]> git.99rst.org Git - git.git/log
git.git
21 years agoMerge with gitk.
Junio C Hamano [Mon, 8 Aug 2005 00:10:20 +0000 (17:10 -0700)]
Merge with gitk.

This merges commit
    fa4da7b32e1197fa5032db6a2de7749868253f8b
from gitk into our head commit
    6b7242aa1acc3c7835f80522914ffc4b2e789a29

Sincerely,
    jit-merge command.

21 years agoBetter graph line details display and expand history coverage.
Paul Mackerras [Sun, 7 Aug 2005 23:47:22 +0000 (09:47 +1000)]
Better graph line details display and expand history coverage.

Now the history remembers when we have clicked on a graph line
and when we have asked for a diff between two commits, as well
as when we have displayed a commit.

The display when you click on a graph line now uses clickable
SHA1 IDs instead of the embedded "Go" buttons.  Also made the
IDs clickable in the header for a diff between two commits.

21 years ago[PATCH] gitk "parent information" in commit window
Linus Torvalds [Sun, 7 Aug 2005 20:58:56 +0000 (13:58 -0700)]
[PATCH] gitk "parent information" in commit window

This adds a useful "Parent:" line to the git commit information window.

It looks something like this (from the infamous octopus merge):

Author: Junio C Hamano <redacted>  2005-05-05 16:16:54
Committer: Junio C Hamano <redacted>  2005-05-05 16:16:54
Parent: fc54a9c30ccad3fde5890d2c0ca2e2acc0848fbc  (Update git-apply-patch-script ...)
Parent: 9e30dd7c0ecc9f10372f31539d0122db97418353  (Make git-prune-script executa ...)
Parent: c4b83e618f1df7d8ecc9392fa40e5bebccbe6b5a  (Do not write out new index if ...)
Parent: 660265909fc178581ef327076716dfd3550e6e7b  (diff-cache shows differences  ...)
Parent: b28858bf65d4fd6d8bb070865518ec43817fe7f3  (Update diff engine for symlin ...)

    Octopus merge of the following five patches.

      Update git-apply-patch-script for symbolic links.
      Make git-prune-script executable again.
      Do not write out new index if nothing has changed.
      diff-cache shows differences for unmerged paths without --cache.
      Update diff engine for symlinks stored in the cache.

Signed-off-by: Junio C Hamano <redacted>
where all the parent commit ID's are clickable, because the new lines are
added as part of the "comment" string, and thus the regular clickability
thing will match them automatically.

I think this is good. And my random-tcl-monkey-skills are clearly getting
better (although it's perfectly possible that somebody who actually knows
what he is doing would have done things differently).

Signed-off-by: Linus Torvalds <redacted>
Signed-off-by: Paul Mackerras <redacted>
21 years agoMake the tag view look like commit header.
Kay Sievers [Sun, 7 Aug 2005 22:02:39 +0000 (00:02 +0200)]
Make the tag view look like commit header.

21 years agoMerge with master.
Junio C Hamano [Sun, 7 Aug 2005 21:12:38 +0000 (14:12 -0700)]
Merge with master.

This merges commit
    6b7242aa1acc3c7835f80522914ffc4b2e789a29
from master into our head commit
    f29eaed12fc07bc5b2e4d1a302136a26ff4488ce

Sincerely,
    jit-merge command.

21 years agoMerge with gitk.
Junio C Hamano [Sun, 7 Aug 2005 21:11:08 +0000 (14:11 -0700)]
Merge with gitk.

This merges commit
    d6e8149bc53de57cab7cc0d9aff6c2341f3a7efa
from gitk into our head commit
    200e7fbbac82b6665f8175af0e375fbef3568c30

Sincerely,
    jit-merge command.

