Users often need to be able to push the current branch so that they
can publish their recent changes to anyone they are collaborating
with on the project.  Associating a keyboard action with this will
make it easier for keyboard-oriented users to quickly activate the
push features.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

We actually only want our F5/M1-R keystroke bound in the main window.
Within a browser/blame/console window pressing these keys should not
execute the rescan action.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

Pressing the escape key while in the merge dialog cancels the merge
and correctly unlocks the index.  Unfortunately this is not true of
the Cancel button, using it closes the dialog but does not release
the index lock, rendering git-gui frozen until you restart it.  We
now properly release the index lock when the Cancel button is used.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

On 'Visualize ...', a gitk process is started.  Since it is run in the
background, catching a possible startup error doesn't work, and the error
output goes to the console git-gui is started from.  The most probable
startup error is that gitk is not installed; so before trying to start,
check for the existence of the gitk program, and popup an error message
unless it's found.

This was noticed and reported by Paul Wise through
 http://bugs.debian.org/429810Signed-off-by: NGerrit Pape <pape@smarden.org>
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

The Tools/Migrate menu option is a hack just for me.  Yes, that's
right, git-gui has a hidden feature that really only works for me,
and the users that I support within my day-job's great firewall.
The menu option is not supported outside of that environment.

In the past we only enabled Tools/Migrate if our special local
script 'gui-miga' existed in the proper location, and if there
was a special '.pvcsrc' in the top level of the working directory.
This latter test for the '.pvcsrc' file is now failing, as the file
was removed from all Git repositories due to changes made to other
tooling within the great firewall's realm.

I have changed the test to only work on Cygwin, and only if the
special 'gui-miga' is present.  This works around the configuration
changes made recently within the great firewall's realm, but really
this entire Tools/Migrate thing should be abstracted out into some
sort of plugin system so other users can extend git-gui as they need.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

Johannes Sixt reported that MinGW/MSYS does not have a nice.exe to
drop the priority of a child process when it gets spawned.  So we
have to avoid trying to start `git blame` through nice when we are
on Windows and do not have Cygwin available to us.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

Junio asked that we don't force the user to have a valid X11 server
configured in $DISPLAY just to obtain the output of `git gui version`.
This makes sense, the user may be an automated tool that is running
without an X server available to it, such as a build script or other
sort of package management system.  Or it might just be a user working
in a non-GUI environment and wondering "what version of git-gui do I
have installed?".

Tcl has a lot of warts, but one of its better ones is that a comment
can be continued to the next line by escaping the LF that would have
ended the comment using a backslash-LF sequence.  In the past we have
used this trick to escape away the 'exec wish' that is actually a Bourne
shell script and keep Tcl from executing it.

I'm using that feature here to comment out the Bourne shell script and
hide it from the Tcl engine.  Except now our Bourne shell script is a
few lines long and checks to see if it should print the version, or not.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

The blame viewer is composed of two different areas, the file
area on top and the commit area on the bottom.  If users are
trying to shift the focus it is probably because they want to
shift from one area to the other, so we just setup Tab and
Shift-Tab to jump from the one half to the other in a cycle.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

Mark Levedahl <mlevedahl@gmail.com> noted that installation on Cygwin
to /usr/bin can cause problems with the automatic guessing of our
library location.  The problem is that installation to /usr/bin
means we actually have:

  /usr/bin   = c:\cygwin\bin
  /usr/share = c:\cygwin\usr\share

So git-gui guesses that its library should be found within the
c:\cygwin\share directory, as that is where it should be relative
to the script itself in c:\cygwin\bin.

In my first version of this patch I tried to use `cygpath` to resolve
/usr/bin and /usr/share to test that they were in the same relative
locations, but that didn't work out correctly as we were actually
testing /usr/share against itself, so it always was equal, and we
always used relative paths.  So my original solution was quite wrong.

Mark suggested we just always disable relative behavior on Cygwin,
because of the complexity of the mount mapping problem, so that's
all I'm doing.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

Because Tk does not assure us the order that it will process
children in before it destroys the main toplevel we cannot safely
save our geometry data during a "bind . <Destroy>" event binding.
The geometry may have already changed as a result of a one or
more children being removed from the layout.  This was pointed
out in gitk by Mark Levedahl, and patched over there by commit
b6047c5a.

So we now also use "wm protocol . WM_DELETE_WINDOW" to detect when
the window is closed by the user, and forward that close event to
our main do_quit routine.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

Apparently git-commit.sh (the command line commit user interface in
core Git) always gives precedence to the prior commit's message if
`commit --amend` is used and a $GIT_DIR/MERGE_MSG file also exists.

We actually were doing the same here in git-gui, but the amended
message got lost if $GIT_DIR/MERGE_MSG already existed because
we started a rescan immediately after loading the prior commit's
body into the edit buffer.  When that happened the rescan found
MERGE_MSG existed and replaced the commit message buffer with the
contents of that file.  This meant the user never saw us pick up
the commit message of the prior commit we are about to replace.

