Re: Errors/warnings from make

[Graham Percival]

On Wed, Jun 15, 2011 at 05:15:38PM +0100, Phil Holmes wrote:
> langdefs.py: warning: lilypond-doc gettext domain not found.
> langdefs.py: warning: lilypond-doc gettext domain not found.

Something to do with translations, but other than that I dunno.

> out/lilysong:26: DeprecationWarning: The popen2 module is
> deprecated.  Use the subprocess module.
>  import popen2
> out/lilysong:26: DeprecationWarning: The popen2 module is
> deprecated.  Use the subprocess module.
>  import popen2

Related to running code written for python 2.3 on python 2.6.
Could be fixed fairly easily by a python person, but such a fix
would require python x.y instead of 2.3.  I'm happy to increase
the build requirements, but it would also require an updated
package in GUB, and there's a ton of python patches.

ETA: 30 hours for moving GUB python from 2.4.2 to 2.6 or 2.7.
I am absolutely no volunteering.

> /home/phil/lilypond-git/flower/file-name.cc: In function
> 'std::string get_working_directory()':
> /home/phil/lilypond-git/flower/file-name.cc:100: warning: ignoring
> return value of 'char* getcwd(char*, size_t)', declared with
> attribute warn_unused_result

standard C or C++ error.  I fixed one of those in my phd project
just a few hours ago.  :)

> ar: creating out/library.a
> /home/phil/lilypond-git/lily/glissando-engraver.cc: In member
> function 'void
> Glissando_engraver::acknowledge_note_column(Grob_info)':
> /home/phil/lilypond-git/lily/glissando-engraver.cc:124: warning:
> comparison between signed and unsigned integer expressions

those are never good, but they're often not bad either.  Other
than cluttering up the output.

> Are any of these a problem?

None of them are good, but nothing looks urgent.  I'd give them a
Low in the tracker, if you thought it was worth adding them there.

> There are also a load of lines similar to:
> 
> download.itexi:545: warning: @image file
> `lilypond/pictures/Learning_Win7_Open_Menu.txt' (for text)
> unreadable: No such file or directory.
> 
> - just the filename is different.  Anyone know about these?

I see them a lot, but I don't know why.  They're used for
makeinfo, and my initial guess is that either those files get
created after makeinfo is called, or else makeinfo doesn't have
the right -I include dir to point to where those files exist.

when producing an @image in the info version of texinfo (i.e.
makeinfo(1) ) when you're compiling in text-only mode, we use
those .

Cheers,
- Graham