* [PATCH] user-manual: add advanced topic "bisecting merges"
From: Steffen Prohaska @ 2007-11-04 9:16 UTC (permalink / raw)
To: git; +Cc: Steffen Prohaska
This commits adds a discussion of the challenge of bisecting
merge commits to the user manual. The text is slightly
adapted from a mail by Junio C Hamano <gitster@pobox.com>
to the mailing list
<http://marc.info/?l=git&m=119403257315527&w=2>.
The discussion is added to "Exploring git history" in a
sub-section titled "Advanced topics". The discussion requires
detailed knowledge about git. It is assumed that the reader will
skip advanced topics on first reading. At least the text suggest
to do so.
Signed-off-by: Steffen Prohaska <prohaska@zib.de>
---
Documentation/user-manual.txt | 89 +++++++++++++++++++++++++++++++++++++++++
1 files changed, 89 insertions(+), 0 deletions(-)
Junio's discussion was enlightening for me. I think it's a good idea to add
such discussions to the user manual. It was not obvious to me where to place
the discussion in the current structure of the manual. So I added "Advanced
topics".
Is a sub-section "Advanced topics" a good idea? Any better suggestions?
Steffen
diff --git a/Documentation/user-manual.txt b/Documentation/user-manual.txt
index d99adc6..480e7c1 100644
--- a/Documentation/user-manual.txt
+++ b/Documentation/user-manual.txt
@@ -934,6 +934,95 @@ Figuring out why this works is left as an exercise to the (advanced)
student. The gitlink:git-log[1], gitlink:git-diff-tree[1], and
gitlink:git-hash-object[1] man pages may prove helpful.
+[[history-advanced-topics]]
+Advanced topics
+---------------
+This section covers advanced topics that typically require more
+knowledge about git than the manual presented to this point.
+
+You may want to skip the section at first reading, and come back
+later when you have a better understanding of git.
+
+[[bisect-merges]]
+Why bisecting merge commits can be harder than bisecting linear history
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+The following text is based upon an email by Junio C. Hamano to
+the git mailing list
+(link:http://marc.info/?l=git&m=119403257315527&w=2[link:http://marc.info/?l=git&m=119403257315527&w=2]).
+It was slightly adapted for this manual.
+
+Bisecting merges can be challenging due to the complexity of
+changes introduced at a merge. Bisecting through merges is not a
+technical problem. The real problem is what to do when the
+culprit turns out to be a merge commit. How to spot what really
+is wrong, and figure out how to fix. The problem is not for the
+tool but for the human, and it is real.
+
+Imagine this history.
+
+................................................
+ ---Z---o---X---...---o---A---C---D
+ \ /
+ o---o---Y---...---o---B
+................................................
+
+Suppose that on the upper development line, the meaning of one
+of the functions existed at Z was changed at commit X. The
+commits from Z leading to A change both the function's
+implementation and all calling sites that existed at Z, as well
+as new calling sites they add, to be consistent. There is no
+bug at A.
+
+Suppose in the meantime the lower development line somebody
+added a new calling site for that function at commit Y. The
+commits from Z leading to B all assume the old semantics of that
+function and the callers and the callee are consistent with each
+other. There is no bug at B, either.
+
+You merge to create C. There is no textual conflict with this
+three way merge, and the result merges cleanly. You bisect
+this, because you found D is bad and you know Z was good. Your
+bisect will find that C (merge) is broken. Understandably so,
+as at C, the new calling site of the function added by the lower
+branch is not converted to the new semantics, while all the
+other calling sites that already existed at Z would have been
+converted by the merge. The new calling site has semantic
+adjustment needed, but you do not know that yet. You need to
+find out that is the cause of the breakage by looking at the
+merge commit C and the history leading to it.
+
+How would you do that?
+
+Both "git diff A C" and "git diff B C" would be an enormous patch.
+Each of them essentially shows the whole change on each branch
+since they diverged. The developers may have well behaved to
+create good commits that follow the "commit small, commit often,
+commit well contained units" mantra, and each individual commit
+leading from Z to A and from Z to B may be easy to review and
+understand, but looking at these small and easily reviewable
+steps alone would not let you spot the breakage. You need to
+have a global picture of what the upper branch did (and
+among many, one of them is to change the semantics of that
+particular function) and look first at the huge "diff A C"
+(which shows the change the lower branch introduces), and see if
+that huge change is consistent with what have been done between
+Z and A.
+
+If you linearlize the history by rebasing the lower branch on
+top of upper, instead of merging, the bug becomes much easier to
+find and understand. Your history would instead be:
+
+................................................................
+ ---Z---o---X--...---o---A---o---o---Y*--...---o---B*--D*
+................................................................
+
+and there is a single commit Y* between A and B* that introduced
+the new calling site that still uses the old semantics of the
+function, even though that was already modified at X. "git show
+Y*" will be a much smaller patch than "git diff A C" and it is
+much easier to deal with.
+
+
[[Developing-with-git]]
Developing with git
===================
--
1.5.3.4.464.ge1bc2
^ permalink raw reply related
* Re: [PATCH] status&commit: Teach them to show commits of modified submodules.
From: Yin Ping @ 2007-11-04 9:22 UTC (permalink / raw)
To: Junio C Hamano; +Cc: git
In-Reply-To: <7vfxzoyz7f.fsf@gitster.siamese.dyndns.org>
On 11/3/07, Junio C Hamano <gitster@pobox.com> wrote:
> "Yin Ping" <pkufranky@gmail.com> writes:
>
> > On 11/3/07, Junio C Hamano <gitster@pobox.com> wrote:
> >> How does this work when you are a toplevel developer and do not
> >> have the submodule cloned and checked out?
> >>
> >> Our code should treat having the submodule directory and not
> >> having it when there is a mode 160000 entry in the index equally
> >> likely. Cloning and checking-out is _not_ the norm (nor the
> >> exception).
> >>
> > When submodule is not cheched out, it is never modified.
>
> How so? Can't you update the index alone?
>
Sorry i have been away and come back just now.
I think it's a rare case that the submodule entry in the index is
different from HEAD when the submodule is not checked out.
This may only happen when git-reset --soft is called. However, it may
be a wrong operation to commit the changed submodule index entry
while the submodule is never checked out.
There is another similar case, 'git pull' update the submodule entry
in both HEAD and index, but the submodule checked out is not updated.
In this case, 'git-log since..until' in the submodule directory will
fail since 'until' commit is not checked out yet. I think it may be
also a user mistake.
In both case, i think the user should be notified about the
inconsistence. My patch handle this by two warning messages as follows
(where $name is module name)
+ cd $name >&/dev/null || { echo " Warning:
fail to chdir to $name" && exit; }
+ if echo "$left$right" | grep 'unknown
revision' >&/dev/null
+ then
+ echo " Warning: $name is not a
repository or dosn't contains commit $headone/$indexone"
+ else
--
franky
^ permalink raw reply
* Re: [PATCH] status&commit: Teach them to show commits of modified submodules.
From: Yin Ping @ 2007-11-04 9:25 UTC (permalink / raw)
To: Junio C Hamano; +Cc: git
In-Reply-To: <7vfxzoyz7f.fsf@gitster.siamese.dyndns.org>
sorry, this is the reformated message of last one
On 11/3/07, Junio C Hamano <gitster@pobox.com> wrote:
> "Yin Ping" <pkufranky@gmail.com> writes:
>
> > On 11/3/07, Junio C Hamano <gitster@pobox.com> wrote:
> >> How does this work when you are a toplevel developer and do not
> >> have the submodule cloned and checked out?
> >>
> >> Our code should treat having the submodule directory and not
> >> having it when there is a mode 160000 entry in the index equally
> >> likely. Cloning and checking-out is _not_ the norm (nor the
> >> exception).
> >>
> > When submodule is not cheched out, it is never modified.
>
> How so? Can't you update the index alone?
>
Sorry i have been away and come back just now.
I think it's a rare case that the submodule entry in the index is
different from HEAD when the submodule is not checked out.
This may only happen when git-reset --soft is called. However, it may
be a wrong operation to commit the changed submodule index entry
while the submodule is never checked out.
There is another similar case, 'git pull' update the submodule entry
in both HEAD and index, but the submodule checked out is not updated.
In this case, 'git-log since..until' in the submodule directory will
fail since 'until' commit is not checked out yet. I think it may be
also a user mistake.
In both case, i think the user should be notified about the
inconsistence. My patch handle this by two warning messages as follows
(where $name is module name)
+ cd $name >&/dev/null || { echo " Warning:
fail to chdir to $name" && exit; }
+ if echo "$left$right" | grep 'unknown
revision' >&/dev/null
+ then
+ echo " Warning: $name is not a
repository or doesn't contains commit $headone/$indexone"
+ else
--
franky
^ permalink raw reply
* [PATCH] git-reset: add -q option to operate quietly
From: Gerrit Pape @ 2007-11-04 9:37 UTC (permalink / raw)
To: git, Junio C Hamano
Many git commands have a -q option to suppress output to stdout, let's
have it for git-reset too.