21 years ago(revert local fix)
Junio C Hamano [Sun, 7 Aug 2005 21:11:02 +0000 (14:11 -0700)]
(revert local fix)

Signed-off-by: Junio C Hamano <redacted>
21 years agoadd README
Kay Sievers [Sun, 7 Aug 2005 20:39:58 +0000 (22:39 +0200)]
add README

Two files are better than one. :)

21 years agov236
Kay Sievers [Sun, 7 Aug 2005 18:29:03 +0000 (20:29 +0200)]
v236

21 years agov235
Kay Sievers [Sun, 7 Aug 2005 18:28:53 +0000 (20:28 +0200)]
v235

21 years agov234
Kay Sievers [Sun, 7 Aug 2005 18:28:42 +0000 (20:28 +0200)]
v234

21 years agov233
Kay Sievers [Sun, 7 Aug 2005 18:28:33 +0000 (20:28 +0200)]
v233

21 years agov232
Kay Sievers [Sun, 7 Aug 2005 18:28:01 +0000 (20:28 +0200)]
v232

21 years agov229
Kay Sievers [Sun, 7 Aug 2005 18:27:49 +0000 (20:27 +0200)]
v229

21 years agov227
Kay Sievers [Sun, 7 Aug 2005 18:27:38 +0000 (20:27 +0200)]
v227

21 years agov225
Kay Sievers [Sun, 7 Aug 2005 18:27:27 +0000 (20:27 +0200)]
v225

21 years agov220
Kay Sievers [Sun, 7 Aug 2005 18:27:18 +0000 (20:27 +0200)]
v220

21 years agov206
Kay Sievers [Sun, 7 Aug 2005 18:26:49 +0000 (20:26 +0200)]
v206

21 years agov205
Kay Sievers [Sun, 7 Aug 2005 18:26:38 +0000 (20:26 +0200)]
v205

21 years agov203
Kay Sievers [Sun, 7 Aug 2005 18:26:27 +0000 (20:26 +0200)]
v203

21 years agov165
Kay Sievers [Sun, 7 Aug 2005 18:26:12 +0000 (20:26 +0200)]
v165

21 years agov164
Kay Sievers [Sun, 7 Aug 2005 18:26:03 +0000 (20:26 +0200)]
v164

21 years agov163
Kay Sievers [Sun, 7 Aug 2005 18:25:54 +0000 (20:25 +0200)]
v163

21 years agov160
Kay Sievers [Sun, 7 Aug 2005 18:25:42 +0000 (20:25 +0200)]
v160

21 years agov157
Kay Sievers [Sun, 7 Aug 2005 18:25:27 +0000 (20:25 +0200)]
v157

21 years agov154
Kay Sievers [Sun, 7 Aug 2005 18:24:51 +0000 (20:24 +0200)]
v154

21 years agov152
Kay Sievers [Sun, 7 Aug 2005 18:24:43 +0000 (20:24 +0200)]
v152

21 years agov150
Kay Sievers [Sun, 7 Aug 2005 18:24:35 +0000 (20:24 +0200)]
v150

21 years agov149
Kay Sievers [Sun, 7 Aug 2005 18:24:01 +0000 (20:24 +0200)]
v149

21 years agov148
Kay Sievers [Sun, 7 Aug 2005 18:23:49 +0000 (20:23 +0200)]
v148

21 years agov145
Kay Sievers [Sun, 7 Aug 2005 18:23:35 +0000 (20:23 +0200)]
v145

21 years agov143
Kay Sievers [Sun, 7 Aug 2005 18:23:24 +0000 (20:23 +0200)]
v143

21 years agov142
Kay Sievers [Sun, 7 Aug 2005 18:23:12 +0000 (20:23 +0200)]
v142

21 years agov136
Kay Sievers [Sun, 7 Aug 2005 18:22:53 +0000 (20:22 +0200)]
v136

21 years agov133
Kay Sievers [Sun, 7 Aug 2005 18:22:44 +0000 (20:22 +0200)]
v133

21 years agov125
Kay Sievers [Sun, 7 Aug 2005 18:21:46 +0000 (20:21 +0200)]
v125

21 years agov121
Kay Sievers [Sun, 7 Aug 2005 18:21:34 +0000 (20:21 +0200)]
v121

