von Zweigbergk <martin.von.zweigbergk@gmail.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Running "show origin of this line" currently fails when the the work
tree is not the parent of the git directory.  Fix it by feeding
git-blame paths relative to $GIT_WORK_TREE instead of "$GIT_DIR/..".
Signed-off-by: NMartin von Zweigbergk <martin.von.zweigbergk@gmail.com>
Signed-off-by: NPaul Mackerras <paulus@samba.org>