This was asked for by Joey Hess through
http://bugs.debian.org/444933
Signed-off-by: Gerrit Pape <pape@smarden.org>
---
Documentation/git-reset.txt | 7 +++++--
builtin-reset.c | 14 ++++++++++----
2 files changed, 15 insertions(+), 6 deletions(-)
diff --git a/Documentation/git-reset.txt b/Documentation/git-reset.txt
index 87afa6f..050e4ea 100644
--- a/Documentation/git-reset.txt
+++ b/Documentation/git-reset.txt
@@ -8,8 +8,8 @@ git-reset - Reset current HEAD to the specified state
SYNOPSIS
--------
[verse]
-'git-reset' [--mixed | --soft | --hard] [<commit>]
-'git-reset' [--mixed] <commit> [--] <paths>...
+'git-reset' [--mixed | --soft | --hard] [-q] [<commit>]
+'git-reset' [--mixed] [-q] <commit> [--] <paths>...
DESCRIPTION
-----------
@@ -45,6 +45,9 @@ OPTIONS
switched to. Any changes to tracked files in the working tree
since <commit> are lost.
+-q::
+ Be quiet, only report errors.
+
<commit>::
Commit to make the current HEAD.
diff --git a/builtin-reset.c b/builtin-reset.c
index e1dc31e..dd301df 100644
--- a/builtin-reset.c
+++ b/builtin-reset.c
@@ -18,7 +18,7 @@
#include "tree.h"
static const char builtin_reset_usage[] =
-"git-reset [--mixed | --soft | --hard] [<commit-ish>] [ [--] <paths>...]";
+"git-reset [--mixed | --soft | --hard] [-q] [<commit-ish>] [ [--] <paths>...]";
static char *args_to_str(const char **argv)
{
@@ -173,7 +173,7 @@ static const char *reset_type_names[] = { "mixed", "soft", "hard", NULL };
int cmd_reset(int argc, const char **argv, const char *prefix)
{
- int i = 1, reset_type = NONE, update_ref_status = 0;
+ int i = 1, reset_type = NONE, update_ref_status = 0, quiet = 0;
const char *rev = "HEAD";
unsigned char sha1[20], *orig = NULL, sha1_orig[20],
*old_orig = NULL, sha1_old_orig[20];
@@ -185,7 +185,7 @@ int cmd_reset(int argc, const char **argv, const char *prefix)
reflog_action = args_to_str(argv);
setenv("GIT_REFLOG_ACTION", reflog_action, 0);
- if (i < argc) {
+ while (i < argc) {
if (!strcmp(argv[i], "--mixed")) {
reset_type = MIXED;
i++;
@@ -198,6 +198,12 @@ int cmd_reset(int argc, const char **argv, const char *prefix)
reset_type = HARD;
i++;
}
+ else if (!strcmp(argv[i], "-q")) {
+ quiet = 1;
+ i++;
+ }
+ else
+ break;
}
if (i < argc && argv[i][0] != '-')
@@ -258,7 +264,7 @@ int cmd_reset(int argc, const char **argv, const char *prefix)
switch (reset_type) {
case HARD:
- if (!update_ref_status)
+ if (!update_ref_status && !quiet)
print_new_head_line(commit);
break;
case SOFT: /* Nothing else to do. */
--
1.5.3.5
^ permalink raw reply related
* Re: What's cooking in git.git (topics)
From: Jakub Narebski @ 2007-11-04 9:43 UTC (permalink / raw)
To: git
In-Reply-To: <7vr6j6ve90.fsf@gitster.siamese.dyndns.org>
[Cc: Junio C Hamano <gitster@pobox.com>, git @vger.kernel.org]
Junio C Hamano wrote:
> * jn/gitweb (Sat Nov 3 00:41:20 2007 +0100) 9 commits
Now that I have sned those patches ;-) I have a few doubts about them
> + gitweb: Use config file for repository description and URLs
> + gitweb: Read repo config using 'git config -z -l'
I'd like some comments on that series, preferably by someone better
with Perl than me, but I think this is a good change performance wise.
> + gitweb: Add tests for overriding gitweb config with repo config
More tests is always good.
> + gitweb: Use href(-replay=>1, action=>...) to generate alternate
> views
> + gitweb: Use href(-replay=>1, page=>...) to generate pagination
> links
> + gitweb: Easier adding/changing parameters to current URL
Now I'm not so sure about this, because it changes semantics of "next page"
and alternate view links: after this series they count from current
version, not from the displayed version.
But perhaps those doubts are unnecessary...
> + gitweb: Remove CGI::Carp::set_programname() call from t9500 gitweb
> test
This removes unnecessary code, which can cause mysterious errors.
> + gitweb: Add 'status_str' to parse_difftree_raw_line output
> + gitweb: Always set 'from_file' and 'to_file' in
> parse_difftree_raw_line
This simplifies gitweb code, and I think there aren't any issues with those
patches.
> Will push these out to 'master' over the weekend.
Thanks.
--
Jakub Narebski
Warsaw, Poland
ShadeHawk on #git
^ permalink raw reply
* Re: [PATCH] git-format-patch: Number patches when there are more than one
From: Andreas Ericsson @ 2007-11-04 9:49 UTC (permalink / raw)
To: Johannes Schindelin; +Cc: Mike Hommey, git, spearce
In-Reply-To: <Pine.LNX.4.64.0711040023000.4362@racer.site>
Johannes Schindelin wrote:
> Hi,
>
> On Sat, 3 Nov 2007, Andreas Ericsson wrote:
>
>> Every piece of DWIM can be translated as "do what I didn't say". If you
>> had to say it, it wouldn't be DWIM after all.
>
> At some point, though, it becomes a "Do what _I_ (and nobody else does, or
> at least what a substantial part of the rest of the world did not) mean",
> and that point is reached here and now.
>
Every piece of dwimmery is suggested by someone and usually backed by a few,
so it's likely wanted by many. Your ultra-conserative "works for *me*, so
shut up and stop trying to change things" means git really has to be flawless.
For you, that is. For others, there are still things to improve. I'd
appreciate if you could losen the "MY playpen" a little though, as it takes
an awful lot of energy from the constructive discussions.
> Wow. I did not think I had to explain _that_.
>
Dito.
--
Andreas Ericsson andreas.ericsson@op5.se
OP5 AB www.op5.se
Tel: +46 8-230225 Fax: +46 8-230231
^ permalink raw reply
* Re: [PATCH] status&commit: Teach them to show commits of modified submodules.
From: Yin Ping @ 2007-11-04 9:56 UTC (permalink / raw)
To: Junio C Hamano; +Cc: git
In-Reply-To: <46dff0320711040125v111c75davb49e9822537f4b19@mail.gmail.com>
On 11/4/07, Yin Ping <pkufranky@gmail.com> wrote:
> sorry, this is the reformated message of last one
Oh, I'm beaten my the mail client (gmail client will automatically
wrap line to width 80, which is not suitable for patch). Just ignore
my last messge and bear the wrapped patch line. Sorry for the
inconvience again
--
franky
^ permalink raw reply
* Re: [PATCH] git-fetch: more terse fetch output
From: Steffen Prohaska @ 2007-11-04 9:56 UTC (permalink / raw)
To: Nicolas Pitre
Cc: Mike Hommey, Linus Torvalds, Junio C Hamano, git, Shawn O. Pearce,
Jeff King
In-Reply-To: <alpine.LFD.0.9999.0711031645350.21255@xanadu.home>
On Nov 3, 2007, at 9:50 PM, Nicolas Pitre wrote:
> On Sat, 3 Nov 2007, Mike Hommey wrote:
>
>> On Sat, Nov 03, 2007 at 04:30:27PM -0400, Nicolas Pitre wrote:
>>> On Sat, 3 Nov 2007, Linus Torvalds wrote:
>>>
>>>>
>>>>
>>>> On Sat, 3 Nov 2007, Nicolas Pitre wrote:
>>>>>
>>>>> Receiving objects: 100% (5439/5439), 1.60 MiB | 636 KiB/s, done.
>>>>
>>>> I mostly like this, but can we please just use "MB/kB" instead of
>>>> "MiB/KiB"?
>>>>
>>>> I hope it was some kind of joke on crazy EU bureaucrats that
>>>> just wasn't
>>>> caught in time.
>>>
>>> I don't care either ways. In fact my own preference is for MB/
>>> kB, but
>>> if I had used that first I'm sure someone else would have asked
>>> for the
>>> purist notations.
>>
>> As far as you don't claim 1MB is 1024KB, it's okay.
>
> [ heh, I knew someone would say something ]
>
> Yes, to me, 1MB is 1024 KB. Always been, until those idiotic hard
> disk
> manufacturers decided to redefine the common interpretation of what
> everyone else used to consider what a MB is just to boost their
> marketing claims.
I believe it doesn't matter what prefix we use for a
_progress indicator_, as long as we use the same prefix for
the the amount already transferred and the bandwidth. A precise
language doesn't matter here.
Here is a short excerpt from a discussion of the standard.
I haven't downloaded the full document "IEEE Trial-Use Standard
for Prefixes for Binary Multiples" but copied from Wikipedia <http://
en.wikipedia.org/wiki/Binary_prefix>:
“This standard is prepared with two goals in mind: (1) to
preserve the SI prefixes as unambiguous decimal multipliers and
(2) to provide alternative prefixes for those cases where binary
multipliers are needed. The first goal affects the general
public, the wide audience of technical and nontechnical persons
who use computers without much concern for their construction
or inner working. These persons will normally interpret kilo,
mega, etc., in their proper decimal sense. The second goal
speaks to specialists—the prefixes for binary multiples make
it possible for persons who work in the information sciences
to communicate with precision.”