21 years agov118
Kay Sievers [Sun, 7 Aug 2005 18:21:23 +0000 (20:21 +0200)]
v118

21 years agov107
Kay Sievers [Sun, 7 Aug 2005 18:21:04 +0000 (20:21 +0200)]
v107

21 years agov089
Kay Sievers [Sun, 7 Aug 2005 18:20:20 +0000 (20:20 +0200)]
v089

21 years agov088
Kay Sievers [Sun, 7 Aug 2005 18:20:07 +0000 (20:20 +0200)]
v088

21 years agov085
Kay Sievers [Sun, 7 Aug 2005 18:19:56 +0000 (20:19 +0200)]
v085

21 years agov082
Kay Sievers [Sun, 7 Aug 2005 18:19:45 +0000 (20:19 +0200)]
v082

21 years agov080
Kay Sievers [Sun, 7 Aug 2005 18:18:44 +0000 (20:18 +0200)]
v080

21 years agov078
Kay Sievers [Sun, 7 Aug 2005 18:18:30 +0000 (20:18 +0200)]
v078

21 years agov077
Kay Sievers [Sun, 7 Aug 2005 18:18:13 +0000 (20:18 +0200)]
v077

21 years agov073
Kay Sievers [Sun, 7 Aug 2005 18:18:01 +0000 (20:18 +0200)]
v073

21 years agov071
Kay Sievers [Sun, 7 Aug 2005 18:17:50 +0000 (20:17 +0200)]
v071

21 years agov070
Kay Sievers [Sun, 7 Aug 2005 18:17:42 +0000 (20:17 +0200)]
v070

21 years agov064
Kay Sievers [Sun, 7 Aug 2005 18:17:19 +0000 (20:17 +0200)]
v064

21 years agov063
Kay Sievers [Sun, 7 Aug 2005 18:17:09 +0000 (20:17 +0200)]
v063

21 years agov062
Kay Sievers [Sun, 7 Aug 2005 18:17:00 +0000 (20:17 +0200)]
v062

21 years agov057
Kay Sievers [Sun, 7 Aug 2005 18:16:07 +0000 (20:16 +0200)]
v057

21 years agov056
Kay Sievers [Sun, 7 Aug 2005 18:15:57 +0000 (20:15 +0200)]
v056

21 years agov055
Kay Sievers [Sun, 7 Aug 2005 18:15:44 +0000 (20:15 +0200)]
v055

21 years agov053
Kay Sievers [Sun, 7 Aug 2005 18:14:48 +0000 (20:14 +0200)]
v053

rename gitweb.pl to gitweb.cgi

21 years agov049
Kay Sievers [Sun, 7 Aug 2005 18:13:11 +0000 (20:13 +0200)]
v049

21 years agov048
Kay Sievers [Sun, 7 Aug 2005 18:13:02 +0000 (20:13 +0200)]
v048

21 years agov043
Kay Sievers [Sun, 7 Aug 2005 18:12:11 +0000 (20:12 +0200)]
v043

21 years agov042
Kay Sievers [Sun, 7 Aug 2005 18:09:33 +0000 (20:09 +0200)]
v042

21 years agov041
Kay Sievers [Sun, 7 Aug 2005 18:08:29 +0000 (20:08 +0200)]
v041

21 years agov041
Kay Sievers [Sun, 7 Aug 2005 18:08:03 +0000 (20:08 +0200)]
v041

21 years agov035
Kay Sievers [Sun, 7 Aug 2005 18:06:09 +0000 (20:06 +0200)]
v035

21 years agov031
Kay Sievers [Sun, 7 Aug 2005 18:05:55 +0000 (20:05 +0200)]
v031

21 years agov027
Kay Sievers [Sun, 7 Aug 2005 18:05:44 +0000 (20:05 +0200)]
v027

21 years agov026
Kay Sievers [Sun, 7 Aug 2005 18:05:32 +0000 (20:05 +0200)]
v026

21 years agov025
Kay Sievers [Sun, 7 Aug 2005 18:05:15 +0000 (20:05 +0200)]
v025

21 years agov021
Kay Sievers [Sun, 7 Aug 2005 18:03:52 +0000 (20:03 +0200)]
v021

21 years agov021
Kay Sievers [Sun, 7 Aug 2005 18:03:14 +0000 (20:03 +0200)]
v021