Johannes Sixt <J.Sixt@eudaptics.com> found this bug in git-gui by
running `git cherry-pick -n $someid` and then trying to amend the
prior commit in git-gui, thus combining the contents of $someid
with the contents of HEAD, and reusing the commit message of HEAD,
not $someid.  With the recent changes to make cherry-pick use the
$GIT_DIR/MERGE_MSG file Johannes saw git-gui pick up the message
of $someid, not HEAD.  Now we always use HEAD if we are amending.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

Earlier git.git applied a large "war on whitespace" patch that was
created using 'apply --whitespace=strip'.  Unfortunately a few of
git-gui's own files got caught in the mix and were also cleaned up.
That was a6080a0a.

This patch is needed in git-gui.git to reapply those exact same
changes here, otherwise our version generator script is unable to
obtain our version number from git-describe when we are hosted in
the git.git repository.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

The main window's diff header bar background switched from orange
to gold recently, and I liked the effect it had on readability of
the text.  Since I wanted the blame viewer to match, here it is.

Though this probably should be a user defined color, or at least
a constant somewhere that everyone can reference.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

Usually when you are looking at blame annotations for a region of
a file you are more interested in why something was originally
done then why it is here now.  This is because most of the time
when we get original annotation data we are looking at a simple
refactoring performed to better organize code, not to change its
semantic meaning or function.  Reorganizations are sometimes of
interest, but not usually.

We now show the original commit data first in the tooltip.  This
actually looks quite nice as the original commit will usually have an
author date prior to the current (aka move/copy) annotation's commit,
so the two commits will now tend to appear in chronological order.

I also found myself to always be clicking on the line of interest
in the file column but I always wanted the original tracking data
and not the move/copy data.  So I changed our default commit from
$asim_data (the simple move/copy annotation) to the more complex
$amov_data (the -M -C -C original annotation).
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

It feels wrong to call the -M -C -C annotations "move/copy tracking"
as they are actually the original locations.  So I'm relabeling
the status bar to show "copy/move tracking annotations" for the
current file (no -M -C -C) as that set of annotations tells us who
put the hunk here (who moved/copied it).  I'm now calling the -M
-C -C pass "original location annotations" as that's what we're
really digging for.

I also tried to clarify some of the text in the hover tooltip.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

To prevent neighboring lines that are different commits from using
the same background color we now use 3 colors and assign them
by selecting the color that is not used before or after the line
in question.  We still color "on the fly" as we receive hunks from
git-blame, but we delay our color decisions until we are getting
the original location data (the slower -M -C -C pass) as that is
usually more fine-grained than the current location data.

Credit goes to Martin Waitz for the tri-coloring concept.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

When the user clicks on a commit link within one of the columns
in the blame viewer we now jump them not just to that commit/file
pair but also to the line of the original file.  This saves the
user a lot of time, as they don't need to search through the new
file data for the chunk they were previously looking at.

We also restore the prior view when the user clicks the back button
to return to a pior commit/file pair that they were looking at.

Turned out this was quite tricky to get working in Tk.  Every time
I tried to jump the text widgets to the correct locations by way
of the "yview moveto" or "see" subcommands Tk performed the change
until the current event finished dispatching, and then reset the
views back to 0, making the change never take place.  Forcing Tk
to run the pending events before we jump the UI resolves the issue.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

If we have commit data from both the simple blame and the
rename/move tracking blame and they differ than there is a
bigger story to tell.  We now include data from both commits
so that the user can see that this link as moved, who moved
it, and where it originated from.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

We now perform two passes over any input file given to the blame
viewer.  Our first pass is a quick "git-blame" with no options,
getting the details of how each line arrived into this file.  We
are specifically ignoring/omitting the rename detection logic as
this first pass is to determine why things got into the state they
are in.

Once the first pass is complete and is displayed in the UI we run
a second pass, using the much more CPU intensive "-M -C -C" options
to perform extensive rename/movement detection.  The output of this
second pass is shown in a different column, allowing the user to see
for any given line how it got to be, and if it came from somewhere
else, where that is.

This is actually very instructive when run on our own lib/branch.tcl
script.  That file grew recently out of a very large block of code
in git-gui.sh.  The first pass shows when I created that file, while
the second pass shows the original commit information.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

If the user clicks on a line region that we haven't yet received
an annotation for from git-blame we show them "Loading annotation".
But I don't want the user to confuse this loading message with a
commit whose first line is "Loading annotation" and think we messed
up our display somehow.  Since we never use italics for anything
else, I'm going with the idea that italic slant can be used to show
data is missing/elided out at the time being.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

Calling the commit message pane $w_cmit is a tad confusing when
we also have the $w_cgrp column that shows the abbreviated SHA-1s.

