1 %{ // -*- mode: c++; c-file-style: "linux"; indent-tabs-mode: t -*-
3 This file is part of LilyPond, the GNU music typesetter.
5 Copyright (C) 1996--2012 Han-Wen Nienhuys <hanwen@xs4all.nl>
6 Jan Nieuwenhuizen <janneke@gnu.org>
8 LilyPond is free software: you can redistribute it and/or modify
9 it under the terms of the GNU General Public License as published by
10 the Free Software Foundation, either version 3 of the License, or
11 (at your option) any later version.
13 LilyPond is distributed in the hope that it will be useful,
14 but WITHOUT ANY WARRANTY; without even the implied warranty of
15 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 GNU General Public License for more details.
18 You should have received a copy of the GNU General Public License
19 along with LilyPond. If not, see <http://www.gnu.org/licenses/>.
22 /* Mode and indentation are at best a rough approximation based on TAB
23 * formatting (reasonable for compatibility with unspecific editor
24 * modes as Flex modes are hard to find) and need manual correction
25 * frequently. Without a reasonably dependable way of formatting a
26 * Flex file sensibly, there is little point in trying to fix the
27 * inconsistent state of indentation.
33 after making a change to the lexer rules, run
34 flex -b <this lexer file>
37 contains no backup states, but only the reminder
38 Compressed tables always back up.
39 (don-t forget to rm lex.yy.cc :-)
48 /* Flex >= 2.5.29 fix; FlexLexer.h's multiple include bracing breaks
49 when building the actual lexer. */
56 #include "context-def.hh"
57 #include "duration.hh"
58 #include "international.hh"
59 #include "interval.hh"
60 #include "lily-guile.hh"
61 #include "lily-lexer.hh"
62 #include "lily-parser.hh"
63 #include "lilypond-version.hh"
66 #include "music-function.hh"
67 #include "parse-scm.hh"
70 #include "source-file.hh"
71 #include "std-string.hh"
72 #include "string-convert.hh"
79 #define isatty HORRIBLEKLUDGE
81 void strip_trailing_white (string&);
82 void strip_leading_white (string&);
83 string lyric_fudge (string s);
84 SCM lookup_markup_command (string s);
85 SCM lookup_markup_list_command (string s);
86 bool is_valid_version (string s);
89 #define start_quote() do { \
90 yy_push_state (quote); \
94 #define yylval (*lexval_)
96 #define yylloc (*lexloc_)
98 #define YY_USER_ACTION add_lexed_char (YYLeng ());
101 SCM scan_fraction (string);
102 SCM (* scm_parse_error_handler) (void *);
112 %option yyclass="Lily_lexer"
114 %option never-interactive
131 /* The strategy concerning multibyte characters is to accept them but
132 * call YYText_utf8 for patterns that might contain them, in order to
133 * get a single code path responsible for flagging non-UTF-8 input:
134 * Patterns for accepting only valid UTF-8 without backing up are
135 * really hard to do and complex, and if nice error messages are
136 * wanted, one would need patterns catching the invalid input as well.
138 * Since editors and operating environments don't necessarily behave
139 * reasonably in the presence of mixed encodings, we flag encoding
140 * errors also in identifiers, comments, and strings where it would be
141 * conceivable to just transparently work with the byte string. But
142 * the whole point of caring about UTF-8 in here at all is too avoid
143 * stranger errors later when input passes into backends or log files
144 * or console output or error messages.