Binary multiplier make it possible to communicate with
precision. If you use the binary prefixes it is absolutely
clear what you are talking about. The meaning of the old
prefixes depend on the context.
But all this does not matter here. We are talking about a
progress indicator. Precision doesn't matter. The user wants
to be sure that something is happening and get an idea of how
much longer it will take to finish the fetch. An approximation
is sufficient.
What matters though is saving two characters, because the line
width is limited. So lets take "MB, kB".
Steffen
^ permalink raw reply
* [PATCH qgit] Add support for --early-output option of git log command
From: Marco Costalba @ 2007-11-04 10:25 UTC (permalink / raw)
To: Git Mailing List; +Cc: Linus Torvalds
With this option 'git log' outputs the first commits
as soon as they are ready in order to be able to show at
least something quickly, even if the full output may take
longer to generate.
At the end of the reordering correct revisions are re-sent.
This avoid the user to wait in front a blank screen for
several seconds in case of big repos and cold chaches.
STGit repositories are still not correctly handled, and
file history (annotation) does not uses, still, this new
--early-output feature.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
---
Patch to apply on top of current qgit-2.0 at
git://git.kernel.org/pub/scm/qgit/qgit4.git
src/common.h | 2 +-
src/git.cpp | 12 ++++++++----
src/git.h | 3 ++-
src/git_startup.cpp | 39 +++++++++++++++----
4 files changed, 46 insertions(+), 10 deletions(-)
diff --git a/src/common.h b/src/common.h
index 198348e..de3cb7d 100644
--- a/src/common.h
+++ b/src/common.h
@@ -289,7 +289,7 @@ namespace QGit {
class Rev {
// prevent implicit C++ compiler defaults
Rev();
- Rev(const Rev&);
+// Rev(const Rev&);
Rev& operator=(const Rev&);
public:
Rev(const QByteArray& b, uint s, int idx, int* next, bool withDiff)
diff --git a/src/git.cpp b/src/git.cpp
index d273c60..ef9b627 100644
--- a/src/git.cpp
+++ b/src/git.cpp
@@ -73,9 +73,11 @@ const QString FileHistory::sha(int row) const {
return (row < 0 || row >= _rowCnt ? "" : revOrder.at(row));
}
-void FileHistory::clear() {
+void FileHistory::clear(bool earlyOutput) {
+
+ if (!earlyOutput)
+ git->cancelDataLoading(this);
- git->cancelDataLoading(this);
qDeleteAll(revs);
revs.clear();
revOrder.clear();
@@ -83,9 +85,11 @@ void FileHistory::clear() {
lns->clear();
fNames.clear();
curFNames.clear();
- qDeleteAll(rowData);
- rowData.clear();
+ if (!earlyOutput) {
+ qDeleteAll(rowData);
+ rowData.clear();
+ }
if (testFlag(REL_DATE_F)) {
_secs = QDateTime::currentDateTime().toTime_t();
_headerInfo[4] = "Last Change";
diff --git a/src/git.h b/src/git.h
index de014e0..92879fb 100644
--- a/src/git.h
+++ b/src/git.h
@@ -27,7 +27,7 @@ Q_OBJECT
public:
FileHistory(QObject* parent, Git* git);
~FileHistory();
- void clear();
+ void clear(bool earlyOutput = false);
const QString sha(int row) const;
int row(SCRef sha) const;
const QStringList fileNames() const { return fNames; }
@@ -251,6 +251,7 @@ private:
bool startParseProc(SCList initCmd, FileHistory* fh, SCRef buf);
bool tryFollowRenames(FileHistory* fh);
bool populateRenamedPatches(SCRef sha, SCList nn, FileHistory* fh,
QStringList* on, bool bt);
+ void doEarlyOutput(Rev* rev, int* start);
int addChunk(FileHistory* fh, const QByteArray& ba, int ofs);
void parseDiffFormat(RevFile& rf, SCRef buf);
void parseDiffFormatLine(RevFile& rf, SCRef line, int parNum);
diff --git a/src/git_startup.cpp b/src/git_startup.cpp
index 3faa059..df272fc 100644
--- a/src/git_startup.cpp
+++ b/src/git_startup.cpp
@@ -492,7 +492,9 @@ bool Git::startRevList(SCList args, FileHistory* fh) {
the file deletion revision.
*/
initCmd << QString("-r -m -p --full-index").split(' ');
- }
+ } else
+ initCmd << QString("--early-output");
+
return startParseProc(initCmd + args, fh, QString());
}
@@ -839,13 +841,38 @@ void Git::loadFileNames() {
indexTree();
}
+void Git::doEarlyOutput(Rev* rev, int* start) {
+
+ delete rev;
+ *start += QString("Final output:\n").length();
+
+ Rev* cl = NULL;
+ const Rev* r = revLookup(ZERO_SHA);
+ if (r)
+ cl = new Rev(*r); // copy working dir revision
+
+ revData->clear(true); // keep row QByteArray data
+
+ if (cl) { // re-add working dir revision
+ revData->revs.insert(ZERO_SHA, cl);
+ revData->revOrder.append(ZERO_SHA);
+ }
+}
+
int Git::addChunk(FileHistory* fh, const QByteArray& ba, int start) {
RevMap& r = fh->revs;
int nextStart;
+ Rev* rev;
- // only here we create a new rev
- Rev* rev = new Rev(ba, start, fh->revOrder.count(), &nextStart,
!isMainHistory(fh));
+ do {
+ // only here we create a new rev
+ rev = new Rev(ba, start, fh->revOrder.count(), &nextStart,
!isMainHistory(fh));
+
+ if (nextStart == -2)
+ doEarlyOutput(rev, &start);
+
+ } while (nextStart == -2);
if (nextStart == -1) { // half chunk detected
delete rev;
@@ -1331,6 +1358,7 @@ int Rev::indexData(bool quick, bool withDiff) const {
/*
This is what 'git log' produces:
+ - a possible one line with "Final output:\n" in case of --early-output option
- one line with "commit" + sha + an arbitrary amount of parent's sha, in case
of a merge in file history the line terminates with "(from <sha of parent>)"
- one line with "log size" + len of this record
@@ -1346,9 +1374,12 @@ int Rev::indexData(bool quick, bool withDiff) const {
- a terminating '\0'
*/
int last = ba.size() - 1;
- if (start > last)
+ if (start > last) // offset 'start' points to the char after "commit "
return -1;
+ if (uint(ba.at(start) == 'u'))
+ return -2; // "Final output:", let caller handle this
+
// take in account --boundary and --left-right options
startOfs = uint(ba.at(start) == '-' || ba.at(start) == '<' ||
ba.at(start) == '>');
boundary = startOfs && ba.at(start) == '-';
--
1.5.3.5.532.g5c38-dirty
^ permalink raw reply related
* [PATCH 02/10] Update git-sh-setup(1) to allow transparent use of git-rev-parse --parseopt
From: Pierre Habouzit @ 2007-11-04 10:30 UTC (permalink / raw)
To: gitster, Junio C Hamano; +Cc: git, Pierre Habouzit
In-Reply-To: <1194172262-1563-2-git-send-email-madcoder@debian.org>
If you set OPTIONS_SPEC, git-sh-setups uses git-rev-parse --parseopt
automatically.
It also diverts usage to re-exec $0 with the -h option as parse-options.c
will catch that.
If you need git-rev-parse --parseopt to keep the `--` the user may have
passed to your command, just set OPTIONS_KEEPDASHDASH to a non empty value.
Signed-off-by: Pierre Habouzit <madcoder@debian.org>
---
git-sh-setup.sh | 48 ++++++++++++++++++++++++++++++------------------
1 files changed, 30 insertions(+), 18 deletions(-)
diff --git a/git-sh-setup.sh b/git-sh-setup.sh
index 86d7d4c..e1cf885 100755
--- a/git-sh-setup.sh
+++ b/git-sh-setup.sh
@@ -16,9 +16,36 @@ die() {
exit 1
}
-usage() {
- die "Usage: $0 $USAGE"
-}
+if test -n "$OPTIONS_SPEC"; then
+ usage() {
+ exec "$0" -h
+ }
+
+ parseopt_extra=
+ [ -n "$OPTIONS_KEEPDASHDASH" ] &&
+ parseopt_extra="$parseopt_extra --keep-dashdash"
+
+ eval `echo "$OPTIONS_SPEC" | git rev-parse --parseopt $parseopt_extra -- "$@" || echo exit $?`
+else
+ usage() {
+ die "Usage: $0 $USAGE"
+ }
+
+ if [ -z "$LONG_USAGE" ]
+ then
+ LONG_USAGE="Usage: $0 $USAGE"
+ else
+ LONG_USAGE="Usage: $0 $USAGE
+
+$LONG_USAGE"
+ fi
+
+ case "$1" in
+ -h|--h|--he|--hel|--help)
+ echo "$LONG_USAGE"
+ exit
+ esac
+fi
set_reflog_action() {
if [ -z "${GIT_REFLOG_ACTION:+set}" ]
@@ -91,21 +118,6 @@ get_author_ident_from_commit () {
LANG=C LC_ALL=C sed -ne "$pick_author_script"
}
-if [ -z "$LONG_USAGE" ]
-then
- LONG_USAGE="Usage: $0 $USAGE"
-else
- LONG_USAGE="Usage: $0 $USAGE
-
-$LONG_USAGE"
-fi
-
-case "$1" in
- -h|--h|--he|--hel|--help)
- echo "$LONG_USAGE"
- exit
-esac
-
# Make sure we are in a valid repository of a vintage we understand.
if [ -z "$SUBDIRECTORY_OK" ]
then
--
1.5.3.5.1509.g66d41
^ permalink raw reply related
* [PATCH 03/10] Migrate git-clean.sh to use git-rev-parse --parseopt.
