]> git.donarmstrong.com Git - lilypond.git/log
lilypond.git
8 years agoRelease: update news. release/2.19.21-1
Phil Holmes [Sun, 24 May 2015 12:12:20 +0000 (13:12 +0100)]
Release: update news.

8 years agoPO: update template.
Phil Holmes [Sun, 24 May 2015 12:12:20 +0000 (13:12 +0100)]
PO: update template.

8 years agoRelease: bump VERSION_DEVEL.
Phil Holmes [Sun, 24 May 2015 12:12:20 +0000 (13:12 +0100)]
Release: bump VERSION_DEVEL.

8 years agoRemove snippets from /new that are now in the LSR
Phil Holmes [Sat, 23 May 2015 14:34:46 +0000 (15:34 +0100)]
Remove snippets from /new that are now in the LSR

8 years agoAdds snippet showing transcription of mensural music to modern
Phil Holmes [Wed, 20 May 2015 10:03:55 +0000 (11:03 +0100)]
Adds snippet showing transcription of mensural music to modern

8 years agoSome updates to the ancient music section
Phil Holmes [Tue, 19 May 2015 09:43:57 +0000 (10:43 +0100)]
Some updates to the ancient music section

8 years agoIssue 4365: non-member unsmob<T> replaces T::unsmob and T::is_smob
Dan Eble [Sat, 2 May 2015 02:21:43 +0000 (22:21 -0400)]
Issue 4365: non-member unsmob<T> replaces T::unsmob and T::is_smob

Hide Smob_base<>::is_smob() to prevent the misleading situation of
D::is_smob(S) returning true when S represents an ancestor of D rather
than a D.  This issue was previously worked around by overriding
is_smob in several derived smob classes; that is no longer needed.

LY_ASSERT_SMOB() now returns a smob pointer.  This is not used yet,
but is available for future use to avoid unsmobbing twice.

Smob_base<>::smob_p() is hidden because there doesn't seem to be a
need for it to be public.

Files in lily/include were edited by hand.  Files in lily itself were
processed with the following sed script.

s/\([A-Za-z_]\+\)::derived_unsmob/unsmob<\1>/g
s/derived_unsmob *<\([A-Za-z_]\+\)>/unsmob<\1>/g
s/\([A-Za-z_]\+\)::unsmob/unsmob<\1>/g
s/\([A-Za-z_]\+\)::is_smob/unsmob<\1>/g
s/Input \*\([A-Za-z_]\+\) = unsmob /Input *\1 = unsmob<Input> /g
s/LY_ASSERT_SMOB (\([^)]*\))$/LY_ASSERT_SMOB (\1);/g

8 years agoRemove unneeded include of grob.hh in unpure-pure-container.cc
David Kastrup [Fri, 22 May 2015 13:15:12 +0000 (15:15 +0200)]
Remove unneeded include of grob.hh in unpure-pure-container.cc

8 years agoIssue 4400: Rework LY_DECLARE_SMOB_PROC to declare a member function
David Kastrup [Sun, 17 May 2015 10:04:32 +0000 (12:04 +0200)]
Issue 4400: Rework LY_DECLARE_SMOB_PROC to declare a member function

This is nicer to work with.  LY_DECLARE_SMOB_PROC now needs to
additionally get the class name for declaring the smob procedure: this
is astonishingly hard to avoid.  Moving the initialization out of
Smob_base<Super>::init into a separate routine smob_proc_init defined by
LY_DECLARE_SMOB_PROC considerably simplifies the initialization.

8 years agoIssue 4401: Remove unnecessary static casts on function member pointers
David Kastrup [Sun, 17 May 2015 14:14:09 +0000 (16:14 +0200)]
Issue 4401: Remove unnecessary static casts on function member pointers

The commits

6d391dc04f0faa3248d64aa36faf38cf9e4e5fa2 Even more typecasting of member function pointers in smobs.tcc
30203dfe5c45fb9c6255512ba0f8366b322d303f Avoid comparing pointers-to-member of covariant types

were done as hotfixes (apparently without associated tracker issues) to
get the current implementation of smobs to compile with old versions of
GCC in GUB as of Oct 2014.  The attempts to make these versions of GCC
work were ultimately unsuccessful and GUB was upgraded to newer versions
of GCC.

It does not make sense to retain the workarounds that did not really
help.

8 years agoIssue 4398/3: Move the protection of all_fonts_global to Protected_scm
David Kastrup [Thu, 14 May 2015 18:50:57 +0000 (20:50 +0200)]
Issue 4398/3: Move the protection of all_fonts_global to Protected_scm

Previously, this abused the protect/unprotect mechanism only intended
for temporary protection.

