patch::: 1.1.23.jcn8: tfm reader

[lilypond.git] / TODO

Features you cannot find in the doco as working, should be mentioned
here.

This is an assorted collection of stuff that will be done, might be
done, or is an idea that I want to think about

Most of the items are marked in the code as well, with full explanation. 
grep for TODO and ugh/ugr/urg


* check out PS points vs. LilyPond points (PS = 1/72 inch, LilyPond=1/72.27)

* Instrument names in the margin. Example: 
  input/test/hara-kiri.ly. The text should of course
  be to the left of the staff lines.

* There is a bug in pl 22. Try /input/test/slurs.ly and 
  you will find '#.#' instead of real numbers at some
  places in the empedded ps code.

* (Annoying but not a bug:) Slurs that are broken at line 
  breaks often extend too far into the margin, both to the 
  left and to the right. How do you adjust it?

* SkipBars=0 doesn't work. My score contains lots of 
  multibar rests, which should be written out bar by bar
  in the score (where the 'harakiri' mechanism doesn't
  remove full staff lines that only contain rests).
  In the parts, the multibar rests should be split
  by the rehearsal marks I've included every here and 
  there.
  In short, I want the multibar rests to work as they
  used to do before pl 20 but preferably with a nice
  centered rest symbol. 
        * junk backlinks?

        * junk text{sharp,flat,etc}

        * fix convert-mudela manpage

        * automatic staff selection.

        * decimal point in \paper {}

        > I have changed Mark_engraver to use the G_... classes.
        > My implementation has the following shortcomings at the
        > moment.
        > 
        > - markScriptPadding is not supported, since G_staff_side_item does not
        >   support padding.
        > 
        > - markBreakPriority is not supported. 
        > 
        > - Mostly, you want the marks aligned with the bar lines. I don't 
        >   really understand how to implement that feature.
        > 
        > - You get lots of warnings about missing 'number' and 'Large'


        *{  \voiceone <a'4.*2/3  d''4.*2/3 fis''4.*2/3>  [g''8 fis'' e''] d''4 |}
         and=20
         { \voicetwo  d'4 d'2.*2/3 } spacing

        * tied notes for MIDI

        * sharp /flat on trills (in MIDI ??)

        * scm-ify \property values.
- The direction of textual scripts ('^' or '_') is ignored.
        * move class Lookup {} into scm

        * read tfms too.

        * update mi2mu for lilypond 1.1

BUGS:
        * collisions/voices \voiceone \voicetwo are broken; see 
          input/praeludium-fuga-E.ly
          mutopia/J.S.Bach/wtk1-fugue2.ly

        * fix: standchen.

        * fix:  cello suites

        * fix:  text on rests: rediculous dims

        * fix extender

        * fix fingering 'a-5': font direction broken *again*

        * fix braces

        * ly2dvi barfs on linewidth = -1.

        * fix midi output:
           - default duration?  duration must be not entered
             explicitely on first note.

        * turn slope-damping on/off

        * c4 \! \> c4

        * tremolo stuff

        * gallina barlines.

        * fix height of / collisions with lyrics (chords), 
           see input/test/vertical-text.ly; input/test/repeat.ly

        * repeat/volta:
          - one volta spanner per score

        * latex bla.tex broken (titles / \lilyfooter stuff?)

        * check/set minimum slur/tie length

        * msgfmt -o check?

        * \breathmark TeX macro 

        * catch GUILE errors?

        * add new glyphs to font.ly

        * formatting of input stuff. 

        * \notes{ a \< b \cr } vs \notes{ a \< b \! }

        * if possible, it might be nice for a warning to appear if someone does
        \translator with no name and without assigning it to an
        identifier.  

        * space after bars?

        * 'hinterfleisch' before bar (e.g. wtk1-fugue2)?

        * \type Voice \times 2/3 { [c8 c16 c16 c16 c16] }

        * fix singleStaffBracket

        * repeat bars: need distance after ":|" and before "|:"

        Summary of minor spelling irregularities:
        -  capitalization/use of underscores in property names

        * fix SkipBars  -> skipBars

        * \shape 1st dim skipped? 