From: Pierre Habouzit @ 2007-11-04 10:30 UTC (permalink / raw)
To: gitster, Junio C Hamano; +Cc: git, Pierre Habouzit
In-Reply-To: <1194172262-1563-3-git-send-email-madcoder@debian.org>
Also minor consistency tweaks in how errors are caught.
Signed-off-by: Pierre Habouzit <madcoder@debian.org>
---
git-clean.sh | 38 ++++++++++++++++++++------------------
1 files changed, 20 insertions(+), 18 deletions(-)
diff --git a/git-clean.sh b/git-clean.sh
index 4491738..6959433 100755
--- a/git-clean.sh
+++ b/git-clean.sh
@@ -3,16 +3,21 @@
# Copyright (c) 2005-2006 Pavel Roskin
#
-USAGE="[-d] [-f] [-n] [-q] [-x | -X] [--] <paths>..."
-LONG_USAGE='Clean untracked files from the working directory
- -d remove directories as well
- -f override clean.requireForce and clean anyway
- -n don'\''t remove anything, just show what would be done
- -q be quiet, only report errors
- -x remove ignored files as well
- -X remove only ignored files
+OPTIONS_SPEC="\
+git-clean [options] <paths>...
+
+Clean untracked files from the working directory
+
When optional <paths>... arguments are given, the paths
-affected are further limited to those that match them.'
+affected are further limited to those that match them.
+--
+d remove directories as well
+f override clean.requireForce and clean anyway
+n don't remove anything, just show what would be done
+q be quiet, only report errors
+x remove ignored files as well
+X remove only ignored files"
+
SUBDIRECTORY_OK=Yes
. git-sh-setup
require_work_tree
@@ -55,23 +60,20 @@ do
shift
break
;;
- -*)
- usage
- ;;
*)
- break
+ usage # should not happen
+ ;;
esac
shift
done
if [ "$disabled" = true ]; then
- echo "clean.requireForce set and -n or -f not given; refusing to clean"
- exit 1
+ die "clean.requireForce set and -n or -f not given; refusing to clean"
fi
-case "$ignored,$ignoredonly" in
- 1,1) usage;;
-esac
+if [ "$ignored,$ignoredonly" = "1,1" ]; then
+ die "-x and -X cannot be set together"
+fi
if [ -z "$ignored" ]; then
excl="--exclude-per-directory=.gitignore"
--
1.5.3.5.1509.g66d41
^ permalink raw reply related
* ph/parseopt-sh reloaded
From: Pierre Habouzit @ 2007-11-04 10:30 UTC (permalink / raw)
To: gitster; +Cc: git
Okay here is a fixed series wrt the security issue Junio raised.
Instead of the old PARSEOPT_OPTS variable, I now only have
OPTIONS_KEEPDASHDASH to be set to non empty if you want to add
--keep-dashdash. The reason for that is that I dislike that every single
git-rev-parse --parseopt user had to do a `PASREOPT_OPTS=` at the
begining of each script, it's error prone, and ugly.
PARSEOPT_OPTS was an overkill as it wasn't really used for anything else
than OPTIONS_KEEPDASHDASH, and if it has to be used for more, it'll be
easy to extend the specification parser to take options on stdin rather
than through parameters.
I also removed the PARSEOPT_OPTS from git-clone.sh as it was a spurious
use, I don't intend users to override this variable, it's indeed an
internal that changes git-rev-parse --parseopt behaviour in a
incompatible way for the scripts that uses it, it should not be
user-tweakable anyway.
The 10 patch series (and not 11, I forgot about 7 when I incrementally
sent the previous one) is fetcheable from my repository:
git://git.madism.org/git.git on branch ph/parseopt-sh
(ph/parseopt has the remaining patches that are problematic right now
either because of the small change -h vs. -H or the patches that
conflicts with git-fetch series right now, but in the spirit this is
definitely a ph/parseopt series).
Cheers,
^ permalink raw reply
* [PATCH 04/10] Migrate git-clone to use git-rev-parse --parseopt
From: Pierre Habouzit @ 2007-11-04 10:30 UTC (permalink / raw)
To: gitster, Junio C Hamano; +Cc: git, Pierre Habouzit
In-Reply-To: <1194172262-1563-4-git-send-email-madcoder@debian.org>
Signed-off-by: Pierre Habouzit <madcoder@debian.org>
---
git-clone.sh | 102 +++++++++++++++++++++++++++++++++-------------------------
1 files changed, 58 insertions(+), 44 deletions(-)
diff --git a/git-clone.sh b/git-clone.sh
index 0ea3c24..52c5601 100755
--- a/git-clone.sh
+++ b/git-clone.sh
@@ -8,15 +8,36 @@
# See git-sh-setup why.
unset CDPATH
+OPTIONS_SPEC="\
+git-clone [options] <repo> [<dir>]
+--
+n,no-checkout don't create a checkout
+bare create a bare repository
+naked create a bare repository
+l,local to clone from a local repository
+no-hardlinks don't use local hardlinks, always copy
+s,shared setup as a shared repository
+template= path to the template directory
+q,quiet be quiet
+reference= reference repository
+o,origin= use <name> instead of 'origin' to track upstream
+u,upload-pack= path to git-upload-pack on the remote
+depth= create a shallow clone of that depth
+
+use-separate-remote compatibility, do not use
+no-separate-remote compatibility, do not use"
+
die() {
echo >&2 "$@"
exit 1
}
usage() {
- die "Usage: $0 [--template=<template_directory>] [--reference <reference-repo>] [--bare] [-l [-s]] [-q] [-u <upload-pack>] [--origin <name>] [--depth <n>] [-n] <repo> [<dir>]"
+ exec "$0" -h
}
+eval `echo "$OPTIONS_SPEC" | git rev-parse --parseopt -- "$@" || echo exit $?`
+
get_repo_base() {
(
cd "`/bin/pwd`" &&
@@ -106,64 +127,57 @@ depth=
no_progress=
local_explicitly_asked_for=
test -t 1 || no_progress=--no-progress
-while
- case "$#,$1" in
- 0,*) break ;;
- *,-n|*,--no|*,--no-|*,--no-c|*,--no-ch|*,--no-che|*,--no-chec|\
- *,--no-check|*,--no-checko|*,--no-checkou|*,--no-checkout)
- no_checkout=yes ;;
- *,--na|*,--nak|*,--nake|*,--naked|\
- *,-b|*,--b|*,--ba|*,--bar|*,--bare) bare=yes ;;
- *,-l|*,--l|*,--lo|*,--loc|*,--loca|*,--local)
- local_explicitly_asked_for=yes
- use_local_hardlink=yes ;;
- *,--no-h|*,--no-ha|*,--no-har|*,--no-hard|*,--no-hardl|\
- *,--no-hardli|*,--no-hardlin|*,--no-hardlink|*,--no-hardlinks)
- use_local_hardlink=no ;;
- *,-s|*,--s|*,--sh|*,--sha|*,--shar|*,--share|*,--shared)
- local_shared=yes; ;;
- 1,--template) usage ;;
- *,--template)
+
+while test $# != 0
+do
+ case "$1" in
+ -n|--no-checkout)
+ no_checkout=yes ;;
+ --naked|--bare)
+ bare=yes ;;
+ -l|--local)
+ local_explicitly_asked_for=yes
+ use_local_hardlink=yes
+ ;;
+ --no-hardlinks)
+ use_local_hardlink=no ;;
+ -s|--shared)
+ local_shared=yes ;;
+ --template)
shift; template="--template=$1" ;;
- *,--template=*)
- template="$1" ;;
- *,-q|*,--quiet) quiet=-q ;;
- *,--use-separate-remote) ;;
- *,--no-separate-remote)
+ -q|--quiet)
+ quiet=-q ;;
+ --use-separate-remote|--no-separate-remote)
die "clones are always made with separate-remote layout" ;;
- 1,--reference) usage ;;
- *,--reference)
+ --reference)
shift; reference="$1" ;;
- *,--reference=*)
- reference=`expr "z$1" : 'z--reference=\(.*\)'` ;;
- *,-o|*,--or|*,--ori|*,--orig|*,--origi|*,--origin)
- case "$2" in
+ -o,--origin)
+ shift;
+ case "$1" in
'')
usage ;;
*/*)
- die "'$2' is not suitable for an origin name"
+ die "'$1' is not suitable for an origin name"
esac
- git check-ref-format "heads/$2" ||
- die "'$2' is not suitable for a branch name"
+ git check-ref-format "heads/$1" ||
+ die "'$1' is not suitable for a branch name"
test -z "$origin_override" ||
die "Do not give more than one --origin options."
origin_override=yes
- origin="$2"; shift
+ origin="$1"
;;
- 1,-u|1,--upload-pack) usage ;;
- *,-u|*,--upload-pack)
+ -u|--upload-pack)
shift
upload_pack="--upload-pack=$1" ;;
- *,--upload-pack=*)
- upload_pack=--upload-pack=$(expr "z$1" : 'z-[^=]*=\(.*\)') ;;
- 1,--depth) usage;;
- *,--depth)
+ --depth)
+ shift
+ depth="--depth=$1" ;;
+ --)
shift
- depth="--depth=$1";;
- *,-*) usage ;;
- *) break ;;
+ break ;;
+ *)
+ usage ;;
esac
-do
shift
done
--
1.5.3.5.1509.g66d41
^ permalink raw reply related
* [PATCH 01/10] Add a parseopt mode to git-rev-parse to bring parse-options to shell scripts.
