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})*
158 REAL ({INT}\.{N}*)|(-?\.{N}+)
159 STRICTREAL {UNSIGNED}\.{UNSIGNED}
161 HORIZONTALWHITE [ \t]
167 BOM_UTF8 \357\273\277
173 // swallow and ignore carriage returns
176 <extratoken>{ANY_CHAR} {
177 /* Generate a token without swallowing anything */
179 /* First unswallow the eaten character */
180 add_lexed_char (-YYLeng ());
183 /* produce requested token */
184 int type = scm_to_int (scm_caar (extra_tokens_));
185 yylval = scm_cdar (extra_tokens_);
186 extra_tokens_ = scm_cdr (extra_tokens_);
187 if (scm_is_null (extra_tokens_))
193 <extratoken><<EOF>> {
194 /* Generate a token without swallowing anything */
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 /* Use the trailing context feature. Otherwise, the BOM will not be
207 found if the file starts with an identifier definition. */
208 <INITIAL,chords,lyrics,figures,notes>{BOM_UTF8}/.* {
209 if (this->lexloc_->line_number () != 1 || this->lexloc_->column_number () != 0)
211 LexerWarning (_ ("stray UTF-8 BOM encountered").c_str ());
214 debug_output (_ ("Skipping UTF-8 BOM"));
217 <INITIAL,chords,figures,incl,lyrics,markup,notes>{
219 yy_push_state (longcomment);
221 %[^{\n\r][^\n\r]*[\n\r]? {
222 (void) YYText_utf8 ();
231 <INITIAL,notes,figures,chords,markup>{
237 <INITIAL,chords,lyrics,notes,figures>\\version{WHITE}* {
238 yy_push_state (version);
240 <INITIAL,chords,lyrics,notes,figures>\\sourcefilename{WHITE}* {
241 yy_push_state (sourcefilename);
243 <INITIAL,chords,lyrics,notes,figures>\\sourcefileline{WHITE}* {
244 yy_push_state (sourcefileline);
246 <version>\"[^""]*\" { /* got the version number */
247 string s (YYText_utf8 () + 1);
248 s = s.substr (0, s.rfind ('\"'));
252 SCM top_scope = scm_car (scm_last_pair (scopes_));
253 scm_module_define (top_scope, ly_symbol2scm ("version-seen"), SCM_BOOL_T);
255 if (!is_valid_version (s)) {
256 yylval = SCM_UNSPECIFIED;
260 <sourcefilename>\"[^""]*\" {
261 string s (YYText_utf8 () + 1);
262 s = s.substr (0, s.rfind ('\"'));
265 this->here_input().get_source_file ()->name_ = s;
266 message (_f ("Renaming input to: `%s'", s.c_str ()));
267 progress_indication ("\n");
268 scm_module_define (scm_car (scopes_),
269 ly_symbol2scm ("input-file-name"),
274 <sourcefileline>{INT} {
276 sscanf (YYText (), "%d", &i);
279 this->here_input ().get_source_file ()->set_line (here_input ().start (), i);
282 <version>{ANY_CHAR} {
283 LexerError (_ ("quoted string expected after \\version").c_str ());
286 <sourcefilename>{ANY_CHAR} {
287 LexerError (_ ("quoted string expected after \\sourcefilename").c_str ());
290 <sourcefileline>{ANY_CHAR} {
291 LexerError (_ ("integer expected after \\sourcefileline").c_str ());
296 (void) YYText_utf8 ();
299 (void) YYText_utf8 ();
307 <INITIAL,chords,lyrics,notes,figures>\\maininput {
311 main_input_level_ = include_stack_.size ();
312 is_main_input_ = true;
314 yy_push_state (maininput);
315 yy_push_state (state);
318 LexerError (_ ("\\maininput not allowed outside init files").c_str ());