Also turns All_font_metrics into a Scheme Smob.

8 years agoIssue 4398/2: Move the protection of global_translator_dict to Protected_scm
David Kastrup [Thu, 14 May 2015 18:50:57 +0000 (20:50 +0200)]
Issue 4398/2: Move the protection of global_translator_dict to Protected_scm

Previously, this abused the protect/unprotect mechanism only intended
for temporary protection.

8 years agoIssue 4398/1: Move the protection of default_global_scale to Protected_scm
David Kastrup [Thu, 14 May 2015 18:50:57 +0000 (20:50 +0200)]
Issue 4398/1: Move the protection of default_global_scale to Protected_scm

Previously, this abused the protect/unprotect mechanism only intended
for temporary protection.

8 years agoIssue 4397: Arrange Protected_scm variables into single list
David Kastrup [Fri, 15 May 2015 20:21:10 +0000 (22:21 +0200)]
Issue 4397: Arrange Protected_scm variables into single list

Since we need to allocate one cons cell per protected variable anyway,
we can just arrange them into a single list, making it possible to

a) just require a single scm_permanent_object call, namely for the head

b) have a representation able to return all protected elements, should
we require this at some later point of time

8 years agoAdjust beam subdivision to only occur at baseMoment
Carl Sorensen [Thu, 7 May 2015 17:39:33 +0000 (11:39 -0600)]
Adjust beam subdivision to only occur at baseMoment

8 years agoFretboards: markup strings in dot positions
Thomas Morley [Mon, 11 May 2015 20:43:51 +0000 (22:43 +0200)]
Fretboards: markup strings in dot positions

Issue 4120

Fixes disappeared strings, restores the behaviour before the fix
for Issue 2752

Does not scale those strings, though, in order to fix the regression first
and do other feature requests for issue 4120 in a follow up.

8 years agoIssue 2787: Sanitize usage of -DDEBUG, -DNDEBUG and assert
David Kastrup [Tue, 12 May 2015 17:01:57 +0000 (19:01 +0200)]
Issue 2787: Sanitize usage of -DDEBUG, -DNDEBUG and assert

The compiler option -DNDEBUG is no longer being used: -DNDEBUG disables
the assert function, and assert is essentially stating that the program
cannot useful continue if the assertion is not met.  -DNDEBUG is
basically an option for compiling an application to a limited amount of
ROM when aborting with a diagnostic is not preferable to crashing.

This is not the case for LilyPond.  So expensive debugging options now
are enabled with -DDEBUG instead.  There is a new configure option
--enable-checking defaulting to "off" for this now.

At the current point of time, setting --disable-optimising also has the
effect of enabling the checks: this will be retained until Patchy has
been adapted to using --enable-checking.

8 years agoIssue 4394: NR: use -@code{s}/-@code{-sharp} for alteration suffix description
David Kastrup [Wed, 13 May 2015 08:55:45 +0000 (10:55 +0200)]
Issue 4394: NR: use -@code{s}/-@code{-sharp} for alteration suffix description

The previously used -s/--sharp in the pitch alteration suffix table did
not render in a sufficiently recognizable manner, making it hard to
guess that the proper note name would be `a-sharp' rather than `asharp'.

8 years agoIssue 4389: Add aliases fflat et al for f-flat et al in bagpipe.ly
David Kastrup [Mon, 11 May 2015 15:47:46 +0000 (17:47 +0200)]
Issue 4389: Add aliases fflat et al for f-flat et al in bagpipe.ly

While the name conversion of issue 4209 is also done for bagpipe music,
as opposed to "normal" music the respective note names have no short
names and thus can be expected to be used in melodies rather than just
in isolated places like \key commands.

As the change is consequently much more likely to inconvenience bagpipe
players, aliases for the old names are added to bagpipe.ly.

8 years agoIssue 4388: Changes: long english notenames switched from aflat to a-flat
David Kastrup [Mon, 11 May 2015 15:41:02 +0000 (17:41 +0200)]
Issue 4388: Changes: long english notenames switched from aflat to a-flat

This change had been issue 4209.

8 years agoPart_combine_iterator: remove unused parameter
Dan Eble [Mon, 18 May 2015 12:53:15 +0000 (08:53 -0400)]
Part_combine_iterator: remove unused parameter

Somehow I overlooked the warning in the patch for Issue 4385.