From: Pierre Habouzit @ 2007-11-04 10:30 UTC (permalink / raw)
To: gitster, Junio C Hamano; +Cc: git, Pierre Habouzit
In-Reply-To: <1194172262-1563-1-git-send-email-madcoder@debian.org>
Signed-off-by: Pierre Habouzit <madcoder@debian.org>
---
Documentation/git-rev-parse.txt | 75 ++++++++++++++++++++++-
builtin-rev-parse.c | 126 +++++++++++++++++++++++++++++++++++++++
2 files changed, 199 insertions(+), 2 deletions(-)
diff --git a/Documentation/git-rev-parse.txt b/Documentation/git-rev-parse.txt
index 4758c33..6811656 100644
--- a/Documentation/git-rev-parse.txt
+++ b/Documentation/git-rev-parse.txt
@@ -23,6 +23,13 @@ distinguish between them.
OPTIONS
-------
+--parseopt::
+ Use `git-rev-parse` in option parsing mode (see PARSEOPT section below).
+
+--keep-dash-dash::
+ Only meaningful in `--parseopt` mode. Tells the option parser to echo
+ out the first `--` met instead of skipping it.
+
--revs-only::
Do not output flags and parameters not meant for
`git-rev-list` command.
@@ -288,10 +295,74 @@ Here are a handful examples:
C^@ I J F
F^! D G H D F
+PARSEOPT
+--------
+
+In `--parseopt` mode, `git-rev-parse` helps massaging options to bring to shell
+scripts the same facilities C builtins have. It works as an option normalizer
+(e.g. splits single switches aggregate values), a bit like `getopt(1)` does.
+
+It takes on the standard input the specification of the options to parse and
+understand, and echoes on the standard ouput a line suitable for `sh(1)` `eval`
+to replace the arguments with normalized ones. In case of error, it ouputs
+usage on the standard error stream, and exits with code 129.
+
+Input Format
+~~~~~~~~~~~~
+
+`git-ref-parse --parseopt` input format is fully text based. It has two parts,
+separated by a line that contains only `--`. The lines before (should be more
+than one) are used for the usage. The lines after describe the options.
+
+Each line of options has this format:
+
+------------
+<opt_spec><arg_spec>? SP+ help LF
+------------
+
+`<opt_spec>`::
+ its format is the short option character, then the long option name
+ separated by a comma. Both parts are not required, though at least one
+ is necessary. `h,help`, `dry-run` and `f` are all three correct
+ `<opt_spec>`.
+
+`<arg_spec>`::
+ an `<arg_spec>` tells the option parser if the option has an argument
+ (`=`), an optionnal one (`?` though its use is discouraged) or none
+ (no `<arg_spec>` in that case).
+
+The rest of the line after as many spaces up to the ending line feed is used
+as the help associated to the option.
+
+Blank lines are ignored, and lines that don't match this specification are used
+as option group headers (start the line with a space to purposely create such
+lines).
+
+Example
+~~~~~~~
+
+------------
+OPTS_SPEC="\
+some-command [options] <args>...
+
+some-command does foo and bar !
+--
+h,help show the help
+
+foo some nifty option --foo
+bar= some cool option --bar with an argument
+
+ An option group Header
+C? option C with an optionnal argument"
+
+eval `echo "$OPTS_SPEC" | git-rev-parse --parseopt -- "$@" || echo exit $?`
+------------
+
+
Author
------
-Written by Linus Torvalds <torvalds@osdl.org> and
-Junio C Hamano <junkio@cox.net>
+Written by Linus Torvalds <torvalds@osdl.org> .
+Junio C Hamano <junkio@cox.net> and Pierre Habouzit <madcoder@debian.org>
Documentation
--------------
diff --git a/builtin-rev-parse.c b/builtin-rev-parse.c
index 8d78b69..054519b 100644
--- a/builtin-rev-parse.c
+++ b/builtin-rev-parse.c
@@ -8,6 +8,7 @@
#include "refs.h"
#include "quote.h"
#include "builtin.h"
+#include "parse-options.h"
#define DO_REVS 1
#define DO_NOREV 2
@@ -209,6 +210,128 @@ static int try_difference(const char *arg)
return 0;
}
+static int parseopt_dump(const struct option *o, const char *arg, int unset)
+{
+ struct strbuf *parsed = o->value;
+ if (unset)
+ strbuf_addf(parsed, " --no-%s", o->long_name);
+ else if (o->short_name)
+ strbuf_addf(parsed, " -%c", o->short_name);
+ else
+ strbuf_addf(parsed, " --%s", o->long_name);
+ if (arg) {
+ strbuf_addch(parsed, ' ');
+ sq_quote_buf(parsed, arg);
+ }
+ return 0;
+}
+
+static const char *skipspaces(const char *s)
+{
+ while (isspace(*s))
+ s++;
+ return s;
+}
+
+static int cmd_parseopt(int argc, const char **argv, const char *prefix)
+{
+ static int keep_dashdash = 0;
+ static char const * const parseopt_usage[] = {
+ "git-rev-parse --parseopt [options] -- [<args>...]",
+ NULL
+ };
+ static struct option parseopt_opts[] = {
+ OPT_BOOLEAN(0, "keep-dashdash", &keep_dashdash,
+ "keep the `--` passed as an arg"),
+ OPT_END(),
+ };
+
+ struct strbuf sb, parsed;
+ const char **usage = NULL;
+ struct option *opts = NULL;
+ int onb = 0, osz = 0, unb = 0, usz = 0;
+
+ strbuf_init(&parsed, 0);
+ strbuf_addstr(&parsed, "set --");
+ argc = parse_options(argc, argv, parseopt_opts, parseopt_usage,
+ PARSE_OPT_KEEP_DASHDASH);
+ if (argc < 1 || strcmp(argv[0], "--"))
+ usage_with_options(parseopt_usage, parseopt_opts);
+
+ strbuf_init(&sb, 0);
+ /* get the usage up to the first line with a -- on it */
+ for (;;) {
+ if (strbuf_getline(&sb, stdin, '\n') == EOF)
+ die("premature end of input");
+ ALLOC_GROW(usage, unb + 1, usz);
+ if (!strcmp("--", sb.buf)) {
+ if (unb < 1)
+ die("no usage string given before the `--' separator");
+ usage[unb] = NULL;
+ break;
+ }
+ usage[unb++] = strbuf_detach(&sb, NULL);
+ }
+
+ /* parse: (<short>|<short>,<long>|<long>)[=?]? SP+ <help> */
+ while (strbuf_getline(&sb, stdin, '\n') != EOF) {
+ const char *s;
+ struct option *o;
+
+ if (!sb.len)
+ continue;
+
+ ALLOC_GROW(opts, onb + 1, osz);
+ memset(opts + onb, 0, sizeof(opts[onb]));
+
+ o = &opts[onb++];
+ s = strchr(sb.buf, ' ');
+ if (!s || *sb.buf == ' ') {
+ o->type = OPTION_GROUP;
+ o->help = xstrdup(skipspaces(s));
+ continue;
+ }
+
+ o->type = OPTION_CALLBACK;
+ o->help = xstrdup(skipspaces(s));
+ o->value = &parsed;
+ o->callback = &parseopt_dump;
+ switch (s[-1]) {
+ case '=':
+ s--;
+ break;
+ case '?':
+ o->flags = PARSE_OPT_OPTARG;
+ s--;
+ break;
+ default:
+ o->flags = PARSE_OPT_NOARG;
+ break;
+ }
+
+ if (s - sb.buf == 1) /* short option only */
+ o->short_name = *sb.buf;
+ else if (sb.buf[1] != ',') /* long option only */
+ o->long_name = xmemdupz(sb.buf, s - sb.buf);
+ else {
+ o->short_name = *sb.buf;
+ o->long_name = xmemdupz(sb.buf + 2, s - sb.buf - 2);
+ }
+ }
+ strbuf_release(&sb);
+
+ /* put an OPT_END() */
+ ALLOC_GROW(opts, onb + 1, osz);
+ memset(opts + onb, 0, sizeof(opts[onb]));
+ argc = parse_options(argc, argv, opts, usage,
+ keep_dashdash ? PARSE_OPT_KEEP_DASHDASH : 0);
+
+ strbuf_addf(&parsed, " --");
+ sq_quote_argv(&parsed, argv, argc, 0);
+ puts(parsed.buf);
+ return 0;
+}
+
int cmd_rev_parse(int argc, const char **argv, const char *prefix)
{
int i, as_is = 0, verify = 0;
@@ -216,6 +339,9 @@ int cmd_rev_parse(int argc, const char **argv, const char *prefix)
git_config(git_default_config);
+ if (argc > 1 && !strcmp("--parseopt", argv[1]))
+ return cmd_parseopt(argc - 1, argv + 1, prefix);
+
for (i = 1; i < argc; i++) {
const char *arg = argv[i];
--
1.5.3.5.1509.g66d41
^ permalink raw reply related
* [PATCH 06/10] Migrate git-merge.sh to use git-rev-parse --parseopt
From: Pierre Habouzit @ 2007-11-04 10:30 UTC (permalink / raw)
To: gitster, Junio C Hamano; +Cc: git, Pierre Habouzit
In-Reply-To: <1194172262-1563-6-git-send-email-madcoder@debian.org>
Signed-off-by: Pierre Habouzit <madcoder@debian.org>
---
git-merge.sh | 125 ++++++++++++++++++++++++---------------------------------
1 files changed, 53 insertions(+), 72 deletions(-)
diff --git a/git-merge.sh b/git-merge.sh
index b9f0519..d19bfc2 100755
--- a/git-merge.sh
+++ b/git-merge.sh
@@ -3,7 +3,18 @@
# Copyright (c) 2005 Junio C Hamano
#
-USAGE='[-n] [--summary] [--[no-]commit] [--[no-]squash] [--[no-]ff] [-s <strategy>] [-m=<merge-message>] <commit>+'
+OPTIONS_SPEC="\
+git-merge [options] <remote>...