321 <INITIAL,chords,lyrics,figures,notes>\\include {
322 yy_push_state (incl);
324 <incl>\"[^""]*\" { /* got the include file name */
325 string s (YYText_utf8 ()+1);
326 s = s.substr (0, s.rfind ('"'));
328 new_input (s, sources_);
331 <incl>\\{BLACK}*{WHITE}? { /* got the include identifier */
332 string s = YYText_utf8 () + 1;
333 strip_trailing_white (s);
334 if (s.length () && (s[s.length () - 1] == ';'))
335 s = s.substr (0, s.length () - 1);
337 SCM sid = lookup_identifier (s);
338 if (scm_is_string (sid)) {
339 new_input (ly_scm2string (sid), sources_);
342 string msg (_f ("wrong or undefined identifier: `%s'", s ));
344 LexerError (msg.c_str ());
345 SCM err = scm_current_error_port ();
346 scm_puts ("This value was found in the table: ", err);
347 scm_display (sid, err);
350 <incl>(\$|#) { // scm for the filename
352 Input hi = here_input();
354 SCM sval = ly_parse_scm (hi.start (), &n, hi,
355 be_safe_global && is_main_input_, parser_);
356 sval = eval_scm (sval);
358 for (int i = 0; i < n; i++)
362 char_count_stack_.back () += n;
364 if (scm_is_string (sval)) {
365 new_input (ly_scm2string (sval), sources_);
368 LexerError (_ ("string expected after \\include").c_str ());
369 if (sval != SCM_UNDEFINED) {
370 SCM err = scm_current_error_port ();
371 scm_puts ("This value was found instead: ", err);
372 scm_display (sval, err);
377 <incl,version,sourcefilename>\"[^""]* { // backup rule
378 LexerError (_ ("end quote missing").c_str ());
382 /* Flex picks the longest matching pattern including trailing
383 * contexts. Without the backup pattern, r-. does not trigger the
384 * {RESTNAME} rule but rather the {WORD}/[-_] rule coming later,
385 * needed for avoiding backup states.
388 <chords,notes,figures>{RESTNAME}/[-_] | // pseudo backup rule
389 <chords,notes,figures>{RESTNAME} {
390 char const *s = YYText ();
391 yylval = scm_from_locale_string (s);
394 <chords,notes,figures>q/[-_] | // pseudo backup rule
395 <chords,notes,figures>q {
396 yylval = SCM_UNSPECIFIED;
397 return CHORD_REPETITION;
400 <chords,notes,figures>R/[-_] | // pseudo backup rule
401 <chords,notes,figures>R {
402 yylval = SCM_UNSPECIFIED;
403 return MULTI_MEASURE_REST;
405 <INITIAL,chords,figures,lyrics,markup,notes># { //embedded scm
407 Input hi = here_input();
409 SCM sval = ly_parse_scm (hi.start (), &n, hi,
410 be_safe_global && is_main_input_, parser_);
412 if (sval == SCM_UNDEFINED)
415 for (int i = 0; i < n; i++)
419 char_count_stack_.back () += n;
425 <INITIAL,chords,figures,lyrics,markup,notes>\$ { //immediate scm
427 Input hi = here_input();
429 SCM sval = ly_parse_scm (hi.start (), &n, hi,
430 be_safe_global && is_main_input_, parser_);
432 for (int i = 0; i < n; i++)
436 char_count_stack_.back () += n;
438 sval = eval_scm (sval, '$');
440 int token = scan_scm_id (sval);
441 if (!scm_is_eq (yylval, SCM_UNSPECIFIED))
445 <INITIAL,notes,lyrics>{
447 yylval = SCM_UNSPECIFIED;
448 return DOUBLE_ANGLE_OPEN;
451 yylval = SCM_UNSPECIFIED;
452 return DOUBLE_ANGLE_CLOSE;
458 yylval = SCM_UNSPECIFIED;
462 yylval = SCM_UNSPECIFIED;