        * fix marks/barnumbers/marginal notes

        * broken scripts:
        lbheel = \script { "bheel" 0 0 -1  0 0 }
        rbheel = \script { "bheel" 0 0 1 0 0 }
        lbtoe = \script { "btoe" 0 0 -1 0 0 }
        rbtoe = \script { "btoe" 0 0 1 0 0 }
        lfheel = \script { "fheel" 0 0 -1  0 0 }
        rfheel = \script { "fheel" 0 0 1 0 0 }
        lftoe = \script { "ftoe" 0 0 -1 0 0 }
        rftoe = \script { "ftoe" 0 0 1 0 0 }
        
        and also
        
        portato= \script { "portato" 0 -1 0 1 0 }

STUFF
        * uniformise recent feta contributions.

        * check out legal/(c) matters for scores.

        * check out GCC signatures?

        * Align_element::padding  ?

        * uniformise property names.

        * ydirection <-> yDirection

        * typo checks on property names?

        * use streambufs and iostream
        to provide IO handling for TeX stream, mudela stream, data-file.

        * seriously buffer TeX output (do profile of writing .5 mb TeX file.)

        * strip EXEs before installing

        * zip target for binary windows dist (JBR)

        * junking \skip req in lyrics

        * Language:
        - \type -> \context ?
        - \translator -> ?
        - fix \partial 
        - \bla {} vs \bla ; 
        - mix engraver hacking with music ?
        - \once\property  KEY = VAL
        - \addtranslator, \removetranslator
        - junk ^ and _ for scripts
        - junk _ for lyrics.
        - abstract grammar.

        * percussion note heads

        * mi2mu empty staffs.

        * midi_instrument -> midiInstrument

        * horizontal centering of dynamics 

        * \font\fontA=feta20.afm (.afm?)

        * gzip RH manpage

        * stable make/config stuff

        * $DEPENDENCIES_OUTPUT support

        * fix weird behaviour when .AFMs/.SCMs not found.

        * Xdvi zooming ?! Try to understand GS garbage collection.
                gs: Error: /undefined in draw_beam
                gs: Operand stack:

        * fix vertical alignment and stafflines

        * GrandStaff needs more work -- I want a single word
        `harpsichord' to the left of the grandstaff, not one on each
        stave.  (Organ staff -- with separate pedal -- but common
        properties may be something to think about.  Of course for
        organ you want to be able to give stop indications on the way
        through, so the \property Voice.Instrument would be a stop,
        and \property GrandStaff.instrument would be PipeOrgan...)

        * revise the Score_priority_align_engraver concept.  It sucks. 

        * make new VoiceOne, VoiceTwo, VoiceThree contexts with
ydirection and hshift preset

        * *.yo: fix pod manpage layout legacy

        * text-items clash with stems/beams

        * --include, -I option for ly2dvi (pass on to lily)

        * fix placement of beam-less abbrev

        * \meter 3/4;
        <{\voiceone cis4. cis8 cis4 | cis4 cis cis | r1 }
         {\voicetwo cis,4 r r | r2. | r1 }> |
        }

        * try to use template<Type> iso MACRO(Type)

        - NEWS: ... Musical_pitch (analogous to Duration and Rhythmic_req)
          think about, analogous to pitch: 
           * { a4 a16 } c <-- duration of c?
           * <a4 b8> c <-- duration of c?

        * \tempo 4. = 90;

        * fix audio-items (don't use (input) requests, but copy data)

        * make sure all ex's have a mudela-version

        * do rest-collisions for Multi_measure_rests as well.

        * split error/warning in error/warning/non_fatal_error

        * add a Duration_convert member to Duration_iter to set parameters.
        Junk global duration settings. 
 
        * deps for ly2dvi

        * minimum length second part broken tie
 
        * en-,discouraged linebreaking: 
          - handle DISALLOW < penalty > FORCE
          - discourage breaking of slurs

        * fix variable define/lookup parser global<->paper

        * fix title/instrument placements; see input/test/title.ly

        * <\voiceone c4. \voicetwo c4>

        * add full cello suites to web site (kom op, Maarten!)

        * Rethink Time_description
          * \cadenza , \meter, \grouping should all be \properties

        * rename
          - measure -> bar
          - abbrev -> tremolo
          - abbreviation-beam -> (measured/unmeasured) tremolo
          - partial -> anacrouse

        * shared lib

        * robustify Beam (in case of no stems, 32nd notes)

        * decentralised spring factory

        * move paper vars into engraver properties

        * check for groff / troff/ nroff et

        * more intelligent file searching
        - make LilyPond RPM fully relocatable

        * nicen examples:
        - break-up heavily-nested score blocks

        * disable spaces in TeX stuff
        
        * handle ^C for tmp/file creation.