151 WORD {A}([-_]{A}|{A})*
153 /* SPECIAL category is for every letter that needs to get passed to
154 * the parser rather than being redefinable by the user */
155 SPECIAL [-+*/=<>{}!?_^'',.:]
161 REAL ({INT}\.{N}*)|(-?\.{N}+)
162 STRICTREAL {UNSIGNED}\.{UNSIGNED}
164 HORIZONTALWHITE [ \t]
170 BOM_UTF8 \357\273\277
176 // swallow and ignore carriage returns
179 <extratoken>{ANY_CHAR} {
180 /* Generate a token without swallowing anything */
182 /* First unswallow the eaten character */
183 add_lexed_char (-YYLeng ());
186 /* produce requested token */
187 int type = scm_to_int (scm_caar (extra_tokens_));
188 yylval = scm_cdar (extra_tokens_);
189 extra_tokens_ = scm_cdr (extra_tokens_);
190 if (scm_is_null (extra_tokens_))
196 <extratoken><<EOF>> {
197 /* Generate a token without swallowing anything */
199 /* produce requested token */
200 int type = scm_to_int (scm_caar (extra_tokens_));
201 yylval = scm_cdar (extra_tokens_);
202 extra_tokens_ = scm_cdr (extra_tokens_);
203 if (scm_is_null (extra_tokens_))
209 /* Use the trailing context feature. Otherwise, the BOM will not be
210 found if the file starts with an identifier definition. */
211 <INITIAL,chords,lyrics,figures,notes>{BOM_UTF8}/.* {
212 if (this->lexloc_->line_number () != 1 || this->lexloc_->column_number () != 0)
214 LexerWarning (_ ("stray UTF-8 BOM encountered").c_str ());
217 debug_output (_ ("Skipping UTF-8 BOM"));
220 <INITIAL,chords,figures,incl,lyrics,markup,notes>{
222 yy_push_state (longcomment);
224 %[^{\n\r][^\n\r]*[\n\r]? {
225 (void) YYText_utf8 ();
234 <INITIAL,notes,figures,chords,markup>{
240 <INITIAL,chords,lyrics,notes,figures>\\version{WHITE}* {
241 yy_push_state (version);
243 <INITIAL,chords,lyrics,notes,figures>\\sourcefilename{WHITE}* {
244 yy_push_state (sourcefilename);
246 <INITIAL,chords,lyrics,notes,figures>\\sourcefileline{WHITE}* {
247 yy_push_state (sourcefileline);
249 <version>\"[^""]*\" { /* got the version number */
250 string s (YYText_utf8 () + 1);
251 s = s.substr (0, s.rfind ('\"'));
255 SCM top_scope = scm_car (scm_last_pair (scopes_));
256 scm_module_define (top_scope, ly_symbol2scm ("version-seen"), SCM_BOOL_T);
258 if (!is_valid_version (s)) {
259 yylval = SCM_UNSPECIFIED;
263 <sourcefilename>\"[^""]*\" {
264 string s (YYText_utf8 () + 1);
265 s = s.substr (0, s.rfind ('\"'));
268 this->here_input().get_source_file ()->name_ = s;
269 message (_f ("Renaming input to: `%s'", s.c_str ()));
270 progress_indication ("\n");
271 scm_module_define (scm_car (scopes_),
272 ly_symbol2scm ("input-file-name"),
277 <sourcefileline>{INT} {
279 sscanf (YYText (), "%d", &i);
282 this->here_input ().get_source_file ()->set_line (here_input ().start (), i);
285 <version>{ANY_CHAR} {
286 LexerError (_ ("quoted string expected after \\version").c_str ());
289 <sourcefilename>{ANY_CHAR} {
290 LexerError (_ ("quoted string expected after \\sourcefilename").c_str ());
293 <sourcefileline>{ANY_CHAR} {
294 LexerError (_ ("integer expected after \\sourcefileline").c_str ());
299 (void) YYText_utf8 ();
302 (void) YYText_utf8 ();
310 <INITIAL,chords,lyrics,notes,figures>\\maininput {
314 main_input_level_ = include_stack_.size ();
315 is_main_input_ = true;
317 yy_push_state (maininput);
318 yy_push_state (state);