469 yylval = SCM_UNSPECIFIED;
473 yylval = SCM_UNSPECIFIED;
477 yylval = SCM_UNSPECIFIED;
483 {WORD}/[-_] | // backup rule
485 return scan_bare_word (YYText_utf8 ());
488 {COMMAND}/[-_] | // backup rule
490 return scan_escaped_word (YYText_utf8 () + 1);
493 yylval = scan_fraction (YYText ());
497 yylval = scm_c_read_string (YYText ());
500 {UNSIGNED}/[/.] | // backup rule
502 yylval = scm_c_read_string (YYText ());
506 yylval = scm_c_read_string (YYText () + 1);
513 char c = escaped_char (YYText ()[1]);
514 yylval = scm_cons (scm_from_locale_stringn (&c, 1),
518 yylval = scm_cons (scm_from_locale_string (YYText_utf8 ()),
525 /* yylval is union. Must remember STRING before setting SCM*/
527 yylval = scm_string_concatenate_reverse (yylval,
534 yylval = scm_cons (scm_from_locale_string (YYText ()),
544 yylval = scan_fraction (YYText ());
548 yylval = scm_c_read_string (YYText ());
551 {UNSIGNED}/[/.] | // backup rule
553 yylval = scm_c_read_string (YYText ());
556 {COMMAND}/[-_] | // backup rule
558 return scan_escaped_word (YYText_utf8 () + 1);
560 /* Characters needed to express durations, assignments, barchecks */
562 yylval = SCM_UNSPECIFIED;
565 [^$#{}\"\\ \t\n\r\f0-9]+ {
567 string s (YYText_utf8 ());
568 yylval = SCM_UNSPECIFIED;
574 yylval = ly_string2scm (s);
578 /* This should really just cover {} */
580 yylval = SCM_UNSPECIFIED;
581 return YYText ()[0]; // above catches all multibytes.
585 {WORD}/[-_] | // backup rule
587 return scan_bare_word (YYText_utf8 ());
589 {COMMAND}/[-_] | // backup rule
591 return scan_escaped_word (YYText_utf8 () + 1);
594 yylval = scan_fraction (YYText ());
597 {UNSIGNED}/\/ | // backup rule
599 yylval = scm_c_read_string (YYText ());
603 yylval = SCM_UNSPECIFIED;
607 yylval = SCM_UNSPECIFIED;
611 yylval = SCM_UNSPECIFIED;
615 yylval = SCM_UNSPECIFIED;
619 yylval = SCM_UNSPECIFIED;
623 yylval = SCM_UNSPECIFIED;
624 return YYText ()[0]; // WORD catches all multibyte.
631 yylval = SCM_UNSPECIFIED;
634 {COMMAND}/[-_] | // backup rule
636 string str (YYText_utf8 () + 1);
638 int token_type = MARKUP_FUNCTION;
639 SCM s = lookup_markup_command (str);
641 // lookup-markup-command returns a pair with the car
642 // being the function to call, and the cdr being the
643 // call signature specified to define-markup-command,
644 // a list of predicates.
646 if (!scm_is_pair (s)) {
647 // If lookup-markup-command was not successful, we
648 // try lookup-markup-list-command instead.
649 // If this fails as well, we just scan and return
651 s = lookup_markup_list_command (str);
653 token_type = MARKUP_LIST_FUNCTION;
655 return scan_escaped_word (str);
658 // If the list of predicates is, say,
659 // (number? number? markup?), then tokens
660 // EXPECT_MARKUP EXPECT_SCM EXPECT_SCM EXPECT_NO_MORE_ARGS
661 // will be generated. Note that we have to push them
662 // in reverse order, so the first token pushed in the
663 // loop will be EXPECT_NO_MORE_ARGS.
667 // yylval now contains the function to call as token
668 // value (for token type MARKUP_FUNCTION or
669 // MARKUP_LIST_FUNCTION).
