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. */
55 #include "context-def.hh"
56 #include "duration.hh"
57 #include "international.hh"
58 #include "interval.hh"
59 #include "lily-guile.hh"
60 #include "lily-lexer.hh"
61 #include "lily-parser.hh"
62 #include "lilypond-version.hh"
65 #include "music-function.hh"
66 #include "parse-scm.hh"
69 #include "source-file.hh"
70 #include "std-string.hh"
73 #include "lily-imports.hh"
80 #define isatty HORRIBLEKLUDGE
82 void strip_trailing_white (string&);
83 void strip_leading_white (string&);
84 string lyric_fudge (string s);
85 SCM lookup_markup_command (string s);
86 SCM lookup_markup_list_command (string s);
87 bool is_valid_version (string s);
90 #define start_quote() do { \
91 yy_push_state (quote); \
96 The inside of \"violin1" is marked by commandquote mode
99 #define start_command_quote() do { \
100 yy_push_state (commandquote); \
104 #define yylval (*lexval_)
106 #define yylloc (*lexloc_)
108 #define YY_USER_ACTION add_lexed_char (YYLeng ());
111 SCM scan_fraction (string);
112 SCM (* scm_parse_error_handler) (void *);
122 %option yyclass="Lily_lexer"
124 %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 /* Use the trailing context feature. Otherwise, the BOM will not be
190 found if the file starts with an identifier definition. */
191 <INITIAL,chords,lyrics,figures,notes>{BOM_UTF8}/.* {
192 if (lexloc_->line_number () != 1 || lexloc_->column_number () != 0)
194 LexerWarning (_ ("stray UTF-8 BOM encountered").c_str ());
197 debug_output (_ ("Skipping UTF-8 BOM"));
200 <INITIAL,chords,figures,incl,lyrics,markup,notes>{
202 yy_push_state (longcomment);
204 %[^{\n\r][^\n\r]*[\n\r]? {
205 (void) YYText_utf8 ();
214 <INITIAL,notes,figures,chords,markup>{
220 <INITIAL,chords,lyrics,notes,figures>\\version{WHITE}* {
221 yy_push_state (version);
223 <INITIAL,chords,lyrics,notes,figures>\\sourcefilename{WHITE}* {
224 yy_push_state (sourcefilename);
226 <INITIAL,chords,lyrics,notes,figures>\\sourcefileline{WHITE}* {
227 yy_push_state (sourcefileline);
229 <version>\"[^""]*\" { /* got the version number */
230 string s (YYText_utf8 () + 1);
231 s = s.substr (0, s.rfind ('\"'));
235 SCM top_scope = scm_car (scm_last_pair (scopes_));
236 scm_module_define (top_scope, ly_symbol2scm ("version-seen"), SCM_BOOL_T);
238 if (!is_valid_version (s)) {
239 yylval = SCM_UNSPECIFIED;
243 <sourcefilename>\"[^""]*\" {
244 string s (YYText_utf8 () + 1);
245 s = s.substr (0, s.rfind ('\"'));
248 here_input().get_source_file ()->name_ = s;
249 message (_f ("Renaming input to: `%s'", s.c_str ()));
250 progress_indication ("\n");
251 scm_module_define (scm_car (scopes_),
252 ly_symbol2scm ("input-file-name"),
257 <sourcefileline>{INT} {
259 sscanf (YYText (), "%d", &i);
262 here_input ().get_source_file ()->set_line (here_input ().start (), i);
265 <version>{ANY_CHAR} {
266 LexerError (_ ("quoted string expected after \\version").c_str ());
269 <sourcefilename>{ANY_CHAR} {
270 LexerError (_ ("quoted string expected after \\sourcefilename").c_str ());
273 <sourcefileline>{ANY_CHAR} {
274 LexerError (_ ("integer expected after \\sourcefileline").c_str ());
279 (void) YYText_utf8 ();
282 (void) YYText_utf8 ();
290 <INITIAL,chords,lyrics,notes,figures>\\maininput {
294 main_input_level_ = include_stack_.size ();
295 is_main_input_ = true;
