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); \
95 The inside of \"violin1" is marked by commandquote mode
98 #define start_command_quote() do { \
99 yy_push_state (commandquote); \
103 #define yylval (*lexval_)
105 #define yylloc (*lexloc_)
107 #define YY_USER_ACTION add_lexed_char (YYLeng ());
110 SCM scan_fraction (string);
111 SCM (* scm_parse_error_handler) (void *);
121 %option yyclass="Lily_lexer"
123 %option never-interactive
141 /* The strategy concerning multibyte characters is to accept them but
142 * call YYText_utf8 for patterns that might contain them, in order to
143 * get a single code path responsible for flagging non-UTF-8 input:
144 * Patterns for accepting only valid UTF-8 without backing up are
145 * really hard to do and complex, and if nice error messages are
146 * wanted, one would need patterns catching the invalid input as well.
148 * Since editors and operating environments don't necessarily behave
149 * reasonably in the presence of mixed encodings, we flag encoding
150 * errors also in identifiers, comments, and strings where it would be
151 * conceivable to just transparently work with the byte string. But
152 * the whole point of caring about UTF-8 in here at all is too avoid
153 * stranger errors later when input passes into backends or log files
154 * or console output or error messages.
161 WORD {A}([-_]{A}|{A})*
163 /* SPECIAL category is for every letter that needs to get passed to
164 * the parser rather than being redefinable by the user */
165 SPECIAL [-+*/=<>{}!?_^'',.:]
171 REAL ({INT}\.{N}*)|(-?\.{N}+)
172 STRICTREAL {UNSIGNED}\.{UNSIGNED}
174 HORIZONTALWHITE [ \t]
180 BOM_UTF8 \357\273\277
186 // swallow and ignore carriage returns
189 <extratoken>{ANY_CHAR} {
190 /* Generate a token without swallowing anything */
192 /* First unswallow the eaten character */
193 add_lexed_char (-YYLeng ());
196 /* produce requested token */
197 int type = scm_to_int (scm_caar (extra_tokens_));
198 yylval = scm_cdar (extra_tokens_);
199 extra_tokens_ = scm_cdr (extra_tokens_);
200 if (scm_is_null (extra_tokens_))
206 <extratoken><<EOF>> {
207 /* Generate a token without swallowing anything */
209 /* produce requested token */
210 int type = scm_to_int (scm_caar (extra_tokens_));
211 yylval = scm_cdar (extra_tokens_);
212 extra_tokens_ = scm_cdr (extra_tokens_);
213 if (scm_is_null (extra_tokens_))
219 /* Use the trailing context feature. Otherwise, the BOM will not be
220 found if the file starts with an identifier definition. */
221 <INITIAL,chords,lyrics,figures,notes>{BOM_UTF8}/.* {
222 if (this->lexloc_->line_number () != 1 || this->lexloc_->column_number () != 0)
224 LexerWarning (_ ("stray UTF-8 BOM encountered").c_str ());
227 debug_output (_ ("Skipping UTF-8 BOM"));
230 <INITIAL,chords,figures,incl,lyrics,markup,notes>{
232 yy_push_state (longcomment);
234 %[^{\n\r][^\n\r]*[\n\r]? {
235 (void) YYText_utf8 ();
244 <INITIAL,notes,figures,chords,markup>{
250 <INITIAL,chords,lyrics,notes,figures>\\version{WHITE}* {
251 yy_push_state (version);
253 <INITIAL,chords,lyrics,notes,figures>\\sourcefilename{WHITE}* {
254 yy_push_state (sourcefilename);
256 <INITIAL,chords,lyrics,notes,figures>\\sourcefileline{WHITE}* {