321 LexerError (_ ("\\maininput not allowed outside init files").c_str ());
324 <INITIAL,chords,lyrics,figures,notes>\\include {
325 yy_push_state (incl);
327 <incl>\"[^""]*\" { /* got the include file name */
328 string s (YYText_utf8 ()+1);
329 s = s.substr (0, s.rfind ('"'));
331 new_input (s, sources_);
334 <incl>\\{BLACK}*{WHITE}? { /* got the include identifier */
335 string s = YYText_utf8 () + 1;
336 strip_trailing_white (s);
337 if (s.length () && (s[s.length () - 1] == ';'))
338 s = s.substr (0, s.length () - 1);
340 SCM sid = lookup_identifier (s);
341 if (scm_is_string (sid)) {
342 new_input (ly_scm2string (sid), sources_);
345 string msg (_f ("wrong or undefined identifier: `%s'", s ));
347 LexerError (msg.c_str ());
348 SCM err = scm_current_error_port ();
349 scm_puts ("This value was found in the table: ", err);
350 scm_display (sid, err);
353 <incl>(\$|#) { // scm for the filename
355 Input hi = here_input();
357 SCM sval = ly_parse_scm (hi.start (), &n, hi,
358 be_safe_global && is_main_input_, parser_);
359 sval = eval_scm (sval);
361 for (int i = 0; i < n; i++)
365 char_count_stack_.back () += n;
367 if (scm_is_string (sval)) {
368 new_input (ly_scm2string (sval), sources_);
371 LexerError (_ ("string expected after \\include").c_str ());
372 if (sval != SCM_UNDEFINED) {
373 SCM err = scm_current_error_port ();
374 scm_puts ("This value was found instead: ", err);
375 scm_display (sval, err);
380 <incl,version,sourcefilename>\"[^""]* { // backup rule
381 LexerError (_ ("end quote missing").c_str ());
385 /* Flex picks the longest matching pattern including trailing
386 * contexts. Without the backup pattern, r-. does not trigger the
387 * {RESTNAME} rule but rather the {WORD}/[-_] rule coming later,
388 * needed for avoiding backup states.
391 <chords,notes,figures>{RESTNAME}/[-_] | // pseudo backup rule
392 <chords,notes,figures>{RESTNAME} {
393 char const *s = YYText ();
394 yylval = scm_from_locale_string (s);
397 <chords,notes,figures>q/[-_] | // pseudo backup rule
398 <chords,notes,figures>q {
399 yylval = SCM_UNSPECIFIED;
400 return CHORD_REPETITION;
403 <chords,notes,figures>R/[-_] | // pseudo backup rule
404 <chords,notes,figures>R {
405 yylval = SCM_UNSPECIFIED;
406 return MULTI_MEASURE_REST;
408 <INITIAL,chords,figures,lyrics,markup,notes># { //embedded scm
410 Input hi = here_input();
412 SCM sval = ly_parse_scm (hi.start (), &n, hi,
413 be_safe_global && is_main_input_, parser_);
415 if (sval == SCM_UNDEFINED)
418 for (int i = 0; i < n; i++)
422 char_count_stack_.back () += n;
428 <INITIAL,chords,figures,lyrics,markup,notes>\$ { //immediate scm
430 Input hi = here_input();
432 SCM sval = ly_parse_scm (hi.start (), &n, hi,
433 be_safe_global && is_main_input_, parser_);
435 for (int i = 0; i < n; i++)
439 char_count_stack_.back () += n;
441 sval = eval_scm (sval, '$');
443 int token = scan_scm_id (sval);
444 if (!scm_is_eq (yylval, SCM_UNSPECIFIED))
448 <INITIAL,notes,lyrics>{
450 yylval = SCM_UNSPECIFIED;
451 return DOUBLE_ANGLE_OPEN;
454 yylval = SCM_UNSPECIFIED;
455 return DOUBLE_ANGLE_CLOSE;
461 yylval = SCM_UNSPECIFIED;
465 yylval = SCM_UNSPECIFIED;
472 yylval = SCM_UNSPECIFIED;
476 yylval = SCM_UNSPECIFIED;
480 yylval = SCM_UNSPECIFIED;
484 yylval = SCM_UNSPECIFIED;
488 yylval = SCM_UNSPECIFIED;
