[pacman-dev] compiling pacman and python wrapper
I would like to create a python wrapper to access to pacman features, and so I started to take a look about how it works.. The first thing that I noted is that the INSTALL instructions are a bit confusing. Because it says: --8<---------------cut here---------------start------------->8--- Briefly, the shell commands `./configure; make; make install' should configure, build, and install this package. The following instructions are generic. Run `./configure --help` for specific options. --8<---------------cut here---------------end--------------->8--- But actually there is no "configure" until autoconf is run, but that is explained only later. So maybe just inverting the order might be fine, first talking about autoconf and configure.ac and after talking about configure && make. Then I was actually trying to compile it myself but I get on configure: --8<---------------cut here---------------start------------->8--- configure: creating ./config.status config.status: error: cannot find input file: `lib/libalpm/Makefile.in' --8<---------------cut here---------------end--------------->8--- (from the git repository). And about my original idea, is anyone interested? From what I've understood I think I might use SWIG using the file alpm.h as header and export all the symbols in a nice Python wrapper, does it make sense?
On 22/01/12 16:33, Andrea Crotti wrote:
I would like to create a python wrapper to access to pacman features, and so I started to take a look about how it works..
The first thing that I noted is that the INSTALL instructions are a bit confusing. Because it says:
--8<---------------cut here---------------start------------->8--- Briefly, the shell commands `./configure; make; make install' should configure, build, and install this package. The following instructions are generic. Run `./configure --help` for specific options. --8<---------------cut here---------------end--------------->8---
But actually there is no "configure" until autoconf is run, but that is explained only later. So maybe just inverting the order might be fine, first talking about autoconf and configure.ac and after talking about configure && make.
Then I was actually trying to compile it myself but I get on configure:
--8<---------------cut here---------------start------------->8--- configure: creating ./config.status config.status: error: cannot find input file: `lib/libalpm/Makefile.in' --8<---------------cut here---------------end--------------->8---
(from the git repository).
And about my original idea, is anyone interested? From what I've understood I think I might use SWIG using the file alpm.h as header and export all the symbols in a nice Python wrapper, does it make sense?
If i understand you correctly, you want to use pacman with python so why don't you look at PYALPM. ( Libalpm bindings for Python 3) -- Jelle van der Waa
On Sun, Jan 22, 2012 at 9:33 AM, Andrea Crotti
I would like to create a python wrapper to access to pacman features, and so I started to take a look about how it works..
The first thing that I noted is that the INSTALL instructions are a bit confusing. Because it says:
--8<---------------cut here---------------start------------->8--- Briefly, the shell commands `./configure; make; make install' should configure, build, and install this package. The following instructions are generic. Run `./configure --help` for specific options. --8<---------------cut here---------------end--------------->8---
But actually there is no "configure" until autoconf is run, but that is explained only later. So maybe just inverting the order might be fine, first talking about autoconf and configure.ac and after talking about configure && make. ./autogen.sh if you are building from git and not a tarball.
-Dan
On 01/22/2012 04:56 PM, Dan McGee wrote:
./autogen.sh if you are building from git and not a tarball. -Dan
Ok thanks, the point was just that it was not written anywhere, maybe it would be good to add a section INSTALL FROM GIT. I can figure it out of course but it might be good to have it somewhere.. Or some other projects have a separate file, for example INSTALL.git which might not ship with the tarball for example.
On Sun, Jan 22, 2012 at 07:22:44PM +0000, Andrea Crotti wrote:
On 01/22/2012 04:56 PM, Dan McGee wrote:
./autogen.sh if you are building from git and not a tarball. -Dan
Ok thanks, the point was just that it was not written anywhere, maybe it would be good to add a section INSTALL FROM GIT. I can figure it out of course but it might be good to have it somewhere..
Or some other projects have a separate file, for example INSTALL.git which might not ship with the tarball for example.
On 01/22/2012 07:23 PM, Dave Reisner wrote:
Thanks I didn't see that, but in general if I want to play with some code I don't really install with yaourt, I just clone/checkout the repository, and the information about how to build should be in the repository itself, not somewhere else, my 2 cents.. Not a big deal of course..
On Sun, Jan 22, 2012 at 1:39 PM, Andrea Crotti
On 01/22/2012 07:23 PM, Dave Reisner wrote:
Thanks I didn't see that, but in general if I want to play with some code I don't really install with yaourt, I just clone/checkout the repository, and the information about how to build should be in the repository itself, not somewhere else, my 2 cents..
Not a big deal of course..
Having an autogen or autogen.sh script ends to be a pretty standard convention in the C/autoconf program world. -Dan
participants (4)
-
Andrea Crotti
-
Dan McGee
-
Dave Reisner
-
Jelle van der Waa