257 yy_push_state (sourcefileline);
259 <version>\"[^""]*\" { /* got the version number */
260 string s (YYText_utf8 () + 1);
261 s = s.substr (0, s.rfind ('\"'));
265 SCM top_scope = scm_car (scm_last_pair (scopes_));
266 scm_module_define (top_scope, ly_symbol2scm ("version-seen"), SCM_BOOL_T);
268 if (!is_valid_version (s)) {
269 yylval = SCM_UNSPECIFIED;
273 <sourcefilename>\"[^""]*\" {
274 string s (YYText_utf8 () + 1);
275 s = s.substr (0, s.rfind ('\"'));
278 this->here_input().get_source_file ()->name_ = s;
279 message (_f ("Renaming input to: `%s'", s.c_str ()));
280 progress_indication ("\n");
281 scm_module_define (scm_car (scopes_),
282 ly_symbol2scm ("input-file-name"),
287 <sourcefileline>{INT} {
289 sscanf (YYText (), "%d", &i);
292 this->here_input ().get_source_file ()->set_line (here_input ().start (), i);
295 <version>{ANY_CHAR} {
296 LexerError (_ ("quoted string expected after \\version").c_str ());
299 <sourcefilename>{ANY_CHAR} {
300 LexerError (_ ("quoted string expected after \\sourcefilename").c_str ());
303 <sourcefileline>{ANY_CHAR} {
304 LexerError (_ ("integer expected after \\sourcefileline").c_str ());
309 (void) YYText_utf8 ();
312 (void) YYText_utf8 ();
320 <INITIAL,chords,lyrics,notes,figures>\\maininput {
324 main_input_level_ = include_stack_.size ();
325 is_main_input_ = true;
327 yy_push_state (maininput);
328 yy_push_state (state);
331 LexerError (_ ("\\maininput not allowed outside init files").c_str ());
334 <INITIAL,chords,lyrics,figures,notes>\\include {
335 yy_push_state (incl);
337 <incl>\"[^""]*\" { /* got the include file name */
338 string s (YYText_utf8 ()+1);
339 s = s.substr (0, s.rfind ('"'));
341 new_input (s, sources_);
344 <incl>\\{BLACK}*{WHITE}? { /* got the include identifier */
345 string s = YYText_utf8 () + 1;
346 strip_trailing_white (s);
347 if (s.length () && (s[s.length () - 1] == ';'))
348 s = s.substr (0, s.length () - 1);
350 SCM sid = lookup_identifier (s);
351 if (scm_is_string (sid)) {
352 new_input (ly_scm2string (sid), sources_);
355 string msg (_f ("wrong or undefined identifier: `%s'", s ));
357 LexerError (msg.c_str ());
358 SCM err = scm_current_error_port ();
359 scm_puts ("This value was found in the table: ", err);
360 scm_display (sid, err);
363 <incl>(\$|#) { // scm for the filename
365 Input hi = here_input();
367 SCM sval = ly_parse_scm (hi.start (), &n, hi,
368 be_safe_global && is_main_input_, parser_);
369 sval = eval_scm (sval);
371 for (int i = 0; i < n; i++)
375 char_count_stack_.back () += n;
377 if (scm_is_string (sval)) {
378 new_input (ly_scm2string (sval), sources_);
381 LexerError (_ ("string expected after \\include").c_str ());
382 if (sval != SCM_UNDEFINED) {
383 SCM err = scm_current_error_port ();
384 scm_puts ("This value was found instead: ", err);
385 scm_display (sval, err);
390 <incl,version,sourcefilename>\"[^""]* { // backup rule
391 LexerError (_ ("end quote missing").c_str ());
395 /* Flex picks the longest matching pattern including trailing
396 * contexts. Without the backup pattern, r-. does not trigger the
397 * {RESTNAME} rule but rather the {WORD}/[-_] rule coming later,
398 * needed for avoiding backup states.