        
        * dots & rest collisions.

        * documentation
        - introduction?
        - info?
        - LaTeX?
        - more manpages?

        * versioning for Feta

        * the warning about the negative slur/tie length appears twice
        which is irritating.

        * The `3' in the meter key is a one or two pixels too thin (at 600dpi)
        in the middle:

        * the bracket is ugly (wings are too parabolic, should be more circular)

        * better hshift (dots, distance, head dependent)

        * clefs (AG): The "8" should appear closer to the actual clef,
          touching it.

        * put errorlevel in Input class

        * junk nesting slurs

        * integrate midi-elts from mi2mu and lily?

        * score-bar-scripts
        FMR: Free memory read:
          * This is occurring while in:
                Bar_column::do_substitute_dependency(Score_element*,Score_element*) [bar
        -column.cc:29]
                Score_element::unlink() [score-elem.cc:383]
                Paper_score::set_breaking(const Array<Column_x_positions>&) [p-score.cc:
        150]
                Paper_score::calc_breaking() [p-score.cc:189]
                Paper_score::process() [p-score.cc:203]
                Score::run_translator(Music_output_def*) [score.cc:117]
          * Reading 4 bytes from 0x65d3a0 in the heap.
          * Address 0x65d3a0 is 160 bytes into a freed  block at 0x65d300 of 208 bytes.
          * This block was allocated from:
                malloc         [rtlib.o]
                __builtin_new  [libgcc.a]
                Span_score_bar::clone()const [span-score-bar.hh:20]
                Item::copy_breakable_items() [item.cc:69]
                Item::do_breakable_col_processing() [item.cc:89]
                Score_element::calcalute_dependencies(int,int,void(Score_element::*)()*)
         [score-elem.cc:171]
          * There have been 22 frees since this block was freed from:
                free           [rtlib.o]
                __builtin_delete [libgcc.a]
                Span_score_bar::~Span_score_bar() [span-score-bar.hh:18]
                Paper_score::set_breaking(const Array<Column_x_positions>&) [p-score.cc:
        156]
                Paper_score::calc_breaking() [p-score.cc:189]
                Paper_score::process() [p-score.cc:203]
        Postprocessing elements... 
        TeX output to score-bar-scripts.tex...


3RD PARTY BUGS:

        * make GCC warn about ctor that leaves member vars  uninitialised.

        * redhat (v?) graphical install bomb-out?

        * GNU diff 2.7: diff -rN does not see a new directory with empty file

        * GNU patch 2.1: <must find simple example> if patch chokes on a
          hunk, or on allegedly unsorted hunks in a diff, it may stop 
          without a trace of failure (create .rej file, print error or 
          dump core) and just ignore any further diffs.

        * glibc 2.0:

          f = fopen ("/dev/null", "r")
          assert (feof (f))

3RD PARTY PROJECTS:

        * GNU indent: handle C++

        * GNU patch 
          - 'double-fix' mode: ignore identical fix, rather than suggest 
            to revert patch direction when (more or less? exactly) the same 
            thing is fixed

PROJECTS

        * Scripts:
        - fix position of bar number
        - stack scripts in chords:
          % fingering:        footing:
            < a-4 c-2 g-1 >   < c-\ltoe e-\lheel >
        - script-spacing
        - slur parts as a script
        - stack parameterised scripts for fingering and footing:

        * convert-mudela:
        - automatically add  a \version if not present.
        - better parsing
        - make smarter
        - convert-mudela -e empty.ly

        * emacs mudela-mode
        -  should handle block comments too.
        -  handle lexer modes (\header, \melodic, \lyric) etc.
        -  indentation
        -  notenames?
        -  fontlock: \melodic \melodic

        * use properties for:
        - plet?
        - Staff_sym size
        - default tempo.
        - font size
        - bartypes (get rid of derived Span_bar_engravers)
        - cadenza mode? 

        * page handling:
        - PS output
        - optimal pagebreaking.

        * Unicode support? -> man 7 unicode
        - 16 bit Strings
        - -u switch
        - detect of 16/8 bit files
        - examples to go with it.

