Git development
 help / color / mirror / Atom feed
* Re: Simple questions on GIT usage.
From: Randal L. Schwartz @ 2005-11-14 16:37 UTC (permalink / raw)
  To: Andreas Ericsson
In-Reply-To: <4378B687.3060701@op5.se>

>>>>> "Andreas" == Andreas Ericsson <ae@op5.se> writes:

Andreas> man git-branch

Andreas> It will tell you about its -d and -D options.

It will?  My manpage has no -d or -D option for git-branch.

And I've been looking for something like that too.  If I want to
abandon a what-if branch, how do I make sure that the branch and all
objects are deleted?

-- 
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<merlyn@stonehenge.com> <URL:http://www.stonehenge.com/merlyn/>
Perl/Unix/security consulting, Technical writing, Comedy, etc. etc.
See PerlTraining.Stonehenge.com for onsite and open-enrollment Perl training!

^ permalink raw reply

* Libification update
From: Matthias Urlichs @ 2005-11-14 16:31 UTC (permalink / raw)
  To: git

I've merged my current library-ification effort with Junio's current HEAD.

URL: http://netz.smurf.noris.de/git/git.git#libize

There's a Python interface, written using Pyrex. You can now walk object
trees natively, with Python:

>>> import git
>>> e=git.env()
>>> db=e.objdb
>>> c=db.lookup("HEAD").parent[0] # "HEAD^" will work too
>>> t=c.tree
>>> t.entries[0].name
'.gitignore'
>>> t["debian"].entries[0].name
'changelog'
>>> t[".gitignore"].data[0:5]
'/git\n'
>>> t[".gitignore"].data.readline()
'/git\n'    # this will work tomorrow (I hope)

I have not implemented much beyond that; keeping the code in sync is
work, so doing more will have to wait until this gets merged.

I did play with SWIG a bit. It would have simplified interfacing
languages other than Python to git. Unfortunately, there were problems: 

- refering to objects SWIG did not allocate itself is a nontrivial
  exercise
- so is keeping the Python-visible namespace clean
  (unless you're using C++ ... probably not something we'd want to subject
  the git core to ...)
- SWIG's error reporting is *scary*.
  In comparison, Pyrex mostly does the right thing.

If somebody wants to start working on a SWIG/perl/tcl/... interface
for the library, be my guest. ;-)

-- 
Matthias Urlichs   |   {M:U} IT Design @ m-u-it.de   |  smurf@smurf.noris.de
Disclaimer: The quote was selected randomly. Really. | http://smurf.noris.de
 - -
Ask the person next to you.

^ permalink raw reply

* Re: Simple questions on GIT usage.
From: Franck @ 2005-11-14 16:36 UTC (permalink / raw)
  To: Git Mailing List; +Cc: Andreas Ericsson
In-Reply-To: <cda58cb80511140746w2f691fb8y@mail.gmail.com>

Andreas

Could you cc me when you're responding to my questions since I'm not
subscribed to the GIT mailing list for now...thanks

git-branch -d option is not documented on online documentation at:
http://www.kernel.org/pub/software/scm/git/docs/
That's why I didn't find it by my self...

Regarding the second question, your feeling is wrong.  What you
described is well documented in the git tutorial. There's no "hidden"
question behind the one I posted in my first email. So the question is
still: Is there a way to get information about the working repository
? For instance if I create a branch which is not based on the current
HEAD position of master branch but instead is based on a previous
poisition of master branch.  Then later I want to know how I create
this branch, how can I get this information ?

Thanks
--
               Franck

^ permalink raw reply

* Re: ./configure script prototype
From: Fernando J. Pereda @ 2005-11-14 16:32 UTC (permalink / raw)
  To: Git Mailing List
In-Reply-To: <20051114132956.GT30496@pasky.or.cz>

[-- Attachment #1: Type: text/plain, Size: 871 bytes --]

On Mon, Nov 14, 2005 at 02:29:56PM +0100, Petr Baudis wrote:
| Dear diary, on Mon, Nov 14, 2005 at 01:18:52PM CET, I got a letter
| where Andreas Ericsson <ae@op5.se> said that...
| > 		--prefix=*)
| > 			prefix=${1##*=}
| 
| Aren't those heavy bashisms?

As far as I know, it should be supported by a POSIX 'sh'[1]:

---8<---
${parameter##word}

Remove Largest Prefix Pattern. The word shall be expanded to produce a
pattern. The parameter expansion shall then result in parameter, with
the largest portion of the prefix matched by the pattern deleted.
---8<---

Whether other shells support it or not... I don't know.