21 years agov021
Kay Sievers [Sun, 7 Aug 2005 18:02:47 +0000 (20:02 +0200)]
v021

21 years agov021
Kay Sievers [Sun, 7 Aug 2005 18:02:33 +0000 (20:02 +0200)]
v021

21 years agov020
Kay Sievers [Sun, 7 Aug 2005 18:00:05 +0000 (20:00 +0200)]
v020

21 years agov016
Kay Sievers [Sun, 7 Aug 2005 17:59:41 +0000 (19:59 +0200)]
v016

21 years agov014
Kay Sievers [Sun, 7 Aug 2005 17:59:24 +0000 (19:59 +0200)]
v014

21 years agov006
Kay Sievers [Sun, 7 Aug 2005 17:58:12 +0000 (19:58 +0200)]
v006

21 years agov005
Kay Sievers [Sun, 7 Aug 2005 17:57:58 +0000 (19:57 +0200)]
v005

21 years agov004
Kay Sievers [Sun, 7 Aug 2005 17:56:59 +0000 (19:56 +0200)]
v004

21 years agov003
Kay Sievers [Sun, 7 Aug 2005 17:56:44 +0000 (19:56 +0200)]
v003

21 years agov003
Kay Sievers [Sun, 7 Aug 2005 17:56:10 +0000 (19:56 +0200)]
v003

21 years agov002
Kay Sievers [Sun, 7 Aug 2005 17:55:05 +0000 (19:55 +0200)]
v002

21 years agov001
Kay Sievers [Sun, 7 Aug 2005 17:54:31 +0000 (19:54 +0200)]
v001

21 years agov000
Kay Sievers [Sun, 7 Aug 2005 17:53:54 +0000 (19:53 +0200)]
v000

21 years agov000
Kay Sievers [Sun, 7 Aug 2005 17:52:52 +0000 (19:52 +0200)]
v000

21 years agofirst working version
Kay Sievers [Sun, 7 Aug 2005 17:49:46 +0000 (19:49 +0200)]
first working version

21 years agoUse lf translation rather than binary when reading commit data.
Paul Mackerras [Sun, 7 Aug 2005 10:01:24 +0000 (20:01 +1000)]
Use lf translation rather than binary when reading commit data.

The effect of this is that it allows Tcl to do the locale-specific
conversion of the input data to its internal unicode representation.
That means that commit messages in Russian or other languages should
be displayed correctly now (according to the locale that is in effect.)

21 years agoFix build rules for debian package.
Junio C Hamano [Sun, 7 Aug 2005 05:29:21 +0000 (22:29 -0700)]
Fix build rules for debian package.

Run install-tools target to install the tools to accept e-mail
patches.  Also clean up the main Makefile a bit.

Signed-off-by: Junio C Hamano <redacted>
21 years agoChange cursor to a hand cursor when over a SHA1 ID link.
Paul Mackerras [Sun, 7 Aug 2005 05:27:57 +0000 (15:27 +1000)]
Change cursor to a hand cursor when over a SHA1 ID link.

This is based on suggestions by Jeff Epler and Linus Torvalds, but
extended so that we do the switching between the watch cursor and
the normal cursor correctly as well.

Also fixed a bug pointed out by Junio Hamano - I wasn't incrementing
the link number (duh!).

21 years agoFix RPM build that omitted templates and tools.
Junio C Hamano [Sun, 7 Aug 2005 03:38:13 +0000 (20:38 -0700)]
Fix RPM build that omitted templates and tools.