671 push_extra_token(EXPECT_NO_MORE_ARGS);
673 for (; scm_is_pair(s); s = scm_cdr(s)) {
674 SCM predicate = scm_car(s);
676 if (predicate == ly_lily_module_constant ("markup-list?"))
677 push_extra_token(EXPECT_MARKUP_LIST);
678 else if (predicate == ly_lily_module_constant ("markup?"))
679 push_extra_token(EXPECT_MARKUP);
681 push_extra_token(EXPECT_SCM, predicate);
685 [^$#{}\"\\ \t\n\r\f]+ {
686 string s (YYText_utf8 ());
688 yylval = ly_string2scm (s);
692 yylval = SCM_UNSPECIFIED;
693 return YYText ()[0]; // Above is catchall for multibyte
697 <longcomment><<EOF>> {
698 LexerError (_ ("EOF found inside a comment").c_str ());
703 LexerError (_ ("EOF found inside string").c_str ());
708 yylval = SCM_UNSPECIFIED;
711 is_main_input_ = include_stack_.size () > main_input_level_;
714 main_input_level_ = 0;
716 if (YYSTATE != maininput)
718 LexerError (_ ("Unfinished main input").c_str ());
721 } while (YYSTATE != maininput);
725 if (!close_input () || !is_main_input_)
726 /* Returns YY_NULL */
729 else if (!close_input ())
730 /* Returns YY_NULL */
734 <maininput>{ANY_CHAR} {
735 while (include_stack_.size () > main_input_level_
742 {WORD}/[-_] | // backup rule
744 return scan_bare_word (YYText_utf8 ());
746 {COMMAND}/[-_] | // backup rule
748 return scan_escaped_word (YYText_utf8 () + 1);
753 yylval = scan_fraction (YYText ());
757 -{UNSIGNED} | // backup rule
759 yylval = scm_c_read_string (YYText ());
763 {UNSIGNED}/\/ | // backup rule
765 yylval = scm_c_read_string (YYText ());
771 yylval = SCM_UNSPECIFIED;
775 -/\. | // backup rule
777 yylval = SCM_UNSPECIFIED;
781 <INITIAL,notes,figures>. {
782 yylval = SCM_UNSPECIFIED;
786 <INITIAL,lyrics,notes,figures>\\. {
787 yylval = SCM_UNSPECIFIED;
788 char c = YYText ()[1];
792 return E_ANGLE_CLOSE;
796 return E_EXCLAMATION;
802 return E_BRACKET_OPEN;
806 return E_BRACKET_CLOSE;
818 string msg = _f ("invalid character: `%s'", YYText_utf8 ());
819 LexerError (msg.c_str ());
820 yylval = SCM_UNSPECIFIED;
821 return '%'; // Better not return half a utf8 character.
826 /* Make the lexer generate a token of the given type as the next token.
827 TODO: make it possible to define a value for the token as well */
829 Lily_lexer::push_extra_token (int token_type, SCM scm)
831 if (scm_is_null (extra_tokens_))
833 if (YY_START != extratoken)
834 hidden_state_ = YY_START;
835 yy_push_state (extratoken);
837 extra_tokens_ = scm_acons (scm_from_int (token_type), scm, 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 bool extra = (YYSTATE == extratoken);
880 SCM p = scm_assq (alist, pitchname_tab_stack_);
885 if (scm_is_false (p))
886 p = scm_cons (alist, alist_to_hashq (alist));
887 pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_);
888 yy_push_state (notes);
891 hidden_state_ = YYSTATE;
892 yy_push_state (extratoken);
897 Lily_lexer::pop_state ()
899 bool extra = (YYSTATE == extratoken);
904 if (YYSTATE == notes || YYSTATE == chords)
905 pitchname_tab_stack_ = scm_cdr (pitchname_tab_stack_);
910 hidden_state_ = YYSTATE;
911 yy_push_state (extratoken);
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 (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 (last_input_));
942 if (sid != SCM_UNDEFINED)
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);
954 Lily_lexer::scan_scm_id (SCM sid)
956 if (is_music_function (sid))
958 int funtype = SCM_FUNCTION;
962 SCM s = get_music_function_signature (sid);
963 SCM cs = scm_car (s);
965 if (scm_is_pair (cs))
970 if (scm_is_eq (cs, ly_lily_module_constant ("ly:music?")))