8 years agoRevert "Issue 3229: Prefer \relative { ... } over \relative x'' { ... }"
David Kastrup [Sun, 17 May 2015 22:02:12 +0000 (00:02 +0200)]
Revert "Issue 3229: Prefer \relative { ... } over \relative x'' { ... }"

This reverts commit 94644b3a701073c99eabb88d95b20cd3f3ac5af0.

8 years agoRun scripts/auxiliar/update-with-convert-ly.sh
David Kastrup [Sun, 17 May 2015 17:28:06 +0000 (19:28 +0200)]
Run scripts/auxiliar/update-with-convert-ly.sh

Afterwards revert all changes in Documentation/snippets/*.ly (thus
retaining Documentation/snippets/new), then run
scripts/auxiliar/makelsr.py, leaving only the files in
Documentation/snippets/new and their versions in Documentation/snippets
converted: if we don't want to retain the automatic conversion rule, we
only want to keep those files which do not get reconverted on reimport.

8 years agoIssue 3229: Prefer \relative { ... } over \relative x'' { ... }
David Kastrup [Wed, 6 Mar 2013 23:23:16 +0000 (00:23 +0100)]
Issue 3229: Prefer \relative { ... } over \relative x'' { ... }

As a followup to the commits for issue 3245, this conversion rule
replaces uses of \relative with reference pitch by those without
reference pitch for a number of patterns where convert-ly can detect the
first actual note reliably.

The rule is prepared to skip over files containing \include or \language
with a non-c-using notename language since the patterns do not work for
them.

However, this part of the rule is disabled in this commit as it turns
out that the LilyPond code base converts well (and more thoroughly)
without it.

8 years agoAdd flip-stencil function
Paul Morris [Tue, 5 May 2015 03:12:52 +0000 (23:12 -0400)]
Add flip-stencil function

Issue 4370
Patch 2/2

8 years agoFix variable name in make-path-stencil
Paul Morris [Tue, 5 May 2015 03:18:15 +0000 (23:18 -0400)]
Fix variable name in make-path-stencil

Issue 4370 Patch 1/2

To follow the convention that xxx?
indicates a predicate.

8 years agoRun scripts/auxiliar/update-with-convert-ly.sh
David Kastrup [Sun, 17 May 2015 17:19:07 +0000 (19:19 +0200)]
Run scripts/auxiliar/update-with-convert-ly.sh

Also some whitespace fixes.

8 years agoIssue 4384/2: Undo effects of \shiftOff convert-ly rule
David Kastrup [Mon, 11 May 2015 14:15:54 +0000 (16:15 +0200)]
Issue 4384/2: Undo effects of \shiftOff convert-ly rule

Includes the effects of several manual operations in consequence

8 years agoIssue 4384: Remove convert-ly rule \shiftOff -> \undo\shiftOn
David Kastrup [Sun, 10 May 2015 12:17:48 +0000 (14:17 +0200)]
Issue 4384: Remove convert-ly rule \shiftOff -> \undo\shiftOn

For the rationale, see
<URL:https://codereview.appspot.com/190500043/#msg1>

The presence of the rule will cause similarly surprising effects to
users as it would to the current state of documentation translations.

Even if it were properly mentioned in the diagnostic description of
the conversion rule, people would rarely look closely enough to
notice.

8 years agoIssue 4386: Part_combine_iterator: move mmrest handling outside the
Dan Eble [Thu, 7 May 2015 12:07:33 +0000 (08:07 -0400)]
Issue 4386: Part_combine_iterator: move mmrest handling outside the
part-routing state machine

8 years agoIssue 4385: Part_combine_iterator: remove residue related to marks
Dan Eble [Mon, 11 May 2015 02:02:34 +0000 (22:02 -0400)]
Issue 4385: Part_combine_iterator: remove residue related to marks

This could have been removed when the generation of mark events was
moved from C++ to Scheme in aae8f9ee8f62e8a57f4f60a60e639a040571f37b,
but it went unnoticed at that time.

8 years agoIssue 4374 / 4: Use mkstemp for intermediate ps files
Masamichi Hosoda [Sat, 9 May 2015 13:49:14 +0000 (22:49 +0900)]
Issue 4374 / 4: Use mkstemp for intermediate ps files

When backend is `ps' (default),
intermediate ps files are created by mkstemp.

The following issues are fixed.

Filename issue (to compile lines.ly, gs_init.ly etc.)
Intermediate file issue (to overwrite and delete .ps file)

8 years agoIssue 4374 / 3: Add copy-binary-file procedure
Masamichi Hosoda [Sat, 9 May 2015 13:29:14 +0000 (22:29 +0900)]
Issue 4374 / 3: Add copy-binary-file procedure

Procedure "copy-binary-file" is
to copy a binary file correctly on any platforms.

