[PATCH 0 of 6] Build documentation using distutils - v3

[Dan Villiom Podlaski Christiansen]

On 26 Nov 2009, at 23:00, Mads Kiilerich wrote:

> Dan Villiom Podlaski Christiansen wrote, On 11/26/2009 08:48 PM:
>> This is an updated version of my distutils-docutils-marriage
>> patch. The primary changes are:
> 
> It doesn't apply cleanly to crew default - what should it be applied to?
> 
> applying setup_install_translation_files_as_package_data
> applying setup_expand_configuration_variables_in_data_paths
> applying setup_add_missing_share_directory_to_install_data_on_unix_platforms
> applying setup_install_some_useful_data_files
> applying setup_generate_build__install_documentation_using_distutils__docutils
> patching file doc/Makefile
> Hunk #1 FAILED at 0
> 1 out of 1 hunks FAILED -- saving rejects to file doc/Makefile.rej

I just updated to crew tip, and it still applies just fine here. What are you using to patch the file? The failing hunk is simply a deletion of the file, so it should be unlikely to fail if you are patching with Mercurial.

Please note that the patches are in Git format.

>> * The package data is now installed as actual package data instead of
>>   regular data.
>> * Both documentation&  manual locations are fully customisable, and
>>   the values may refer to distutils configuration variables.
>> * The RPM build specification has been updated to use distutils
>>   exclusively. In theory, it works; I haven't been able to test it.
> 
> rpmbuild fails with:
> 
> + python setup.py install --root /home/mk/hg/rpmbuild/BUILDROOT/mercurial-1.4+51-ea7e8def4911.i386 --prefix /usr --install-doc /doc/- --install-man /usr/share/man
> running install
> error: error in command line: command 'install' has no such option 'install_man'
> error: Bad exit status from /var/tmp/rpm-tmp.l0yV2r (%install)

The ordering of the files in the patch is ‘Makefile’, ‘contrib/mercurial.spec’, ‘doc/Makefile’ & ‘setup.py’. Perhaps ‘setup.py’ wasn't touched after the failure to patch ‘doc/Makefile’?

> (I wonder where $install_data and $dist_name-$dist_version comes from, and I guess that the explicit installation of shell completion needs change too - but I haven't got so far that that is a problem.)

All three variable strings should be passed unmodified to setup.py, for distutils to expand. I don't know how *.spec files work; should I escape them with a backslash?

--

Dan Villiom Podlaski Christiansen
danchr at gmail.com

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 1943 bytes
Desc: not available
URL: <http://selenic.com/pipermail/mercurial-devel/attachments/20091126/664c64ba/attachment.bin>