489 return E_EXCLAMATION;
492 yylval = SCM_UNSPECIFIED;
496 yylval = SCM_UNSPECIFIED;
502 {WORD}/[-_] | // backup rule
504 return scan_bare_word (YYText_utf8 ());
507 {COMMAND}/[-_] | // backup rule
509 return scan_escaped_word (YYText_utf8 () + 1);
512 yylval = scan_fraction (YYText ());
516 yylval = scm_c_read_string (YYText ());
519 {UNSIGNED}/[/.] | // backup rule
521 yylval = scm_c_read_string (YYText ());
525 yylval = scm_c_read_string (YYText () + 1);
532 char c = escaped_char (YYText ()[1]);
533 yylval = scm_cons (scm_from_locale_stringn (&c, 1),
537 yylval = scm_cons (scm_from_locale_string (YYText_utf8 ()),
544 /* yylval is union. Must remember STRING before setting SCM*/
546 yylval = scm_string_concatenate_reverse (yylval,
553 yylval = scm_cons (scm_from_locale_string (YYText ()),
563 yylval = scan_fraction (YYText ());
567 yylval = scm_c_read_string (YYText ());
570 {UNSIGNED}/[/.] | // backup rule
572 yylval = scm_c_read_string (YYText ());
575 {COMMAND}/[-_] | // backup rule
577 return scan_escaped_word (YYText_utf8 () + 1);
580 // UTF-8 already covered by COMMAND
581 return scan_shorthand (YYText ());
583 /* Characters needed to express durations, assignments */
585 yylval = SCM_UNSPECIFIED;
588 [^$#{}\"\\ \t\n\r\f0-9]+ {
590 string s (YYText_utf8 ());
591 yylval = SCM_UNSPECIFIED;
597 yylval = ly_string2scm (s);
601 /* This should really just cover {} */
603 yylval = SCM_UNSPECIFIED;
608 {WORD}/[-_] | // backup rule
610 return scan_bare_word (YYText_utf8 ());
612 {COMMAND}/[-_] | // backup rule
614 return scan_escaped_word (YYText_utf8 () + 1);
617 yylval = scan_fraction (YYText ());
620 {UNSIGNED}/\/ | // backup rule
622 yylval = scm_c_read_string (YYText ());
626 yylval = SCM_UNSPECIFIED;
630 yylval = SCM_UNSPECIFIED;
634 yylval = SCM_UNSPECIFIED;
638 yylval = SCM_UNSPECIFIED;
642 yylval = SCM_UNSPECIFIED;
650 yylval = SCM_UNSPECIFIED;
653 {COMMAND}/[-_] | // backup rule
655 string str (YYText_utf8 () + 1);
657 int token_type = MARKUP_FUNCTION;
658 SCM s = lookup_markup_command (str);
660 // lookup-markup-command returns a pair with the car
661 // being the function to call, and the cdr being the
662 // call signature specified to define-markup-command,
663 // a list of predicates.
665 if (!scm_is_pair (s)) {
666 // If lookup-markup-command was not successful, we
667 // try lookup-markup-list-command instead.
668 // If this fails as well, we just scan and return
670 s = lookup_markup_list_command (str);
672 token_type = MARKUP_LIST_FUNCTION;
674 return scan_escaped_word (str);
677 // If the list of predicates is, say,
678 // (number? number? markup?), then tokens
679 // EXPECT_MARKUP EXPECT_SCM EXPECT_SCM EXPECT_NO_MORE_ARGS
680 // will be generated. Note that we have to push them
681 // in reverse order, so the first token pushed in the
682 // loop will be EXPECT_NO_MORE_ARGS.
686 // yylval now contains the function to call as token
687 // value (for token type MARKUP_FUNCTION or
688 // MARKUP_LIST_FUNCTION).
690 push_extra_token(EXPECT_NO_MORE_ARGS);
692 for (; scm_is_pair(s); s = scm_cdr(s)) {
693 SCM predicate = scm_car(s);
695 if (predicate == ly_lily_module_constant ("markup-list?"))
696 push_extra_token(EXPECT_MARKUP_LIST);
697 else if (predicate == ly_lily_module_constant ("markup?"))