Many many thanks go to Chris Wright and H. Peter Anvin whose
help were essential to get me going this build.

Signed-off-by: Junio C Hamano <redacted>
21 years agoMerge with master.
Junio C Hamano [Sun, 7 Aug 2005 03:54:43 +0000 (20:54 -0700)]
Merge with master.

This merges commit
    bfe19f876cb20bea606e1a698030c017f31965c1
from master into our head commit
    7a59013290bb26b80c997e5698757a735f1a92e0

Sincerely,
    jit-merge command.

21 years ago[PATCH] Extend "git reset" to take a reset point
Linus Torvalds [Sun, 7 Aug 2005 01:01:03 +0000 (18:01 -0700)]
[PATCH] Extend "git reset" to take a reset point

This was triggered by a query by Sam Ravnborg, and extends "git reset" to
reset the index and the .git/HEAD pointer to an arbitrarily named point.

For example

git reset HEAD^

will just reset the current HEAD to its own parent - leaving the working
directory untouched, but effectively un-doing the top-most commit. You
might want to do this if you realize after you committed that you made a
mistake that you want to fix up: reset your HEAD back to its previous
state, fix up the working directory and re-do the commit.

If you want to totally un-do the commit (and reset your working directory
to that point too), you'd first use "git reset HEAD^" to reset to the
parent, and then do a "git checkout -f" to reset the working directory
state to that point in time too.

Signed-off-by: Linus Torvalds <redacted>
Signed-off-by: Junio C Hamano <redacted>
21 years agoGIT 0.99.4 (release candidate)
Junio C Hamano [Sat, 6 Aug 2005 20:10:43 +0000 (13:10 -0700)]
GIT 0.99.4 (release candidate)

This is my first attempt to adjust Debian and RPM to pass
prefix, to prepare the 0.99.4 release.

It updates debian/rules and git-core.spec.in to properly pass
prefix when building binary packages.  It also updates
debian/changelog to make the resulting binary package name
0.99.4; this is not needed on the RPM side (it takes the version
number from the main Makefile).

Signed-off-by: Junio C Hamano <redacted>
21 years agoRedo the templates generation and installation.
Junio C Hamano [Sat, 6 Aug 2005 19:50:14 +0000 (12:50 -0700)]
Redo the templates generation and installation.

Per discussion with people interested in binary packaging,
change the default template location from /etc/git-core to
/usr/share/git-core hierarchy.  If a user wants to run git
before installing for whatever reason, in addition to adding
$src to the PATH environment variable, git-init-db can be run
with --template=$src/templates/blt/ parameter.

Signed-off-by: Junio C Hamano <redacted>
21 years agogitk proposed fix: handle more than one SHA1 links.
Junio C Hamano [Sat, 6 Aug 2005 16:28:31 +0000 (09:28 -0700)]
gitk proposed fix: handle more than one SHA1 links.

Signed-off-by: Junio C Hamano <redacted>
21 years agosend-pack: allow generic sha1 expression on the source side.
Junio C Hamano [Sat, 6 Aug 2005 17:12:03 +0000 (10:12 -0700)]
send-pack: allow generic sha1 expression on the source side.

This extends the source side semantics to match what Linus
suggested.

An example:

    $ git-send-pack kernel.org:/pub/scm/git/git.git pu^^:master pu

    would allow me to push the current pu into pu, and the
    commit two commits before it into master, on my public
    repository.

The revised rule for updating remote heads is as follows.

 $ git-send-pack [--all] <remote> [<ref>...]

 - When no <ref> is specified:

   - with '--all', it is the same as specifying the full refs/*
     path for all local refs;

   - without '--all', it is the same as specifying the full
     refs/* path for refs that exist on both ends;

 - When one or more <ref>s are specified:

   - a single token <ref> (i.e. no colon) must be a pattern that
     tail-matches refs/* path for an existing local ref.  It is
     an error for the pattern to match no local ref, or more
     than one local refs.  The matching ref is pushed to the
     remote end under the same name.