        * Spring_spacer:
        - write a faster Spring_spacer (without matrices if possible)
        - relate energybound to linelen unitspace fontsize etc.
        - used fixed point fp?

        * \header
        - output header info to MIDI too.

        * a musical dictionary. See Documentation/vocabulary-*, other
        languages:
        - explanations
        - italian
        - german
        - dutch
        - swedish
         .. ?

        * chords
        - guitar chords (fret diagrams)
        - other chord name styles (american(?) style)
        - basso continuo
        - MIDI output?

        * Debug logs:
        - derive lily_stream, Dstream, texstream from ostream?
        - indentable stream as baseclass for TeX stream, lily stream, Dstream.
        - use dstream feature in mi2mu
        - integrate IO checking into stream class

        * input converters
        - NIFF?
        - ABC? 
        - SMDL?

        * add to MIDI output:
        - tempo change
        - ornaments (and trills?)
        - repeat/volta
        - slurs
        - accents
        - dynamics
        - account for rhythmic position in measure
        - etc.

        * grace notes
        - make separate class for Grace_*_engravers 
        (which is at Voice_gravs level)
        - make encapsulated spacing problems.

        * accidentals per Voice_group

        * Output class, handles : (smallish)
        - help text /(c) notice?
        - version line
        - warning /errors/progress
        - abort on error/warning; exit status
        - quiet/ignore-version options
        - logfile output

        * piano staff

        * elaborate Staff_side baseclass:
        - scripts
        - text
        - slur start/end

        * PS 
        - speed up PS code.
        - PS lines to .eps files
        - write custom prolog


        * Output an output format independent (ofi) typeset file; and 
          make ofi2 TeX, MusiXTex, Display PostScript, PostScript,
          Ascii... interpreters. (difficult)

        - NIFF ?
        - Common Music Notation
        - PostScript output (esp. Beams, Slurs, etc)
        - Move all Score_elems, into libmutype.a
        - SGML

        * Documentation
        - Doc (C) stuff of sheet music
        - a better format?  SGML?  TeX?  Info?
        - a better tutorial

        * more spanners (smallish)
        - Glissando
        - trill

        * Rewrite Stem, Beam and Rhythmic_grouping (hairy)
        - [c8. c32 c32]
        - doc on Rhythmic_grouping
        - general shaving on Beam and Stem
        - use plet grouping
        - separate visual from rhythmic info even more
        - beams over bars

        * lines:
        - Ledger lines, should be separate item: Ledger_lines, Ledger_lines
        - set different line sizes / thicknesses
        - beam stem showthrough, inter beam sizes (french vs german style)

        * Collisions
        - left/right note balls should be handled by Collision:
        < \multi 2; { \stem 1; <b c> } { <f as b> } >

        * Keys:
        - merge key{item} & localkey{item}?
        - spacing, national styles?

        * Support for TAB
        - redo Notehead to allow digits as heads
        - add a \tablature {} lexer mode for easy entering

        * should adjust stemlength for flag number.

        * breaking:
        - caching breakpoints
        - saving them
        - specify number of lines

        * markers

INPUTLANGUAGE

        * c4 4 4 4 for c4 c4 c4 c4?

        * neatly (and readably) expressing concurrent
        voices

        * <c f a>4 ?

        * Viola mode?

                @c c g e g 
        for
                c g es g, 

                @A c g e g
        for
                cis gis e gis

        * relative mode for pitch entering

        * configure pitch_byte

        * rest name configurable

SMALLISH PROJECTS

        * limit memory usage; the full coriolan takes up too much resources.

        * fix confusing naming in Engraver vs Engraver_group_engraver (context)

        * progress when creating MIDI elts.

        * A range for context errors (eg. mark both { and }. )

        * lyric in staff  (sharpsharp in staff, text below)

        * half sharps/flats

        * write Dynamic_line (to group dynamics horizontally)

        * use Real for all y positions.
        - use translate_axis( .. ,Y_AXIS)  iso separate variables.

        * half-sharps, half-flats

        * adaptive accidental spacing.

        * handle EOF graciously in error messages.

        * declare notenametab?

        * midi esp.: use I32 iso int where 32 bits are needed (or assumed...)

        * stafftypes: voice names/ instrument names.

        * lily \tempo

        * C meter iso 4/4, idem for alla breve, etc.