297 yy_push_state (maininput);
298 yy_push_state (state);
301 LexerError (_ ("\\maininput not allowed outside init files").c_str ());
304 <INITIAL,chords,lyrics,figures,notes>\\include {
305 yy_push_state (incl);
307 <incl>\"[^""]*\" { /* got the include file name */
308 string s (YYText_utf8 ()+1);
309 s = s.substr (0, s.rfind ('"'));
311 new_input (s, sources_);
314 <incl>\\{BLACK}*{WHITE}? { /* got the include identifier */
315 string s = YYText_utf8 () + 1;
316 strip_trailing_white (s);
317 if (s.length () && (s[s.length () - 1] == ';'))
318 s = s.substr (0, s.length () - 1);
320 SCM sid = lookup_identifier (s);
321 if (scm_is_string (sid)) {
322 new_input (ly_scm2string (sid), sources_);
325 string msg (_f ("wrong or undefined identifier: `%s'", s ));
327 LexerError (msg.c_str ());
328 SCM err = scm_current_error_port ();
329 scm_puts ("This value was found in the table: ", err);
330 scm_display (sid, err);
333 <incl>(\$|#) { // scm for the filename
334 Input hi = here_input();
336 SCM sval = ly_parse_scm (hi, be_safe_global && is_main_input_, parser_);
337 sval = eval_scm (sval, hi);
338 int n = hi.end () - hi.start ();
340 for (int i = 0; i < n; i++)
344 char_count_stack_.back () += n;
346 if (scm_is_string (sval)) {
347 new_input (ly_scm2string (sval), sources_);
350 LexerError (_ ("string expected after \\include").c_str ());
351 if (!SCM_UNBNDP (sval)) {
352 SCM err = scm_current_error_port ();
353 scm_puts ("This value was found instead: ", err);
354 scm_display (sval, err);
359 <incl,version,sourcefilename>\"[^""]* { // backup rule
360 LexerError (_ ("end quote missing").c_str ());
364 /* Flex picks the longest matching pattern including trailing
365 * contexts. Without the backup pattern, r-. does not trigger the
366 * {RESTNAME} rule but rather the {WORD}/[-_] rule coming later,
367 * needed for avoiding backup states.
370 <chords,notes,figures>{RESTNAME}/[-_] | // pseudo backup rule
371 <chords,notes,figures>{RESTNAME} {
372 char const *s = YYText ();
373 yylval = scm_from_ascii_string (s);
376 <chords,notes,figures>q/[-_] | // pseudo backup rule
377 <chords,notes,figures>q {
378 yylval = SCM_UNSPECIFIED;
379 return CHORD_REPETITION;
382 <chords,notes,figures>R/[-_] | // pseudo backup rule
383 <chords,notes,figures>R {
384 yylval = SCM_UNSPECIFIED;
385 return MULTI_MEASURE_REST;
387 <INITIAL,chords,figures,lyrics,markup,notes># { //embedded scm
388 Input hi = here_input();
390 SCM sval = ly_parse_scm (hi, be_safe_global && is_main_input_, parser_);
392 if (SCM_UNBNDP (sval))
395 int n = hi.end () - hi.start ();
396 for (int i = 0; i < n; i++)
400 char_count_stack_.back () += n;
406 <INITIAL,chords,figures,lyrics,markup,notes>\$ { //immediate scm
407 Input hi = here_input();
409 SCM sval = ly_parse_scm (hi, be_safe_global && is_main_input_, parser_);
411 int n = hi.end () - hi.start ();
413 for (int i = 0; i < n; i++)
417 char_count_stack_.back () += n;
419 sval = eval_scm (sval, hi, '$');
421 int token = scan_scm_id (sval);
422 if (!scm_is_eq (yylval, SCM_UNSPECIFIED))
426 <INITIAL,notes,lyrics,chords>{
428 yylval = SCM_UNSPECIFIED;
429 return DOUBLE_ANGLE_OPEN;
432 yylval = SCM_UNSPECIFIED;
433 return DOUBLE_ANGLE_CLOSE;
437 <INITIAL,notes,chords>{
439 yylval = SCM_UNSPECIFIED;
443 yylval = SCM_UNSPECIFIED;
450 yylval = SCM_UNSPECIFIED;
454 yylval = SCM_UNSPECIFIED;
