Re: lilypond-book is hosed

[Graham Percival]

On Thu, Dec 24, 2009 at 12:05:32AM +0100, John Mandereau wrote:
> Le mercredi 23 décembre 2009 à 22:45 +0000, Graham Percival a écrit :
> > Remember why we did this: the regtest filenames weren't matching
> > up between versions because the [options] changed.
> 
> Sorry for having been idle at that time, but regtest comparison
> shouldn't prevent us from changing lilypond-book fragment options.

Yes; that's why we changed the hash calculation from self.ly() to
self.full_ly()  -- produced lily-abcd1234 filename would not
depend on the fragment options, so those could be changed without
disturbing the version-to-version comparisons.

> >   I have a nicer
> > solution, though: for the regtests (and *only* the regtests), use
> > the filename instead of any hash function.
> 
> You mean using the filename in regtest comparison? If this isn't
> already the case, this would be an excellent idea; I see no good reason
> why it would need to rely on hashes that are part of generated snippet
> filenames. \sourcefilename tag should be reliable enough to identify a
> regression test name from HTML output plus xx/lily-xxxxxxxx.ly files.

I mean not using xx/lily-xxxxxxx.ly files for the regtests.
Instead, use files of the form:
  out/lybook-db/regtest/xxxxxxxxxxxx.ly
where the filename is used directly.  Like:
  out/lybook-db/regtest/ancient-accidental.ly
  out/lybook-db/regtest/slur-broken.ly
  out/lybook-db/regtest/tuplet-numbers-avoid-slurs.ly

(or whatever the actual regtest filenames are)

> > Yes, this would mean that any modified regtest will show up on the
> > regtest comparison -- but I think that's a feature, not a bug.
> > The bug meisters *should* be made aware of any changed regtest[1],
> > even if both versions look perfect.
> 
> This would be extremely cumbersome to check.

?  All that will happen is that if somebody modifies a regtest
(like adding an additional bar to slur-broken.ly), that regtest
will show up in the regtest comparison for the next version.  On
average I think that adds 3-4 images per release?

> Anyway, I'm not looking at this issue in the very near future; I rely on
> somebody more informed than me to handle this issue or give me precise
> instructions to hack a known place in a known particular script.

The regtest naming issue should be fairly simple: add a new option
to lilypond-book that turns off the snippet hash calculation.
Wherever hash(self.ly()) is called, check the option and return
the filename if it's set.  Modify the stepmake/stepmake/ or
input/regression/GNUmakefile to add this option.  Etc.

If you're not intersted, please add this to the issue tracker, and
I'll tackle it once I'm done the website.

Cheers,
- Graham