[Bash-completion-devel] How to build the autocompletion from git repo?

Igor Murzov intergalactic.anonymous at gmail.com
Wed Jun 4 06:55:15 UTC 2014


On Tue, 3 Jun 2014 13:03:50 +0300
Ville Skyttä <ville.skytta at iki.fi> wrote:

> On Tue, Jun 3, 2014 at 1:49 AM, Igor Murzov
> <intergalactic.anonymous at gmail.com> wrote:
> > On Wed, 2 Apr 2014 17:07:34 +0300
> >>
> >> "autoreconf -i" is a good first try with just about any project using autotools.
> >
> > Not everyone knows how to use autotools, so I would agree that
> > having autogen.sh is a good idea, even if it contains only
> > one line :)
> 
> I have no idea why someone who doesn't know autotools or what (s)he's
> doing after cloning a standardly set up repository for an open source
> project would know what to do when finding a script by that name, or
> why he would look for a script in the first place. Some kind of a
> short mention in README's INSTALLATION section would sound better to
> me, but... meh.

It's just a common buildflow that is described in numerous
articles like "how to build software from source in three
easy steps" on the internet. The process is simple:
1. Run `make all` to build the program and then `make install`
   to install it.
2. Have no Makefile? Run `./configure` first.
3. Have no configure? Run `./autogen.sh`.
4. Have no autogen.sh? Read INSTALL file carefully or ask
   for assistance on the ML or IRC :)
People don't want to learn autotools if everything they want
is to compile new version. If you want to prevent people
from going to step 4, make 'em stop on step 3 :)


-- Igor



More information about the Bash-completion-devel mailing list