Current state of automatic footnotes.

Current state of automatic footnotes.

[MikeSol, 2011-06-06 14:32:24]

Hey all,

Here is the current state of automatic footnotes.  Some things to note:

(1)  It nixes the two-pass algorithm to account for annotation width but
provides enough information to create a two (or many) pass algorithm entirely
from Scheme.  Said algorithm can go in the LSR and/or docs.

(2)  The "text" feature of FootnoteItem and FootnoteSpanner no longer kludgily
tacks on a stencil - it remains pure text.

Cheers,
MS

[Bertrand Bordage, 2011-06-18 13:54:16]

Hi Mike!
There is a problem with your patch, I can not compile until the end, even after
"git clean -fxd." This crashes when building internals.texi
And there is still the same spacing issues with "number_raise" and notes higher
than 9...

Thanks,
Bertrand

http://codereview.appspot.com/4580041/diff/2001/lily/page-layout-problem.cc
File lily/page-layout-problem.cc (right):

http://codereview.appspot.com/4580041/diff/2001/lily/page-layout-problem.cc#n...
lily/page-layout-problem.cc:209: annotation->translate_axis
(footnote_stencil->extent (Y_AXIS)[UP] + number_raise -
annotation->extent(Y_AXIS)[UP], Y_AXIS);
Looks like the vertical spacing problem comes from here.
Why not just :
annotation->translate_axis (number_raise, Y_AXIS);
?
Or with a Scheme interface that allows users to chose between fixed and variable
height ?
Same comment for line 239.

[Graham Percival (old account), 2011-06-21 09:17:25]

I can confirm the "cannot compile with this patch applied".

[MikeSol, 2011-06-22 16:27:47]

On 2011/06/21 09:17:25, Graham Percival wrote:
> I can confirm the "cannot compile with this patch applied".

Should apply cleanly - please confirm & leave your comments!

Cheers,
MS

[Neil Puttock, 2011-06-22 22:02:15]

Hi Mike,

I'll take a closer look at this at the weekend.

Cheers,
Neil

http://codereview.appspot.com/4580041/diff/10001/input/regression/footnote-au...
File input/regression/footnote-auto-numbering-page-reset.ly (right):

http://codereview.appspot.com/4580041/diff/10001/input/regression/footnote-au...
input/regression/footnote-auto-numbering-page-reset.ly:3: texidoc = "Lilypond
does footnotes."
needs a more descriptive texidoc (same applies to other tests)

http://codereview.appspot.com/4580041/diff/10001/lily/balloon.cc
File lily/balloon.cc (right):

http://codereview.appspot.com/4580041/diff/10001/lily/balloon.cc#newcode136
lily/balloon.cc:136: "annotation-whiteout "
needs adding to scm/define-grob-properties.scm

http://codereview.appspot.com/4580041/diff/10001/ly/music-functions-init.ly
File ly/music-functions-init.ly (right):

http://codereview.appspot.com/4580041/diff/10001/ly/music-functions-init.ly#n...
ly/music-functions-init.ly:364: \footnoteGrob $grob-name $offset \markup { "" }
$footnote
I haven't tested this, but could you use \markup \null instead?

http://codereview.appspot.com/4580041/diff/10001/ly/music-functions-init.ly#n...
ly/music-functions-init.ly:392: 'text (markup "")
'text (make-null-markup) ?

http://codereview.appspot.com/4580041/diff/10001/scm/lily-library.scm
File scm/lily-library.scm (right):

http://codereview.appspot.com/4580041/diff/10001/scm/lily-library.scm#newcode749
scm/lily-library.scm:749: (define-public (random-string pool n)
remove (since you're now using gensym)

http://codereview.appspot.com/4580041/diff/10001/scm/lily-library.scm#newcode763
scm/lily-library.scm:763: (define-public (random-lowercase-string n)
remove

[pkx166h, 2011-06-25 07:15:39]

Hello,

I get an error/seg fault when I try to make:

(sorry of this a bit verbose)

make runs 'for a bit' then I get

--snip--
ezh --header=texidoc --header=texidoccs --header=texidocde --header=texidoces
--header=texidocfr --header=texidochu --header=texidocit --header=texidocja
--header=texidocnl --header=texidoczh -dcheck-internal-types -ddump-signatures
-danti-alias-factor=2' --output=./out --format=texi --verbose  --skip-lily-check
--lily-output-dir /home/james/lilypond-git/out/lybook-db
--redirect-lilypond-output snippets.tely
cp -p contributor.texi out/contributor.texi
cp -p web.texi out/web.texi
LANG= makeinfo --enable-encoding -I /home/james/lilypond-git/Documentation -I.
-I./out --output=out/lilypond-usage.info out/usage.texi
LANG= makeinfo --enable-encoding -I /home/james/lilypond-git/Documentation -I.
-I./out --output=out/lilypond-contributor.info out/contributor.texi
LANG= makeinfo --enable-encoding -I /home/james/lilypond-git/Documentation -I.
-I./out --output=out/lilypond-internals.info out/internals.texi
LANG= makeinfo --enable-encoding -I /home/james/lilypond-git/Documentation -I.
-I./out --output=out/lilypond-web.info out/web.texi
Segmentation fault
make[1]: *** [out/lilypond-internals.info] Error 139
make[1]: *** Waiting for unfinished jobs....
lilypond-book.py (GNU LilyPond) 2.15.3
Reading snippets.tely...
Dissecting...
Writing snippets...
--snip--

then it carries on 'as normal' and ends in

--snip--

Compiling /home/james/lilypond-git/Documentation/out/template.texi...
/home/james/lilypond-git/Documentation/out/template.texi is up to date.
rm out/weblinks.itexi
make[1]: Leaving directory `/home/james/lilypond-git/Documentation'
make: *** [all] Error 2

--snip--

Hopefully Neil will verify this (in case I have not done something right).