+git-merge [options] <msg> HEAD <remote>
+--
+summary show a diffstat at the end of the merge
+n,no-summary don't show a diffstat at the end of the merge
+squash create a single commit instead of doing a merge
+commit perform a commit if the merge sucesses (default)
+ff allow fast forward (default)
+s,strategy= merge strategy to use
+m,message= message to be used for the merge commit (if any)
+"
SUBDIRECTORY_OK=Yes
. git-sh-setup
@@ -132,72 +143,47 @@ merge_name () {
fi
}
-parse_option () {
- case "$1" in
- -n|--n|--no|--no-|--no-s|--no-su|--no-sum|--no-summ|\
- --no-summa|--no-summar|--no-summary)
- show_diffstat=false ;;
- --summary)
- show_diffstat=t ;;
- --sq|--squ|--squa|--squas|--squash)
- allow_fast_forward=t squash=t no_commit=t ;;
- --no-sq|--no-squ|--no-squa|--no-squas|--no-squash)
- allow_fast_forward=t squash= no_commit= ;;
- --c|--co|--com|--comm|--commi|--commit)
- allow_fast_forward=t squash= no_commit= ;;
- --no-c|--no-co|--no-com|--no-comm|--no-commi|--no-commit)
- allow_fast_forward=t squash= no_commit=t ;;
- --ff)
- allow_fast_forward=t squash= no_commit= ;;
- --no-ff)
- allow_fast_forward=false squash= no_commit= ;;
- -s=*|--s=*|--st=*|--str=*|--stra=*|--strat=*|--strate=*|\
- --strateg=*|--strategy=*|\
- -s|--s|--st|--str|--stra|--strat|--strate|--strateg|--strategy)
- case "$#,$1" in
- *,*=*)
- strategy=`expr "z$1" : 'z-[^=]*=\(.*\)'` ;;
- 1,*)
- usage ;;
- *)
- strategy="$2"
- shift ;;
- esac
- case " $all_strategies " in
- *" $strategy "*)
- use_strategies="$use_strategies$strategy " ;;
- *)
- die "available strategies are: $all_strategies" ;;
- esac
- ;;
- -m=*|--m=*|--me=*|--mes=*|--mess=*|--messa=*|--messag=*|--message=*)
- merge_msg=`expr "z$1" : 'z-[^=]*=\(.*\)'`
- have_message=t
- ;;
- -m|--m|--me|--mes|--mess|--messa|--messag|--message)
- shift
- case "$#" in
- 1) usage ;;
- esac
- merge_msg="$1"
- have_message=t
- ;;
- -*) usage ;;
- *) return 1 ;;
- esac
- shift
- args_left=$#
-}
-
parse_config () {
- while test $# -gt 0
- do
- parse_option "$@" || usage
- while test $args_left -lt $#
- do
+ while test $# != 0; do
+ case "$1" in
+ -n|--no-summary)
+ show_diffstat=false ;;
+ --summary)
+ show_diffstat=t ;;
+ --squash)
+ allow_fast_forward=t squash=t no_commit=t ;;
+ --no-squash)
+ allow_fast_forward=t squash= no_commit= ;;
+ --commit)
+ allow_fast_forward=t squash= no_commit= ;;
+ --no-commit)
+ allow_fast_forward=t squash= no_commit=t ;;
+ --ff)
+ allow_fast_forward=t squash= no_commit= ;;
+ --no-ff)
+ allow_fast_forward=false squash= no_commit= ;;
+ -s|--strategy)
+ shift
+ case " $all_strategies " in
+ *" $1 "*)
+ use_strategies="$use_strategies$1 " ;;
+ *)
+ die "available strategies are: $all_strategies" ;;
+ esac
+ ;;
+ -m|--message)
shift
- done
+ merge_msg="$1"
+ have_message=t
+ ;;
+ --)
+ shift
+ break ;;
+ *) usage ;;
+ esac
+ shift
done
+ args_left=$#
}
test $# != 0 || usage
@@ -209,17 +195,12 @@ then
mergeopts=$(git config "branch.${branch#refs/heads/}.mergeoptions")
if test -n "$mergeopts"
then
- parse_config $mergeopts
+ parse_config $mergeopts --
fi
fi
-while parse_option "$@"
-do
- while test $args_left -lt $#
- do
- shift
- done
-done
+parse_config "$@"
+while test $args_left -lt $#; do shift; done
if test -z "$show_diffstat"; then
test "$(git config --bool merge.diffstat)" = false && show_diffstat=false
--
1.5.3.5.1509.g66d41
^ permalink raw reply related
* [PATCH 05/10] Migrate git-am.sh to use git-rev-parse --parseopt
From: Pierre Habouzit @ 2007-11-04 10:30 UTC (permalink / raw)
To: gitster, Junio C Hamano; +Cc: git, Pierre Habouzit
In-Reply-To: <1194172262-1563-5-git-send-email-madcoder@debian.org>
Signed-off-by: Pierre Habouzit <madcoder@debian.org>
---
git-am.sh | 93 +++++++++++++++++++++++++++++++-----------------------------
1 files changed, 48 insertions(+), 45 deletions(-)
diff --git a/git-am.sh b/git-am.sh
index 2514d07..2d2b1c6 100755
--- a/git-am.sh
+++ b/git-am.sh
@@ -2,11 +2,25 @@
#
# Copyright (c) 2005, 2006 Junio C Hamano
-USAGE='[--signoff] [--dotest=<dir>] [--keep] [--utf8 | --no-utf8]
- [--3way] [--interactive] [--binary]
- [--whitespace=<option>] [-C<n>] [-p<n>]
- <mbox>|<Maildir>...
- or, when resuming [--skip | --resolved]'
+OPTIONS_SPEC="\
+git-am [options] <mbox>|<Maildir>...