971 funtype = MUSIC_FUNCTION;
972 else if (scm_is_eq (cs, ly_lily_module_constant ("ly:event?")))
973 funtype = EVENT_FUNCTION;
974 else if (ly_is_procedure (cs))
975 funtype = SCM_FUNCTION;
976 else programming_error ("Bad syntax function predicate");
978 push_extra_token (EXPECT_NO_MORE_ARGS);
979 for (s = scm_cdr (s); scm_is_pair (s); s = scm_cdr (s))
981 SCM optional = SCM_UNDEFINED;
984 if (scm_is_pair (cs))
986 optional = SCM_CDR (cs);
990 if (cs == Pitch_type_p_proc)
991 push_extra_token (EXPECT_PITCH);
992 else if (cs == Duration_type_p_proc)
993 push_extra_token (EXPECT_DURATION);
994 else if (ly_is_procedure (cs))
995 push_extra_token (EXPECT_SCM, cs);
998 programming_error ("Function parameter without type-checking predicate");
1001 if (!scm_is_eq (optional, SCM_UNDEFINED))
1002 push_extra_token (EXPECT_OPTIONAL, optional);
1007 return identifier_type (sid);
1011 Lily_lexer::scan_bare_word (string str)
1013 SCM sym = ly_symbol2scm (str.c_str ());
1014 if ((YYSTATE == notes) || (YYSTATE == chords)) {
1015 SCM handle = SCM_BOOL_F;
1016 if (scm_is_pair (pitchname_tab_stack_))
1017 handle = scm_hashq_get_handle (scm_cdar (pitchname_tab_stack_), sym);
1019 if (scm_is_pair (handle)) {
1020 yylval = scm_cdr (handle);
1021 if (unsmob_pitch (yylval))
1022 return (YYSTATE == notes) ? NOTENAME_PITCH : TONICNAME_PITCH;
1023 else if (scm_is_symbol (yylval))
1026 else if ((YYSTATE == chords)
1027 && (handle = scm_hashq_get_handle (chordmodifier_tab_, sym))!= SCM_BOOL_F)
1029 yylval = scm_cdr (handle);
1030 return CHORD_MODIFIER;
1033 yylval = ly_string2scm (str);
1038 Lily_lexer::get_state () const
1040 if (YY_START == extratoken)
1041 return hidden_state_;
1047 Lily_lexer::is_note_state () const
1049 return get_state () == notes;
1053 Lily_lexer::is_chord_state () const
1055 return get_state () == chords;
1059 Lily_lexer::is_lyric_state () const
1061 return get_state () == lyrics;
1065 Lily_lexer::is_figure_state () const
1067 return get_state () == figures;
1070 // The extra_token parameter specifies how to convert multiple values
1071 // into additional tokens. For '#', additional values get pushed as
1072 // SCM_IDENTIFIER. For '$', they get checked for their type and get
1073 // pushed as a corresponding *_IDENTIFIER token. Since the latter
1074 // tampers with yylval, it can only be done from the lexer itself, so
1075 // this function is private.