Cheers,
Ferdy

[1] http://www.opengroup.org/onlinepubs/009695399/utilities/xcu_chap02.html

-- 
Fernando J. Pereda Garcimartin
Gentoo Developer (Alpha,net-mail,mutt,git)
20BB BDC3 761A 4781 E6ED  ED0B 0A48 5B0C 60BD 28D4

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

^ permalink raw reply

* Re: [PATCH] Fix bunch of fd leaks in http-fetch
From: Nick Hengeveld @ 2005-11-13  3:37 UTC (permalink / raw)
  To: Petr Baudis; +Cc: Junio C Hamano, git
In-Reply-To: <20051112195513.GF30496@pasky.or.cz>

On Sat, Nov 12, 2005 at 08:55:13PM +0100, Petr Baudis wrote:

> What about the rest of the leaks?
> 
> Specifically, the one around release_request(), and the one caused by
> re-open()ing local in start_request() when re-calling it on existing
> request.

There should be a close before calling start_request() with an
alternate - start doesn't currently check for an existing fd before it
opens a new one.  There should also be closes for the indexfile and packfile
fds.

So our patches combined should take care of those cases and report on
anything that may have been missed, which doesn't seem like a bad thing.
The extra close in fetch_object shouldn't cause any problems because
it only happens if there is still a valid fd in the request, which
should never happen...

I'm pretty sure that fds are either already closed or were never opened
prior to each call to release_request though.  release is called in the
following circumstances:

1) process_request_queue finds a WAITING request and the sha1 exists
   locally, no need to start() and the fd is never opened
2) fetch_object finds the sha1 exists locally, which means the object
   was already in the repo (same as #1) or prefetch got it loose
   and called finish
3) fetch_object finds an ABORTED request; requests abort when open fails
   or when the request didn't start (in which case the fd was closed
   after setting the request state)
4) fetch_object finds a request that isn't WAITING/ACTIVE/ABORTED
   (ie. COMPLETE);  finish is called when the request state is set
   to COMPLETE except in the aforementioned alternate restart case

-- 
For a successful technology, reality must take precedence over public
relations, for nature cannot be fooled.

^ permalink raw reply

* Re: ./configure script prototype
From: Andreas Ericsson @ 2005-11-14 16:11 UTC (permalink / raw)
  To: Chris Wedgwood; +Cc: Git Mailing List
In-Reply-To: <20051114155119.GA23912@taniwha.stupidest.org>

Chris Wedgwood wrote:
> On Mon, Nov 14, 2005 at 03:32:13PM +0100, Andreas Ericsson wrote:
> 
> 
>>Dunno. I only know bash-scripting, really. I'll replace them with
>>some sed thing instead.
> 
> 
> For some people /bin/sh != /bin/bash so you probably want "!/bin/bash"
> instead.

Ach no. The configure script really needs to be portable. POSIX (or SUS 
or some such) define a minimum syntax that any shell must support when 
invoked as /bin/sh.

-- 
Andreas Ericsson                   andreas.ericsson@op5.se
OP5 AB                             www.op5.se
Tel: +46 8-230225                  Fax: +46 8-230231

^ permalink raw reply

* Re: Simple questions on GIT usage.
From: Andreas Ericsson @ 2005-11-14 16:08 UTC (permalink / raw)
  Cc: Git Mailing List
In-Reply-To: <cda58cb80511140746w2f691fb8y@mail.gmail.com>

Franck wrote:
> Hi,
> 
> I'd like to know how to delete an entire branch and all it's related
> objects. Is the following correct to accomplish that ?
> 

man git-branch

It will tell you about its -d and -D options.

> 
> And a second question is: how to get the tree ID of the working
> repository ?


I get a feeling there's a higher-level question behind this one, like 
"How can I get a diff between an updated (but not committed) index and 
the files in the working directory?"

When you have a problem it's usually better to just tell people what 
sort of end-result you're hoping to achieve and what steps you've taken 
to accomplish it so far. Asking for details about something that might 
be wrong thinking from the start won't help you but will still require 
much of our time to answer. I.e. nobody gains anything but the work 
still has to be done.