8 years agoIssue 4374 / 2: Add make-tmpfile procedure
Masamichi Hosoda [Sat, 16 May 2015 11:46:55 +0000 (20:46 +0900)]
Issue 4374 / 2: Add make-tmpfile procedure

Procedure "make-tmpfile" is
to create a unique temporary file using "mkstemp!".

8 years agoIssue 4374 / 1: Fix platform identification
Masamichi Hosoda [Sat, 9 May 2015 08:20:19 +0000 (17:20 +0900)]
Issue 4374 / 1: Fix platform identification

On Cygwin environment,
Guile's uname returns the strings of "CYGWIN_NT-6.3" etc.
It contains unnecessary strings for the platform identification,
such as Windows version.

Therefore, to remove them to return the only "cygwin".

8 years agoIssue 4379: Part_combine_iterator: simplify context substitution
Dan Eble [Sat, 9 May 2015 01:35:04 +0000 (21:35 -0400)]
Issue 4379: Part_combine_iterator: simplify context substitution

Music_iterator::substitute_outlet: do nothing if the new outlet is
identical to the old (a sanity test)

8 years agoIssue 4372: convert-ly misses some bflat -> b-flat
David Kastrup [Sun, 10 May 2015 14:03:42 +0000 (16:03 +0200)]
Issue 4372: convert-ly misses some bflat -> b-flat

In python/convert-rules.py, using '\b' for note name boundaries turns
out to be too audacious since bflat4 is one word.  A proper solution
would be to match the variable wordsyntax and let the match expression
use a replacement function that triggers on the right words.  However,
such a large number of matches would be expensive.  So instead we are
using positive lookahead and lookbehind expressions to get reasonably
good matches.

8 years agoIssue 4380: Some optimizations during dispatching
David Kastrup [Sat, 9 May 2015 18:37:19 +0000 (20:37 +0200)]
Issue 4380: Some optimizations during dispatching

Don't generate hashes multiple times and stop detection of pending grobs
when the first one is encountered.

8 years agorepeat tremolo: robust to zero-duration elements
Keith OHara [Sun, 3 May 2015 05:49:54 +0000 (22:49 -0700)]
repeat tremolo: robust to zero-duration elements

8 years agoIssue 4357/7: Replace translator_listener_record mess with Scheme
David Kastrup [Thu, 30 Apr 2015 15:47:29 +0000 (17:47 +0200)]
Issue 4357/7: Replace translator_listener_record mess with Scheme

8 years agoIssue 4357/6: Extend Callback_wrapper class to extend for translator listening needs
David Kastrup [Wed, 29 Apr 2015 09:25:28 +0000 (11:25 +0200)]
Issue 4357/6: Extend Callback_wrapper class to extend for translator listening needs

8 years agoIssue 4357/5: Remove Scheme listeners as they are just callbacks now.
David Kastrup [Tue, 28 Apr 2015 16:39:57 +0000 (18:39 +0200)]
Issue 4357/5: Remove Scheme listeners as they are just callbacks now.

8 years agoMerge branch 'issue4357' into HEAD
David Kastrup [Tue, 12 May 2015 12:28:23 +0000 (14:28 +0200)]
Merge branch 'issue4357' into HEAD

This avoids admitting a non-compiling commit into the history mainline

8 years agoIssue 4357/4: Reimplement Listener around generic SCM callback and instance
David Kastrup [Tue, 28 Apr 2015 09:19:23 +0000 (11:19 +0200)]
Issue 4357/4: Reimplement Listener around generic SCM callback and instance

8 years agoIssue 4357/3: Run preparatory script replacing listener macros
David Kastrup [Mon, 27 Apr 2015 19:11:31 +0000 (21:11 +0200)]
Issue 4357/3: Run preparatory script replacing listener macros

This script is

for i in $(git grep -l 'IMPLEMENT_LISTENER (')
do
    sed -i '/IMPLEMENT_LISTENER ([^)]*$/{N;s/\n\s*/ /}' $i
    sedscript=$(git grep -h 'IMPLEMENT_LISTENER (' $i|
       sed 's|^.*IMPLEMENT_LISTENER (\([^,]*\), \([^)]*\)).*$|s/GET_LISTENER (\\([_a-z]\\+->\\)\\?\2)/\\1GET_LISTENER (\1, \2)/|')
    sed -i "$sedscript" $i
done
sed -i '/IMPLEMENT_LISTENER (/d' $(git grep -l 'IMPLEMENT_LISTENER (')
sed -i 's/DECLARE_LISTENER (\([^)]*\));\?/void \1 (SCM);/' $(git grep -l 'DECLARE_LISTENER (')