1078 Lily_lexer::eval_scm (SCM readerdata, char extra_token)
1080 SCM sval = SCM_UNDEFINED;
1082 if (!SCM_UNBNDP (readerdata))
1084 sval = ly_eval_scm (scm_car (readerdata),
1085 *unsmob_input (scm_cdr (readerdata)),
1086 be_safe_global && is_main_input_,
1090 if (SCM_UNBNDP (sval))
1093 return SCM_UNSPECIFIED;
1096 if (extra_token && SCM_VALUESP (sval))
1098 sval = scm_struct_ref (sval, SCM_INUM0);
1100 if (scm_is_pair (sval)) {
1101 for (SCM p = scm_reverse (scm_cdr (sval));
1105 SCM v = scm_car (p);
1106 if (Music *m = unsmob_music (v))
1108 if (!unsmob_input (m->get_property ("origin")))
1109 m->set_spot (override_input (last_input_));
1113 switch (extra_token) {
1115 token = scan_scm_id (v);
1116 if (!scm_is_eq (yylval, SCM_UNSPECIFIED))
1117 push_extra_token (token, yylval);
1120 push_extra_token (SCM_IDENTIFIER, v);
1124 sval = scm_car (sval);
1126 sval = SCM_UNSPECIFIED;
1129 if (Music *m = unsmob_music (sval))
1131 if (!unsmob_input (m->get_property ("origin")))
1132 m->set_spot (override_input (last_input_));
1138 /* Check for valid UTF-8 that has no overlong or surrogate codes and
1139 is in the range 0-0x10ffff */
1142 Lily_lexer::YYText_utf8 ()
1144 const char * const p = YYText ();
1145 for (int i=0; p[i];) {
1146 if ((p[i] & 0xff) < 0x80) {
1150 int oldi = i; // start of character
1151 int more = 0; // # of followup bytes, 0 if bad
1152 switch (p[i++] & 0xff) {
1153 // 0xc0 and 0xc1 are overlong prefixes for
1154 // 0x00-0x3f and 0x40-0x7f respectively, bad.
1155 case 0xc2: // 0x80-0xbf
1156 case 0xc3: // 0xc0-0xff
1157 case 0xc4: // 0x100-0x13f
1158 case 0xc5: // 0x140-0x17f
1159 case 0xc6: // 0x180-0x1bf
1160 case 0xc7: // 0x1c0-0x1ff
1161 case 0xc8: // 0x200-0x23f
1162 case 0xc9: // 0x240-0x27f
1163 case 0xca: // 0x280-0x2bf
1164 case 0xcb: // 0x2c0-0x2ff
1165 case 0xcc: // 0x300-0x33f
1166 case 0xcd: // 0x340-0x37f
1167 case 0xce: // 0x380-0x3bf
1168 case 0xcf: // 0x3c0-0x3ff
1169 case 0xd0: // 0x400-0x43f
1170 case 0xd1: // 0x440-0x47f
1171 case 0xd2: // 0x480-0x4bf
1172 case 0xd3: // 0x4c0-0x4ff
1173 case 0xd4: // 0x500-0x53f
1174 case 0xd5: // 0x540-0x57f
1175 case 0xd6: // 0x580-0x5bf
1176 case 0xd7: // 0x5c0-0x5ff
1177 case 0xd8: // 0x600-0x63f
1178 case 0xd9: // 0x640-0x67f
1179 case 0xda: // 0x680-0x6bf
1180 case 0xdb: // 0x6c0-0x6ff
1181 case 0xdc: // 0x700-0x73f
1182 case 0xdd: // 0x740-0x77f
1183 case 0xde: // 0x780-0x7bf
1184 case 0xdf: // 0x7c0-0x7ff
1185 more = 1; // 2-byte sequences, 0x80-0x7ff
1188 // don't allow overlong sequences for 0-0x7ff
1189 if ((p[i] & 0xff) < 0xa0)
1191 case 0xe1: // 0x1000-0x1fff
1192 case 0xe2: // 0x2000-0x2fff
1193 case 0xe3: // 0x3000-0x3fff
1194 case 0xe4: // 0x4000-0x4fff
1195 case 0xe5: // 0x5000-0x5fff
1196 case 0xe6: // 0x6000-0x6fff
1197 case 0xe7: // 0x7000-0x7fff
1198 case 0xe8: // 0x8000-0x8fff
1199 case 0xe9: // 0x9000-0x9fff
