Merge from upstream.

* scripts/lilypond-book.py (process_snippets): Argument fix.

* SConstruct:
* buildscripts/builder.py:
* lily/SConscript:
* mf/SConscript:
* Documentation/user/SConscript:  More SCons fixes.

* lily/font-config.cc (init_fontconfig): Only initialize if
global cache_file is found.

* Documentation/user/instrument-notation.itely,
input/regression/breathing-sign-ancient.ly,
input/test/divisiones.ly: Removed redundant property settings for
vaticana contexts.

* ly/gregorian-init.ly: Minor edit of comments.

    * ly/music-functions-init.ly: Updated \overrideProperty to use the
        new \applyOutput.

*** empty log message ***

* lily/ledger-line-engraver.cc (stop_translation_timestep): new
function. Delay ledger modifying ledgered grobs to
stop_translation_timestep(), to ensure that we're ledgering
according to the new spanner.
(acknowledge_staff_symbol): don't check for non-NULL
staff_sym->get_bound()

* lily/staff-symbol-engraver.cc (start_spanner): set left bound on
creation.

* Documentation/pictures/GNUmakefile (XPM_FILES): src-wildcard for
pictures too.

* Documentation/misc/GNUmakefile (TEXTS): use src-wildcard for
disting MISC files.

(dist-toplevel-txt-files): dist aclocal.m4
directly from srcdir/stepmake/aclocal.m4

* Documentation/misc/GNUmakefile (TEXTS): use src-wildcard for
disting MISC files.

* lily/simple-spacer.cc (get_column_desc): desc -> description.

* lily/simple-spacer.cc (get_line_configuration): add
keep-inside-line rods
(struct Column_desc): add constructor
(struct Rod_desc): add constructor
(get_line_forces): check for forced page breaks here too

* lily/instrument-name-engraver.cc (acknowledge_axis_group): prevent
support/elements cycles between InstrumentNames and AxisGroups

* stepmake/stepmake/topdocs-targets.make (make-txt-files): define
make-txt-files target.

* stepmake/stepmake/toplevel-targets.make (do-top-doc): use
make-txt-files target.

* configure.in (LINK_GXX_STATICALLY): $srcdir for readlink.py

*** empty log message ***

*** empty log message ***

* ly/gregorian-init.ly: Rewrote \versus and \responsum in order to
fix lyrics alignment problems.  They are implemented now as scheme
functions that add the appropriate unicode char to the first lyric
event rather than adding a separate lyric event.

* buildscripts/output-distance.py
(ComparisonData.create_html_result_page): put version/directories
in table header.

* Documentation/topdocs/NEWS.tely (Top): document spacing section
feature.

(Top): document spacing section
feature.

* configure.in (LINK_GXX_STATICALLY): use readlink.py to resolve links.

* buildscripts/readlink.py: add  file.

lily/tuplet-number.cc (print): prevent stencil from being
garbage collected.

(write-system-signature): explicitly
-well, superfluously- close output port.

(main): oops. Add extra argument.

*** empty log message ***

Clarified example from Francisco Vila, thanks!

(main): use compare-XXX for
compare files.
(main): --threshold option, default at 0.30

(paper-system-annotate): fix problem when
annotating an empty system

reinstate old web tar/copying.

* lily/lyric-combine-music-iterator.cc (find_voice): return 0 if
nothing changes. (Erik S)

* lily/percent-repeat-iterator.cc (get_music_list): fix repeat
count. (Erik S)

*** empty log message ***

* scm/define-markup-commands.scm (whiteout): do not force
foreground color of argument markup to black.

* scm/stencil.scm (annotate-y-interval): put arrow dimension at
the left of the arrow, instead of below the arrow name, so that,
when two little arrows are vertically stacked, their dimensions
and name should not overlap. Add a color key parameter.

* scm/paper-system.scm (paper-system-annotate): Annotate
next-space+next-padding instead of next-space. Annotate space
between next-padding and next-space+padding, respectively, and
following system extent and refpoint-Y-extent. Use colors.

* scm/page.scm (annotate-page): translate annotations slightly to
the right.

* buildscripts/output-distance.py
(SystemLink.output_expression_change_count): keep track of changed
details, and dump in details html page.

* lily/figured-bass-position-engraver.cc (acknowledge_slur):
add slurs and ties to support too.