698 push_extra_token(EXPECT_MARKUP);
700 push_extra_token(EXPECT_SCM, predicate);
704 [^$#{}\"\\ \t\n\r\f]+ {
705 string s (YYText_utf8 ());
707 yylval = ly_string2scm (s);
711 yylval = SCM_UNSPECIFIED;
716 <longcomment><<EOF>> {
717 LexerError (_ ("EOF found inside a comment").c_str ());
722 LexerError (_ ("EOF found inside string").c_str ());
727 yylval = SCM_UNSPECIFIED;
730 is_main_input_ = include_stack_.size () > main_input_level_;
733 main_input_level_ = 0;
735 if (YYSTATE != maininput)
737 LexerError (_ ("Unfinished main input").c_str ());
740 } while (YYSTATE != maininput);
742 extra_tokens_ = SCM_EOL;
745 if (!close_input () || !is_main_input_)
746 /* Returns YY_NULL */
749 else if (!close_input ())
750 /* Returns YY_NULL */
754 <maininput>{ANY_CHAR} {
755 while (include_stack_.size () > main_input_level_
762 {WORD}/[-_] | // backup rule
764 return scan_bare_word (YYText_utf8 ());
766 {COMMAND}/[-_] | // backup rule
768 return scan_escaped_word (YYText_utf8 () + 1);
773 yylval = scan_fraction (YYText ());
777 -{UNSIGNED} | // backup rule
779 yylval = scm_c_read_string (YYText ());
783 {UNSIGNED}/\/ | // backup rule
785 yylval = scm_c_read_string (YYText ());
790 -/\. { // backup rule
791 yylval = SCM_UNSPECIFIED;
795 <INITIAL,chords,lyrics,figures,notes>{SPECIAL} {
796 yylval = SCM_UNSPECIFIED;
800 <INITIAL,chords,lyrics,figures,notes>{SHORTHAND} {
801 return scan_shorthand (YYText_utf8 ()); // should not be utf-8
805 string msg = _f ("invalid character: `%s'", YYText_utf8 ());
806 LexerError (msg.c_str ());
807 yylval = SCM_UNSPECIFIED;
808 return '%'; // Better not return half a utf8 character.
813 /* Make the lexer generate a token of the given type as the next token.
814 TODO: make it possible to define a value for the token as well */
816 Lily_lexer::push_extra_token (int token_type, SCM scm)
818 if (scm_is_null (extra_tokens_))
820 if (YY_START != extratoken)
821 hidden_state_ = YY_START;
822 yy_push_state (extratoken);
824 extra_tokens_ = scm_acons (scm_from_int (token_type), scm, extra_tokens_);
828 Lily_lexer::push_chord_state (SCM alist)
830 SCM p = scm_assq (alist, pitchname_tab_stack_);
832 if (scm_is_false (p))
833 p = scm_cons (alist, alist_to_hashq (alist));
834 pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_);
835 yy_push_state (chords);
839 Lily_lexer::push_figuredbass_state ()
841 yy_push_state (figures);
845 Lily_lexer::push_initial_state ()
847 yy_push_state (INITIAL);
851 Lily_lexer::push_lyric_state ()
853 yy_push_state (lyrics);
857 Lily_lexer::push_markup_state ()
859 yy_push_state (markup);
863 Lily_lexer::push_note_state (SCM alist)
865 bool extra = (YYSTATE == extratoken);
867 SCM p = scm_assq (alist, pitchname_tab_stack_);
872 if (scm_is_false (p))
873 p = scm_cons (alist, alist_to_hashq (alist));
874 pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_);
875 yy_push_state (notes);
878 hidden_state_ = YYSTATE;
879 yy_push_state (extratoken);
884 Lily_lexer::pop_state ()
886 bool extra = (YYSTATE == extratoken);
891 if (YYSTATE == notes || YYSTATE == chords)
892 pitchname_tab_stack_ = scm_cdr (pitchname_tab_stack_);
894 // don't cross the maininput threshold
895 if (YYSTATE != maininput)
899 hidden_state_ = YYSTATE;
900 yy_push_state (extratoken);
905 Lily_lexer::identifier_type (SCM sid)
907 int k = try_special_identifiers (&yylval , sid);
908 return k >= 0 ? k : SCM_IDENTIFIER;
913 Lily_lexer::scan_escaped_word (const string &str)
915 // use more SCM for this.