8 years agoIssue 4357/2: Make dispatchers callable for dispatching
David Kastrup [Wed, 22 Apr 2015 12:44:15 +0000 (14:44 +0200)]
Issue 4357/2: Make dispatchers callable for dispatching

8 years agoIssue 4357/1: Move self_scm_ into non-templated Smob_core base class.
David Kastrup [Fri, 17 Apr 2015 10:55:56 +0000 (12:55 +0200)]
Issue 4357/1: Move self_scm_ into non-templated Smob_core base class.

8 years agoIssue 4376: Add LY_ASSERT_DERIVED_SMOB macro
David Kastrup [Wed, 6 May 2015 22:30:18 +0000 (00:30 +0200)]
Issue 4376: Add LY_ASSERT_DERIVED_SMOB macro

When LY_ASSERT_SMOB is called with a class that is actually derived from
some smobified base class providing is_smob and unsmob, the resulting
check and error message are inaccurate.  This patch uses dynamic_cast to
implement a separate LY_ASSERT_DERIVED_SMOB attaining sensible behavior
in that case.  When no derivation is involved, the performance impact is
minimal.

8 years agoLSR updates: fixes errored change from issue 3981
Phil Holmes [Tue, 12 May 2015 11:44:39 +0000 (12:44 +0100)]
LSR updates: fixes errored change from issue 3981

8 years agoUpdate incipit snippets
Phil Holmes [Tue, 5 May 2015 12:02:50 +0000 (13:02 +0100)]
Update incipit snippets

9 years agoIssue 4378: Rewrite Protected_scm for static lifetime requirements
David Kastrup [Fri, 8 May 2015 09:04:01 +0000 (11:04 +0200)]
Issue 4378: Rewrite Protected_scm for static lifetime requirements

The use of Protected_scm is required where the per-instance object
marking mechanisms don't work.  This patch changes its semantics to
cater exclusively for the use case of permanent variable life time (as
opposed to permanent value life time).  It no longer employs the
reversible scm_gc_protect_object but rather uses scm_permanent_object
for protecting one cons cell as its principal storage mechanism.

It is also ensured that construction/destruction (which happen at very
early/late phases of program execution) do not require a working memory
subsystem.

9 years agoAvoid gratuitous use of Protected_scm copy constructor
David Kastrup [Fri, 8 May 2015 09:34:32 +0000 (11:34 +0200)]
Avoid gratuitous use of Protected_scm copy constructor

9 years agoIssue 4377: Run a script for using derived_unsmob where obvious
David Kastrup [Thu, 7 May 2015 19:22:20 +0000 (21:22 +0200)]
Issue 4377: Run a script for using derived_unsmob where obvious

Script is:

matched="[^()]*"
for i in 1 2 3 4 5
do
    matched="\\(($matched)\\|[^()]\\)*"
done
filelist="$(git grep -l '\<dynamic_cast[^>]*> ([_a-zA-Z]*::unsmob\>')"
typelist="$(sed -n 's/^.*\<dynamic_cast\s*<\([_a-zA-Z]\+\)\s*\*> (\([_a-zA-Z]*\)::unsmob\>.*$/\1/p' $filelist | sort -u )"
for typ in $typelist
do
    # crazy: we have unsmob definitions that are not even used
    if git grep -q "$typ::unsmob" || git grep -q "\\<$typ"'\s*\*\s*unsmob\s*(SCM'
    then
echo "There already is $typ::unsmob"
sed -i '/unsmob (SCM/,/;/!s/\<dynamic_cast\s*<\('"$typ"'\)\s*\*> (\([_a-zA-Z]*\)::unsmob\s*(\('"$matched"'\))\s*)/\1::unsmob (\3)/g' $filelist
    else
sed -i 's/\<dynamic_cast\s*<\('"$typ"'\)\s*\*> (\([_a-zA-Z]*\)::unsmob\s*(\('"$matched"'\))\s*)/derived_unsmob<\1> (\3)/g' $filelist
    fi
done

The awkward bit at the start of the script is for matching matched
parentheses.  This replaces the construct

    dynamic_cast<T *>(xxx::unsmob (yyy))

with

    derived_unsmob<T> (yyy)

where appropriate.  If T::unsmob already exists, it is used
instead (apart from inside of its own definition, of course).