So w_cmit -> w_cviewer, as it is the "commit viewer"; and
w_cgrp -> w_amov as it is the "annotated commit + move tracking"
column.  Also changed line_data -> amov_data, as that list is
exactly the results shown in w_amov.

Why call the column "move tracking"?  Because this column holds
data from "git blame -M -C".  I'm considering adding an additional
column that holds the data from "git blame" without -M/-C, showing
who did the copy/move, and when they did it.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

If the user runs the blame viewer on a working directory file
instead of a specific commit-ish then we have no value for the
commit SHA1 or the summary line; this causes the history menu
to get an empty entry at the very bottom.  We now look for this
odd case and call the meny entry "Working Directory".
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

The Tcl list datatype is significantly faster to work with than
the array type, especially if our indexes are a consecutive set
of numbers, like say line numbers in a file.

This rather large change reorganizes the internal data structure
of the blame viewer to use a proper Tcl list for the annotation
information about a line.  Each line is given its own list within
the larger line_data list, where the indexes correspond to various
facts about that particular line.

The interface does seem to be more responsive this way, with less
time required by Tcl to process blame, and to switch to another
version of the same file.  It could just be a placebo effect, but
either way most Tcl experts perfer lists for this type of work over
arrays.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

The code to handle our three different text widgets is a bit
on the messy side as we issue the same command on all three
widgets one at a time.  Adding (or removing) columns from the
viewer is messy, as a lot of locations need to have the new
column added into the sequence, or removed from it.

We also now delete the tags we create for each commit when
we switch to display another "commit:path" pair.  This way the
text viewer doesn't get bogged down with a massive number of tags
as we traverse through history.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

The array variable "order" used to be used to tell us in what
order each commit was received in.  Recent changes have removed
that need for an ordering and the "order" array is now just a
boolean 'do we have that commit yet' flag.

The colors were moved to fields, so they appear inside of the
blame viewer instance.  This keeps two different concurrently
running blame viewers from stepping on each other's ordering
of the colors in group_colors.

Most of the other fields were moved around a little bit so
that they are organized by major category and value lifespan.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

This list used to store the commits in the order we received
them in.  I originally was using it to update the colors of
the commit before and the commit after the current commit,
but since that interface concept turned out to be horribly
ugly and has been removed we no longer need this list.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

After we finish reading a chunk of data from the file stream
we know how many digits we need in the line number column to
show the current maximum line number.  If our line number column
isn't wide enough, we should expand it out to the correct width.

Any file over our default allowance of 5 digits (99,999 lines)
is so large that the slight UI "glitch" when we widen the column
out is trivial compared to the time it will take Git to fully do
the annotations.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

Most source code files are under 9,999 lines of text, so using a
field width of 5 characters meant that we should have had one char
padding on the left edge (because we right-justify the line number).
Unfortunately when I added the right margin earlier (when I removed
the padding) I ate into the extra character's space, losing the left
margin.  This put the line numbers too close to the commit column in
any file with more than 999 lines in it.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

The colors I originally picked out on a Mac OS X system look a
tad too dark on a Windows 2000 system; the greys are dark enough
to make it difficult to read some lines of text and the green used
to highlight the current commit was also difficult to read text on.

I also added a third grey to the mix, to try and help some files
that wind up with a number of neighboring chunks getting the same
colors.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

On Mac OS X the OS has "features" that like to draw thick black
borders around the text field that has focus.  This is nice if
you want to know where your text is going and are blind as a bat,
but it isn't the best thing to have in a table that is being
faked through the abuse of Tk text widgets.

By setting our takefocus, highlightthickness and padx/y we can
get rid of this border and get our text widgets packed right next
to each other, with no padding between them.  This makes the blame
background color smoothly run across the entire line of commit data,
line number and file content.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

Originally I had placed this loaded column between the line number
and the file line data to help users know if a particular line has
received annotation data or not yet.  This way users would know if
the line(s) they were interested in were ready for viewing, or if
they still had to wait.  It also was an entertaining way for the
user to spend their time waiting for git-blame --incremental to
compute the complete set of annotations.

However it is completely useless now that we show the abbreviated
commit SHA-1 and author initials in the leftmost column.  That area
is empty until we get the annotation data, and as soon as we get it
in we display something there, indicating to the user that there is
now blame data ready.  Further with the tooltips the user is likely
to see the data as soon as it comes in, as they are probably not
keeping their mouse perfectly still.  So I'm removing the field to
save screen space for more useful things, like file content.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

Some commit lines can get really long when users enter a lot of
text without linewrapping (for example).  Rather than letting the
menu get out of control in terms of width we clip the summary to
the first 50+ characters.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

Apparently Tk on Mac OS X won't draw a button with an image using a
transparent background.  Instead it draws the button using some sort
of 3D effect, even though I asked for no relief and no border.  The
background is also not our orange that we expected it to be.