Incidentally, the answer to the quoted question above is (usually)
	git diff-index -p HEAD

Otherwise;
man git (and look in the "IDENTIFIER TERMINOLOGY". I wish I had).

and

for i in /usr/bin/git-*tree*; do man ${i##*/}; done

-- 
Andreas Ericsson                   andreas.ericsson@op5.se
OP5 AB                             www.op5.se
Tel: +46 8-230225                  Fax: +46 8-230231

^ permalink raw reply

* Re: ./configure script prototype
From: Chris Wedgwood @ 2005-11-14 15:51 UTC (permalink / raw)
  To: Andreas Ericsson; +Cc: Git Mailing List
In-Reply-To: <43789FED.7060401@op5.se>

On Mon, Nov 14, 2005 at 03:32:13PM +0100, Andreas Ericsson wrote:

> Dunno. I only know bash-scripting, really. I'll replace them with
> some sed thing instead.

For some people /bin/sh != /bin/bash so you probably want "!/bin/bash"
instead.

^ permalink raw reply

* Simple questions on GIT usage.
From: Franck @ 2005-11-14 15:46 UTC (permalink / raw)
  To: Git Mailing List

Hi,

I'd like to know how to delete an entire branch and all it's related
objects. Is the following correct to accomplish that ?

        # rm -f .git/refs/heads/mybranch
        # git-prune

And a second question is: how to get the tree ID of the working
repository ? I know it's very basic but I can't find out the answer in
any documentations.

Thanks
--
               Franck

^ permalink raw reply

* Re: Can't use gitk.
From: Andreas Ericsson @ 2005-11-14 14:34 UTC (permalink / raw)
  To: Git Mailing List
In-Reply-To: <cda58cb80511140631n46fdce52n@mail.gmail.com>

Franck wrote:
> 2005/11/14, Wayne Scott <wsc9tt@gmail.com>:
> 
> well I found a fix, but I don't know why it actually fixes things :)
> 
> I replaced in .bash_profile
>         PATH=~/bin:$PATH
> by
>         PATH=$HOME/bin:$PATH
> 
> Any ideas ?
> 

~ is only interpolated as $HOME when bash (or /bin/sh) is run as an 
interactive shell. Tab-completion won't work from non-interactive shells 
either.

-- 
Andreas Ericsson                   andreas.ericsson@op5.se
OP5 AB                             www.op5.se
Tel: +46 8-230225                  Fax: +46 8-230231

^ permalink raw reply

* Re: ./configure script prototype
From: Andreas Ericsson @ 2005-11-14 14:32 UTC (permalink / raw)
  To: Git Mailing List
In-Reply-To: <20051114132956.GT30496@pasky.or.cz>

Petr Baudis wrote:
> Dear diary, on Mon, Nov 14, 2005 at 01:18:52PM CET, I got a letter
> where Andreas Ericsson <ae@op5.se> said that...
> 
>>		--prefix=*)
>>			prefix=${1##*=}
> 
> 
> Aren't those heavy bashisms?
> 

Dunno. I only know bash-scripting, really. I'll replace them with some 
sed thing instead.

-- 
Andreas Ericsson                   andreas.ericsson@op5.se
OP5 AB                             www.op5.se
Tel: +46 8-230225                  Fax: +46 8-230231

^ permalink raw reply

* Re: Can't use gitk.
From: Franck @ 2005-11-14 14:31 UTC (permalink / raw)
  To: Wayne Scott; +Cc: Andreas Ericsson, Git Mailing List
In-Reply-To: <59a6e5830511140604o12af40fdp1b05cc84962d5903@mail.gmail.com>

2005/11/14, Wayne Scott <wsc9tt@gmail.com>:
> Probably you need to fix your .profile or .bashrc.   Set your PATH
> unconditionally and not just for interactive shells. In general
> anytime you make some setup in your shell conditional on being
> interactive it is just asking for problems.
>
> Test like this:
>    ssh localhost 'echo $PATH'
>

well I found a fix, but I don't know why it actually fixes things :)

I replaced in .bash_profile
        PATH=~/bin:$PATH
by
        PATH=$HOME/bin:$PATH

Any ideas ?