458 yylval = SCM_UNSPECIFIED;
462 yylval = SCM_UNSPECIFIED;
466 yylval = SCM_UNSPECIFIED;
467 return E_EXCLAMATION;
470 yylval = SCM_UNSPECIFIED;
474 yylval = SCM_UNSPECIFIED;
480 {WORD}/[-_] | // backup rule
482 return scan_bare_word (YYText_utf8 ());
485 start_command_quote ();
487 {COMMAND}/[-_] | // backup rule
489 return scan_escaped_word (YYText_utf8 () + 1);
492 yylval = scan_fraction (YYText ());
496 yylval = scm_c_read_string (YYText ());
499 {UNSIGNED}/[/.] | // backup rule
501 yylval = scm_c_read_string (YYText ());
505 yylval = scm_c_read_string (YYText () + 1);
510 <quote,commandquote>{
512 char c = escaped_char (YYText ()[1]);
513 yylval = scm_cons (scm_from_ascii_stringn (&c, 1),
517 yylval = scm_cons (scm_from_utf8_string (YYText_utf8 ()),
522 /* yylval is union. Must remember STRING before setting SCM*/
524 yylval = scm_string_concatenate_reverse (yylval,
528 if (get_state () == commandquote) {
530 return scan_escaped_word (ly_scm2string (yylval));
538 yylval = scm_cons (scm_from_ascii_string (YYText ()),
548 yylval = scan_fraction (YYText ());
552 yylval = scm_c_read_string (YYText ());
555 {UNSIGNED}/[/.] | // backup rule
557 yylval = scm_c_read_string (YYText ());
561 start_command_quote ();
563 {COMMAND}/[-_] | // backup rule
565 return scan_escaped_word (YYText_utf8 () + 1);
568 // UTF-8 already covered by COMMAND
569 return scan_shorthand (YYText ());
571 /* Characters needed to express durations, assignments */
573 yylval = SCM_UNSPECIFIED;
576 [^|*.=$#{}\"\\ \t\n\r\f0-9][^$#{}\"\\ \t\n\r\f0-9]* {
578 string s (YYText_utf8 ());
579 yylval = SCM_UNSPECIFIED;
585 yylval = ly_string2scm (s);
589 /* This should really just cover {} */
591 yylval = SCM_UNSPECIFIED;
596 {WORD}/[-_] | // backup rule
598 return scan_bare_word (YYText_utf8 ());
601 start_command_quote ();
603 {COMMAND}/[-_] | // backup rule
605 return scan_escaped_word (YYText_utf8 () + 1);
608 yylval = scan_fraction (YYText ());
611 {UNSIGNED}/\/ | // backup rule
613 yylval = scm_c_read_string (YYText ());
617 yylval = SCM_UNSPECIFIED;
621 yylval = SCM_UNSPECIFIED;
625 yylval = SCM_UNSPECIFIED;
629 yylval = SCM_UNSPECIFIED;
633 yylval = SCM_UNSPECIFIED;
641 yylval = SCM_UNSPECIFIED;
645 yylval = SCM_UNSPECIFIED;
649 start_command_quote ();
651 {COMMAND}/[-_] | // backup rule
653 string str (YYText_utf8 () + 1);
655 int token_type = MARKUP_FUNCTION;
656 SCM s = lookup_markup_command (str);
658 // lookup-markup-command returns a pair with the car
659 // being the function to call, and the cdr being the
660 // call signature specified to define-markup-command,
661 // a list of predicates.
663 if (!scm_is_pair (s)) {
664 // If lookup-markup-command was not successful, we
665 // try lookup-markup-list-command instead.
666 // If this fails as well, we just scan and return
668 s = lookup_markup_list_command (str);
670 token_type = MARKUP_LIST_FUNCTION;
672 return scan_escaped_word (str);
675 // If the list of predicates is, say,
676 // (number? number? markup?), then tokens
677 // EXPECT_MARKUP EXPECT_SCM EXPECT_SCM EXPECT_NO_MORE_ARGS
678 // will be generated. Note that we have to push them
679 // in reverse order, so the first token pushed in the
680 // loop will be EXPECT_NO_MORE_ARGS.
684 // yylval now contains the function to call as token
685 // value (for token type MARKUP_FUNCTION or
686 // MARKUP_LIST_FUNCTION).