* buildscripts/output-distance.py (main): set dest_dir
argument. Add --max-count option.

add note about setting
properties in Staff context.

(acknowledge_slur):
add slurs and ties to support too.

(main): set dest_dir
argument. Add --max-count option.

Trivial fix.

* ly/gregorian-init.ly: Converted to utf-8.  Added memorable
shortcuts for special unicode chars that are useful in chant
notation.  Revised exisiting and added new comments.

tweaks.

*** empty log message ***

*** empty log message ***

* tex/texinfo.cnf: Add support for U+0132 (IJ) and U+0133 (ij).

* tex/texinfo.cnf: Fix typo (\euro -> \minus).

(FileLink): new class. collect
info systems from one .ly file.

Partial fix for @funindex (doesn't break anything).

(sectionheading, q.v.): call \quoteexpand for
@example too. Backportme.

* scm/define-grob-properties.scm: Previous patch contained
spurious line from another patch not yet committed.  Fixed.

Francisco's latest Spanish glossary fixes and Werner's UTF-8 texinfo
patch; thanks both of you!

* scm/define-grob-properties.scm: Add missing doc strings for
ancient notation.

* buildscripts/output-distance.py
(ComparisonData.write_text_result_page): separate the scores, sort
by geo_score.

* buildscripts/output-distance.py (me_path): use argv[0] for
sys.path extension
(ComparisonData.img_cell): use colored borders.

(output-classic-framework): start at 1 for
system signatures.

* buildscripts/output-distance.py (me_path): use argv[0] for
sys.path extension
(ComparisonData.img_cell): use colored borders.

* scm/stencil.scm (write-system-signature): escape newlines too.
(write-system-signature): use (1,-1) interval for empty interval

(write-system-signature): escape newlines too.

*** empty log message ***

*** empty log message ***

Various additions and edits.

(tree-lib-prefix-current):
(tree-share-prefix-current): Use version number in tree, add
`current' symlinks.

* make/mutopia-rules.make ($(outdir)/%.png $(outdir)/%.pdf
$(outdir)/%.ly $(outdir)/%.ps): dump sigs for normal files too.

* stepmake/aclocal.m4: robuster version detection.

robuster version detection.

* stepmake/aclocal.m4: revert version detection.

* buildscripts/output-distance.py
(ComparisonData.create_html_result_page): copy only if file
exists, fixup src paths.

* stepmake/bin/add-html-footer.py (built): junk gulp_file()

* stepmake/aclocal.m4: Robustification for (gcc) version detection.

* cygwin/postinstall-lilypond.sh: Do not fail if info docs aren't
available.  This should be moved to postinstall-lilypond-doc.

simplify local-WWW-post.
remove -type l from find.

simplify local-WWW-post.
remove -type l from find.

simplify local-WWW-post.

(ComparisonData.compare_trees):
fixes. Don't follow symlinks.
(ComparisonData.create_html_result_page): remove previous results.
insert ../python in path.

* buildscripts/output-distance.py (ComparisonData.compare_trees):
fixes. Don't follow symlinks.
(ComparisonData.create_html_result_page): remove previous results.

* scripts/lilypond-book.py (PREAMBLE_LY): spacing fixes.

* config.make.in (datadir): Add datarootdir to silence autoconf.

* stepmake/aclocal.m4: Update for autoconf-2.59d.

Fix some bugs in the dynamic engraver and PostScript backend

(Two-pass vertical spacing): add documentation for two-pass spacing
technique.

Compile fix.

New location for this file (moved from bugs/ CVS).

* lily/note-head.cc: Fixed programming_error message.

(pure-conversion): move pure-callback
further into body.

(default-script-alist): set avoid-slur for turn,
marcato, stopped, thumb.

(process_music): don't
skimp on memory usage.

* lily/beam.cc (set_stem_lengths): do set_stemend for invisible
stems too. Fixes #5

* lily/parenthesis-engraver.cc (acknowledge_grob): set parent of
parenthesis item. This fixes premature Y-extent triggering. Fixes
issue #95.

* VERSION (PATCH_LEVEL): bump version.

Remove empty file.

Info about adding names to pianostaffs and other contexts, thanks Marcus!

Update Finnish translation.