Thanks
--
               Franck

^ permalink raw reply

* Re: Can't use gitk.
From: Franck @ 2005-11-14 14:13 UTC (permalink / raw)
  To: Johannes Schindelin; +Cc: Andreas Ericsson, Git Mailing List
In-Reply-To: <Pine.LNX.4.63.0511141442310.8531@wbgn013.biozentrum.uni-wuerzburg.de>

Hi Johannes

2005/11/14, Johannes Schindelin <Johannes.Schindelin@gmx.de>:
> Hi,
>
> On Mon, 14 Nov 2005, Franck wrote:
>
> > nope every commands I'm running are not run as root. $HOME/bin is in my
> > path.
>
> Is it possible that your wish is stripping away $HOME/bin from your PATH?
> Could you try the following?
>
>         $ wish
>         % git-rev-list
>

That's what I'm thinking too, but what you suggested is failing as
well (I added HEAD parameter to giv-rev-list command thought).

thanks
--
               Franck

^ permalink raw reply

* Re: Can't use gitk.
From: Wayne Scott @ 2005-11-14 14:04 UTC (permalink / raw)
  To: Franck; +Cc: Andreas Ericsson, Git Mailing List
In-Reply-To: <cda58cb80511140600k2c0188c4i@mail.gmail.com>

Probably you need to fix your .profile or .bashrc.   Set your PATH
unconditionally and not just for interactive shells. In general
anytime you make some setup in your shell conditional on being
interactive it is just asking for problems.

Test like this:
   ssh localhost 'echo $PATH'

-Wayne

On 11/14/05, Franck <vagabon.xyz@gmail.com> wrote:
> Andreas
>
> 2005/11/14, Andreas Ericsson <ae@op5.se>:
> > Yes, but wish or some such might strip $HOME/bin away from it.
> >
> > Try doing
> >
> > make clean; make prefix=/usr
> > sudo make prefix=/usr install
> >
> > and see if it works.
>
> it works.
>
> >
> > If it does, some program somewhere is manipulating your $PATH.
>
> It seems that once wish is executed, $PATH does not include anymore my
> $HOME/bin directory. Do you have any idea why ?
>
> Thanks
> --
>                Franck
> -
> To unsubscribe from this list: send the line "unsubscribe git" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>

^ permalink raw reply

* Re: Can't use gitk.
From: Franck @ 2005-11-14 14:00 UTC (permalink / raw)
  To: Andreas Ericsson; +Cc: Git Mailing List
In-Reply-To: <43787E6E.8030702@op5.se>

Andreas

2005/11/14, Andreas Ericsson <ae@op5.se>:
> Yes, but wish or some such might strip $HOME/bin away from it.
>
> Try doing
>
> make clean; make prefix=/usr
> sudo make prefix=/usr install
>
> and see if it works.

it works.

>
> If it does, some program somewhere is manipulating your $PATH.

It seems that once wish is executed, $PATH does not include anymore my
$HOME/bin directory. Do you have any idea why ?

Thanks
--
               Franck

^ permalink raw reply

* Re: Can't use gitk.
From: Johannes Schindelin @ 2005-11-14 13:45 UTC (permalink / raw)
  To: Franck; +Cc: Andreas Ericsson, Git Mailing List
In-Reply-To: <cda58cb80511140355q1add0ba5n@mail.gmail.com>

Hi,

On Mon, 14 Nov 2005, Franck wrote:

> nope every commands I'm running are not run as root. $HOME/bin is in my 
> path.

Is it possible that your wish is stripping away $HOME/bin from your PATH? 
Could you try the following?

	$ wish
	% git-rev-list

(If a lot of SHA1s are whizzing by, you can Ctrl-C it, and I am wrong.)

Hth,
Dscho

^ permalink raw reply

* Re: ./configure script prototype
From: Petr Baudis @ 2005-11-14 13:29 UTC (permalink / raw)
  To: Andreas Ericsson; +Cc: Git Mailing List
In-Reply-To: <437880AC.3040101@op5.se>

Dear diary, on Mon, Nov 14, 2005 at 01:18:52PM CET, I got a letter
where Andreas Ericsson <ae@op5.se> said that...
> 		--prefix=*)
> 			prefix=${1##*=}

Aren't those heavy bashisms?

-- 
				Petr "Pasky" Baudis
Stuff: http://pasky.or.cz/
VI has two modes: the one in which it beeps and the one in which
it doesn't.