688 push_extra_token (here_input (), EXPECT_NO_MORE_ARGS);
690 for (; scm_is_pair(s); s = scm_cdr(s)) {
691 SCM predicate = scm_car(s);
693 if (predicate == Lily::markup_list_p)
694 push_extra_token (here_input (), EXPECT_MARKUP_LIST);
695 else if (predicate == Lily::markup_p)
696 push_extra_token (here_input (), EXPECT_MARKUP);
698 push_extra_token (here_input (), EXPECT_SCM, predicate);
702 [^$#{}\"\\ \t\n\r\f]+ {
703 string s (YYText_utf8 ());
705 yylval = ly_string2scm (s);
709 yylval = SCM_UNSPECIFIED;
714 <longcomment><<EOF>> {
715 LexerError (_ ("EOF found inside a comment").c_str ());
719 <quote,commandquote><<EOF>> {
720 LexerError (_ ("EOF found inside string").c_str ());
725 yylval = SCM_UNSPECIFIED;
728 is_main_input_ = include_stack_.size () > main_input_level_;
731 main_input_level_ = 0;
733 if (YYSTATE != maininput)
735 LexerError (_ ("Unfinished main input").c_str ());
738 } while (YYSTATE != maininput);
740 extra_tokens_ = SCM_EOL;
743 if (!close_input () || !is_main_input_)
744 /* Returns YY_NULL */
747 else if (!close_input ())
748 /* Returns YY_NULL */
752 <maininput>{ANY_CHAR} {
753 while (include_stack_.size () > main_input_level_
760 {WORD}/[-_] | // backup rule
762 return scan_bare_word (YYText_utf8 ());
765 start_command_quote ();
767 {COMMAND}/[-_] | // backup rule
769 return scan_escaped_word (YYText_utf8 () + 1);
774 yylval = scan_fraction (YYText ());
778 -{UNSIGNED} | // backup rule
780 yylval = scm_c_read_string (YYText ());
784 {UNSIGNED}/\/ | // backup rule
786 yylval = scm_c_read_string (YYText ());
791 -/\. { // backup rule
792 yylval = SCM_UNSPECIFIED;
796 <INITIAL,chords,lyrics,figures,notes>{SPECIAL} {
797 yylval = SCM_UNSPECIFIED;
801 <INITIAL,chords,lyrics,figures,notes>{SHORTHAND} {
802 return scan_shorthand (YYText_utf8 ()); // should not be utf-8
806 string msg = _f ("invalid character: `%s'", YYText_utf8 ());
807 LexerError (msg.c_str ());
808 yylval = SCM_UNSPECIFIED;
809 return '%'; // Better not return half a utf8 character.
814 /* Make the lexer generate a token of the given type as the next token.
815 TODO: make it possible to define a value for the token as well */
817 Lily_lexer::push_extra_token (Input const &where, int token_type, SCM scm)
819 extra_tokens_ = scm_cons (scm_cons2 (where.smobbed_copy (),
820 scm_from_int (token_type),
821 scm), extra_tokens_);
825 Lily_lexer::pop_extra_token ()
827 if (scm_is_null (extra_tokens_))
830 /* produce requested token */
831 yylloc = *unsmob<Input> (scm_caar (extra_tokens_));
832 int type = scm_to_int (scm_cadar (extra_tokens_));
833 yylval = scm_cddar (extra_tokens_);
834 extra_tokens_ = scm_cdr (extra_tokens_);
839 Lily_lexer::push_chord_state (SCM alist)
841 SCM p = scm_assq (alist, pitchname_tab_stack_);
843 if (scm_is_false (p))
844 p = scm_cons (alist, alist_to_hashq (alist));
845 pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_);
846 yy_push_state (chords);
850 Lily_lexer::push_figuredbass_state ()
852 yy_push_state (figures);
856 Lily_lexer::push_initial_state ()
858 yy_push_state (INITIAL);
862 Lily_lexer::push_lyric_state ()
864 yy_push_state (lyrics);
868 Lily_lexer::push_markup_state ()
870 yy_push_state (markup);
874 Lily_lexer::push_note_state (SCM alist)
876 SCM p = scm_assq (alist, pitchname_tab_stack_);
878 if (scm_is_false (p))
879 p = scm_cons (alist, alist_to_hashq (alist));
880 pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_);
881 yy_push_state (notes);
885 Lily_lexer::pop_state ()
887 if (YYSTATE == notes || YYSTATE == chords)
888 pitchname_tab_stack_ = scm_cdr (pitchname_tab_stack_);
890 // don't cross the maininput threshold
891 if (YYSTATE != maininput)
897 Lily_lexer::identifier_type (SCM sid)
899 int k = try_special_identifiers (&yylval , sid);
900 return k >= 0 ? k : SCM_IDENTIFIER;
905 Lily_lexer::scan_escaped_word (const string &str)
907 // use more SCM for this.