917 // SCM sym = ly_symbol2scm (str.c_str ());
919 yylval = SCM_UNSPECIFIED;
920 int i = lookup_keyword (str);
925 SCM sid = lookup_identifier (str);
926 if (Music *m = unsmob_music (sid))
928 m->set_spot (override_input (last_input_));
931 if (sid != SCM_UNDEFINED)
932 return scan_scm_id (sid);
934 string msg (_f ("unknown escaped string: `\\%s'", str));
935 LexerError (msg.c_str ());
937 yylval = ly_string2scm (str);
943 Lily_lexer::scan_shorthand (const string &str)
945 SCM sid = lookup_identifier (str);
946 if (Music *m = unsmob_music (sid))
948 m->set_spot (override_input (last_input_));
951 if (sid != SCM_UNDEFINED)
952 return scan_scm_id (sid);
954 string msg (_f ("undefined character or shorthand: %s", str));
955 LexerError (msg.c_str ());
957 yylval = ly_string2scm (str);
963 Lily_lexer::scan_scm_id (SCM sid)
965 if (is_music_function (sid))
967 int funtype = SCM_FUNCTION;
971 SCM s = get_music_function_signature (sid);
972 SCM cs = scm_car (s);
974 if (scm_is_pair (cs))
979 if (scm_is_eq (cs, ly_lily_module_constant ("ly:music?")))
980 funtype = MUSIC_FUNCTION;
981 else if (scm_is_eq (cs, ly_lily_module_constant ("ly:event?")))
982 funtype = EVENT_FUNCTION;
983 else if (ly_is_procedure (cs))
984 funtype = SCM_FUNCTION;
985 else programming_error ("Bad syntax function predicate");
987 push_extra_token (EXPECT_NO_MORE_ARGS);
988 for (s = scm_cdr (s); scm_is_pair (s); s = scm_cdr (s))
990 SCM optional = SCM_UNDEFINED;
993 if (scm_is_pair (cs))
995 optional = SCM_CDR (cs);
999 if (cs == Pitch_type_p_proc)
1000 push_extra_token (EXPECT_PITCH);
1001 else if (cs == Duration_type_p_proc)
1002 push_extra_token (EXPECT_DURATION);
1003 else if (ly_is_procedure (cs))
1004 push_extra_token (EXPECT_SCM, cs);
1007 programming_error ("Function parameter without type-checking predicate");
1010 if (!scm_is_eq (optional, SCM_UNDEFINED))
1011 push_extra_token (EXPECT_OPTIONAL, optional);
1016 return identifier_type (sid);
1020 Lily_lexer::scan_bare_word (const string &str)
1022 SCM sym = ly_symbol2scm (str.c_str ());
1023 if ((YYSTATE == notes) || (YYSTATE == chords)) {
1024 SCM handle = SCM_BOOL_F;
1025 if (scm_is_pair (pitchname_tab_stack_))
1026 handle = scm_hashq_get_handle (scm_cdar (pitchname_tab_stack_), sym);
1028 if (scm_is_pair (handle)) {
1029 yylval = scm_cdr (handle);
1030 if (unsmob_pitch (yylval))
1031 return (YYSTATE == notes) ? NOTENAME_PITCH : TONICNAME_PITCH;
1032 else if (scm_is_symbol (yylval))
1035 else if ((YYSTATE == chords)
1036 && (handle = scm_hashq_get_handle (chordmodifier_tab_, sym))!= SCM_BOOL_F)
1038 yylval = scm_cdr (handle);
1039 return CHORD_MODIFIER;
1042 yylval = ly_string2scm (str);
1047 Lily_lexer::get_state () const
1049 if (YY_START == extratoken)
1050 return hidden_state_;
1056 Lily_lexer::is_note_state () const
1058 return get_state () == notes;
1062 Lily_lexer::is_chord_state () const
1064 return get_state () == chords;
1068 Lily_lexer::is_lyric_state () const
1070 return get_state () == lyrics;
1074 Lily_lexer::is_figure_state () const
1076 return get_state () == figures;
1079 // The extra_token parameter specifies how to convert multiple values
1080 // into additional tokens. For '#', additional values get pushed as
1081 // SCM_IDENTIFIER. For '$', they get checked for their type and get
1082 // pushed as a corresponding *_IDENTIFIER token. Since the latter
1083 // tampers with yylval, it can only be done from the lexer itself, so
1084 // this function is private.
