1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
|
"""
pygments.lexers.lilypond
~~~~~~~~~~~~~~~~~~~~~~~~
Lexer for LilyPond.
:copyright: Copyright 2006-2023 by the Pygments team, see AUTHORS.
:license: BSD, see LICENSE for details.
"""
import re
from pygments.lexer import bygroups, default, inherit, words
from pygments.lexers.lisp import SchemeLexer
from pygments.lexers._lilypond_builtins import (
keywords, pitch_language_names, clefs, scales, repeat_types, units,
chord_modifiers, pitches, music_functions, dynamics, articulations,
music_commands, markup_commands, grobs, translators, contexts,
context_properties, grob_properties, scheme_functions, paper_variables,
header_variables
)
from pygments.token import Token
__all__ = ["LilyPondLexer"]
# In LilyPond, (unquoted) name tokens only contain letters, hyphens,
# and underscores, where hyphens and underscores must not start or end
# a name token.
#
# Note that many of the entities listed as LilyPond built-in keywords
# (in file `_lilypond_builtins.py`) are only valid if surrounded by
# double quotes, for example, 'hufnagel-fa1'. This means that
# `NAME_END_RE` doesn't apply to such entities in valid LilyPond code.
NAME_END_RE = r"(?=\d|[^\w\-]|[\-_][\W\d])"
def builtin_words(names, backslash, suffix=NAME_END_RE):
prefix = r"[\-_^]?"
if backslash == "mandatory":
prefix += r"\\"
elif backslash == "optional":
prefix += r"\\?"
else:
assert backslash == "disallowed"
return words(names, prefix, suffix)
class LilyPondLexer(SchemeLexer):
"""
Lexer for input to LilyPond, a text-based music typesetter.
.. important::
This lexer is meant to be used in conjunction with the ``lilypond`` style.
.. versionadded:: 2.11
"""
name = 'LilyPond'
url = 'https://lilypond.org'
aliases = ['lilypond']
filenames = ['*.ly']
mimetypes = []
flags = re.DOTALL | re.MULTILINE
# Because parsing LilyPond input is very tricky (and in fact
# impossible without executing LilyPond when there is Scheme
# code in the file), this lexer does not try to recognize
# lexical modes. Instead, it catches the most frequent pieces
# of syntax, and, above all, knows about many kinds of builtins.
# In order to parse embedded Scheme, this lexer subclasses the SchemeLexer.
# It redefines the 'root' state entirely, and adds a rule for #{ #}
# to the 'value' state. The latter is used to parse a Scheme expression
# after #.
def get_tokens_unprocessed(self, text):
"""Highlight Scheme variables as LilyPond builtins when applicable."""
for index, token, value in super().get_tokens_unprocessed(text):
if token is Token.Name.Function or token is Token.Name.Variable:
if value in scheme_functions:
token = Token.Name.Builtin.SchemeFunction
elif token is Token.Name.Builtin:
token = Token.Name.Builtin.SchemeBuiltin
yield index, token, value
tokens = {
"root": [
# Whitespace.
(r"\s+", Token.Text.Whitespace),
# Multi-line comments. These are non-nestable.
(r"%\{.*?%\}", Token.Comment.Multiline),
# Simple comments.
(r"%.*?$", Token.Comment.Single),
# End of embedded LilyPond in Scheme.
(r"#\}", Token.Punctuation, "#pop"),
# Embedded Scheme, starting with # ("delayed"),
# or $ (immediate). #@ and and $@ are the lesser known
# "list splicing operators".
(r"[#$]@?", Token.Punctuation, "value"),
# Any kind of punctuation:
# - sequential music: { },
# - parallel music: << >>,
# - voice separator: << \\ >>,
# - chord: < >,
# - bar check: |,
# - dot in nested properties: \revert NoteHead.color,
# - equals sign in assignments and lists for various commands:
# \override Stem.color = red,
# - comma as alternative syntax for lists: \time 3,3,2 4/4,
# - colon in tremolos: c:32,
# - double hyphen and underscore in lyrics: li -- ly -- pond __
# (which must be preceded by ASCII whitespace)
(r"""(?x)
\\\\
| (?<= \s ) (?: -- | __ )
| [{}<>=.,:|]