401 <chords,notes,figures>{RESTNAME}/[-_] | // pseudo backup rule
402 <chords,notes,figures>{RESTNAME} {
403 char const *s = YYText ();
404 yylval = scm_from_locale_string (s);
407 <chords,notes,figures>q/[-_] | // pseudo backup rule
408 <chords,notes,figures>q {
409 yylval = SCM_UNSPECIFIED;
410 return CHORD_REPETITION;
413 <chords,notes,figures>R/[-_] | // pseudo backup rule
414 <chords,notes,figures>R {
415 yylval = SCM_UNSPECIFIED;
416 return MULTI_MEASURE_REST;
418 <INITIAL,chords,figures,lyrics,markup,notes># { //embedded scm
420 Input hi = here_input();
422 SCM sval = ly_parse_scm (hi.start (), &n, hi,
423 be_safe_global && is_main_input_, parser_);
425 if (sval == SCM_UNDEFINED)
428 for (int i = 0; i < n; i++)
432 char_count_stack_.back () += n;
438 <INITIAL,chords,figures,lyrics,markup,notes>\$ { //immediate scm
440 Input hi = here_input();
442 SCM sval = ly_parse_scm (hi.start (), &n, hi,
443 be_safe_global && is_main_input_, parser_);
445 for (int i = 0; i < n; i++)
449 char_count_stack_.back () += n;
451 sval = eval_scm (sval, '$');
453 int token = scan_scm_id (sval);
454 if (!scm_is_eq (yylval, SCM_UNSPECIFIED))
458 <INITIAL,notes,lyrics>{
460 yylval = SCM_UNSPECIFIED;
461 return DOUBLE_ANGLE_OPEN;
464 yylval = SCM_UNSPECIFIED;
465 return DOUBLE_ANGLE_CLOSE;
471 yylval = SCM_UNSPECIFIED;
475 yylval = SCM_UNSPECIFIED;
482 yylval = SCM_UNSPECIFIED;
486 yylval = SCM_UNSPECIFIED;
490 yylval = SCM_UNSPECIFIED;
494 yylval = SCM_UNSPECIFIED;
498 yylval = SCM_UNSPECIFIED;
499 return E_EXCLAMATION;
502 yylval = SCM_UNSPECIFIED;
506 yylval = SCM_UNSPECIFIED;
512 {WORD}/[-_] | // backup rule
514 return scan_bare_word (YYText_utf8 ());
517 start_command_quote ();
519 {COMMAND}/[-_] | // backup rule
521 return scan_escaped_word (YYText_utf8 () + 1);
524 yylval = scan_fraction (YYText ());
528 yylval = scm_c_read_string (YYText ());
531 {UNSIGNED}/[/.] | // backup rule
533 yylval = scm_c_read_string (YYText ());
537 yylval = scm_c_read_string (YYText () + 1);
542 <quote,commandquote>{
544 char c = escaped_char (YYText ()[1]);
545 yylval = scm_cons (scm_from_locale_stringn (&c, 1),
549 yylval = scm_cons (scm_from_locale_string (YYText_utf8 ()),
554 /* yylval is union. Must remember STRING before setting SCM*/
556 yylval = scm_string_concatenate_reverse (yylval,
560 if (get_state () == commandquote) {
562 return scan_escaped_word (ly_scm2string (yylval));
570 yylval = scm_cons (scm_from_locale_string (YYText ()),
580 yylval = scan_fraction (YYText ());
584 yylval = scm_c_read_string (YYText ());
587 {UNSIGNED}/[/.] | // backup rule
589 yylval = scm_c_read_string (YYText ());
593 start_command_quote ();
595 {COMMAND}/[-_] | // backup rule
597 return scan_escaped_word (YYText_utf8 () + 1);
600 // UTF-8 already covered by COMMAND
601 return scan_shorthand (YYText ());
603 /* Characters needed to express durations, assignments */
605 yylval = SCM_UNSPECIFIED;
608 [^|*.=$#{}\"\\ \t\n\r\f0-9][^$#{}\"\\ \t\n\r\f0-9]* {
610 string s (YYText_utf8 ());
611 yylval = SCM_UNSPECIFIED;
617 yylval = ly_string2scm (s);
621 /* This should really just cover {} */
623 yylval = SCM_UNSPECIFIED;
628 {WORD}/[-_] | // backup rule
630 return scan_bare_word (YYText_utf8 ());
633 start_command_quote ();
635 {COMMAND}/[-_] | // backup rule
637 return scan_escaped_word (YYText_utf8 () + 1);
640 yylval = scan_fraction (YYText ());