* mf/feta-bolletjes.mf (solfa_quarter_width): use
solfa_base_notewidth as unit for non-DO-notes too.

* input/mutopia/W.A.Mozart/mozart-hrn-3.ly: update email.

(Top): simpler, more clearer override.

*** empty log message ***

* python/lilylib.py (progress): Minor fix.

* scripts/lilypond-book.py (Lilypond_file_snippet.my_system):
write snippet-names binary.

* scripts/lilypond-book.py (bindir): use os.pathsep for prepending
to $PATH.

* scm/define-grobs.scm (all-grob-descriptions): use callback to
calc dot count.

* scm/output-lib.scm (string-number::calc-text): new function.
(note-head::calc-duration-log): new function.

* scm/output-lib.scm (fingering::calc-text): new function.

* python/lilylib.py (system): use os.system for mingw32

* scripts/lilypond-book.py (bindir): use os.pathsep for prepending
to $PATH.

* scripts/lilypond-book.py (bindir): Use os.path.join to add the
path, in order to get correct path separators on all platforms.

* scm/backend-library.scm: update email address.

* scm/define-grobs.scm (all-grob-descriptions): use callback to
calc dot count.

* scm/output-lib.scm (string-number::calc-text): new function.
(note-head::calc-duration-log): new function.

* scm/output-lib.scm (fingering::calc-text): new function.

* Documentation/topdocs/NEWS.tely (Top): add duration-log override
example.

* scm/output-lib.scm (string-number::calc-text): new function.
(note-head::calc-duration-log): new function.

* scm/output-lib.scm (fingering::calc-text): new function.

(process_music): idem.

* scm/define-grobs.scm (all-grob-descriptions): use callback to
calc dot count.

* lily/dots-engraver.cc (class Dots_engraver): new engraver:
handle dot creation separately.

* lily/tab-note-heads-engraver.cc (process_music): idem.

* lily/drum-note-engraver.cc (process_music): idem.

* lily/note-heads-engraver.cc (process_music): idem.

* lily/rest-engraver.cc: don't set duration-log.

* scm/output-lib.scm (string-number::calc-text): new function.
(note-head::calc-duration-log): new function.

* lily/fingering-engraver.cc (make_script): remove 'text setting.

* scm/output-lib.scm (fingering::calc-text): new function.

* scripts/lilypond-book.py: Fix typo gs-load-font -> gs-load-fonts
 and include-eps-font -> include-eps-fonts

use internalBarNumber

* lily/multi-measure-rest-engraver.cc: use internalBarNumber

* lily/timing-translator.cc: list properties.
(start_translation_timestep): update internalBarNumber too.

* scm/define-context-properties.scm
(all-user-translation-properties): add internalBarNumber.

*** empty log message ***

(link_file): verbosity for OSError.

#include cleanup.

add assert to
scm_or_str2symbol(SCM).

*** empty log message ***

*** empty log message ***

* lily/spacing-spanner.cc (musical_column_spacing): handle
musical->nonmusical spacing case, where there is no spacing wish.

* VERSION: release 2.9.20

* lily/pango-font.cc (text_stencil): allow 'tight' bounds.
(pango_item_string_stencil): use logical rect horizontally (to
preserve spaces), ink_rect vertically (for time signatures).

* lily/font-metric.cc (word_stencil): new interface function

* input/regression/utf-8-mixed-text.ly: new file.

* lily/percent-repeat-engraver.cc (listen_percent): remove
spurious warning.

* lily/phrasing-slur-engraver.cc (acknowledge_script): only ack
non-dynamic scripts. Fixes #30.

* lily/pango-font.cc (pango_item_string_stencil): use
logical_rect. This prevents spaces after words from disappearing,
issue #72.

(acknowledge_script): only ack
non-dynamic scripts. Fixes #30.

* lily/slur.cc (outside_slur_callback): make offset_scm optional.

* lily/beam.cc (rest_collision_callback): make prev_offset optional.

* lily/grob-closure.cc (chain_offset_callback): don't pass 0 but
SCM_UNDEFINED for non-existent data.

* lily/side-position-interface.cc (general_side_position): allow
optional current_offset argument for chaining, so combinations of side-position
and outside slur callback don't add up.  Fixes issue #92.

(pango_item_string_stencil): use
logical_rect. This prevents spaces after words from disappearing.