More about Abjad

How it works

How does Python suddenly know what musical notes are? And how to make a musical score?

Use Python’s dir() built-in to get a sense of the answer:

>>> dir()
['Accelerando', 'Articulation', 'Beam', 'Chord', 'Clef', 'Container', 'Context',
'Crescendo', 'Decrescendo', 'Duration', 'Dynamic', 'Fermata', 'Fraction',
'Glissando', 'Hairpin', 'KeySignature', 'LilyPondCommand', 'Markup', 'Measure',
'MultimeasureRest', 'Multiplier', 'NamedPitch', 'Note', 'Offset', 'Ratio',
'Rest', 'Ritardando', 'Score', 'Sequence', 'Skip', 'Slur', 'Staff',
'StaffGroup', 'Tempo', 'Tie', 'TimeSignature', 'Timespan', 'Tuplet', 'Voice',
'__builtins__', '__cached__', '__doc__', '__file__', '__loader__', '__name__',
'__package__', '__path__', '__spec__', '__version__', '__version_info__',
'abctools', 'abjad_configuration', 'agenttools', 'attach', 'commandlinetools',
'datastructuretools', 'demos', 'detach', 'docs', 'documentationtools',
'durationtools', 'exceptiontools', 'expressiontools', 'ext', 'f', 'graph',
'indicatortools', 'inspect_', 'instrumenttools', 'ipythontools', 'iterate',
'label', 'lilypondfiletools', 'lilypondnametools', 'lilypondparsertools', 'ly',
'markuptools', 'mathtools', 'metertools', 'mutate', 'new', 'override', 'parse',
'patterntools', 'persist', 'pitchtools', 'play', 'print_function',
'quantizationtools', 'rhythmmakertools', 'rhythmtreetools', 'schemetools',
'scoretools', 'select', 'selectiontools', 'selectortools', 'sequence',
'sequencetools', 'set_', 'show', 'spannertools', 'stringtools', 'systemtools',
'templatetools', 'timespantools', 'tonalanalysistools', 'topleveltools']

Calling from abjad import * causes Python to load hundreds or thousands of lines of Abjad’s code into the global namespace for you to use. Abjad’s code is organized into a collection of several dozen different score-related packages. These packages comprise hundreds of classes that model things like notes and rests and more than a thousand functions that let you do things like transpose music or change the way beams look in your score.

Inspecting output

Use dir() to take a look at the contents of the systemtools package:

>>> dir(systemtools)
['AbjadConfiguration', 'BenchmarkScoreMaker', 'Configuration',
'FilesystemState', 'ForbidUpdate', 'FormatSpecification', 'IOManager',
'ImportManager', 'LilyPondFormatBundle', 'LilyPondFormatManager', 'Memoize',
'NullContextManager', 'ProgressIndicator', 'RedirectedStreams',
'StorageFormatAgent', 'StorageFormatSpecification', 'TemporaryDirectory',
'TemporaryDirectoryChange', 'TestCase', 'TestManager', 'Timer', 'UpdateManager',
'WellformednessManager', '__builtins__', '__cached__', '__doc__', '__file__',
'__loader__', '__name__', '__package__', '__path__', '__spec__',
'_documentation_section']

The systemtools package implements I/O functions that help you work with the files you create in Abjad.

Use systemtools.open_last_ly() to see the last LilyPond input file created in Abjad:

% Abjad revision 12452
% 2013-10-22 13:32

\version "2.17.3"
\language "english"

\header {
    tagline = \markup {  }
}

\score {
    c'4
}

Notice:

  1. Abjad inserts two lines of %-prefixed comments at the top of the LilyPond files it creates.
  2. Abjad includes version and language commands automatically.
  3. Abjad includes a special abjad.scm file resident somewhere on your computer.
  4. Abjad includes dummy LilyPond header.
  5. Abjad includes a one-note score expression similar to the one you created in the last tutorial.

When you called show(note) Abjad created the LilyPond input file shown above. Abjad then called LilyPond on that .ly file to create a PDF.

(Quit your text editor in the usual way to return to the Python interpreter.)

Now use systemtools.open_last_log() to see the output LilyPond created as it ran:

GNU LilyPond 2.17.3
Processing `7721.ly'
Parsing...
Interpreting music...
Preprocessing graphical objects...
Finding the ideal number of pages...
Fitting music on 1 page...
Drawing systems...
Layout output to `7721.ps'...
Converting to `./7721.pdf'...
Success: compilation successfully completed

This will look familiar from the previous tutorial where we created a LilyPond file by hand.

(Quit your text editor in the usual way to return to the Python interpreter.)