909 // SCM sym = ly_symbol2scm (str.c_str ());
911 yylval = SCM_UNSPECIFIED;
912 int i = lookup_keyword (str);
917 SCM sid = lookup_identifier (str);
918 if (Music *m = unsmob<Music> (sid))
920 m->set_spot (override_input (here_input ()));
923 if (!SCM_UNBNDP (sid))
924 return scan_scm_id (sid);
926 string msg (_f ("unknown escaped string: `\\%s'", str));
927 LexerError (msg.c_str ());
929 yylval = ly_string2scm (str);
935 Lily_lexer::scan_shorthand (const string &str)
937 SCM sid = lookup_identifier (str);
938 if (Music *m = unsmob<Music> (sid))
940 m->set_spot (override_input (here_input ()));
943 if (!SCM_UNBNDP (sid))
944 return scan_scm_id (sid);
946 string msg (_f ("undefined character or shorthand: %s", str));
947 LexerError (msg.c_str ());
949 yylval = ly_string2scm (str);
955 Lily_lexer::scan_scm_id (SCM sid)
957 if (Music_function *fun = unsmob<Music_function> (sid))
959 int funtype = SCM_FUNCTION;
963 SCM s = fun->get_signature ();
964 SCM cs = scm_car (s);
966 if (scm_is_pair (cs))
971 if (scm_is_eq (cs, Lily::ly_music_p))
972 funtype = MUSIC_FUNCTION;
973 else if (scm_is_eq (cs, Lily::ly_event_p))
974 funtype = EVENT_FUNCTION;
975 else if (ly_is_procedure (cs))
976 funtype = SCM_FUNCTION;
977 else programming_error ("Bad syntax function predicate");
979 push_extra_token (here_input (), EXPECT_NO_MORE_ARGS);
980 for (s = scm_cdr (s); scm_is_pair (s); s = scm_cdr (s))
982 SCM optional = SCM_UNDEFINED;
985 if (scm_is_pair (cs))
987 optional = SCM_CDR (cs);
991 if (ly_is_procedure (cs))
992 push_extra_token (here_input (), EXPECT_SCM, cs);
995 programming_error ("Function parameter without type-checking predicate");
998 if (!scm_is_eq (optional, SCM_UNDEFINED))
999 push_extra_token (here_input (), EXPECT_OPTIONAL, optional);
1004 return identifier_type (sid);
1008 Lily_lexer::scan_bare_word (const string &str)
1010 SCM sym = ly_symbol2scm (str.c_str ());
1011 if ((YYSTATE == notes) || (YYSTATE == chords)) {
1012 SCM handle = SCM_BOOL_F;
1013 if (scm_is_pair (pitchname_tab_stack_))
1014 handle = scm_hashq_get_handle (scm_cdar (pitchname_tab_stack_), sym);
1016 if (scm_is_pair (handle)) {
1017 yylval = scm_cdr (handle);
1018 if (unsmob<Pitch> (yylval))
1019 return (YYSTATE == notes) ? NOTENAME_PITCH : TONICNAME_PITCH;
1020 else if (scm_is_symbol (yylval))
1023 else if ((YYSTATE == chords)
1024 && scm_is_true (handle = scm_hashq_get_handle (chordmodifier_tab_, sym)))
1026 yylval = scm_cdr (handle);
1027 return CHORD_MODIFIER;
1030 yylval = ly_string2scm (str);
1035 Lily_lexer::get_state () const
1041 Lily_lexer::is_note_state () const
1043 return get_state () == notes;
1047 Lily_lexer::is_chord_state () const
1049 return get_state () == chords;
1053 Lily_lexer::is_lyric_state () const
1055 return get_state () == lyrics;
1059 Lily_lexer::is_figure_state () const
1061 return get_state () == figures;
1064 // The extra_token parameter specifies how to convert multiple values
1065 // into additional tokens. For '#', additional values get pushed as
1066 // SCM_IDENTIFIER. For '$', they get checked for their type and get
1067 // pushed as a corresponding *_IDENTIFIER token. Since the latter
1068 // tampers with yylval, it can only be done from the lexer itself, so
1069 // this function is private.