^ permalink raw reply

* Re: http protocol, cloning git.git, fails (too many open files)
From: wa1ter @ 2005-11-14 13:09 UTC (permalink / raw)
  To: git
In-Reply-To: <20051114010940.GQ30496@pasky.or.cz>

On Mon, 14 Nov 2005, Petr Baudis wrote:

> Date: Mon, 14 Nov 2005 02:09:40 +0100
> From: Petr Baudis <pasky@suse.cz>
> Cc: git@vger.kernel.org
> Newsgroups: gmane.comp.version-control.git
> Subject: Re: http protocol, cloning git.git, fails (too many open files)
>
> Dear diary, on Mon, Nov 14, 2005 at 01:42:33AM CET, I got a letter
> where wa1ter@myrealbox.com said that...

> > So, what does cogito do differently...

> Cogito uses the same logic and underlying GIT fetching commands for
> cloning as well as for fetching, while git-clone uses some custom logic
> for cloning ("dumb http clone")...

BTW, I just noticed one big difference between git clone and cg-clone:

git clone stores the URL in .git/remotes/origin while cg-clone puts
it in .git/branches/origin.  When I do a cg-update after a git clone
then cg-update can't find the URL.

^ permalink raw reply

* Re: ./configure script prototype
From: Andreas Ericsson @ 2005-11-14 12:18 UTC (permalink / raw)
  To: Git Mailing List
In-Reply-To: <43788078.4040403@op5.se>

[-- Attachment #1: Type: text/plain, Size: 612 bytes --]

Andreas Ericsson wrote:
> I've started writing up a configuration script to move some of the logic 
> out of the Makefile and make it a bit easier to enable/disable certain 
> stuff.
> 
> I'm not exactly fluent in what sh flavours accept what syntax and the 
> uname -m output is a bit of a mystery for some architectures (PowerPC, 
> notably) so the SHA1 method selection stuff might not work.
> 
> Thoughts? Comments? Patches?
> 

Attachments? ;)

-- 
Andreas Ericsson                   andreas.ericsson@op5.se
OP5 AB                             www.op5.se
Tel: +46 8-230225                  Fax: +46 8-230231

[-- Attachment #2: configure --]
[-- Type: text/plain, Size: 5978 bytes --]

#!/bin/sh

# VERSION must be here so we know where to put the git_libexec_dir later on
VERSION=0.99.9.GIT

# some default values
default_prefix=$HOME
enable_http=Yes
enable_expat=Yes
without_openssl=0

tar=$(which gtar 2>/dev/null || which tar 2>/dev/null)
install=$(which ginstall 2>/dev/null || which install 2>/dev/null)

function usage ()
{
cat << EOF
Usage: $0 [VAR=VALUE]...

Installation directories:
 --prefix=PREFIX         files go in directories under PREFIX [$default_prefix]
 --exec-prefix=EPREFIX   executables go in directories under EPREFIX [PREFIX]
 --bindir=DIR            user executables [EPREFIX/bin]
 --datadir=DIR           read-only architecture-independent data [PREFIX/share]
 --mandir=DIR            man-page base directory [\$datadir/man]

git specific stuff and where to install it:
 --git-exec-path=DIR     git executables path [*currently* \$bindir]
 --git-python-dir=DIR    git python scripts [\$datadir/git/python]
 --git-template-dir=DIR  git templates [\$datadir/git/templates]

Other compilation options:
 --disable-http          disable http transport support [enable]
 --without-expat         build without expat (disables http-push) [enable]
 --with-curl=DIR         path to curl installation [/usr]
 --with-expat=DIR        path to libexpat installation [/usr]
 --with-sha1=METHOD      possible options are MOZILLA_SHA1, PPC_SHA1, ARM_SHA1
                         and OPENSSL_SHA1 [autodetect / OPENSSL_SHA1]
 --with-shell=PATH       path to bourne-like shell [/bin/sh]
 --with-perl=PATH        path to perl [/usr/bin/perl]
 --with-python=PATH      path to python [/usr/bin/python]
EOF
	exit 0
}

# for debugging, really
: ${outfile=config.mak}

# Try to keep these in the same order as the help output above.
while [ $# -gt 0 ]; do
	case "$1" in
		--help*)
			usage
		;;
		--prefix=*)
			prefix=${1##*=}
		;;
		--exec-prefix=*)
			exec_prefix=${1##*=}
		;;
		--bindir=*)
			bindir=${1##*=}
		;;
		--datadir=*)
			datadir=${1##*=}
		;;
		--mandir=*)
			mandir=${1##*=}
		;;
		--git-exec-path=*)
			git_exec_path=${1##*=}
		;;
		--git-python-dir=*)
			git_python_dir=${1##*=}
		;;
		--git-template-dir=*)
			git_template_dir=${1##*=}
		;;
		--disable-http*)
			enable_http=No
			enable_expat=No
		;;
		--without-expat*)
			enable_expat=No
		;;
		--with-curl=*)
			curl_dir=${1##*=}
		;;
		--with-expat=*)
			expat_dir=${1##*=}
		;;
		--with-sha1=*)
			sha1=${1##*=}
		;;
		--with-shell=*)
			shell=${1##*=}
		;;
		--with-perl=*)
			perl=${1##*=}
		;;
		--with-python=*)
			python=${1##*=}
		;;
	esac
	shift
done