643 {UNSIGNED}/\/ | // backup rule
645 yylval = scm_c_read_string (YYText ());
649 yylval = SCM_UNSPECIFIED;
653 yylval = SCM_UNSPECIFIED;
657 yylval = SCM_UNSPECIFIED;
661 yylval = SCM_UNSPECIFIED;
665 yylval = SCM_UNSPECIFIED;
673 yylval = SCM_UNSPECIFIED;
677 start_command_quote ();
679 {COMMAND}/[-_] | // backup rule
681 string str (YYText_utf8 () + 1);
683 int token_type = MARKUP_FUNCTION;
684 SCM s = lookup_markup_command (str);
686 // lookup-markup-command returns a pair with the car
687 // being the function to call, and the cdr being the
688 // call signature specified to define-markup-command,
689 // a list of predicates.
691 if (!scm_is_pair (s)) {
692 // If lookup-markup-command was not successful, we
693 // try lookup-markup-list-command instead.
694 // If this fails as well, we just scan and return
696 s = lookup_markup_list_command (str);
698 token_type = MARKUP_LIST_FUNCTION;
700 return scan_escaped_word (str);
703 // If the list of predicates is, say,
704 // (number? number? markup?), then tokens
705 // EXPECT_MARKUP EXPECT_SCM EXPECT_SCM EXPECT_NO_MORE_ARGS
706 // will be generated. Note that we have to push them
707 // in reverse order, so the first token pushed in the
708 // loop will be EXPECT_NO_MORE_ARGS.
712 // yylval now contains the function to call as token
713 // value (for token type MARKUP_FUNCTION or
714 // MARKUP_LIST_FUNCTION).
716 push_extra_token(EXPECT_NO_MORE_ARGS);
718 for (; scm_is_pair(s); s = scm_cdr(s)) {
719 SCM predicate = scm_car(s);
721 if (predicate == ly_lily_module_constant ("markup-list?"))
722 push_extra_token(EXPECT_MARKUP_LIST);
723 else if (predicate == ly_lily_module_constant ("markup?"))
724 push_extra_token(EXPECT_MARKUP);
726 push_extra_token(EXPECT_SCM, predicate);
730 [^$#{}\"\\ \t\n\r\f]+ {
731 string s (YYText_utf8 ());
733 yylval = ly_string2scm (s);
737 yylval = SCM_UNSPECIFIED;
742 <longcomment><<EOF>> {
743 LexerError (_ ("EOF found inside a comment").c_str ());
747 <quote,commandquote><<EOF>> {
748 LexerError (_ ("EOF found inside string").c_str ());
753 yylval = SCM_UNSPECIFIED;
756 is_main_input_ = include_stack_.size () > main_input_level_;
759 main_input_level_ = 0;
761 if (YYSTATE != maininput)
763 LexerError (_ ("Unfinished main input").c_str ());
766 } while (YYSTATE != maininput);
768 extra_tokens_ = SCM_EOL;
771 if (!close_input () || !is_main_input_)
772 /* Returns YY_NULL */
775 else if (!close_input ())
776 /* Returns YY_NULL */
780 <maininput>{ANY_CHAR} {
781 while (include_stack_.size () > main_input_level_
788 {WORD}/[-_] | // backup rule
790 return scan_bare_word (YYText_utf8 ());
793 start_command_quote ();
795 {COMMAND}/[-_] | // backup rule
797 return scan_escaped_word (YYText_utf8 () + 1);
802 yylval = scan_fraction (YYText ());
806 -{UNSIGNED} | // backup rule
808 yylval = scm_c_read_string (YYText ());
812 {UNSIGNED}/\/ | // backup rule
814 yylval = scm_c_read_string (YYText ());
819 -/\. { // backup rule
820 yylval = SCM_UNSPECIFIED;
824 <INITIAL,chords,lyrics,figures,notes>{SPECIAL} {
825 yylval = SCM_UNSPECIFIED;
829 <INITIAL,chords,lyrics,figures,notes>{SHORTHAND} {
830 return scan_shorthand (YYText_utf8 ()); // should not be utf-8
834 string msg = _f ("invalid character: `%s'", YYText_utf8 ());
835 LexerError (msg.c_str ());
836 yylval = SCM_UNSPECIFIED;
837 return '%'; // Better not return half a utf8 character.
842 /* Make the lexer generate a token of the given type as the next token.
843 TODO: make it possible to define a value for the token as well */
845 Lily_lexer::push_extra_token (int token_type, SCM scm)
847 if (scm_is_null (extra_tokens_))
849 if (YY_START != extratoken)
850 hidden_state_ = YY_START;
851 yy_push_state (extratoken);
853 extra_tokens_ = scm_acons (scm_from_int (token_type), scm, extra_tokens_);
857 Lily_lexer::push_chord_state (SCM alist)
859 SCM p = scm_assq (alist, pitchname_tab_stack_);
861 if (scm_is_false (p))
862 p = scm_cons (alist, alist_to_hashq (alist));
863 pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_);
864 yy_push_state (chords);
868 Lily_lexer::push_figuredbass_state ()
870 yy_push_state (figures);
874 Lily_lexer::push_initial_state ()
876 yy_push_state (INITIAL);
880 Lily_lexer::push_lyric_state ()
882 yy_push_state (lyrics);
886 Lily_lexer::push_markup_state ()
888 yy_push_state (markup);
892 Lily_lexer::push_note_state (SCM alist)
894 bool extra = (YYSTATE == extratoken);
896 SCM p = scm_assq (alist, pitchname_tab_stack_);
901 if (scm_is_false (p))
902 p = scm_cons (alist, alist_to_hashq (alist));
903 pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_);
904 yy_push_state (notes);
907 hidden_state_ = YYSTATE;
908 yy_push_state (extratoken);
913 Lily_lexer::pop_state ()
915 bool extra = (YYSTATE == extratoken);
920 if (YYSTATE == notes || YYSTATE == chords)
921 pitchname_tab_stack_ = scm_cdr (pitchname_tab_stack_);
923 // don't cross the maininput threshold
924 if (YYSTATE != maininput)
928 hidden_state_ = YYSTATE;
929 yy_push_state (extratoken);
934 Lily_lexer::identifier_type (SCM sid)
936 int k = try_special_identifiers (&yylval , sid);
937 return k >= 0 ? k : SCM_IDENTIFIER;
942 Lily_lexer::scan_escaped_word (const string &str)
944 // use more SCM for this.
946 // SCM sym = ly_symbol2scm (str.c_str ());
948 yylval = SCM_UNSPECIFIED;
949 int i = lookup_keyword (str);
954 SCM sid = lookup_identifier (str);
955 if (Music *m = unsmob_music (sid))
957 m->set_spot (override_input (here_input ()));
960 if (sid != SCM_UNDEFINED)
961 return scan_scm_id (sid);
963 string msg (_f ("unknown escaped string: `\\%s'", str));
964 LexerError (msg.c_str ());
966 yylval = ly_string2scm (str);
972 Lily_lexer::scan_shorthand (const string &str)
974 SCM sid = lookup_identifier (str);
975 if (Music *m = unsmob_music (sid))
977 m->set_spot (override_input (here_input ()));
980 if (sid != SCM_UNDEFINED)
981 return scan_scm_id (sid);
983 string msg (_f ("undefined character or shorthand: %s", str));
984 LexerError (msg.c_str ());
986 yylval = ly_string2scm (str);
992 Lily_lexer::scan_scm_id (SCM sid)
994 if (is_music_function (sid))
996 int funtype = SCM_FUNCTION;
1000 SCM s = get_music_function_signature (sid);
1001 SCM cs = scm_car (s);
1003 if (scm_is_pair (cs))
1008 if (scm_is_eq (cs, ly_lily_module_constant ("ly:music?")))