1200 case 0xea: // 0xa000-0xafff
1201 case 0xeb: // 0xb000-0xbfff
1202 case 0xec: // 0xc000-0xcfff
1203 more = 2; // 3-byte sequences, 0x7ff-0xcfff
1205 case 0xed: // 0xd000-0xdfff
1206 // Don't allow surrogate codes 0xd800-0xdfff
1207 if ((p[i] & 0xff) >= 0xa0)
1209 case 0xee: // 0xe000-0xefff
1210 case 0xef: // 0xf000-0xffff
1211 more = 2; // 3-byte sequences,
1212 // 0xd000-0xd7ff, 0xe000-0xffff
1215 // don't allow overlong sequences for 0-0xffff
1216 if ((p[i] & 0xff) < 0x90)
1218 case 0xf1: // 0x40000-0x7ffff
1219 case 0xf2: // 0x80000-0xbffff
1220 case 0xf3: // 0xc0000-0xfffff
1221 more = 3; // 4-byte sequences, 0x10000-0xfffff
1224 // don't allow more than 0x10ffff
1225 if ((p[i] & 0xff) >= 0x90)
1227 more = 3; // 4-byte sequence, 0x100000-0x10ffff
1231 // check that all continuation bytes are valid
1233 if ((p[i++] & 0xc0) != 0x80)
1239 Input h = here_input ();
1240 h.set (h.get_source_file (), h.start () + oldi, h.start () + i);
1241 h.warning (_ ("non-UTF-8 input").c_str ());
1248 urg, belong to string (_convert)
1249 and should be generalised
1252 strip_leading_white (string&s)
1255 for (; i < s.length (); i++)
1256 if (!isspace (s[i]))
1263 strip_trailing_white (string&s)
1265 ssize i = s.length ();
1267 if (!isspace (s[i]))
1270 s = s.substr (0, i + 1);
1275 Lilypond_version oldest_version ("2.7.38");
1279 is_valid_version (string s)
1281 Lilypond_version current ( MAJOR_VERSION "." MINOR_VERSION "." PATCH_LEVEL );
1282 Lilypond_version ver (s);
1283 if (int (ver) < oldest_version)
1285 non_fatal_error (_f ("file too old: %s (oldest supported: %s)", ver.to_string (), oldest_version.to_string ()));
1286 non_fatal_error (_ ("consider updating the input with the convert-ly script"));
1292 non_fatal_error (_f ("program too old: %s (file requires: %s)", current.to_string (), ver.to_string ()));
1303 lyric_fudge (string s)
1307 while ((i = s.find ('_', i)) != string::npos)
1315 Convert "NUM/DEN" into a '(NUM . DEN) cons.
1318 scan_fraction (string frac)
1320 ssize i = frac.find ('/');
1321 string left = frac.substr (0, i);
1322 string right = frac.substr (i + 1, (frac.length () - i + 1));
1324 int n = String_convert::dec2int (left);
1325 int d = String_convert::dec2int (right);
1326 return scm_cons (scm_from_int (n), scm_from_int (d));
1330 lookup_markup_command (string s)
1332 SCM proc = ly_lily_module_constant ("lookup-markup-command");
1333 return scm_call_1 (proc, ly_string2scm (s));
1337 lookup_markup_list_command (string s)
1339 SCM proc = ly_lily_module_constant ("lookup-markup-list-command");
1340 return scm_call_1 (proc, ly_string2scm (s));
1343 /* Shut up lexer warnings. */
1363 silence_lexer_warnings ()
1365 (void) yy_start_stack_ptr;
1366 (void) yy_start_stack_depth;
1367 (void) yy_start_stack;
1368 (void) yy_push_state;
1369 (void) yy_pop_state;
1370 (void) yy_top_state;
1371 (void) silence_lexer_warnings;