[ "$prefix" ] || prefix="$HOME"
[ "$libdir" ] || libdir="$prefix/lib"
[ "$bindir" ] || bindir="$prefix/bin"
[ "$datadir" ] || datadir="$prefix/share"
[ "$mandir" ] || mandir="$datadir/man"
[ "$git_exec_path" ] || git_exec_path="$bindir"
[ "$git_python_dir"] || git_python_dir=$datadir/git/python
[ "$git_template_dir" ] || git_template_dir=$datadir/git/templates

[ "$shell" ] || shell=/bin/sh
if [ -z "$perl" ]; then
	perl=$(which perl 2>/dev/null || perl=/usr/bin/perl)
fi
if [ -z "$python" ]; then
	python=$(which python 2>/dev/null || python=/usr/bin/python)
fi

function need () {
	echo "Need $@"
	for n in "$@"; do
		echo "NEEDS_$n = YesPlease" >> $outfile
	done
}

function no () {
	echo "Working around missing $@"
	for n in "$@"; do
		echo "NO_$n = YesPlease" >> $outfile
	done
}

function cflag () {
	echo "ALL_CFLAGS += $@" >> $outfile
}

# for now this is just an alias
alias ldflag=cflag

#function ldflag () {
#	echo "ALL_LDFLAGS += $@" >> $outfile
#}

sys=$(uname -s)
case $sys in
	Darwin)
		need SSL_WITH_CRYPTO LIBICONV
		## fink
		cflag -I/sw/include
		ldflag -L/sw/lib
		## darwinports
		cflag -I/opt/local/include
		ldflag -L/opt/local/lib
	;;
	SunOS)
		need SOCKET NSL LIBICONV
		no STRCASESTR
		cflag -D__EXTENSIONS__
		[ "$shell" = /bin/sh -a -x /bin/bash ] && shell=/bin/bash
	;;
	OpenBSD)
		need LIBICONV
		no STRCASESTR
		cflag -I/usr/local/include
		ldflag -L/usr/local/lib
	;;
esac

os=$(uname -o)
case $os in
	Cygwin)
		need LIBICONV
		no STRCASESTR IPV6
		cflag -DUSE_SYMLINK_HEAD=0
		echo "X = .exe" >> $outfile
	;;
esac

if [ -z "$sha1" ]; then
	mach=$(uname -m)
	case $mach in
		*arm*)
			sha1=ARM_SHA1
		;;
		i*86)
			sha1=OPENSSL_SHA1
		;;
		# Is this uname -m output for PPC??
		PPC|PowerPC)
			sha1=PPC_SHA1
		;;
		*)
			sha1=MOZILLA_SHA1
		;;
	esac
fi

cat << EOF > $outfile
GIT_VERSION := $VERSION

# Installation destinations
prefix = $prefix
bindir = $bindir
template_dir = $git_template_dir
GIT_EXEC_PATH = $git_exec_path
CFG_GIT_EXEC_PATH := $git_exec_path
GIT_PYTHON_DIR = $git_python_dir

# SHA1 method
$sha1 = YesPlease

# paths to system tools
SHELL_PATH = $shell
PERL_PATH = $perl
PYTHON_PATH = $python
TAR = $tar
INSTALL = $install

# Compilation options, library paths etc. etc.
EOF


# http transport support
if [ "$enable_http" = Yes ]; then
	echo "# HTTP transport is enabled" >> $outfile
	echo 'PROGRAMS += git-http-fetch$X' >> $outfile
	if [ "$curl_dir" ]; then
		cflag -I$curl_dir/include
		echo "CURL_LIBCURL = -L$curl_dir/lib"
	fi
	echo "CURL_LIBCURL += -lcurl" >> $outfile

	if [ "$enable_expat" = Yes ]; then
		echo "# expat (push over http) is enabled" >> $outfile
		echo 'PROGRAMS += git-http-push$X' >> $outfile

		if [ "$expat_dir" ]; then
			cflag -I$expat_dir/include
			echo "EXPAT_LIBEXPAT = -L$expat_dir/lib" >> $outfile
		fi
		echo "EXPAT_LIBEXPAT += -lexpat" >> $outfile
	fi
fi

cat << EOF

Configuration summary:
----------------------

  Compilation options:
    SHA1 method       : $sha1
    HTTP transport    : $enable_http
    HTTP push support : $enable_expat