1072 Lily_lexer::eval_scm (SCM readerdata, Input hi, char extra_token)
1074 SCM sval = SCM_UNDEFINED;
1076 if (!SCM_UNBNDP (readerdata))
1078 sval = ly_eval_scm (readerdata,
1080 be_safe_global && is_main_input_,
1084 if (SCM_UNBNDP (sval))
1087 return SCM_UNSPECIFIED;
1090 if (extra_token && SCM_VALUESP (sval))
1092 sval = scm_struct_ref (sval, SCM_INUM0);
1094 if (scm_is_pair (sval)) {
1095 for (SCM p = scm_reverse (scm_cdr (sval));
1099 SCM v = scm_car (p);
1100 if (Music *m = unsmob<Music> (v))
1102 if (!unsmob<Input> (m->get_property ("origin")))
1103 m->set_spot (override_input (here_input ()));
1107 switch (extra_token) {
1109 token = scan_scm_id (v);
1110 if (!scm_is_eq (yylval, SCM_UNSPECIFIED))
1111 push_extra_token (here_input (),
1115 push_extra_token (here_input (),
1120 sval = scm_car (sval);
1122 sval = SCM_UNSPECIFIED;
1125 if (Music *m = unsmob<Music> (sval))
1127 if (!unsmob<Input> (m->get_property ("origin")))
1128 m->set_spot (override_input (here_input ()));
1134 /* Check for valid UTF-8 that has no overlong or surrogate codes and
1135 is in the range 0-0x10ffff */
1138 Lily_lexer::YYText_utf8 ()
1140 const char * const p = YYText ();
1141 for (int i=0; p[i];) {
1142 if ((p[i] & 0xff) < 0x80) {
1146 int oldi = i; // start of character
1147 int more = 0; // # of followup bytes, 0 if bad
1148 switch (p[i++] & 0xff) {
1149 // 0xc0 and 0xc1 are overlong prefixes for
1150 // 0x00-0x3f and 0x40-0x7f respectively, bad.
1151 case 0xc2: // 0x80-0xbf
1152 case 0xc3: // 0xc0-0xff
1153 case 0xc4: // 0x100-0x13f
1154 case 0xc5: // 0x140-0x17f
1155 case 0xc6: // 0x180-0x1bf
1156 case 0xc7: // 0x1c0-0x1ff
1157 case 0xc8: // 0x200-0x23f
1158 case 0xc9: // 0x240-0x27f
1159 case 0xca: // 0x280-0x2bf
1160 case 0xcb: // 0x2c0-0x2ff
1161 case 0xcc: // 0x300-0x33f
1162 case 0xcd: // 0x340-0x37f
1163 case 0xce: // 0x380-0x3bf
1164 case 0xcf: // 0x3c0-0x3ff
1165 case 0xd0: // 0x400-0x43f
1166 case 0xd1: // 0x440-0x47f
1167 case 0xd2: // 0x480-0x4bf
1168 case 0xd3: // 0x4c0-0x4ff
1169 case 0xd4: // 0x500-0x53f
1170 case 0xd5: // 0x540-0x57f
1171 case 0xd6: // 0x580-0x5bf
1172 case 0xd7: // 0x5c0-0x5ff
1173 case 0xd8: // 0x600-0x63f
1174 case 0xd9: // 0x640-0x67f
1175 case 0xda: // 0x680-0x6bf
1176 case 0xdb: // 0x6c0-0x6ff
1177 case 0xdc: // 0x700-0x73f
1178 case 0xdd: // 0x740-0x77f
1179 case 0xde: // 0x780-0x7bf
1180 case 0xdf: // 0x7c0-0x7ff
1181 more = 1; // 2-byte sequences, 0x80-0x7ff
1184 // don't allow overlong sequences for 0-0x7ff
1185 if ((p[i] & 0xff) < 0xa0)
1187 case 0xe1: // 0x1000-0x1fff
1188 case 0xe2: // 0x2000-0x2fff
1189 case 0xe3: // 0x3000-0x3fff
1190 case 0xe4: // 0x4000-0x4fff
1191 case 0xe5: // 0x5000-0x5fff
1192 case 0xe6: // 0x6000-0x6fff
1193 case 0xe7: // 0x7000-0x7fff
1194 case 0xe8: // 0x8000-0x8fff