James

[Graham Percival (old account), 2011-06-25 16:38:08]

On 2011/06/25 07:15:39, J_lowe wrote:
> I get an error/seg fault when I try to make:

I can confirm this.

[mike_apollinemike.com, 2011-06-26 09:27:03]

On Jun 25, 2011, at 6:38 PM, percival.music.ca@gmail.com wrote:

> On 2011/06/25 07:15:39, J_lowe wrote:
>> I get an error/seg fault when I try to make:
> 
> I can confirm this.
> 
> 

Fixed.  I had accidentally kept "annotation-whiteout" in balloon.cc whereas I
had scrubbed it in all the other files.

One problem is that, after internals.texi is built, I get:

programming error: Parsed object should be dead: static scm_unused_struct*
Prob::mark_smob(scm_unused_struct*)
continuing, cross fingers

I'm not sure what I did to make this happen.  It doesn't interfere w/ the
compilation process, so the patch is still reviewable, but it'd need to be found
and scrubbed before being pushed.
I'm guessing that I'm forgetting to unprotect something that I'm creating.  If
anyone runs across something in my code that creates new events/probs in an
unusual way (i.e. a clone) or anything else that'd cause this, please let me
know!

Cheers,
MS

[mike_apollinemike.com, 2011-06-26 10:03:16]

On Jun 26, 2011, at 11:26 AM, mike@apollinemike.com wrote:

> On Jun 25, 2011, at 6:38 PM, percival.music.ca@gmail.com wrote:
> 
>> On 2011/06/25 07:15:39, J_lowe wrote:
>>> I get an error/seg fault when I try to make:
>> 
>> I can confirm this.
>> 
>> 
> 
> Fixed.  I had accidentally kept "annotation-whiteout" in balloon.cc whereas I
had scrubbed it in all the other files.
> 
> One problem is that, after internals.texi is built, I get:
> 
> programming error: Parsed object should be dead: static scm_unused_struct*
Prob::mark_smob(scm_unused_struct*)
> continuing, cross fingers
> 
> I'm not sure what I did to make this happen.  It doesn't interfere w/ the
compilation process, so the patch is still reviewable, but it'd need to be found
and scrubbed before being pushed.
> I'm guessing that I'm forgetting to unprotect something that I'm creating.  If
anyone runs across something in my code that creates new events/probs in an
unusual way (i.e. a clone) or anything else that'd cause this, please let me
know!
> 

Just a follow-up: I ran all of the footnote related regtests and this error did
not come up.  So it may arise from something else.

Cheers,
MS

[MikeSol, 2011-07-07 08:08:00]

On 2011/06/26 10:03:16, mike_apollinemike.com wrote:
> On Jun 26, 2011, at 11:26 AM, mailto:mike@apollinemike.com wrote:
> 
> > On Jun 25, 2011, at 6:38 PM, mailto:percival.music.ca@gmail.com wrote:
> > 
> >> On 2011/06/25 07:15:39, J_lowe wrote:
> >>> I get an error/seg fault when I try to make:
> >> 
> >> I can confirm this.
> >> 
> >> 
> > 
> > Fixed.  I had accidentally kept "annotation-whiteout" in balloon.cc whereas
I
> had scrubbed it in all the other files.
> > 
> > One problem is that, after internals.texi is built, I get:
> > 
> > programming error: Parsed object should be dead: static scm_unused_struct*
> Prob::mark_smob(scm_unused_struct*)
> > continuing, cross fingers
> > 
> > I'm not sure what I did to make this happen.  It doesn't interfere w/ the
> compilation process, so the patch is still reviewable, but it'd need to be
found
> and scrubbed before being pushed.
> > I'm guessing that I'm forgetting to unprotect something that I'm creating. 
If
> anyone runs across something in my code that creates new events/probs in an
> unusual way (i.e. a clone) or anything else that'd cause this, please let me
> know!
> > 
> 
> Just a follow-up: I ran all of the footnote related regtests and this error
did
> not come up.  So it may arise from something else.
> 
> Cheers,
> MS
> 

Passes regtests.

[MikeSol, 2011-07-28 14:57:58]

Many thanks to everyone for their help on this.

Pushed as 233aad0ba9781e43424c4e77a859e42b660210e6.

Cheers,
MS

[Neil Puttock, 2011-07-29 21:37:51]

On 28 July 2011 15:57,  <mtsolo@gmail.com> wrote:
> Many thanks to everyone for their help on this.
>
> Pushed as 233aad0ba9781e43424c4e77a859e42b660210e6.

Hi Mike, can you look at my comments from a month ago please?  I
believe some of them are still relevant.

Thanks,
Neil