  System tools locations:
    shell   : $shell
    Perl    : $perl
    Python  : $python
    tar     : $tar
    install : $install

  Installation paths:
    bindir utilities : $bindir
    templates        : $git_template_dir
    python scripts   : $git_python_dir
    Everything else  : $git_exec_path

Please see config.mak and top of Makefile for ways to tweak the build further.
EOF

^ permalink raw reply

* ./configure script prototype
From: Andreas Ericsson @ 2005-11-14 12:18 UTC (permalink / raw)
  To: Git Mailing List

I've started writing up a configuration script to move some of the logic 
out of the Makefile and make it a bit easier to enable/disable certain 
stuff.

I'm not exactly fluent in what sh flavours accept what syntax and the 
uname -m output is a bit of a mystery for some architectures (PowerPC, 
notably) so the SHA1 method selection stuff might not work.

Thoughts? Comments? Patches?

-- 
Andreas Ericsson                   andreas.ericsson@op5.se
OP5 AB                             www.op5.se
Tel: +46 8-230225                  Fax: +46 8-230231

^ permalink raw reply

* Re: Can't use gitk.
From: Andreas Ericsson @ 2005-11-14 12:09 UTC (permalink / raw)
  To: Franck; +Cc: Git Mailing List
In-Reply-To: <cda58cb80511140355q1add0ba5n@mail.gmail.com>

Franck wrote:
> 2005/11/14, Andreas Ericsson <ae@op5.se>:
> 
>> The default
>>installation directory for git is $HOME/bin which might not necessarily
>>be in the $PATH once "wish" and friends have had their say.
> 
> 
> nope every commands I'm running are not run as root. $HOME/bin is in my path.
> 

Yes, but wish or some such might strip $HOME/bin away from it.

Try doing

make clean; make prefix=/usr
sudo make prefix=/usr install

and see if it works.

If it does, some program somewhere is manipulating your $PATH.
Otherwise I'm at my wits end. Perhaps someone more clever than myself 
will take a stab at answering tonight when it's daytime in the US.

-- 
Andreas Ericsson                   andreas.ericsson@op5.se
OP5 AB                             www.op5.se
Tel: +46 8-230225                  Fax: +46 8-230231

^ permalink raw reply

* Re: Can't use gitk.
From: Franck @ 2005-11-14 11:55 UTC (permalink / raw)
  To: Andreas Ericsson; +Cc: Git Mailing List
In-Reply-To: <43787831.3030404@op5.se>

2005/11/14, Andreas Ericsson <ae@op5.se>:
> Franck wrote:
> > Hi Andrea
>
> Not sure why, but 85% of all people replying to me by email refer to me
> as "Andrea". Sort of cute I suppose, although I'm somewhat lacking in
> the TnA department.

oops, sorry Andreas for that ! I use to spelling this name without 's'
that's why I did the mistake I think.

> This seems to imply you're running git-rev-list as root. The default
> installation directory for git is $HOME/bin which might not necessarily
> be in the $PATH once "wish" and friends have had their say.

nope every commands I'm running are not run as root. $HOME/bin is in my path.

> > nope I just did: "make; make install". I tried what you suggested but
> > it's still failing the same way...
>
> So your binaries are installed in $HOME/bin then?

yes.

> Are you running all commands as the same user?
>

yes

> You could try installing from RPM. It should work nicely for all users.

well, I would prefer to use gt repository instead since GIT is in
active developpement.

Thanks
--
               Franck

^ permalink raw reply

* Re: Can't use gitk.
From: Andreas Ericsson @ 2005-11-14 11:42 UTC (permalink / raw)
  To: Franck; +Cc: Git Mailing List
In-Reply-To: <cda58cb80511140322v686e4ee0g@mail.gmail.com>

Franck wrote:
> Hi Andrea
> 

Not sure why, but 85% of all people replying to me by email refer to me 
as "Andrea". Sort of cute I suppose, although I'm somewhat lacking in 
the TnA department.