9 years agoDefine derived_unsmob<class T> function
David Kastrup [Thu, 7 May 2015 19:21:54 +0000 (21:21 +0200)]
Define derived_unsmob<class T> function

This is a shortcut for dynamic_cast<T *>(T::unsmob (...)) in order to
provide an unsmob function for derived classes.

9 years agoRelease: bump VERSION.
Phil Holmes [Sun, 10 May 2015 20:29:58 +0000 (21:29 +0100)]
Release: bump VERSION.

9 years agoRelease: bump Welcome versions. release/2.19.20-1
Phil Holmes [Sun, 10 May 2015 12:25:52 +0000 (13:25 +0100)]
Release: bump Welcome versions.

9 years agoRelease: update news.
Phil Holmes [Sun, 10 May 2015 12:25:52 +0000 (13:25 +0100)]
Release: update news.

9 years agoPO: update template.
Phil Holmes [Sun, 10 May 2015 12:25:51 +0000 (13:25 +0100)]
PO: update template.

9 years agoRelease: bump VERSION_DEVEL.
Phil Holmes [Sun, 10 May 2015 12:25:51 +0000 (13:25 +0100)]
Release: bump VERSION_DEVEL.

9 years agoRe-run makelsr
James Lowe [Sun, 10 May 2015 11:53:11 +0000 (12:53 +0100)]
Re-run makelsr

A renamed snippet from the
previous checkin had
had not been removed cleanly
and so was breaking make

9 years agoRun makelsr
James Lowe [Sun, 10 May 2015 11:02:25 +0000 (12:02 +0100)]
Run makelsr

For commit
0129dea8eff59c10ba6e295f6f2cd48083fc5296

9 years agoDoc: NR section 3.5.x MIDI file creation tidy up
James Lowe [Thu, 19 Mar 2015 13:14:19 +0000 (13:14 +0000)]
Doc: NR section 3.5.x MIDI file creation tidy up

Issue 2877

I feel we're constantly updating a
'messy' chapter that doesn't really
follow many of the CG guidelines and
so could be better organized.

Apart from some @cindex or @seealso
references no new information and
nothing technical has been removed
but the information has been
reorganized so that it is (I hope)
grouped more logically.

Some snippets have been created from
the original text as they were
inappropriate for the NR.

The @lilypond examples have been simplified
and turned into @examples instead as the
engraving output is largely irrelevant
in a section about MIDI; saving space.

9 years agoUpdate NR font size section (Issue 3370)
Phil Holmes [Sun, 3 May 2015 13:51:36 +0000 (14:51 +0100)]
Update NR font size section (Issue 3370)

9 years agoPartcombiner documentation (Issue 4307)
Phil Holmes [Sat, 2 May 2015 14:16:46 +0000 (15:16 +0100)]
Partcombiner documentation (Issue 4307)

9 years agoDocument incipit command
Phil Holmes [Mon, 4 May 2015 11:19:53 +0000 (12:19 +0100)]
Document incipit command

9 years agoIssue 4356: Part combiner: generate mark events in Scheme (not C++)
Dan Eble [Sat, 18 Apr 2015 03:01:55 +0000 (23:01 -0400)]
Issue 4356: Part combiner: generate mark events in Scheme (not C++)

9 years agoUnify @documentencoding
Jean-Charles Malahieude [Fri, 8 May 2015 17:26:46 +0000 (19:26 +0200)]
Unify @documentencoding

before switching from texi2html to texi2any:
 utf-8 -> UTF-8 everywhere

9 years agoDoc: updates translation status
Jean-Charles Malahieude [Fri, 8 May 2015 16:36:13 +0000 (18:36 +0200)]
Doc: updates translation status

9 years agoDoc-fr: updates LM and NR
Jean-Charles Malahieude [Fri, 8 May 2015 16:20:07 +0000 (18:20 +0200)]
Doc-fr: updates LM and NR

9 years agoIssue 4369: silence warning in page-breaking.cc
Dan Eble [Mon, 4 May 2015 23:46:53 +0000 (19:46 -0400)]
Issue 4369: silence warning in page-breaking.cc

9 years agoIssue 4373: Start up GUILE type system before anything else
David Kastrup [Wed, 6 May 2015 09:55:18 +0000 (11:55 +0200)]
Issue 4373: Start up GUILE type system before anything else

This permits using our datatypes in the general Scheme initialization
phase without crashing or order dependencies.

