Re: issues 2266 and 1721

2012-04-18 Thread Phil Holmes 
- Original Message - 
From: Jean-Charles Malahieude lily...@orange.fr

To: Phil Holmes m...@philholmes.net
Cc: Lily Bugs bug-lilypond@gnu.org; lilypond-devel 
lilypond-de...@gnu.org

Sent: Tuesday, April 17, 2012 7:46 PM
Subject: Re: issues 2266 and 1721


Le 15/04/2012 14:56, Phil Holmes disait :

- Original Message - From: Jean-Charles Malahieude
lily...@orange.fr
To: Phil Holmes m...@philholmes.net
Cc: Lily Bugs bug-lilypond@gnu.org; lilypond-devel
lilypond-de...@gnu.org
Sent: Saturday, April 14, 2012 5:46 PM
Subject: Re: issues 2266 and 1721



It might have something to do with the way both glossary and snippet
are built: I just noticed, on fresh make doc on staging (as of
312f7ebc83ec9fb8cbbddfcf78b65a8502c16ab2 ), that
music-glossary.splittexi.log weight is 0 byte, but
snippets.splittexi.log is 190.9 Kio (1618 lines).



I reckon I now know why this is happening. We get the same error if the
text of pitches.itely is cut down to:

@node Hauteurs
@section Hauteurs
@translationof Pitches

Morceaux choisis :
@rlsrnamed{Pitches,Hauteurs}.

The error goes away if @translationof Pitches is changed to
@translationof Pitcher. So I think @rlsrnamed{Pitches,Hauteurs} is
translated automatically to @rlsrnamed{Hauteurs,Hauteurs} and the system
can't then find the node Hauteurs.

What do you think?



This might make one point.
But what is still strange is:
in fr/learning/common-notation.itely I read

@node Altérations et armure
@subsection Altérations et armure
@translationof Accidentals and key signatures


and, in fr/notation/pitches.itely,
@rlearning{Altérations et armure}

which points to
  learning/accidentals-and-key-signatures.fr.html
never to learning/Altérations-et-armure.fr.html which by the way doesn't
exist.

So, why is it tempted to translate the base file name only when dealing
with snippets?

Don't forget I've no skill in programming; I'm just trying to understand
what is going on.  Would it be possible that the origin being the macros
at the head of snippets.tely?  or something weird in conjunction with
extract_texi_filenames.py?

I think that if everything goes well,
Documentation/snippets.splittexi.log should not weight that much with
all those lines like
** `Pitches' is up for `Pitches: Separating key cancellations from key
signature changes', but has no menu entry for this node
** No node following `Pitches: Separating key cancellations from key
signature changes' in menu, but `Pitches: Transposing pitches with
minimum accidentals (Smart transpose)' follows in sectionning

Cheers,
Jean-Charles
===

I have no idea _why_ these things happen: I was simply showing a possible 
reason for those errors occurring.  It seemed to me a possible way of fixing 
the errors.




--
Phil Holmes



___
bug-lilypond mailing list
bug-lilypond@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-lilypond

Re: issues 2266 and 1721

2012-04-18 Thread Jean-Charles Malahieude 

Le 18/04/2012 16:29, Phil Holmes disait :

- Original Message - From: Jean-Charles Malahieude
I think that if everything goes well,
Documentation/snippets.splittexi.log should not weight that much with
all those lines like
** `Pitches' is up for `Pitches: Separating key cancellations from key
signature changes', but has no menu entry for this node
** No node following `Pitches: Separating key cancellations from key
signature changes' in menu, but `Pitches: Transposing pitches with
minimum accidentals (Smart transpose)' follows in sectionning



I have no idea _why_ these things happen: I was simply showing a
possible reason for those errors occurring. It seemed to me a possible
way of fixing the errors.



One more thing:
I wanted to know what happens when trying to build a translated 
Snippets, which generates out-www/.../snippets/pitches.fr.html with 
unnumberedsubsec translated as soon as exists a doctitlefr, and I notice 
that Documentation/fr/snippets.splittexi.log weights exactly the same as 
the English version with its 1618 lines.


The only thing that differs is the file-name part (take into account 
that Altering the length of beamed stems doesn't have a texidocfr so 
remains in English):


** `Pitches: Adding ambitus per voice' doesn't appear in menus
** `Pitches' is up for `Pitches: Adding ambitus per voice', but has no 
menu entry for this node
** No node following `Pitches: Adding ambitus per voice' in menu, but 
`Pitches: Altering the length of beamed stems' follows in sectionning


vs.

** `Pitches: Un ambitus par voix' doesn't appear in menus
** `Pitches' is up for `Pitches: Un ambitus par voix', but has no menu 
entry for this node
** No node following `Pitches: Un ambitus par voix' in menu, but 
`Pitches: Altering the length of beamed stems' follows in sectionning



Oh, wait a moment! I just realize this is absolutely normal, since there 
is no sub-menu in this file. So, just forget those 1618 lines!


All of this might just come from the macroization of the noding, mixed 
between snippets.tely and extract_texi_filename.py


Cheers,
Jean-Charles

___
bug-lilypond mailing list
bug-lilypond@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-lilypond