1087 Lily_lexer::eval_scm (SCM readerdata, char extra_token)
1089 SCM sval = SCM_UNDEFINED;
1091 if (!SCM_UNBNDP (readerdata))
1093 sval = ly_eval_scm (scm_car (readerdata),
1094 *unsmob_input (scm_cdr (readerdata)),
1095 be_safe_global && is_main_input_,
1099 if (SCM_UNBNDP (sval))
1102 return SCM_UNSPECIFIED;
1105 if (extra_token && SCM_VALUESP (sval))
1107 sval = scm_struct_ref (sval, SCM_INUM0);
1109 if (scm_is_pair (sval)) {
1110 for (SCM p = scm_reverse (scm_cdr (sval));
1114 SCM v = scm_car (p);
1115 if (Music *m = unsmob_music (v))
1117 if (!unsmob_input (m->get_property ("origin")))
1118 m->set_spot (override_input (last_input_));
1122 switch (extra_token) {
1124 token = scan_scm_id (v);
1125 if (!scm_is_eq (yylval, SCM_UNSPECIFIED))
1126 push_extra_token (token, yylval);
1129 push_extra_token (SCM_IDENTIFIER, v);
1133 sval = scm_car (sval);
1135 sval = SCM_UNSPECIFIED;
1138 if (Music *m = unsmob_music (sval))
1140 if (!unsmob_input (m->get_property ("origin")))
1141 m->set_spot (override_input (last_input_));
1147 /* Check for valid UTF-8 that has no overlong or surrogate codes and
1148 is in the range 0-0x10ffff */
1151 Lily_lexer::YYText_utf8 ()
1153 const char * const p = YYText ();
1154 for (int i=0; p[i];) {
1155 if ((p[i] & 0xff) < 0x80) {
1159 int oldi = i; // start of character
1160 int more = 0; // # of followup bytes, 0 if bad
1161 switch (p[i++] & 0xff) {
1162 // 0xc0 and 0xc1 are overlong prefixes for
1163 // 0x00-0x3f and 0x40-0x7f respectively, bad.
1164 case 0xc2: // 0x80-0xbf
1165 case 0xc3: // 0xc0-0xff
1166 case 0xc4: // 0x100-0x13f
1167 case 0xc5: // 0x140-0x17f
1168 case 0xc6: // 0x180-0x1bf
1169 case 0xc7: // 0x1c0-0x1ff
1170 case 0xc8: // 0x200-0x23f
1171 case 0xc9: // 0x240-0x27f
1172 case 0xca: // 0x280-0x2bf
1173 case 0xcb: // 0x2c0-0x2ff
1174 case 0xcc: // 0x300-0x33f
1175 case 0xcd: // 0x340-0x37f
1176 case 0xce: // 0x380-0x3bf
1177 case 0xcf: // 0x3c0-0x3ff
1178 case 0xd0: // 0x400-0x43f
1179 case 0xd1: // 0x440-0x47f
1180 case 0xd2: // 0x480-0x4bf
1181 case 0xd3: // 0x4c0-0x4ff
1182 case 0xd4: // 0x500-0x53f
1183 case 0xd5: // 0x540-0x57f
1184 case 0xd6: // 0x580-0x5bf
1185 case 0xd7: // 0x5c0-0x5ff
1186 case 0xd8: // 0x600-0x63f
1187 case 0xd9: // 0x640-0x67f
1188 case 0xda: // 0x680-0x6bf
1189 case 0xdb: // 0x6c0-0x6ff
1190 case 0xdc: // 0x700-0x73f
1191 case 0xdd: // 0x740-0x77f
1192 case 0xde: // 0x780-0x7bf
1193 case 0xdf: // 0x7c0-0x7ff
1194 more = 1; // 2-byte sequences, 0x80-0x7ff
1197 // don't allow overlong sequences for 0-0x7ff
1198 if ((p[i] & 0xff) < 0xa0)
1200 case 0xe1: // 0x1000-0x1fff
1201 case 0xe2: // 0x2000-0x2fff
1202 case 0xe3: // 0x3000-0x3fff
1203 case 0xe4: // 0x4000-0x4fff
1204 case 0xe5: // 0x5000-0x5fff
1205 case 0xe6: // 0x6000-0x6fff
1206 case 0xe7: // 0x7000-0x7fff
1207 case 0xe8: // 0x8000-0x8fff
1208 case 0xe9: // 0x9000-0x9fff