   - <src>:<dst> can have different cases.  <src> is first tried
     as the tail-matching pattern for refs/* path.

     - If more than one matches are found, it is an error.

     - If one match is found, <dst> must either match no remote
       ref and start with "refs/", or match exactly one remote
       ref.  That remote ref is updated with the sha1 value
       obtained from the <src> sha1.

     - If no match is found, it is given to get_extended_sha1();
       it is an error if get_extended_sha1() does not find an
       object name.  If it succeeds, <dst> must either match
       no remote ref and start with "refs/" or match exactly
       one remote ref.  That remote ref is updated with the sha1
       value.

Signed-off-by: Junio C Hamano <redacted>
21 years agosend-pack: allow the same source to be pushed more than once.
Junio C Hamano [Sat, 6 Aug 2005 17:16:52 +0000 (10:16 -0700)]
send-pack: allow the same source to be pushed more than once.

The revised code accidentally inherited the restriction that a
reference can be pushed only once, only because the original did
not allow renaming.  This is no longer necessary so lift it.

Signed-off-by: Junio C Hamano <redacted>
21 years ago[PATCH] Make git-sh-setup-script do what it was supposed to do
Linus Torvalds [Sat, 6 Aug 2005 17:04:50 +0000 (10:04 -0700)]
[PATCH] Make git-sh-setup-script do what it was supposed to do

Duh. A missing && meant that half the tests that git-sh-setup-script were
_meant_ to do were actually totally ignored.

In particular, the git sanity checking ended up only testing that the
GIT_OBJECT_DIRECTORY was sane, not that GIT_DIR itself was..

Signed-off-by: Linus Torvalds <redacted>
Signed-off-by: Junio C Hamano <redacted>
21 years agoMerge with gitk.
Junio C Hamano [Sat, 6 Aug 2005 16:24:15 +0000 (09:24 -0700)]
Merge with gitk.

This merges commit
    d698206c12a4680a92c5f4894f0345dc7dcfe62a
from gitk into our head commit
    2c6e4771959dbe8116f39587d912f1215c06cd0d

Sincerely,
    jit-merge command.

21 years agoAdd forward and back buttons and make SHA1 IDs clickable links.
Paul Mackerras [Sat, 6 Aug 2005 12:06:06 +0000 (22:06 +1000)]
Add forward and back buttons and make SHA1 IDs clickable links.

When we display the commit message in the details pane, any string
of 40 [0-9a-f] characters that corresponds to a SHA1 ID that we
know about gets turned into a clickable link, and displayed in
blue and underlined.

We now keep a history of commits that we have looked at, and we
have forward and back buttons for moving within the history list.

21 years ago[PATCH] Assorted documentation patches
Johannes Schindelin [Fri, 5 Aug 2005 15:05:02 +0000 (17:05 +0200)]
[PATCH] Assorted documentation patches

[jc: Johannes spent time and effort to see how consistent our
use of terminilogy is, and as a byproduct made these corrections
not related to the terminology unification.  I really appreciate
it.]

Signed-off-by: Johannes Schindelin <redacted>
Signed-off-by: Junio C Hamano <redacted>
21 years ago[PATCH] git-commit-script fix for degenerated merge
Johannes Schindelin [Fri, 5 Aug 2005 15:34:38 +0000 (17:34 +0200)]
[PATCH] git-commit-script fix for degenerated merge

If merging results in an unchanged tree, git-commit-script should not
complain that there's nothing to commit.

Also, add "[--all]" to usage().

[jc: usually there is no reason to record an unchanging merge,
but this code path is triggered only when there is a nontrivial
merge that needed to be resolved by hand, and we should be able
to record the fact that these two tree heads are dealt with as a
regular two-parent commit in order to help later merges.]

Signed-off-by: Junio C Hamano <redacted>
git clone https://git.99rst.org/PROJECT