9 years agoIssue 3687 (part 4): Update regression tests to use \compressMMRests
Trevor Daniels [Fri, 1 May 2015 21:04:51 +0000 (22:04 +0100)]
Issue 3687 (part 4): Update regression tests to use \compressMMRests

9 years agoIssue 3687 (part 3) Update snippets to use \compressMMRests
Trevor Daniels [Fri, 1 May 2015 21:17:10 +0000 (22:17 +0100)]
Issue 3687 (part 3) Update snippets to use \compressMMRests

9 years agoIssue 3687 (part 2): Amend docs to use \compressMMRests
Trevor Daniels [Fri, 1 May 2015 21:56:47 +0000 (22:56 +0100)]
Issue 3687 (part 2): Amend docs to use \compressMMRests

  Amend the examples in the LM and NR to use the \compressMMRests
  music function rather than the predefs \compressFullBarRests and
  \expandFullBarRests to avoid advocating the use of techniques
  which may have undesirable side effects.

9 years agoIssue 3687 (part 1): Provide a compress MM function for MM rests only
Trevor Daniels [Tue, 28 Apr 2015 22:50:31 +0000 (23:50 +0100)]
Issue 3687 (part 1): Provide a compress MM function for MM rests only

  Multi-measure rests were previously compressed (i.e. all bars
  following the first were removed) by using the \compressFullBarRests
  predef.  This simply turned on 'skipBars, which causes bar lines at
  the ends of bars which contain either no rhythmic event or contain
  a rhythmic event with a duration extending beyond the bar to be removed.
  Such bars are generated by notes with durations longer than a bar,
  like a1*3, or by shorter notes which extend beyond the bar.

  This patch provides a function which also uses 'skipBars, but with
  a filter so that 'skipBars is activated only when an actual MM rest
  is encountered and deactivated by the first following rhythmic event.

9 years agoIssue 3687 (part 0): run convert-ly
Trevor Daniels [Wed, 29 Apr 2015 15:19:05 +0000 (16:19 +0100)]
Issue 3687 (part 0): run convert-ly

  Run convert-ly on the doc files to be changed to bring them
  up to date.

9 years agoRevert "Issue 4360: Reorganize smob initialization to make it more reliable"
David Kastrup [Tue, 5 May 2015 11:00:43 +0000 (13:00 +0200)]
Revert "Issue 4360: Reorganize smob initialization to make it more reliable"

This reverts commit 90d3233da87b372e02690978f89b2add3863fba6.

9 years agoFix lilypond-invoke-editor's temp dir
Masamichi Hosoda [Mon, 4 May 2015 09:59:44 +0000 (18:59 +0900)]
Fix lilypond-invoke-editor's temp dir

Issue 4359

To use environment variable TMP and TEMP is
Windows-only conventions.
The POSIX convention is
to look in the TMPDIR.

9 years agoIssue 4358: Set DynamicLineSpanner direction for \partcombine.
Dan Eble [Tue, 28 Apr 2015 23:59:44 +0000 (19:59 -0400)]
Issue 4358: Set DynamicLineSpanner direction for \partcombine.

This reverses a difference in the regression test
part-combine-force.ly shown in Issue 4348.

9 years agoFix issue 4355 -- broken beam subdivision
Carl Sorensen [Sat, 25 Apr 2015 03:07:51 +0000 (21:07 -0600)]
Fix issue 4355 -- broken beam subdivision

When beams are subdivided, the number of beams joining the subdivisions
should match the length of the subdivision.  Previously, the subdivisions
were joined by a single beam.

9 years agoLSR local import
Phil Holmes [Sat, 2 May 2015 17:20:59 +0000 (18:20 +0100)]
LSR local import

9 years agoAdd rest styles snippet to snippets/new
Phil Holmes [Sat, 2 May 2015 17:18:22 +0000 (18:18 +0100)]
Add rest styles snippet to snippets/new

9 years agoIssue 4360: Reorganize smob initialization to make it more reliable
David Kastrup [Thu, 30 Apr 2015 18:48:09 +0000 (20:48 +0200)]
Issue 4360: Reorganize smob initialization to make it more reliable

This change is a nuisance in that it requires an explicit call to
ADD_SMOB_INIT for each Scheme-related class.

However, this setup guarantees that at the point Scheme execution
commences, all predicates of the likes of "ly:grob?" will be present.
It also guarantees that if a particular type should seem required during
_other_ initializations that happen to depend on the type and get
executed earlier, it will be provided early.