1209 case 0xea: // 0xa000-0xafff
1210 case 0xeb: // 0xb000-0xbfff
1211 case 0xec: // 0xc000-0xcfff
1212 more = 2; // 3-byte sequences, 0x7ff-0xcfff
1214 case 0xed: // 0xd000-0xdfff
1215 // Don't allow surrogate codes 0xd800-0xdfff
1216 if ((p[i] & 0xff) >= 0xa0)
1218 case 0xee: // 0xe000-0xefff
1219 case 0xef: // 0xf000-0xffff
1220 more = 2; // 3-byte sequences,
1221 // 0xd000-0xd7ff, 0xe000-0xffff
1224 // don't allow overlong sequences for 0-0xffff
1225 if ((p[i] & 0xff) < 0x90)
1227 case 0xf1: // 0x40000-0x7ffff
1228 case 0xf2: // 0x80000-0xbffff
1229 case 0xf3: // 0xc0000-0xfffff
1230 more = 3; // 4-byte sequences, 0x10000-0xfffff
1233 // don't allow more than 0x10ffff
1234 if ((p[i] & 0xff) >= 0x90)
1236 more = 3; // 4-byte sequence, 0x100000-0x10ffff
1240 // check that all continuation bytes are valid
1242 if ((p[i++] & 0xc0) != 0x80)
1248 Input h = here_input ();
1249 h.set (h.get_source_file (), h.start () + oldi, h.start () + i);
1250 h.warning (_ ("non-UTF-8 input").c_str ());
1257 urg, belong to string (_convert)
1258 and should be generalised
1261 strip_leading_white (string&s)
1264 for (; i < s.length (); i++)
1265 if (!isspace (s[i]))
1272 strip_trailing_white (string&s)
1274 ssize i = s.length ();
1276 if (!isspace (s[i]))
1279 s = s.substr (0, i + 1);
1284 Lilypond_version oldest_version ("2.7.38");
1288 is_valid_version (string s)
1290 Lilypond_version current ( MAJOR_VERSION "." MINOR_VERSION "." PATCH_LEVEL );
1291 Lilypond_version ver (s);
1292 if (int (ver) < oldest_version)
1294 non_fatal_error (_f ("file too old: %s (oldest supported: %s)", ver.to_string (), oldest_version.to_string ()));
1295 non_fatal_error (_ ("consider updating the input with the convert-ly script"));
1301 non_fatal_error (_f ("program too old: %s (file requires: %s)", current.to_string (), ver.to_string ()));
1312 lyric_fudge (string s)
1316 while ((i = s.find ('_', i)) != string::npos)
1324 Convert "NUM/DEN" into a '(NUM . DEN) cons.
1327 scan_fraction (string frac)
1329 ssize i = frac.find ('/');
1330 string left = frac.substr (0, i);
1331 string right = frac.substr (i + 1, (frac.length () - i + 1));
1333 int n = String_convert::dec2int (left);
1334 int d = String_convert::dec2int (right);
1335 return scm_cons (scm_from_int (n), scm_from_int (d));
1339 lookup_markup_command (string s)
1341 SCM proc = ly_lily_module_constant ("lookup-markup-command");
1342 return scm_call_1 (proc, ly_string2scm (s));
1346 lookup_markup_list_command (string s)
1348 SCM proc = ly_lily_module_constant ("lookup-markup-list-command");
1349 return scm_call_1 (proc, ly_string2scm (s));
1352 /* Shut up lexer warnings. */
1372 silence_lexer_warnings ()
1374 (void) yy_start_stack_ptr;
1375 (void) yy_start_stack_depth;
1376 (void) yy_start_stack;
1377 (void) yy_push_state;
1378 (void) yy_pop_state;
1379 (void) yy_top_state;
1380 (void) silence_lexer_warnings;