Re: nano - packaged and ready for some criticism :)

[Charles Wilson <cwilson at ece dot gatech dot edu>]

Robert Collins wrote:


> 3) I'd suggest ( cd /usr/info ; for i in nano.info ; do install-info
> $i --dir-file=dir ; done ) instead of installing _every_ info file in
> the directory, as that could get quite slow :].


I'll take the blame for this.  Two of my recent packages (the experimental 
versions of autoconf and automake wrapper scripts) had postinstall scripts 
that installed *every* info file.  This was a thinko on my part, but it 
seems to have spread...

> However, as there are changes to this just-around-the corner, when Chuck
> and I convince one another :] 


Ha!

> I think it'll be fine as is. Or you could
> change it, as you'll have to recompress the tarball anyway to get the
> updated install script into the source.


> One last thing, and at this point I don't think it matters, you don't
> include info on how to recreate the binary package and source package in
> your README. I don't think this is an issue, as long as you do document
> that before handing the maintenance to someone else (thinking long term
> here).


I think it DOES matter, but once we (the larger we, not just Robert and I) 
end up with a new -src packaging standard that includes autobuild 
scripts/makefiles, the need for this information in the *README* goes away.

Until then, I think that information SHOULD be there, but it's not a 
showstopper.  Just make a note of it, Gareth, and add the info to your next 
regularly scheduled update of the package.

--Chuck