        * textspanner: center position

        *   % toe to bheel   four to five
            %   ^_u            4_5  
            %     -                           
            %  --|x--         --|x--
            %  --|---         --|---
            %  --|---         --|---
          (where "to" is a tiny bow)

        * auxilliary file for caching info.

        * Text_crescendo

        * clean solution for staffsize in items.

        * revise calcideal

IDEAS

        * Output data structures of Mudela in XML/SGML.

        * create libmudela, or liblily_frontend

        * move MIDI stuff (including Quantization) to a ANSI C libmidi library.

        * use an embedded language: Python, Scheme
        for:
                - Score_elems
                - Translators

        * y -dims in internote? Staff-space? X-dimensions?

        * Spacing_request for manually adjusting spacing

        * caching breakpoints

        * used fixedpoint arithmetic for min. energy.

        * move towards incremental algorithms.

        * far fetched language ideas
        - design meta-language like TeX and Metafont
        - programming/macros input

        * versioning stuff (cvt mudela, mudela, etc.)

        * Klavarskribo?

        * dynamic loading of engravers? 

SMOBS:

****************************************
Han-Wen Nienhuys <hanwen@cs.uu.nl> writes:

> mcmanus@IDT.NET writes:
> > > I want different C++ objects (with scheme embedded) to be linked by
> > > C++ pointers.  I want to access these other objects from the Scheme
> > > code, and I want access to these pointers from C++.
> > 
> > You can do this with a  combination of smob's and primitive functions,
> > though it may not be  be exactly what you wanted.   You would have one
> > smob  for each class in  your application, and then  you would write a
> > primitive function that would return the objects  that are linked to a
> > base object.
> 
> And the smob and the C++ class are interlinked?  Like
> 
>       class Foo_class {
>               Foo_smob *smob_ptr_;
>       }
> 
>       struct Foo_smob {
>               Foo_class *class_ptr_
>       }

Usually you can get away without having this interlinked structure.
One piece of information you need when exporting objects to scheme is
the smob tag for your class.  You might store this value (a long) into
a class static variable, or a file static variable.  I'll use a class
static variable in this example.

I typically use code that works like this (untested code ahead):

class Foo_class {
    static long *Foo_smob_tag;
    SCM obj; // representation as a scheme object, description comes later
}

// call this once on startup
void init_Foo_class {
    static scm_smobfuns type_rec;

    type_rec.mark = foo_mark;
    type_rec.free = foo_free;
    type_rec.print = foo_display;
    type_rec.equalp = 0;
    Foo_class::Foo_smob_tag = new int(scm_new_smob(type_rec));
}

When you want to export an object to scheme, you put a pointer to the
object itself in the cdr of the cell, and the tag in the car of the
cell.  The code looks like this:

// Is this a Foo?
static int
foo_p(SCM obj)
{
    return(SCM_NIMP(obj) && SCM_CAR(obj) == Foo_class::Foo_smob_tag);
}

// given a Scheme representation, return, a C++ representation
static Foo_class *
foo_unbox(SCM obj)
{
    return((Foo_class*)SCM_CDR(obj));
}

// the SCM representation of the object is stored inside itself
// this will become useful when destructor gets called
static SCM
foo_box(Foo_class *foo)
{
    SCM_DEFER_INTS;
    SCM_NEWCELL(foo->obj);
    SCM_SETCAR(foo->obj, Foo_class::Foo_smob_tag);
    SCM_SETCDR(foo->obj, (SCM)foo);
    SCM_ALLOW_INTS;
    return foo->obj;
}

> C++ determines life time, so what is the strategy?

What happens now when the destructor gets called?  Lets set the cdr to
NULL, and then check for that in all of the primitive functions that
use your smob.  We'll call this notion 'live'; a scheme object is
'live' if the C++ representation still exists, it's dead if the C++
object is gone.  You can still have references to dead objects in
scheme code; it's just an error to use them.  This idea is stolen from
Emacs's handling of buffers ('buffer-live-p' and friends).

Add another function, foo_live_p:

static int
foo_live_p(SCM obj)
{
    return(foo_p(obj) && SCM_CDR(obj) != NULL);
}

In you destructor, you need to do:

~Foo_class()
{
    SCM_CDR(this->obj, NULL);
}

When writing new primitives, now just check foo_live_p().

I hope the above helps,

-russ


--
"Even if you are on the right track, you'll get run over 
if you just sit there."  
             --Will Rogers (1879-1935)

***************************