> 2005/11/14, Andreas Ericsson <ae@op5.se>:
> 
>>Franck wrote:
>>
>>>Hi,
>>>
>>>I'm trying to use gitk --all command but it fails like this:
>>>
>>>"""
>>>Error executing git-rev-list: couldn't execute "git-rev-list": no such
>>>file or directory
>>>"""
>>>
>>
>>Is git-rev-list in your $PATH?
> 
> 
> yes:
>             # git-rev-list HEAD

This seems to imply you're running git-rev-list as root. The default 
installation directory for git is $HOME/bin which might not necessarily 
be in the $PATH once "wish" and friends have had their say.

>             3f1d5f26509defe2f691f5e94a2848848d1ba3a0
>             cd9a0ae21e015cc25a0ea98169451a59d49c0909
>             538310ad38629a1c1f983d982e52b568bc1dbe79
> 
> 
>>Did you compile (and/or install) with
>>
>>        make NO_OPENSSL=YesPlease
>>        make NO_OPENSSL=YesPlease install
>>
> 
> 
> nope I just did: "make; make install". I tried what you suggested but
> it's still failing the same way...
> 

So your binaries are installed in $HOME/bin then?

> 
>>Do other git commands work?
>>
> 
> 
> yes, at least some other commands like git-diff-xxx, git-checkout....
> 

Are you running all commands as the same user?

> 
>>>I'm running git on a fedora core 3. Can anybody give me some help ?
>>>
>>
>>Trying to, but a bit more verbosity in the error report wouldn't hurt.
>>
> 
> 
> I'm usig git v0.99.9h
> 

You could try installing from RPM. It should work nicely for all users.

-- 
Andreas Ericsson                   andreas.ericsson@op5.se
OP5 AB                             www.op5.se
Tel: +46 8-230225                  Fax: +46 8-230231

^ permalink raw reply

* Re: Can't use gitk.
From: Franck @ 2005-11-14 11:22 UTC (permalink / raw)
  To: Andreas Ericsson; +Cc: Git Mailing List
In-Reply-To: <437870AF.8060607@op5.se>

Hi Andrea

2005/11/14, Andreas Ericsson <ae@op5.se>:
> Franck wrote:
> > Hi,
> >
> > I'm trying to use gitk --all command but it fails like this:
> >
> > """
> > Error executing git-rev-list: couldn't execute "git-rev-list": no such
> > file or directory
> > """
> >
>
> Is git-rev-list in your $PATH?

yes:
            # git-rev-list HEAD
            3f1d5f26509defe2f691f5e94a2848848d1ba3a0
            cd9a0ae21e015cc25a0ea98169451a59d49c0909
            538310ad38629a1c1f983d982e52b568bc1dbe79

> Did you compile (and/or install) with
>
>         make NO_OPENSSL=YesPlease
>         make NO_OPENSSL=YesPlease install
>

nope I just did: "make; make install". I tried what you suggested but
it's still failing the same way...

> Do other git commands work?
>

yes, at least some other commands like git-diff-xxx, git-checkout....

> > I'm running git on a fedora core 3. Can anybody give me some help ?
> >
>
> Trying to, but a bit more verbosity in the error report wouldn't hurt.
>

I'm usig git v0.99.9h

Thanks
--
               Franck

^ permalink raw reply

* Re: Can't use gitk.
From: Andreas Ericsson @ 2005-11-14 11:10 UTC (permalink / raw)
  To: Franck; +Cc: Git Mailing List
In-Reply-To: <cda58cb80511140253s92f28e2g@mail.gmail.com>

Franck wrote:
> Hi,
> 
> I'm trying to use gitk --all command but it fails like this:
> 
> """
> Error executing git-rev-list: couldn't execute "git-rev-list": no such
> file or directory
> """
> 

Is git-rev-list in your $PATH?
Did you compile (and/or install) with

	make NO_OPENSSL=YesPlease
	make NO_OPENSSL=YesPlease install

Do other git commands work?

> I'm running git on a fedora core 3. Can anybody give me some help ?
> 

Trying to, but a bit more verbosity in the error report wouldn't hurt.

-- 
Andreas Ericsson                   andreas.ericsson@op5.se
OP5 AB                             www.op5.se
Tel: +46 8-230225                  Fax: +46 8-230231

^ permalink raw reply


This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox