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--2015 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"
74 #include "lily-imports.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
140 /* The strategy concerning multibyte characters is to accept them but
141 * call YYText_utf8 for patterns that might contain them, in order to
142 * get a single code path responsible for flagging non-UTF-8 input:
143 * Patterns for accepting only valid UTF-8 without backing up are
144 * really hard to do and complex, and if nice error messages are
145 * wanted, one would need patterns catching the invalid input as well.
147 * Since editors and operating environments don't necessarily behave
148 * reasonably in the presence of mixed encodings, we flag encoding
149 * errors also in identifiers, comments, and strings where it would be
150 * conceivable to just transparently work with the byte string. But
151 * the whole point of caring about UTF-8 in here at all is too avoid
152 * stranger errors later when input passes into backends or log files
153 * or console output or error messages.
160 SYMBOL {A}([-_]{A}|{A})*
162 /* SPECIAL category is for every letter that needs to get passed to
163 * the parser rather than being redefinable by the user */
164 SPECIAL [-+*/=<>{}!?_^'',.:]
170 REAL ({INT}\.{N}*)|(-?\.{N}+)
171 STRICTREAL {UNSIGNED}\.{UNSIGNED}
173 HORIZONTALWHITE [ \t]
179 BOM_UTF8 \357\273\277
185 // swallow and ignore carriage returns
188 /* Use the trailing context feature. Otherwise, the BOM will not be
189 found if the file starts with an identifier definition. */
190 <INITIAL,chords,lyrics,figures,notes>{BOM_UTF8}/.* {
191 if (lexloc_->line_number () != 1 || lexloc_->column_number () != 0)
193 LexerWarning (_ ("stray UTF-8 BOM encountered").c_str ());
196 debug_output (_ ("Skipping UTF-8 BOM"));
199 <INITIAL,chords,figures,incl,lyrics,markup,notes>{
201 yy_push_state (longcomment);
203 %[^{\n\r][^\n\r]*[\n\r]? {
204 (void) YYText_utf8 ();
213 <INITIAL,notes,figures,chords,markup>{
219 <INITIAL,chords,lyrics,notes,figures>\\version{WHITE}* {
220 yy_push_state (version);
222 <INITIAL,chords,lyrics,notes,figures>\\sourcefilename{WHITE}* {
223 yy_push_state (sourcefilename);
225 <INITIAL,chords,lyrics,notes,figures>\\sourcefileline{WHITE}* {
226 yy_push_state (sourcefileline);
228 <version>\"[^""]*\" { /* got the version number */
229 string s (YYText_utf8 () + 1);
230 s = s.substr (0, s.rfind ('\"'));
234 SCM top_scope = scm_car (scm_last_pair (scopes_));
235 scm_module_define (top_scope, ly_symbol2scm ("version-seen"), SCM_BOOL_T);
237 if (!is_valid_version (s)) {
238 yylval = SCM_UNSPECIFIED;
242 <sourcefilename>\"[^""]*\" {
243 string s (YYText_utf8 () + 1);
244 s = s.substr (0, s.rfind ('\"'));
247 here_input().get_source_file ()->name_ = s;
248 message (_f ("Renaming input to: `%s'", s.c_str ()));
249 progress_indication ("\n");
250 scm_module_define (scm_car (scopes_),
251 ly_symbol2scm ("input-file-name"),
256 <sourcefileline>{INT} {
258 sscanf (YYText (), "%d", &i);
261 here_input ().get_source_file ()->set_line (here_input ().start (), i);
264 <version>{ANY_CHAR} {
265 LexerError (_ ("quoted string expected after \\version").c_str ());
268 <sourcefilename>{ANY_CHAR} {
269 LexerError (_ ("quoted string expected after \\sourcefilename").c_str ());
272 <sourcefileline>{ANY_CHAR} {
273 LexerError (_ ("integer expected after \\sourcefileline").c_str ());
278 (void) YYText_utf8 ();
281 (void) YYText_utf8 ();
289 <INITIAL,chords,lyrics,notes,figures>\\maininput {
293 main_input_level_ = include_stack_.size ();
294 is_main_input_ = true;
296 yy_push_state (maininput);
297 yy_push_state (state);
300 LexerError (_ ("\\maininput not allowed outside init files").c_str ());
303 <INITIAL,chords,lyrics,figures,notes>\\include {
304 yy_push_state (incl);
306 <incl>\"[^""]*\" { /* got the include file name */
307 string s (YYText_utf8 ()+1);
308 s = s.substr (0, s.rfind ('"'));
310 new_input (s, sources_);
313 <incl>\\{BLACK}*{WHITE}? { /* got the include identifier */
314 string s = YYText_utf8 () + 1;
315 strip_trailing_white (s);
316 if (s.length () && (s[s.length () - 1] == ';'))
317 s = s.substr (0, s.length () - 1);
319 SCM sid = lookup_identifier (s);
320 if (scm_is_string (sid)) {
321 new_input (ly_scm2string (sid), sources_);
324 string msg (_f ("wrong or undefined identifier: `%s'", s ));
326 LexerError (msg.c_str ());
327 SCM err = scm_current_error_port ();
328 scm_puts ("This value was found in the table: ", err);
329 scm_display (sid, err);
332 <incl>(\$|#) { // scm for the filename
333 Input hi = here_input();
335 SCM sval = ly_parse_scm (hi, be_safe_global && is_main_input_, parser_);
336 sval = eval_scm (sval, hi);
337 int n = hi.end () - hi.start ();
339 for (int i = 0; i < n; i++)
343 char_count_stack_.back () += n;
345 if (scm_is_string (sval)) {
346 new_input (ly_scm2string (sval), sources_);
349 LexerError (_ ("string expected after \\include").c_str ());
350 if (!SCM_UNBNDP (sval)) {
351 SCM err = scm_current_error_port ();
352 scm_puts ("This value was found instead: ", err);
353 scm_display (sval, err);
358 <incl,version,sourcefilename>\"[^""]* { // backup rule
359 LexerError (_ ("end quote missing").c_str ());
363 /* Flex picks the longest matching pattern including trailing
364 * contexts. Without the backup pattern, r-. does not trigger the
365 * {RESTNAME} rule but rather the {SYMBOL}/[-_] rule coming later,
366 * needed for avoiding backup states.
369 <chords,notes,figures>{RESTNAME}/[-_] | // pseudo backup rule
370 <chords,notes,figures>{RESTNAME} {
371 char const *s = YYText ();
372 yylval = scm_from_ascii_string (s);
375 <chords,notes,figures>q/[-_] | // pseudo backup rule
376 <chords,notes,figures>q {
377 yylval = SCM_UNSPECIFIED;
378 return CHORD_REPETITION;
381 <chords,notes,figures>R/[-_] | // pseudo backup rule
382 <chords,notes,figures>R {
383 yylval = SCM_UNSPECIFIED;
384 return MULTI_MEASURE_REST;
386 <INITIAL,chords,figures,lyrics,markup,notes># { //embedded scm
387 Input hi = here_input();
389 SCM sval = ly_parse_scm (hi, be_safe_global && is_main_input_, parser_);
391 if (SCM_UNBNDP (sval))
394 int n = hi.end () - hi.start ();
395 for (int i = 0; i < n; i++)
399 char_count_stack_.back () += n;
405 <INITIAL,chords,figures,lyrics,markup,notes>\$ { //immediate scm
406 Input hi = here_input();
408 SCM sval = ly_parse_scm (hi, be_safe_global && is_main_input_, parser_);
410 int n = hi.end () - hi.start ();
412 for (int i = 0; i < n; i++)
416 char_count_stack_.back () += n;
418 sval = eval_scm (sval, hi, '$');
420 if (YYSTATE == markup && ly_is_procedure (sval))
422 SCM sig = Lily::markup_command_signature (sval);
423 if (scm_is_true (sig))
426 int token = MARKUP_FUNCTION;
427 if (scm_is_true (Lily::markup_list_function_p (sval)))
428 token = MARKUP_LIST_FUNCTION;