+git-am [options] --resolved
+git-am [options] --skip
+--
+d,dotest= use <dir> and not .dotest
+i,interactive= run interactively
+b,binary pass --allo-binary-replacement to git-apply
+3,3way allow fall back on 3way merging if needed
+s,signoff add a Signed-off-by line to the commit message
+u,utf8 recode into utf8 (default)
+k,keep pass -k flagg to git-mailinfo
+whitespace= pass it through git-apply
+C= pass it through git-apply
+p= pass it through git-apply
+resolvemsg= override error message when patch failure occurs
+r,resolved to be used after a patch failure
+skip skip the current patch"
+
. git-sh-setup
set_reflog_action am
require_work_tree
@@ -110,49 +124,38 @@ git_apply_opt=
while test $# != 0
do
case "$1" in
- -d=*|--d=*|--do=*|--dot=*|--dote=*|--dotes=*|--dotest=*)
- dotest=`expr "z$1" : 'z-[^=]*=\(.*\)'`; shift ;;
- -d|--d|--do|--dot|--dote|--dotes|--dotest)
- case "$#" in 1) usage ;; esac; shift
- dotest="$1"; shift;;
-
- -i|--i|--in|--int|--inte|--inter|--intera|--interac|--interact|\
- --interacti|--interactiv|--interactive)
- interactive=t; shift ;;
-
- -b|--b|--bi|--bin|--bina|--binar|--binary)
- binary=t; shift ;;
-
- -3|--3|--3w|--3wa|--3way)
- threeway=t; shift ;;
- -s|--s|--si|--sig|--sign|--signo|--signof|--signoff)
- sign=t; shift ;;
- -u|--u|--ut|--utf|--utf8)
- utf8=t; shift ;; # this is now default
- --no-u|--no-ut|--no-utf|--no-utf8)
- utf8=; shift ;;
- -k|--k|--ke|--kee|--keep)
- keep=t; shift ;;
-
- -r|--r|--re|--res|--reso|--resol|--resolv|--resolve|--resolved)
- resolved=t; shift ;;
-
- --sk|--ski|--skip)
- skip=t; shift ;;
-
- --whitespace=*|-C*|-p*)
- git_apply_opt="$git_apply_opt $1"; shift ;;
-
- --resolvemsg=*)
- resolvemsg=${1#--resolvemsg=}; shift ;;
-
+ -i|--interactive)
+ interactive=t ;;
+ -b|--binary)
+ binary=t ;;
+ -3|--3way)
+ threeway=t ;;
+ -s--signoff)
+ sign=t ;;
+ -u|--utf8)
+ utf8=t ;; # this is now default
+ --no-utf8)
+ utf8= ;;
+ -k|--keep)
+ keep=t ;;
+ -r|--resolved)
+ resolved=t ;;
+ --skip)
+ skip=t ;;
+ -d|--dotest)
+ shift; dotest=$1;;
+ --resolvemsg)
+ shift; resolvemsg=$1 ;;
+ --whitespace)
+ git_apply_opt="$git_apply_opt $1=$2"; shift ;;
+ -C|-p)
+ git_apply_opt="$git_apply_opt $1$2"; shift ;;
--)
- shift; break ;;
- -*)
- usage ;;
+ shift; break ;;
*)
- break ;;
+ usage ;;
esac
+ shift
done
# If the dotest directory exists, but we have finished applying all the
--
1.5.3.5.1509.g66d41
^ permalink raw reply related
* [PATCH 07/10] Migrate git-instaweb.sh to use git-rev-parse --parseopt
From: Pierre Habouzit @ 2007-11-04 10:30 UTC (permalink / raw)
To: gitster, Junio C Hamano; +Cc: git, Pierre Habouzit
In-Reply-To: <1194172262-1563-7-git-send-email-madcoder@debian.org>
Signed-off-by: Pierre Habouzit <madcoder@debian.org>
---
git-instaweb.sh | 73 ++++++++++++++++++++++---------------------------------
1 files changed, 29 insertions(+), 44 deletions(-)
diff --git a/git-instaweb.sh b/git-instaweb.sh
index 95c3e5a..d912bf5 100755
--- a/git-instaweb.sh
+++ b/git-instaweb.sh
@@ -2,9 +2,20 @@
#
# Copyright (c) 2006 Eric Wong
#
-USAGE='[--start] [--stop] [--restart]
- [--local] [--httpd=<httpd>] [--port=<port>] [--browser=<browser>]
- [--module-path=<path> (for Apache2 only)]'
+
+OPTIONS_SPEC="\
+git-instaweb [options] (--start | --stop | --restart)
+--
+l,local only bind on 127.0.0.1
+p,port= the port to bind to
+d,httpd= the command to launch
+b,browser= the browser to launch
+m,module-path= the module path (only needed for apache2)
+ Action
+stop stop the web server
+start start the web server
+restart restart the web server
+"
. git-sh-setup
@@ -78,52 +89,26 @@ do
start_httpd
exit 0
;;
- --local|-l)
+ -l|--local)
local=true
;;
- -d|--httpd|--httpd=*)
- case "$#,$1" in
- *,*=*)
- httpd=`expr "$1" : '-[^=]*=\(.*\)'` ;;
- 1,*)
- usage ;;
- *)
- httpd="$2"
- shift ;;
- esac
+ -d|--httpd)
+ shift
+ httpd="$1"
+ ;;
+ -b|--browser)
+ shift
+ browser="$1"
;;
- -b|--browser|--browser=*)
- case "$#,$1" in
- *,*=*)
- browser=`expr "$1" : '-[^=]*=\(.*\)'` ;;
- 1,*)
- usage ;;
- *)
- browser="$2"
- shift ;;
- esac
+ -p|--port)
+ shift
+ port="$1"
;;
- -p|--port|--port=*)
- case "$#,$1" in
- *,*=*)
- port=`expr "$1" : '-[^=]*=\(.*\)'` ;;
- 1,*)
- usage ;;
- *)
- port="$2"
- shift ;;
- esac
+ -m|--module-path)
+ shift
+ module_path="$1"
;;
- -m|--module-path=*|--module-path)
- case "$#,$1" in
- *,*=*)
- module_path=`expr "$1" : '-[^=]*=\(.*\)'` ;;
- 1,*)
- usage ;;
- *)
- module_path="$2"
- shift ;;
- esac
+ --)
;;
*)
usage
--
1.5.3.5.1509.g66d41
^ permalink raw reply related
* [PATCH 10/10] Migrate git-repack.sh to use git-rev-parse --parseopt
From: Pierre Habouzit @ 2007-11-04 10:31 UTC (permalink / raw)
To: gitster, Junio C Hamano; +Cc: git, Pierre Habouzit
In-Reply-To: <1194172262-1563-10-git-send-email-madcoder@debian.org>
---
git-repack.sh | 23 ++++++++++++++++++-----
1 files changed, 18 insertions(+), 5 deletions(-)
diff --git a/git-repack.sh b/git-repack.sh
index 7220635..4d4840e 100755
--- a/git-repack.sh
+++ b/git-repack.sh
@@ -3,7 +3,21 @@
# Copyright (c) 2005 Linus Torvalds
#
-USAGE='[-a|-A] [-d] [-f] [-l] [-n] [-q] [--max-pack-size=N] [--window=N] [--window-memory=N] [--depth=N]'
+OPTIONS_SPEC="\
+git-repack [options]
+--
+a pack everything in a single pack
+A same as -a, and keep unreachable objects too
+d remove redundant packs, and run git-prune-packed
+f pass --no-reuse-delta to git-pack-objects
+q,quiet be quiet
+l pass --local to git-pack-objects
+ Packing constraints
+window= size of the window used for delta compression
+window-memory= same as the above, but limit memory size instead of entries count
+depth= limits the maximum delta depth
+max-pack-size= maximum size of each packfile
+"
SUBDIRECTORY_OK='Yes'
. git-sh-setup
@@ -20,10 +34,9 @@ do
-q) quiet=-q ;;
-f) no_reuse=--no-reuse-object ;;
-l) local=--local ;;
- --max-pack-size=*) extra="$extra $1" ;;
- --window=*) extra="$extra $1" ;;
- --window-memory=*) extra="$extra $1" ;;
- --depth=*) extra="$extra $1" ;;
+ --max-pack-size|--window|--window-memory|--depth)
+ extra="$extra $1=$2"; shift ;;
+ --) shift; break;;
*) usage ;;
esac
shift
--
1.5.3.5.1509.g66d41
^ permalink raw reply related
* [PATCH 09/10] Migrate git-quiltimport.sh to use git-rev-parse --parseopt
From: Pierre Habouzit @ 2007-11-04 10:31 UTC (permalink / raw)
To: gitster, Junio C Hamano; +Cc: git, Pierre Habouzit
In-Reply-To: <1194172262-1563-9-git-send-email-madcoder@debian.org>
Signed-off-by: Pierre Habouzit <madcoder@debian.org>
---
git-quiltimport.sh | 38 +++++++++++++++-----------------------
1 files changed, 15 insertions(+), 23 deletions(-)
diff --git a/git-quiltimport.sh b/git-quiltimport.sh
index 880c81d..b6c24c8 100755
--- a/git-quiltimport.sh
+++ b/git-quiltimport.sh
@@ -1,5 +1,11 @@
#!/bin/sh
-USAGE='--dry-run --author <author> --patches </path/to/quilt/patch/directory>'
+OPTIONS_SPEC="\
+git-quiltimport [options]
+--
+n,dry-run dry run
+author= author name and email address for patches without any
+patches= path to the quilt series and patches
+"
SUBDIRECTORY_ON=Yes
. git-sh-setup
@@ -8,39 +14,25 @@ quilt_author=""
while test $# != 0
do
case "$1" in
- --au=*|--aut=*|--auth=*|--autho=*|--author=*)
- quilt_author=$(expr "z$1" : 'z-[^=]*\(.*\)')
- shift
- ;;
-
- --au|--aut|--auth|--autho|--author)
- case "$#" in 1) usage ;; esac
+ --author)
shift
quilt_author="$1"
- shift
;;
-
- --dry-run)
- shift
+ -n|--dry-run)
dry_run=1
;;
-
- --pa=*|--pat=*|--patc=*|--patch=*|--patche=*|--patches=*)
- QUILT_PATCHES=$(expr "z$1" : 'z-[^=]*\(.*\)')
- shift
- ;;
-
- --pa|--pat|--patc|--patch|--patche|--patches)
- case "$#" in 1) usage ;; esac
- shift
+ --patches)
QUILT_PATCHES="$1"
shift
;;
-
+ --)
+ shift
+ break;;
*)
- break
+ usage
;;
esac
+ shift
done
# Quilt Author
--
1.5.3.5.1509.g66d41
^ permalink raw reply related
* [PATCH 08/10] Migrate git-checkout.sh to use git-rev-parse --parseopt --keep-dashdash
From: Pierre Habouzit @ 2007-11-04 10:31 UTC (permalink / raw)
To: gitster, Junio C Hamano; +Cc: git, Pierre Habouzit
In-Reply-To: <1194172262-1563-8-git-send-email-madcoder@debian.org>
Also fix some space versus tabs issues.
---
git-checkout.sh | 99 +++++++++++++++++++++++++++----------------------------
1 files changed, 49 insertions(+), 50 deletions(-)
diff --git a/git-checkout.sh b/git-checkout.sh
index 8993920..c00cedd 100755
--- a/git-checkout.sh
+++ b/git-checkout.sh
@@ -1,6 +1,16 @@
#!/bin/sh
-USAGE='[-q] [-f] [-b <new_branch>] [-m] [<branch>] [<paths>...]'
+OPTIONS_KEEPDASHDASH=t
+OPTIONS_SPEC="\
+git-branch [options] [<branch>] [<paths>...]