""", Token.Punctuation),
# Pitches, with optional octavation marks, octave check,
# and forced or cautionary accidental.
(words(pitches, suffix=r"=?[',]*!?\??" + NAME_END_RE), Token.Pitch),
# Strings, optionally with direction specifier.
(r'[\-_^]?"', Token.String, "string"),
# Numbers.
(r"-?\d+\.\d+", Token.Number.Float), # 5. and .5 are not allowed
(r"-?\d+/\d+", Token.Number.Fraction),
# Integers, or durations with optional augmentation dots.
# We have no way to distinguish these, so we highlight
# them all as numbers.
#
# Normally, there is a space before the integer (being an
# argument to a music function), which we check here. The
# case without a space is handled below (as a fingering
# number).
(r"""(?x)
(?<= \s ) -\d+
| (?: (?: \d+ | \\breve | \\longa | \\maxima )
\.* )
""", Token.Number),
# Separates duration and duration multiplier highlighted as fraction.
(r"\*", Token.Number),
# Ties, slurs, manual beams.
(r"[~()[\]]", Token.Name.Builtin.Articulation),
# Predefined articulation shortcuts. A direction specifier is
# required here.
(r"[\-_^][>^_!.\-+]", Token.Name.Builtin.Articulation),
# Fingering numbers, string numbers.
(r"[\-_^]?\\?\d+", Token.Name.Builtin.Articulation),
# Builtins.
(builtin_words(keywords, "mandatory"), Token.Keyword),
(builtin_words(pitch_language_names, "disallowed"), Token.Name.PitchLanguage),
(builtin_words(clefs, "disallowed"), Token.Name.Builtin.Clef),
(builtin_words(scales, "mandatory"), Token.Name.Builtin.Scale),
(builtin_words(repeat_types, "disallowed"), Token.Name.Builtin.RepeatType),
(builtin_words(units, "mandatory"), Token.Number),
(builtin_words(chord_modifiers, "disallowed"), Token.ChordModifier),
(builtin_words(music_functions, "mandatory"), Token.Name.Builtin.MusicFunction),
(builtin_words(dynamics, "mandatory"), Token.Name.Builtin.Dynamic),
# Those like slurs that don't take a backslash are covered above.
(builtin_words(articulations, "mandatory"), Token.Name.Builtin.Articulation),
(builtin_words(music_commands, "mandatory"), Token.Name.Builtin.MusicCommand),
(builtin_words(markup_commands, "mandatory"), Token.Name.Builtin.MarkupCommand),
(builtin_words(grobs, "disallowed"), Token.Name.Builtin.Grob),
(builtin_words(translators, "disallowed"), Token.Name.Builtin.Translator),
# Optional backslash because of \layout { \context { \Score ... } }.
(builtin_words(contexts, "optional"), Token.Name.Builtin.Context),
(builtin_words(context_properties, "disallowed"), Token.Name.Builtin.ContextProperty),
(builtin_words(grob_properties, "disallowed"),
Token.Name.Builtin.GrobProperty,
"maybe-subproperties"),
# Optional backslashes here because output definitions are wrappers
# around modules. Concretely, you can do, e.g.,
# \paper { oddHeaderMarkup = \evenHeaderMarkup }
(builtin_words(paper_variables, "optional"), Token.Name.Builtin.PaperVariable),
(builtin_words(header_variables, "optional"), Token.Name.Builtin.HeaderVariable),
# Other backslashed-escaped names (like dereferencing a
# music variable), possibly with a direction specifier.
(r"[\-_^]?\\.+?" + NAME_END_RE, Token.Name.BackslashReference),
# Definition of a variable. Support assignments to alist keys
# (myAlist.my-key.my-nested-key = \markup \spam \eggs).
(r"""(?x)
(?: [^\W\d] | - )+
(?= (?: [^\W\d] | [\-.] )* \s* = )
""", Token.Name.Lvalue),
# Virtually everything can appear in markup mode, so we highlight
# as text. Try to get a complete word, or we might wrongly lex
# a suffix that happens to be a builtin as a builtin (e.g., "myStaff").
(r"([^\W\d]|-)+?" + NAME_END_RE, Token.Text),
(r".", Token.Text),
],
"string": [
(r'"', Token.String, "#pop"),
(r'\\.', Token.String.Escape),
(r'[^\\"]+', Token.String),
],
"value": [
# Scan a LilyPond value, then pop back since we had a
# complete expression.
(r"#\{", Token.Punctuation, ("#pop", "root")),
inherit,
],
# Grob subproperties are undeclared and it would be tedious
# to maintain them by hand. Instead, this state allows recognizing
# everything that looks like a-known-property.foo.bar-baz as
# one single property name.
"maybe-subproperties": [
(r"\s+", Token.Text.Whitespace),
(r"(\.)((?:[^\W\d]|-)+?)" + NAME_END_RE,
bygroups(Token.Punctuation, Token.Name.Builtin.GrobProperty)),
default("#pop"),
]
}
|