9 years agoIssue 4364: Allow ImageMagick's compare to exit with status 1
David Kastrup [Fri, 1 May 2015 22:35:35 +0000 (00:35 +0200)]
Issue 4364: Allow ImageMagick's compare to exit with status 1

Apparently Ubuntu 15.04 has a version of "compare" that cannot be easily
persuaded to return anything but exit status 1 (which indicates
dissimilar images but no actual error condition) so we allow this in
script/build/output-distance.py in order to keep "make check" from
failing.

This patch is somewhat artless but does the trick.

9 years agopo: updates Danish
Jean-Charles Malahieude [Fri, 1 May 2015 12:08:36 +0000 (14:08 +0200)]
po: updates Danish

9 years agoDoc-fr: update LM and NR
Jean-Charles Malahieude [Fri, 1 May 2015 12:01:55 +0000 (14:01 +0200)]
Doc-fr: update LM and NR

9 years agoweb-fr: updates introduction
Jean-Charles Malahieude [Fri, 1 May 2015 11:58:27 +0000 (13:58 +0200)]
web-fr: updates introduction

9 years agosnippets: too many @ left
Jean-Charles Malahieude [Fri, 1 May 2015 11:24:33 +0000 (13:24 +0200)]
snippets: too many @ left

9 years agoMerge branch 'master' into translation
Jean-Charles Malahieude [Thu, 30 Apr 2015 18:04:11 +0000 (20:04 +0200)]
Merge branch 'master' into translation

9 years agoDoc-ca: fix remaining errors
Jean-Charles Malahieude [Thu, 30 Apr 2015 17:27:21 +0000 (19:27 +0200)]
Doc-ca: fix remaining errors

9 years agoDoc-es: add requested documentencoding
Jean-Charles Malahieude [Thu, 30 Apr 2015 17:25:40 +0000 (19:25 +0200)]
Doc-es: add requested documentencoding

otherwise texinfo barfs and build fails.

9 years agoDoc-de: add language specification where needed
Jean-Charles Malahieude [Thu, 30 Apr 2015 17:23:44 +0000 (19:23 +0200)]
Doc-de: add language specification where needed

This fixes build failure due to texinfo barfing.

9 years agoDoc: typo in @documentencoding
Jean-Charles Malahieude [Thu, 30 Apr 2015 17:21:05 +0000 (19:21 +0200)]
Doc: typo in @documentencoding

coding: utf-8
BUT
@documentencoding UTF-8

9 years agoupdate texinfo.tex from upstream
Jean-Charles Malahieude [Thu, 30 Apr 2015 17:18:57 +0000 (19:18 +0200)]
update texinfo.tex from upstream

Really deal with nonascii-characters in @node and @index
(accented capitalized and ß).

9 years agostem-tremolo-style.ly: Minor doc improvement.
Werner Lemberg [Thu, 30 Apr 2015 06:15:16 +0000 (08:15 +0200)]
stem-tremolo-style.ly: Minor doc improvement.

9 years agoRename makefile variable TMP
Masamichi Hosoda [Tue, 28 Apr 2015 13:29:52 +0000 (14:29 +0100)]
Rename makefile variable TMP

Issue 4353

Rename makefile variable TMP

Windows (cygwin and mingw)
system uses environment
variable TMP for a special
purpose. It can not be used
for other purpose like
makefile variable.

9 years agoUpdated patch for issue 155
Carl Sorensen [Wed, 15 Apr 2015 02:04:50 +0000 (20:04 -0600)]
Updated patch for issue 155

Use Joe Neeman's code to enclose dots and accidentals inside parentheses
around note heads.

Eliminate pure-Y-extent (no longer needed)
Improve documentation string for parenthesis friends
Improve regression test

9 years agoIssue 4351/2: Rename Translator_void_method_ptr to Translator::Callback
David Kastrup [Mon, 20 Apr 2015 16:33:16 +0000 (18:33 +0200)]
Issue 4351/2: Rename Translator_void_method_ptr to Translator::Callback

This is somewhat more descriptive.

9 years agoIssue 4351/1: Rename Engraver_void_function_engraver_grob_info to Translator::Grob_in...
David Kastrup [Mon, 20 Apr 2015 12:41:15 +0000 (14:41 +0200)]
Issue 4351/1: Rename Engraver_void_function_engraver_grob_info to Translator::Grob_info_callback

This is somewhat more descriptive.  Since the general dispatcher
management is done in Translator rather than Engraver, making this
typedef local to Engraver instead would really complicate things.

9 years agoRelease: bump VERSION.
Phil Holmes [Sun, 26 Apr 2015 21:31:59 +0000 (22:31 +0100)]
Release: bump VERSION.