Earlier I had tried this same trick on Windows and it draws the same
way as the button did, so I'm going to switch to the label as that
seems to be more portable.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

If we have two commits right next to each other in the final
file and they were kept as different blocks in the leftmost
column then its probably because the original filename was
different.  To help the user know where they are digging into
when they click on that link we now show the original file in
the tooltip, but to save space we do so only if the original
file is not the same as the file we are currently viewing.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

Consecutive chunks of a file could come from the same commit, but
have different original file names.  Previously we would have put
them into a single group, but then the hyperlink would jump to only
one of the files, and the other would not be accessible.  Now we can
get to the other file too.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

gitweb has long had a feature where the user can click on any
commit the blame display and go visit that commit's information
page.  From the user could go get the blame display for the file
they are tracking, and try to digg through the history of any
part of the code they are interested in seeing.

We now offer somewhat similiar functionality in git-gui.  The 4
digit commit abreviation in the first column of our blame view is
now offered as a hyperlink if the commit isn't the one we are now
viewing the blame output for (as there is no point in linking back
to yourself).  Clicking on that link will stop the current blame
engine (if still running), push the new target commit onto the
history stack, and restart the blame viewer at that commit, using
the "original file name" as supplied by git-blame for that chunk
of the output.

Users can navigate back to a version they had been viewing before
by way of a back button, which offers the prior commits in a popup
menu displayed right below the back button.  I'm always showing the
menu here as the cost of switching between views is very high; you
don't want to jump to a commit you are not interested in looking at
again.

During switches we throw away all data except the cached commit data,
as that is relatively small compared to most source files and their
annotation marks.  Unfortunately throwing this per-file data away in
Tcl seems to take some time; I probably should move the line indexed
arrays to proper lists and use [lindex] rather than the array lookup
(usually lists are faster).

We now start the git-blame process using "nice", so that its priority
will drop hopefully below our own.  If I don't do this the blame engine
gets a lot of CPU under Windows 2000 and the git-gui user interface is
almost non-responsive, even though Tcl is just sitting there waiting
for events.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

Computing the blame records for a large file with a long project
history can take git a while to run; traditionally we have shown
a little meter in the status area of our blame viewer that lets
the user know how many lines have been finished, and how far we
are through the process.

Usually such progress indicators are drawn with a little progress
bar in the window, where the bar shows how much has been completed
and hides itself when the process is complete.  I'm using a very
simple hack to do that: draw a canvas with a filled rectangle.

Of course the time remaining has absolutely no relationship to the
progress meter.  It could take very little time for git-blame to get
the first 90% of the file, and then it could take many times that to
get the remaining 10%.  So the progress meter doesn't really have any
sort of assurances that it relates to the true progress of the work.
But in practice on some ugly history it does seem to hold a reasonable
indicator to the completion status.  Besides, its amusing to watch and
that keeps the user from realizing git is being somewhat slow.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

At one point I tried to present the blame viewer to an audience of
people on a 640 by 480 pixel LCD projector.  This did not work at
all as the top area (the file data) was taking up all of the screen
realestate and the split point was not adjustable by the user.  In
general locking the user into a specific ratio of display is just
not user friendly.

So we now place a split pane control into the middle of our blame
window, so the user can adjust it to their current needs.  If the
window increases (or decreases) in height we assign the difference
to the file data area, as that is generally the area of the window
that users are trying to see more of when they grow the window.

Unfortunately there appears to be a bug in the "pack" layout manager
in Tcl/Tk 8.4.1.  The status bar and the lower commit pane was being
squashed if the window decreased in height.  I think the pack manager
was just not decreasing the size of the panedwindow slave properly if
the main window shrank.  Switching to the "grid" layout manager fixes
the problem, but is slightly uglier setup code.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

Frequently when I'm looking at blocks of code in the blame
viewer I want to know who is the culprit, or who I should
be praising for a job well done.  The tooltips nicely show
this if I mouse over a block, but it doesn't work to get
this detail at a glance.

Since we don't use the leftmost commit column for anything
after the first line within a commit group I'm now tossing
the author's initials into that field, right justified.  It
is quite clearly not a SHA-1 number as we always show the
SHA-1 in lowercase, while we explicitly select only the
uppercase characters from an author's name field, and only
those that are following whitespace.

I'm using initials here over anything else as they are quite
commonly unique within small development teams.  The leading
part of the email address field was out for some of the teams
I work with, as there the email addresses are all of the form
"Givenname.Surname@initech.com".  That will never fit into the
4 characters available.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>

The | in the continued lines of the same commit group as not
easily seen on the left edge; putting a single space in front
of the pipe makes it slightly more visually appealing to me as
I can follow the line down through the group to the next commit
marker.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>