1009 funtype = MUSIC_FUNCTION;
1010 else if (scm_is_eq (cs, ly_lily_module_constant ("ly:event?")))
1011 funtype = EVENT_FUNCTION;
1012 else if (ly_is_procedure (cs))
1013 funtype = SCM_FUNCTION;
1014 else programming_error ("Bad syntax function predicate");
1016 push_extra_token (EXPECT_NO_MORE_ARGS);
1017 for (s = scm_cdr (s); scm_is_pair (s); s = scm_cdr (s))
1019 SCM optional = SCM_UNDEFINED;
1022 if (scm_is_pair (cs))
1024 optional = SCM_CDR (cs);
1028 if (ly_is_procedure (cs))
1029 push_extra_token (EXPECT_SCM, cs);
1032 programming_error ("Function parameter without type-checking predicate");
1035 if (!scm_is_eq (optional, SCM_UNDEFINED))
1036 push_extra_token (EXPECT_OPTIONAL, optional);
1041 return identifier_type (sid);
1045 Lily_lexer::scan_bare_word (const string &str)
1047 SCM sym = ly_symbol2scm (str.c_str ());
1048 if ((YYSTATE == notes) || (YYSTATE == chords)) {
1049 SCM handle = SCM_BOOL_F;
1050 if (scm_is_pair (pitchname_tab_stack_))
1051 handle = scm_hashq_get_handle (scm_cdar (pitchname_tab_stack_), sym);
1053 if (scm_is_pair (handle)) {
1054 yylval = scm_cdr (handle);
1055 if (unsmob_pitch (yylval))
1056 return (YYSTATE == notes) ? NOTENAME_PITCH : TONICNAME_PITCH;
1057 else if (scm_is_symbol (yylval))
1060 else if ((YYSTATE == chords)
1061 && (handle = scm_hashq_get_handle (chordmodifier_tab_, sym))!= SCM_BOOL_F)
1063 yylval = scm_cdr (handle);
1064 return CHORD_MODIFIER;
1067 yylval = ly_string2scm (str);
1072 Lily_lexer::get_state () const
1074 if (YY_START == extratoken)
1075 return hidden_state_;
1081 Lily_lexer::is_note_state () const
1083 return get_state () == notes;
1087 Lily_lexer::is_chord_state () const
1089 return get_state () == chords;
1093 Lily_lexer::is_lyric_state () const
1095 return get_state () == lyrics;
1099 Lily_lexer::is_figure_state () const
1101 return get_state () == figures;
1104 // The extra_token parameter specifies how to convert multiple values
1105 // into additional tokens. For '#', additional values get pushed as
1106 // SCM_IDENTIFIER. For '$', they get checked for their type and get
1107 // pushed as a corresponding *_IDENTIFIER token. Since the latter
1108 // tampers with yylval, it can only be done from the lexer itself, so
1109 // this function is private.
1112 Lily_lexer::eval_scm (SCM readerdata, char extra_token)
1114 SCM sval = SCM_UNDEFINED;
1116 if (!SCM_UNBNDP (readerdata))
1118 sval = ly_eval_scm (scm_car (readerdata),
1119 *unsmob_input (scm_cdr (readerdata)),
1120 be_safe_global && is_main_input_,
1124 if (SCM_UNBNDP (sval))
1127 return SCM_UNSPECIFIED;
1130 if (extra_token && SCM_VALUESP (sval))
1132 sval = scm_struct_ref (sval, SCM_INUM0);
1134 if (scm_is_pair (sval)) {
1135 for (SCM p = scm_reverse (scm_cdr (sval));
1139 SCM v = scm_car (p);
1140 if (Music *m = unsmob_music (v))
1142 if (!unsmob_input (m->get_property ("origin")))
1143 m->set_spot (override_input (here_input ()));
1147 switch (extra_token) {
1149 token = scan_scm_id (v);
1150 if (!scm_is_eq (yylval, SCM_UNSPECIFIED))
1151 push_extra_token (token, yylval);
1154 push_extra_token (SCM_IDENTIFIER, v);
1158 sval = scm_car (sval);
1160 sval = SCM_UNSPECIFIED;
1163 if (Music *m = unsmob_music (sval))
1165 if (!unsmob_input (m->get_property ("origin")))
1166 m->set_spot (override_input (here_input ()));
1172 /* Check for valid UTF-8 that has no overlong or surrogate codes and
1173 is in the range 0-0x10ffff */
1176 Lily_lexer::YYText_utf8 ()
1178 const char * const p = YYText ();
1179 for (int i=0; p[i];) {
1180 if ((p[i] & 0xff) < 0x80) {
1184 int oldi = i; // start of character
1185 int more = 0; // # of followup bytes, 0 if bad
1186 switch (p[i++] & 0xff) {
1187 // 0xc0 and 0xc1 are overlong prefixes for
1188 // 0x00-0x3f and 0x40-0x7f respectively, bad.
1189 case 0xc2: // 0x80-0xbf
1190 case 0xc3: // 0xc0-0xff
1191 case 0xc4: // 0x100-0x13f
1192 case 0xc5: // 0x140-0x17f
1193 case 0xc6: // 0x180-0x1bf
1194 case 0xc7: // 0x1c0-0x1ff
1195 case 0xc8: // 0x200-0x23f
1196 case 0xc9: // 0x240-0x27f
1197 case 0xca: // 0x280-0x2bf
1198 case 0xcb: // 0x2c0-0x2ff
1199 case 0xcc: // 0x300-0x33f
1200 case 0xcd: // 0x340-0x37f
1201 case 0xce: // 0x380-0x3bf
1202 case 0xcf: // 0x3c0-0x3ff
1203 case 0xd0: // 0x400-0x43f
1204 case 0xd1: // 0x440-0x47f
1205 case 0xd2: // 0x480-0x4bf
1206 case 0xd3: // 0x4c0-0x4ff
1207 case 0xd4: // 0x500-0x53f
1208 case 0xd5: // 0x540-0x57f
1209 case 0xd6: // 0x580-0x5bf
1210 case 0xd7: // 0x5c0-0x5ff
1211 case 0xd8: // 0x600-0x63f
1212 case 0xd9: // 0x640-0x67f
1213 case 0xda: // 0x680-0x6bf
1214 case 0xdb: // 0x6c0-0x6ff
1215 case 0xdc: // 0x700-0x73f
1216 case 0xdd: // 0x740-0x77f
1217 case 0xde: // 0x780-0x7bf
1218 case 0xdf: // 0x7c0-0x7ff
1219 more = 1; // 2-byte sequences, 0x80-0x7ff
1222 // don't allow overlong sequences for 0-0x7ff
1223 if ((p[i] & 0xff) < 0xa0)
1225 case 0xe1: // 0x1000-0x1fff
1226 case 0xe2: // 0x2000-0x2fff
1227 case 0xe3: // 0x3000-0x3fff
1228 case 0xe4: // 0x4000-0x4fff
1229 case 0xe5: // 0x5000-0x5fff
1230 case 0xe6: // 0x6000-0x6fff
1231 case 0xe7: // 0x7000-0x7fff
1232 case 0xe8: // 0x8000-0x8fff
1233 case 0xe9: // 0x9000-0x9fff
1234 case 0xea: // 0xa000-0xafff
1235 case 0xeb: // 0xb000-0xbfff
1236 case 0xec: // 0xc000-0xcfff
1237 more = 2; // 3-byte sequences, 0x7ff-0xcfff
1239 case 0xed: // 0xd000-0xdfff
1240 // Don't allow surrogate codes 0xd800-0xdfff
1241 if ((p[i] & 0xff) >= 0xa0)
1243 case 0xee: // 0xe000-0xefff
1244 case 0xef: // 0xf000-0xffff
1245 more = 2; // 3-byte sequences,
1246 // 0xd000-0xd7ff, 0xe000-0xffff
1249 // don't allow overlong sequences for 0-0xffff
1250 if ((p[i] & 0xff) < 0x90)
1252 case 0xf1: // 0x40000-0x7ffff
1253 case 0xf2: // 0x80000-0xbffff
1254 case 0xf3: // 0xc0000-0xfffff
1255 more = 3; // 4-byte sequences, 0x10000-0xfffff
1258 // don't allow more than 0x10ffff
1259 if ((p[i] & 0xff) >= 0x90)
1261 more = 3; // 4-byte sequence, 0x100000-0x10ffff
1265 // check that all continuation bytes are valid
1267 if ((p[i++] & 0xc0) != 0x80)
1273 Input h = here_input ();
1274 h.set (h.get_source_file (), h.start () + oldi, h.start () + i);
1275 h.warning (_ ("non-UTF-8 input").c_str ());
1282 urg, belong to string (_convert)
1283 and should be generalised
1286 strip_leading_white (string&s)
1289 for (; i < s.length (); i++)
1290 if (!isspace (s[i]))
1297 strip_trailing_white (string&s)
1299 ssize i = s.length ();
1301 if (!isspace (s[i]))
1304 s = s.substr (0, i + 1);
1309 Lilypond_version oldest_version ("2.7.38");
1313 is_valid_version (string s)
1315 Lilypond_version current ( MAJOR_VERSION "." MINOR_VERSION "." PATCH_LEVEL );
1316 Lilypond_version ver (s);
1317 if (int (ver) < oldest_version)
1319 non_fatal_error (_f ("file too old: %s (oldest supported: %s)", ver.to_string (), oldest_version.to_string ()));
1320 non_fatal_error (_ ("consider updating the input with the convert-ly script"));
1326 non_fatal_error (_f ("program too old: %s (file requires: %s)", current.to_string (), ver.to_string ()));
1337 lyric_fudge (string s)
1341 while ((i = s.find ('_', i)) != string::npos)
1349 Convert "NUM/DEN" into a '(NUM . DEN) cons.
1352 scan_fraction (string frac)
1354 ssize i = frac.find ('/');
1355 string left = frac.substr (0, i);
1356 string right = frac.substr (i + 1, (frac.length () - i + 1));
1358 int n = String_convert::dec2int (left);
1359 int d = String_convert::dec2int (right);
1360 return scm_cons (scm_from_int (n), scm_from_int (d));
1364 lookup_markup_command (string s)
1366 SCM proc = ly_lily_module_constant ("lookup-markup-command");
1367 return scm_call_1 (proc, ly_string2scm (s));
1371 lookup_markup_list_command (string s)
1373 SCM proc = ly_lily_module_constant ("lookup-markup-list-command");
1374 return scm_call_1 (proc, ly_string2scm (s));
1377 /* Shut up lexer warnings. */
1397 silence_lexer_warnings ()
1399 (void) yy_start_stack_ptr;
1400 (void) yy_start_stack_depth;
1401 (void) yy_start_stack;
1402 (void) yy_push_state;
1403 (void) yy_pop_state;
1404 (void) yy_top_state;
1405 (void) silence_lexer_warnings;