429 push_markup_predicates (sig);
433 int token = scan_scm_id (sval);
434 if (!scm_is_eq (yylval, SCM_UNSPECIFIED))
438 <INITIAL,notes,lyrics,chords>{
440 yylval = SCM_UNSPECIFIED;
441 return DOUBLE_ANGLE_OPEN;
444 yylval = SCM_UNSPECIFIED;
445 return DOUBLE_ANGLE_CLOSE;
449 <INITIAL,notes,chords>{
451 yylval = SCM_UNSPECIFIED;
455 yylval = SCM_UNSPECIFIED;
462 yylval = SCM_UNSPECIFIED;
466 yylval = SCM_UNSPECIFIED;
470 yylval = SCM_UNSPECIFIED;
474 yylval = SCM_UNSPECIFIED;
478 yylval = SCM_UNSPECIFIED;
479 return E_EXCLAMATION;
482 yylval = SCM_UNSPECIFIED;
486 yylval = SCM_UNSPECIFIED;
492 {SYMBOL}/[-_] | // backup rule
494 return scan_bare_word (YYText_utf8 ());
497 start_command_quote ();
499 {COMMAND}/[-_] | // backup rule
501 return scan_escaped_word (YYText_utf8 () + 1);
504 yylval = scan_fraction (YYText ());
508 yylval = scm_c_read_string (YYText ());
511 {UNSIGNED}/[/.] | // backup rule
513 yylval = scm_c_read_string (YYText ());
517 yylval = scm_c_read_string (YYText () + 1);
522 <quote,commandquote>{
524 char c = escaped_char (YYText ()[1]);
525 yylval = scm_cons (scm_from_ascii_stringn (&c, 1),
529 yylval = scm_cons (scm_from_utf8_string (YYText_utf8 ()),
534 /* yylval is union. Must remember STRING before setting SCM*/
536 yylval = scm_string_concatenate_reverse (yylval,
540 if (get_state () == commandquote) {
542 return scan_escaped_word (ly_scm2string (yylval));
550 yylval = scm_cons (scm_from_ascii_string (YYText ()),
560 yylval = scan_fraction (YYText ());
564 yylval = scm_c_read_string (YYText ());
567 {UNSIGNED}/[/.] | // backup rule
569 yylval = scm_c_read_string (YYText ());
573 start_command_quote ();
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][^$#{}\"\\ \t\n\r\f0-9]* {
590 string s (YYText_utf8 ());
591 yylval = SCM_UNSPECIFIED;
597 yylval = ly_string2scm (s);
601 /* This should really just cover {} */
603 yylval = SCM_UNSPECIFIED;
608 {SYMBOL}/[-_] | // backup rule
610 return scan_bare_word (YYText_utf8 ());
613 start_command_quote ();
615 {COMMAND}/[-_] | // backup rule
617 return scan_escaped_word (YYText_utf8 () + 1);
620 yylval = scan_fraction (YYText ());
623 {UNSIGNED}/\/ | // backup rule
625 yylval = scm_c_read_string (YYText ());
629 yylval = SCM_UNSPECIFIED;
633 yylval = SCM_UNSPECIFIED;
637 yylval = SCM_UNSPECIFIED;
641 yylval = SCM_UNSPECIFIED;
645 yylval = SCM_UNSPECIFIED;
653 yylval = SCM_UNSPECIFIED;
657 yylval = SCM_UNSPECIFIED;
661 start_command_quote ();
663 {COMMAND}/[-_] | // backup rule
665 string str (YYText_utf8 () + 1);
667 int token_type = MARKUP_FUNCTION;
668 SCM s = lookup_markup_command (str);
670 // lookup-markup-command returns a pair with the car
671 // being the function to call, and the cdr being the
672 // call signature specified to define-markup-command,
673 // a list of predicates.
675 if (!scm_is_pair (s)) {
676 // If lookup-markup-command was not successful, we
677 // try lookup-markup-list-command instead.
678 // If this fails as well, we just scan and return
680 s = lookup_markup_list_command (str);
682 token_type = MARKUP_LIST_FUNCTION;
684 return scan_escaped_word (str);
687 // If the list of predicates is, say,
688 // (number? number? markup?), then tokens
689 // EXPECT_MARKUP EXPECT_SCM EXPECT_SCM EXPECT_NO_MORE_ARGS
690 // will be generated. Note that we have to push them
691 // in reverse order, so the first token pushed in the
692 // loop will be EXPECT_NO_MORE_ARGS.
696 // yylval now contains the function to call as token
697 // value (for token type MARKUP_FUNCTION or
698 // MARKUP_LIST_FUNCTION).
700 push_markup_predicates (scm_cdr (s));
704 [^$#{}\"\\ \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 ());
721 <quote,commandquote><<EOF>> {
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 {SYMBOL}/[-_] | // backup rule
764 return scan_bare_word (YYText_utf8 ());
767 start_command_quote ();
769 {COMMAND}/[-_] | // backup rule
771 return scan_escaped_word (YYText_utf8 () + 1);
776 yylval = scan_fraction (YYText ());
780 -{UNSIGNED} | // backup rule
782 yylval = scm_c_read_string (YYText ());
786 {UNSIGNED}/\/ | // backup rule
788 yylval = scm_c_read_string (YYText ());
793 -/\. { // backup rule
794 yylval = SCM_UNSPECIFIED;
798 <INITIAL,chords,lyrics,figures,notes>{SPECIAL} {
799 yylval = SCM_UNSPECIFIED;
803 <INITIAL,chords,lyrics,figures,notes>{SHORTHAND} {
804 return scan_shorthand (YYText_utf8 ()); // should not be utf-8
808 string msg = _f ("invalid character: `%s'", YYText_utf8 ());
809 LexerError (msg.c_str ());
810 yylval = SCM_UNSPECIFIED;
811 return '%'; // Better not return half a utf8 character.
816 /* Make the lexer generate a token of the given type as the next token.
817 TODO: make it possible to define a value for the token as well */
819 Lily_lexer::push_extra_token (Input const &where, int token_type, SCM scm)
821 extra_tokens_ = scm_cons (scm_cons2 (where.smobbed_copy (),
822 scm_from_int (token_type),
823 scm), extra_tokens_);
827 Lily_lexer::pop_extra_token ()
829 if (scm_is_null (extra_tokens_))
832 /* produce requested token */
833 yylloc = *unsmob<Input> (scm_caar (extra_tokens_));
834 int type = scm_to_int (scm_cadar (extra_tokens_));
835 yylval = scm_cddar (extra_tokens_);
836 extra_tokens_ = scm_cdr (extra_tokens_);
841 Lily_lexer::push_chord_state (SCM alist)
843 SCM p = scm_assq (alist, pitchname_tab_stack_);
845 if (scm_is_false (p))
846 p = scm_cons (alist, alist_to_hashq (alist));
847 pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_);
848 yy_push_state (chords);
852 Lily_lexer::push_figuredbass_state ()
854 yy_push_state (figures);
858 Lily_lexer::push_initial_state ()
860 yy_push_state (INITIAL);
864 Lily_lexer::push_lyric_state ()
866 yy_push_state (lyrics);
870 Lily_lexer::push_markup_state ()
872 yy_push_state (markup);
876 Lily_lexer::push_note_state (SCM alist)
878 SCM p = scm_assq (alist, pitchname_tab_stack_);
880 if (scm_is_false (p))
881 p = scm_cons (alist, alist_to_hashq (alist));
882 pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_);
883 yy_push_state (notes);
887 Lily_lexer::pop_state ()
889 if (YYSTATE == notes || YYSTATE == chords)
890 pitchname_tab_stack_ = scm_cdr (pitchname_tab_stack_);
892 // don't cross the maininput threshold
893 if (YYSTATE != maininput)
899 Lily_lexer::push_markup_predicates (SCM sig)
901 push_extra_token (here_input (), EXPECT_NO_MORE_ARGS);
902 for (SCM s = sig; scm_is_pair(s); s = scm_cdr(s)) {
903 SCM predicate = scm_car(s);
905 if (scm_is_eq (predicate, SCM (Lily::markup_list_p)))
906 push_extra_token (here_input (), EXPECT_MARKUP_LIST);
907 else if (scm_is_eq (predicate, SCM (Lily::markup_p)))
908 push_extra_token (here_input (), EXPECT_MARKUP);
910 push_extra_token (here_input (), EXPECT_SCM, predicate);
916 Lily_lexer::identifier_type (SCM sid)
918 int k = try_special_identifiers (&yylval , sid);
919 return k >= 0 ? k : SCM_IDENTIFIER;
924 Lily_lexer::scan_escaped_word (const string &str)
926 // use more SCM for this.
928 // SCM sym = ly_symbol2scm (str.c_str ());
930 yylval = SCM_UNSPECIFIED;
931 int i = lookup_keyword (str);
936 SCM sid = lookup_identifier (str);
937 if (Music *m = unsmob<Music> (sid))
939 m->set_spot (override_input (here_input ()));
942 if (!SCM_UNBNDP (sid))
943 return scan_scm_id (sid);
945 string msg (_f ("unknown escaped string: `\\%s'", str));
946 LexerError (msg.c_str ());
948 yylval = ly_string2scm (str);
950 return STRING; // SYMBOL would cause additional processing
954 Lily_lexer::scan_shorthand (const string &str)
956 SCM sid = lookup_identifier (str);
957 if (Music *m = unsmob<Music> (sid))
959 m->set_spot (override_input (here_input ()));
962 if (!SCM_UNBNDP (sid))
963 return scan_scm_id (sid);
965 string msg (_f ("undefined character or shorthand: %s", str));
966 LexerError (msg.c_str ());
968 yylval = ly_string2scm (str);
974 Lily_lexer::scan_scm_id (SCM sid)
976 if (Music_function *fun = unsmob<Music_function> (sid))
978 int funtype = SCM_FUNCTION;
982 SCM s = fun->get_signature ();
983 SCM cs = scm_car (s);
985 if (scm_is_pair (cs))
990 if (scm_is_eq (cs, SCM (Lily::ly_music_p)))
991 funtype = MUSIC_FUNCTION;
992 else if (scm_is_eq (cs, SCM (Lily::ly_event_p)))
993 funtype = EVENT_FUNCTION;
994 else if (ly_is_procedure (cs))
995 funtype = SCM_FUNCTION;
996 else programming_error ("Bad syntax function predicate");
998 push_extra_token (here_input (), EXPECT_NO_MORE_ARGS);
999 for (s = scm_cdr (s); scm_is_pair (s); s = scm_cdr (s))
1001 SCM optional = SCM_UNDEFINED;
1004 if (scm_is_pair (cs))
1006 optional = SCM_CDR (cs);
1010 if (ly_is_procedure (cs))
1011 push_extra_token (here_input (), EXPECT_SCM, cs);
1014 programming_error ("Function parameter without type-checking predicate");
1017 if (!scm_is_eq (optional, SCM_UNDEFINED))
1018 push_extra_token (here_input (), EXPECT_OPTIONAL, optional);
1023 return identifier_type (sid);
1027 Lily_lexer::scan_word (SCM & output, SCM sym)
1029 if ((YYSTATE == notes) || (YYSTATE == chords)) {
1030 SCM handle = SCM_BOOL_F;
1031 if (scm_is_pair (pitchname_tab_stack_))
1032 handle = scm_hashq_get_handle (scm_cdar (pitchname_tab_stack_), sym);
1034 if (scm_is_pair (handle)) {
1035 output = scm_cdr (handle);
1036 if (unsmob<Pitch> (yylval))
1037 return (YYSTATE == notes) ? NOTENAME_PITCH : TONICNAME_PITCH;
1038 else if (scm_is_symbol (yylval))
1041 else if ((YYSTATE == chords)
1042 && scm_is_true (handle = scm_hashq_get_handle (chordmodifier_tab_, sym)))
1044 output = scm_cdr (handle);
1045 return CHORD_MODIFIER;
1048 output = SCM_UNDEFINED;
1053 Lily_lexer::scan_bare_word (const string &str)
1055 int state = scan_word (yylval, ly_symbol2scm (str.c_str ()));
1060 yylval = ly_string2scm (str);
1065 Lily_lexer::get_state () const
1071 Lily_lexer::is_note_state () const
1073 return get_state () == notes;
1077 Lily_lexer::is_chord_state () const
1079 return get_state () == chords;
1083 Lily_lexer::is_lyric_state () const
1085 return get_state () == lyrics;
1089 Lily_lexer::is_figure_state () const
1091 return get_state () == figures;
1094 // The extra_token parameter specifies how to convert multiple values
1095 // into additional tokens. For '#', additional values get pushed as
1096 // SCM_IDENTIFIER. For '$', they get checked for their type and get
1097 // pushed as a corresponding *_IDENTIFIER token. Since the latter
1098 // tampers with yylval, it can only be done from the lexer itself, so
1099 // this function is private.
1102 Lily_lexer::eval_scm (SCM readerdata, Input hi, char extra_token)
1104 SCM sval = SCM_UNDEFINED;
1106 if (!SCM_UNBNDP (readerdata))
1108 sval = ly_eval_scm (readerdata,
1110 be_safe_global && is_main_input_,
1114 if (SCM_UNBNDP (sval))
1117 return SCM_UNSPECIFIED;
1120 if (extra_token && SCM_VALUESP (sval))
1122 sval = scm_struct_ref (sval, SCM_INUM0);
1124 if (scm_is_pair (sval)) {
1125 for (SCM p = scm_reverse (scm_cdr (sval));
1129 SCM v = scm_car (p);
1130 if (Music *m = unsmob<Music> (v))
1132 if (!unsmob<Input> (m->get_property ("origin")))
1133 m->set_spot (override_input (here_input ()));
1137 switch (extra_token) {
1139 token = scan_scm_id (v);
1140 if (!scm_is_eq (yylval, SCM_UNSPECIFIED))
1141 push_extra_token (here_input (),
1145 push_extra_token (here_input (),
1150 sval = scm_car (sval);
1152 sval = SCM_UNSPECIFIED;
1155 if (Music *m = unsmob<Music> (sval))
1157 if (!unsmob<Input> (m->get_property ("origin")))
1158 m->set_spot (override_input (here_input ()));
1164 /* Check for valid UTF-8 that has no overlong or surrogate codes and
1165 is in the range 0-0x10ffff */
1168 Lily_lexer::YYText_utf8 ()
1170 const char * const p = YYText ();
1171 for (int i=0; p[i];) {
1172 if ((p[i] & 0xff) < 0x80) {
1176 int oldi = i; // start of character
1177 int more = 0; // # of followup bytes, 0 if bad
1178 switch (p[i++] & 0xff) {
1179 // 0xc0 and 0xc1 are overlong prefixes for
1180 // 0x00-0x3f and 0x40-0x7f respectively, bad.
1181 case 0xc2: // 0x80-0xbf
1182 case 0xc3: // 0xc0-0xff
1183 case 0xc4: // 0x100-0x13f
1184 case 0xc5: // 0x140-0x17f
1185 case 0xc6: // 0x180-0x1bf
1186 case 0xc7: // 0x1c0-0x1ff
1187 case 0xc8: // 0x200-0x23f
1188 case 0xc9: // 0x240-0x27f
1189 case 0xca: // 0x280-0x2bf
1190 case 0xcb: // 0x2c0-0x2ff
1191 case 0xcc: // 0x300-0x33f
1192 case 0xcd: // 0x340-0x37f
1193 case 0xce: // 0x380-0x3bf
1194 case 0xcf: // 0x3c0-0x3ff
1195 case 0xd0: // 0x400-0x43f
1196 case 0xd1: // 0x440-0x47f
1197 case 0xd2: // 0x480-0x4bf
1198 case 0xd3: // 0x4c0-0x4ff
1199 case 0xd4: // 0x500-0x53f
1200 case 0xd5: // 0x540-0x57f
1201 case 0xd6: // 0x580-0x5bf
1202 case 0xd7: // 0x5c0-0x5ff
1203 case 0xd8: // 0x600-0x63f
1204 case 0xd9: // 0x640-0x67f
1205 case 0xda: // 0x680-0x6bf
1206 case 0xdb: // 0x6c0-0x6ff
1207 case 0xdc: // 0x700-0x73f
1208 case 0xdd: // 0x740-0x77f
1209 case 0xde: // 0x780-0x7bf
1210 case 0xdf: // 0x7c0-0x7ff
1211 more = 1; // 2-byte sequences, 0x80-0x7ff
1214 // don't allow overlong sequences for 0-0x7ff
1215 if ((p[i] & 0xff) < 0xa0)
1217 case 0xe1: // 0x1000-0x1fff
1218 case 0xe2: // 0x2000-0x2fff
1219 case 0xe3: // 0x3000-0x3fff
1220 case 0xe4: // 0x4000-0x4fff
1221 case 0xe5: // 0x5000-0x5fff
1222 case 0xe6: // 0x6000-0x6fff
1223 case 0xe7: // 0x7000-0x7fff
1224 case 0xe8: // 0x8000-0x8fff
1225 case 0xe9: // 0x9000-0x9fff
1226 case 0xea: // 0xa000-0xafff
1227 case 0xeb: // 0xb000-0xbfff
1228 case 0xec: // 0xc000-0xcfff
1229 more = 2; // 3-byte sequences, 0x7ff-0xcfff
1231 case 0xed: // 0xd000-0xdfff
1232 // Don't allow surrogate codes 0xd800-0xdfff
1233 if ((p[i] & 0xff) >= 0xa0)
1235 case 0xee: // 0xe000-0xefff
1236 case 0xef: // 0xf000-0xffff
1237 more = 2; // 3-byte sequences,
1238 // 0xd000-0xd7ff, 0xe000-0xffff
1241 // don't allow overlong sequences for 0-0xffff
1242 if ((p[i] & 0xff) < 0x90)
1244 case 0xf1: // 0x40000-0x7ffff
1245 case 0xf2: // 0x80000-0xbffff
1246 case 0xf3: // 0xc0000-0xfffff
1247 more = 3; // 4-byte sequences, 0x10000-0xfffff
1250 // don't allow more than 0x10ffff
1251 if ((p[i] & 0xff) >= 0x90)
1253 more = 3; // 4-byte sequence, 0x100000-0x10ffff
1257 // check that all continuation bytes are valid
1259 if ((p[i++] & 0xc0) != 0x80)
1265 Input h = here_input ();
1266 h.set (h.get_source_file (), h.start () + oldi, h.start () + i);
1267 h.warning (_ ("non-UTF-8 input").c_str ());
1274 urg, belong to string (_convert)
1275 and should be generalised
1278 strip_leading_white (string&s)
1281 for (; i < s.length (); i++)
1282 if (!isspace (s[i]))
1289 strip_trailing_white (string&s)
1291 ssize i = s.length ();
1293 if (!isspace (s[i]))
1296 s = s.substr (0, i + 1);
1301 Lilypond_version oldest_version ("2.7.38");
1305 is_valid_version (string s)
1307 Lilypond_version current ( MAJOR_VERSION "." MINOR_VERSION "." PATCH_LEVEL );
1308 Lilypond_version ver (s);
1311 non_fatal_error (_f ("Invalid version string \"%s\"", s));
1314 if (ver < oldest_version)
1316 non_fatal_error (_f ("file too old: %s (oldest supported: %s)", ver.to_string (), oldest_version.to_string ()));
1317 non_fatal_error (_ ("consider updating the input with the convert-ly script"));
1323 non_fatal_error (_f ("program too old: %s (file requires: %s)", current.to_string (), ver.to_string ()));
1334 lyric_fudge (string s)
1338 while ((i = s.find ('_', i)) != string::npos)
1346 Convert "NUM/DEN" into a '(NUM . DEN) cons.
1349 scan_fraction (string frac)
1351 ssize i = frac.find ('/');
1352 string left = frac.substr (0, i);
1353 string right = frac.substr (i + 1, (frac.length () - i + 1));
1355 return scm_cons (scm_c_read_string (left.c_str ()),
1356 scm_c_read_string (right.c_str ()));
1360 lookup_markup_command (string s)
1362 return Lily::lookup_markup_command (ly_string2scm (s));
1366 lookup_markup_list_command (string s)
1368 return Lily::lookup_markup_list_command (ly_string2scm (s));
1371 /* Shut up lexer warnings. */
1391 silence_lexer_warnings ()
1393 (void) yy_start_stack_ptr;
1394 (void) yy_start_stack_depth;
1395 (void) yy_start_stack;
1396 (void) yy_push_state;
1397 (void) yy_pop_state;
1398 (void) yy_top_state;
1399 (void) silence_lexer_warnings;