1195 case 0xe9: // 0x9000-0x9fff
1196 case 0xea: // 0xa000-0xafff
1197 case 0xeb: // 0xb000-0xbfff
1198 case 0xec: // 0xc000-0xcfff
1199 more = 2; // 3-byte sequences, 0x7ff-0xcfff
1201 case 0xed: // 0xd000-0xdfff
1202 // Don't allow surrogate codes 0xd800-0xdfff
1203 if ((p[i] & 0xff) >= 0xa0)
1205 case 0xee: // 0xe000-0xefff
1206 case 0xef: // 0xf000-0xffff
1207 more = 2; // 3-byte sequences,
1208 // 0xd000-0xd7ff, 0xe000-0xffff
1211 // don't allow overlong sequences for 0-0xffff
1212 if ((p[i] & 0xff) < 0x90)
1214 case 0xf1: // 0x40000-0x7ffff
1215 case 0xf2: // 0x80000-0xbffff
1216 case 0xf3: // 0xc0000-0xfffff
1217 more = 3; // 4-byte sequences, 0x10000-0xfffff
1220 // don't allow more than 0x10ffff
1221 if ((p[i] & 0xff) >= 0x90)
1223 more = 3; // 4-byte sequence, 0x100000-0x10ffff
1227 // check that all continuation bytes are valid
1229 if ((p[i++] & 0xc0) != 0x80)
1235 Input h = here_input ();
1236 h.set (h.get_source_file (), h.start () + oldi, h.start () + i);
1237 h.warning (_ ("non-UTF-8 input").c_str ());
1244 urg, belong to string (_convert)
1245 and should be generalised
1248 strip_leading_white (string&s)
1251 for (; i < s.length (); i++)
1252 if (!isspace (s[i]))
1259 strip_trailing_white (string&s)
1261 ssize i = s.length ();
1263 if (!isspace (s[i]))
1266 s = s.substr (0, i + 1);
1271 Lilypond_version oldest_version ("2.7.38");
1275 is_valid_version (string s)
1277 Lilypond_version current ( MAJOR_VERSION "." MINOR_VERSION "." PATCH_LEVEL );
1278 Lilypond_version ver (s);
1281 non_fatal_error (_f ("Invalid version string \"%s\"", s));
1284 if (ver < oldest_version)
1286 non_fatal_error (_f ("file too old: %s (oldest supported: %s)", ver.to_string (), oldest_version.to_string ()));
1287 non_fatal_error (_ ("consider updating the input with the convert-ly script"));
1293 non_fatal_error (_f ("program too old: %s (file requires: %s)", current.to_string (), ver.to_string ()));
1304 lyric_fudge (string s)
1308 while ((i = s.find ('_', i)) != string::npos)
1316 Convert "NUM/DEN" into a '(NUM . DEN) cons.
1319 scan_fraction (string frac)
1321 ssize i = frac.find ('/');
1322 string left = frac.substr (0, i);
1323 string right = frac.substr (i + 1, (frac.length () - i + 1));
1325 return scm_cons (scm_c_read_string (left.c_str ()),
1326 scm_c_read_string (right.c_str ()));
1330 lookup_markup_command (string s)
1332 return Lily::lookup_markup_command (ly_string2scm (s));
1336 lookup_markup_list_command (string s)
1338 return Lily::lookup_markup_list_command (ly_string2scm (s));
1341 /* Shut up lexer warnings. */
1361 silence_lexer_warnings ()
1363 (void) yy_start_stack_ptr;
1364 (void) yy_start_stack_depth;
1365 (void) yy_start_stack;
1366 (void) yy_push_state;
1367 (void) yy_pop_state;
1368 (void) yy_top_state;
1369 (void) silence_lexer_warnings;