+--
+b= create a new branch started at <branch>
+l create the new branchs reflog
+track tells if the new branch should track the remote branch
+f proceed even if the index or working tree is not HEAD
+m performa three-way merge on local modifications if needed
+q,quiet be quiet
+"
SUBDIRECTORY_OK=Sometimes
. git-sh-setup
require_work_tree
@@ -20,13 +30,12 @@ quiet=
v=-v
LF='
'
-while [ "$#" != "0" ]; do
- arg="$1"
- shift
- case "$arg" in
- "-b")
- newbranch="$1"
+
+while test $# != 0; do
+ case "$1" in
+ -b)
shift
+ newbranch="$1"
[ -z "$newbranch" ] &&
die "git checkout: -b needs a branch name"
git show-ref --verify --quiet -- "refs/heads/$newbranch" &&
@@ -34,64 +43,54 @@ while [ "$#" != "0" ]; do
git check-ref-format "heads/$newbranch" ||
die "git checkout: we do not like '$newbranch' as a branch name."
;;
- "-l")
+ -l)
newbranch_log=-l
;;
- "--track"|"--no-track")
- track="$arg"
+ --track|--no-track)
+ track="$1"
;;
- "-f")
+ -f)
force=1
;;
-m)
merge=1
;;
- "-q")
+ -q|--quiet)
quiet=1
v=
;;
--)
+ shift
break
;;
- -*)
- usage
- ;;
*)
- if rev=$(git rev-parse --verify "$arg^0" 2>/dev/null)
- then
- if [ -z "$rev" ]; then
- echo "unknown flag $arg"
- exit 1
- fi
- new_name="$arg"
- if git show-ref --verify --quiet -- "refs/heads/$arg"
- then
- rev=$(git rev-parse --verify "refs/heads/$arg^0")
- branch="$arg"
- fi
- new="$rev"
- elif rev=$(git rev-parse --verify "$arg^{tree}" 2>/dev/null)
- then
- # checking out selected paths from a tree-ish.
- new="$rev"
- new_name="$arg^{tree}"
- branch=
- else
- new=
- new_name=
- branch=
- set x "$arg" "$@"
- shift
- fi
- case "$1" in
- --)
- shift ;;
- esac
- break
+ usage
;;
- esac
+ esac
+ shift
done
+arg="$1"
+if rev=$(git rev-parse --verify "$arg^0" 2>/dev/null)
+then
+ [ -z "$rev" ] && die "unknown flag $arg"
+ new_name="$arg"
+ if git show-ref --verify --quiet -- "refs/heads/$arg"
+ then
+ rev=$(git rev-parse --verify "refs/heads/$arg^0")
+ branch="$arg"
+ fi
+ new="$rev"
+ shift
+elif rev=$(git rev-parse --verify "$arg^{tree}" 2>/dev/null)
+then
+ # checking out selected paths from a tree-ish.
+ new="$rev"
+ new_name="$arg^{tree}"
+ shift
+fi
+[ "$1" = "--" ] && shift
+
case "$newbranch,$track" in
,--*)
die "git checkout: --track and --no-track require -b"
@@ -138,8 +137,8 @@ Did you intend to checkout '$@' which can not be resolved as commit?"
git ls-files -- "$@" |
git checkout-index -f -u --stdin
- # Run a post-checkout hook -- the HEAD does not change so the
- # current HEAD is passed in for both args
+ # Run a post-checkout hook -- the HEAD does not change so the
+ # current HEAD is passed in for both args
if test -x "$GIT_DIR"/hooks/post-checkout; then
"$GIT_DIR"/hooks/post-checkout $old $old 0
fi
@@ -294,5 +293,5 @@ fi
# Run a post-checkout hook
if test -x "$GIT_DIR"/hooks/post-checkout; then
- "$GIT_DIR"/hooks/post-checkout $old $new 1
+ "$GIT_DIR"/hooks/post-checkout $old $new 1
fi
--
1.5.3.5.1509.g66d41
^ permalink raw reply related
* Re: Why is --pretty=format: so slow?
From: René Scharfe @ 2007-11-04 10:31 UTC (permalink / raw)
To: git; +Cc: Paul Mackerras, git, Johannes Schindelin
In-Reply-To: <7vmytuvdvq.fsf@gitster.siamese.dyndns.org>
Junio C Hamano schrieb:
> Paul Mackerras <paulus@samba.org> writes:
>
>> Strace seems to indicate that git log is doing at least one sequence
>> of open, fstat64, fcntl64, getdents64 and close for each line of
>> output in the --pretty=format: cases, but not in the other cases.
>
> I bet that is coming from doing find_unique_abbrev() to fill
> fields %h, %t and %p, even when the output format does not ask
> for any of them.
>
> Dscho, can we stop calling find_unique_abbrev() unconditionally
> before being asked?
Incidentally, I'm finishing a patch series to add git-describe style
placeholders. It needs such an optimization even more badly (and has
it). Speeding up other slow placeholder expansions falls out quite
naturally. Please wait just a few minutes..
René
^ permalink raw reply
* Re: [RFC PATCH] Make gitk use --early-output
From: Marco Costalba @ 2007-11-04 10:37 UTC (permalink / raw)
To: Paul Mackerras; +Cc: Linus Torvalds, git
In-Reply-To: <18221.2285.259487.655684@cargo.ozlabs.ibm.com>
On 11/4/07, Paul Mackerras <paulus@samba.org> wrote:
>
> set vnextroot($view) 0
> - set order "--topo-order"
> + set order "--early-output=50"
But --early-output does not imply --topo-order, I guess...
I would think you need _both_ in git log:
git log --early-output --topo-order <...remaining stuff...>
Am I missing something?
Marco
P.S: Why did you choose not let git log (i.e. Linus) to handle the
default number of commits?
"--early-output=50" instead of just "--early-output"
I would say, he added this feature mainly for his personal use, so why
don't let him to tweak git-log defaults to his wishes ;-)
^ permalink raw reply
* gitk graph routing problem
From: Alex Riesen @ 2007-11-04 10:46 UTC (permalink / raw)
To: Paul Mackerras; +Cc: git, Junio C Hamano
To reproduce, try running in git repo:
gitk 02f630448e5d48e..06ea6ba9cf46ef5
Than go some pages (around 5) forward. You should notice system load
going up rapidly. Now try paging back - and graph starts stretching
to the right, to the point nothing fits on the screen anymore.
Gitk as of commit 7388bcbc5431552718dde5c3259d861d2fa75a12 in git
repository. Git - v1.5.3.5-551-g02f6304 (terse fetch output and some
*really* unrelated local modifications. Anyway, the gitk is the same).
Bisect points to or around 6e8c87070306a757c4d7fd2c55cca3a90fe140c7
"gitk: Establish and use global left-to-right ordering for commits".
It loops heavily at this commit and looks broken in all subsequent
commits.
Last know good was 3244729aac7515c. The master of gitk repo is ok too.
^ permalink raw reply
* Re: [PATCH] user-manual: add advanced topic "bisecting merges"
From: Ralf Wildenhues @ 2007-11-04 11:23 UTC (permalink / raw)
To: Steffen Prohaska; +Cc: git
In-Reply-To: <11941677732664-git-send-email-prohaska@zib.de>
Hello Steffen,
A couple of language nits:
* Steffen Prohaska wrote on Sun, Nov 04, 2007 at 10:16:13AM CET:
> +Suppose that on the upper development line, the meaning of one
> +of the functions existed at Z was changed at commit X. The
s/functions/& that/
> +commits from Z leading to A change both the function's
> +implementation and all calling sites that existed at Z, as well
> +as new calling sites they add, to be consistent. There is no
> +bug at A.
[...]
> +You merge to create C. There is no textual conflict with this
> +three way merge, and the result merges cleanly. You bisect
> +this, because you found D is bad and you know Z was good. Your
> +bisect will find that C (merge) is broken. Understandably so,
> +as at C, the new calling site of the function added by the lower
> +branch is not converted to the new semantics, while all the
> +other calling sites that already existed at Z would have been
> +converted by the merge. The new calling site has semantic
> +adjustment needed, but you do not know that yet. You need to
> +find out that is the cause of the breakage by looking at the
s/that/that that/
> +merge commit C and the history leading to it.
[...]
> +If you linearlize the history by rebasing the lower branch on
> +top of upper, instead of merging, the bug becomes much easier to
s/upper/the &/
> +find and understand. Your history would instead be:
[...]
Cheers,
Ralf
^ permalink raw reply
* Re: [RFC PATCH] Make gitk use --early-output
From: Paul Mackerras @ 2007-11-04 11:04 UTC (permalink / raw)
To: Marco Costalba; +Cc: Linus Torvalds, git
In-Reply-To: <e5bfff550711040237s250bcec0iddf1ebdc616e0bbf@mail.gmail.com>
Marco Costalba writes:
> On 11/4/07, Paul Mackerras <paulus@samba.org> wrote:
> >
> > set vnextroot($view) 0
> > - set order "--topo-order"
> > + set order "--early-output=50"
>
> But --early-output does not imply --topo-order, I guess...
Look here in Linus' patch:
+ if (!prefixcmp(arg, "--early-output")) {
+ int count = 100;
+ switch (arg[14]) {
+ case '=':
+ count = atoi(arg+15);
+ /* Fallthrough */
+ case 0:
+ revs->topo_order = 1;
+ revs->early_output = count;
+ continue;
+ }
+ }
So yes, --early-output does imply --topo-order.
> P.S: Why did you choose not let git log (i.e. Linus) to handle the
> default number of commits?
>
> "--early-output=50" instead of just "--early-output"
Because I was thinking of adding a control in the edit/preferences
window for it later on.
Paul.
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox