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
130 /* The strategy concerning multibyte characters is to accept them but
131 * call YYText_utf8 for patterns that might contain them, in order to
132 * get a single code path responsible for flagging non-UTF-8 input:
133 * Patterns for accepting only valid UTF-8 without backing up are
134 * really hard to do and complex, and if nice error messages are
135 * wanted, one would need patterns catching the invalid input as well.
137 * Since editors and operating environments don't necessarily behave
138 * reasonably in the presence of mixed encodings, we flag encoding
139 * errors also in identifiers, comments, and strings where it would be
140 * conceivable to just transparently work with the byte string. But
141 * the whole point of caring about UTF-8 in here at all is too avoid
142 * stranger errors later when input passes into backends or log files
143 * or console output or error messages.
150 WORD {A}([-_]{A}|{A})*
157 REAL ({INT}\.{N}*)|(-?\.{N}+)
159 HORIZONTALWHITE [ \t]
165 BOM_UTF8 \357\273\277
171 // swallow and ignore carriage returns
174 <extratoken>{ANY_CHAR} {
175 /* Generate a token without swallowing anything */
177 /* First unswallow the eaten character */
178 add_lexed_char (-YYLeng ());
181 /* produce requested token */
182 int type = scm_to_int (scm_caar (extra_tokens_));
183 yylval = scm_cdar (extra_tokens_);
184 extra_tokens_ = scm_cdr (extra_tokens_);
185 if (scm_is_null (extra_tokens_))
191 <extratoken><<EOF>> {
192 /* Generate a token without swallowing anything */
194 /* produce requested token */
195 int type = scm_to_int (scm_caar (extra_tokens_));
196 yylval = scm_cdar (extra_tokens_);
197 extra_tokens_ = scm_cdr (extra_tokens_);
198 if (scm_is_null (extra_tokens_))
204 /* Use the trailing context feature. Otherwise, the BOM will not be
205 found if the file starts with an identifier definition. */
206 <INITIAL,chords,lyrics,figures,notes>{BOM_UTF8}/.* {
207 if (this->lexloc_->line_number () != 1 || this->lexloc_->column_number () != 0)
209 LexerWarning (_ ("stray UTF-8 BOM encountered").c_str ());
212 debug_output (_ ("Skipping UTF-8 BOM"));
215 <INITIAL,chords,figures,incl,lyrics,markup,notes>{
217 yy_push_state (longcomment);
219 %[^{\n\r][^\n\r]*[\n\r] {
220 (void) YYText_utf8 ();
222 %[^{\n\r] { // backup rule
223 (void) YYText_utf8 ();
228 (void) YYText_utf8 ();
235 <INITIAL,notes,figures,chords,markup>{
241 <INITIAL,chords,lyrics,notes,figures>\\version{WHITE}* {
242 yy_push_state (version);
244 <INITIAL,chords,lyrics,notes,figures>\\sourcefilename{WHITE}* {
245 yy_push_state (sourcefilename);
247 <INITIAL,chords,lyrics,notes,figures>\\sourcefileline{WHITE}* {
248 yy_push_state (sourcefileline);
250 <version>\"[^""]*\" { /* got the version number */
251 string s (YYText_utf8 () + 1);
252 s = s.substr (0, s.rfind ('\"'));
256 SCM top_scope = scm_car (scm_last_pair (scopes_));
257 scm_module_define (top_scope, ly_symbol2scm ("version-seen"), SCM_BOOL_T);
259 if (!is_valid_version (s)) {
260 yylval = SCM_UNSPECIFIED;
264 <sourcefilename>\"[^""]*\" {
265 string s (YYText_utf8 () + 1);
266 s = s.substr (0, s.rfind ('\"'));
269 this->here_input().get_source_file ()->name_ = s;
270 message (_f ("Renaming input to: `%s'", s.c_str ()));
271 progress_indication ("\n");
272 scm_module_define (scm_car (scopes_),
273 ly_symbol2scm ("input-file-name"),
278 <sourcefileline>{INT} {
280 sscanf (YYText (), "%d", &i);
283 this->here_input ().get_source_file ()->set_line (here_input ().start (), i);
286 <version>{ANY_CHAR} {
287 LexerError (_ ("quoted string expected after \\version").c_str ());
290 <sourcefilename>{ANY_CHAR} {
291 LexerError (_ ("quoted string expected after \\sourcefilename").c_str ());
294 <sourcefileline>{ANY_CHAR} {
295 LexerError (_ ("integer expected after \\sourcefileline").c_str ());
300 (void) YYText_utf8 ();
303 (void) YYText_utf8 ();
311 <INITIAL,chords,lyrics,notes,figures>\\maininput {
315 main_input_level_ = include_stack_.size ();
316 is_main_input_ = true;
319 error (_ ("\\maininput not allowed outside init files"));
322 <INITIAL,chords,lyrics,figures,notes>\\include {
323 yy_push_state (incl);
325 <incl>\"[^""]*\" { /* got the include file name */
326 string s (YYText_utf8 ()+1);
327 s = s.substr (0, s.rfind ('"'));
329 new_input (s, sources_);
332 <incl>\\{BLACK}*{WHITE}? { /* got the include identifier */
333 string s = YYText_utf8 () + 1;
334 strip_trailing_white (s);
335 if (s.length () && (s[s.length () - 1] == ';'))
336 s = s.substr (0, s.length () - 1);
338 SCM sid = lookup_identifier (s);
339 if (scm_is_string (sid)) {
340 new_input (ly_scm2string (sid), sources_);
343 string msg (_f ("wrong or undefined identifier: `%s'", s ));
345 LexerError (msg.c_str ());
346 SCM err = scm_current_error_port ();
347 scm_puts ("This value was found in the table: ", err);
348 scm_display (sid, err);
351 <incl>(\$|#) { // scm for the filename
353 Input hi = here_input();
355 SCM sval = ly_parse_scm (hi.start (), &n, hi,
356 be_safe_global && is_main_input_, parser_);
357 sval = eval_scm (sval);
359 for (int i = 0; i < n; i++)
363 char_count_stack_.back () += n;
365 if (scm_is_string (sval)) {
366 new_input (ly_scm2string (sval), sources_);
369 LexerError (_ ("string expected after \\include").c_str ());
370 if (sval != SCM_UNDEFINED) {
371 SCM err = scm_current_error_port ();
372 scm_puts ("This value was found instead: ", err);
373 scm_display (sval, err);
378 <incl,version,sourcefilename>\"[^""]* { // backup rule
379 error (_ ("end quote missing"));
383 /* Flex picks the longest matching pattern including trailing
384 * contexts. Without the backup pattern, r-. does not trigger the
385 * {RESTNAME} rule but rather the {WORD}/[-_] rule coming later,
386 * needed for avoiding backup states.
389 <chords,notes,figures>{RESTNAME}/[-_] | // pseudo backup rule
390 <chords,notes,figures>{RESTNAME} {
391 char const *s = YYText ();
392 yylval = scm_from_locale_string (s);
395 <chords,notes,figures>q/[-_] | // pseudo backup rule
396 <chords,notes,figures>q {
397 yylval = SCM_UNSPECIFIED;
398 return CHORD_REPETITION;
401 <chords,notes,figures>R/[-_] | // pseudo backup rule
402 <chords,notes,figures>R {
403 yylval = SCM_UNSPECIFIED;
404 return MULTI_MEASURE_REST;
406 <INITIAL,chords,figures,lyrics,markup,notes># { //embedded scm
408 Input hi = here_input();
410 SCM sval = ly_parse_scm (hi.start (), &n, hi,
411 be_safe_global && is_main_input_, parser_);
413 if (sval == SCM_UNDEFINED)
416 for (int i = 0; i < n; i++)
420 char_count_stack_.back () += n;
426 <INITIAL,chords,figures,lyrics,markup,notes>\$ { //immediate scm
428 Input hi = here_input();
430 SCM sval = ly_parse_scm (hi.start (), &n, hi,
431 be_safe_global && is_main_input_, parser_);
433 for (int i = 0; i < n; i++)
437 char_count_stack_.back () += n;
439 sval = eval_scm (sval, '$');
441 int token = scan_scm_id (sval);
442 if (!scm_is_eq (yylval, SCM_UNSPECIFIED))
446 <INITIAL,notes,lyrics>{
448 yylval = SCM_UNSPECIFIED;
449 return DOUBLE_ANGLE_OPEN;
452 yylval = SCM_UNSPECIFIED;
453 return DOUBLE_ANGLE_CLOSE;
459 yylval = SCM_UNSPECIFIED;
463 yylval = SCM_UNSPECIFIED;
470 yylval = SCM_UNSPECIFIED;
474 yylval = SCM_UNSPECIFIED;
478 yylval = SCM_UNSPECIFIED;
484 {WORD}/[-_] | // backup rule
486 return scan_bare_word (YYText_utf8 ());
489 {COMMAND}/[-_] | // backup rule
491 return scan_escaped_word (YYText_utf8 () + 1);
494 yylval = scan_fraction (YYText ());
497 {UNSIGNED}/\/ | // backup rule
499 yylval = scm_c_read_string (YYText ());
503 yylval = scm_c_read_string (YYText () + 1);
510 char c = escaped_char (YYText ()[1]);
511 yylval = scm_cons (scm_from_locale_stringn (&c, 1),
515 yylval = scm_cons (scm_from_locale_string (YYText_utf8 ()),
522 /* yylval is union. Must remember STRING before setting SCM*/
524 yylval = scm_string_concatenate_reverse (yylval,
528 return is_lyric_state () ? LYRICS_STRING : STRING;
531 yylval = scm_cons (scm_from_locale_string (YYText ()),
541 yylval = scan_fraction (YYText ());
544 {UNSIGNED}/\/ | // backup rule
546 yylval = scm_c_read_string (YYText ());
549 {COMMAND}/[-_] | // backup rule
551 return scan_escaped_word (YYText_utf8 () + 1);
553 /* Characters needed to express durations, assignments, barchecks */
555 yylval = SCM_UNSPECIFIED;
558 [^$#{}\"\\ \t\n\r\f0-9]+ {
560 string s (YYText_utf8 ());
561 yylval = SCM_UNSPECIFIED;
567 yylval = ly_string2scm (s);
569 return LYRICS_STRING;
571 /* This should really just cover {} */
573 yylval = SCM_UNSPECIFIED;
574 return YYText ()[0]; // above catches all multibytes.
578 {WORD}/[-_] | // backup rule
580 return scan_bare_word (YYText_utf8 ());
582 {COMMAND}/[-_] | // backup rule
584 return scan_escaped_word (YYText_utf8 () + 1);
587 yylval = scan_fraction (YYText ());
590 {UNSIGNED}/\/ | // backup rule
592 yylval = scm_c_read_string (YYText ());
596 yylval = SCM_UNSPECIFIED;
600 yylval = SCM_UNSPECIFIED;
604 yylval = SCM_UNSPECIFIED;
608 yylval = SCM_UNSPECIFIED;
612 yylval = SCM_UNSPECIFIED;
616 yylval = SCM_UNSPECIFIED;
617 return YYText ()[0]; // WORD catches all multibyte.
624 yylval = SCM_UNSPECIFIED;
627 {COMMAND}/[-_] | // backup rule
629 string str (YYText_utf8 () + 1);
631 int token_type = MARKUP_FUNCTION;
632 SCM s = lookup_markup_command (str);
634 // lookup-markup-command returns a pair with the car
635 // being the function to call, and the cdr being the
636 // call signature specified to define-markup-command,
637 // a list of predicates.
639 if (!scm_is_pair (s)) {
640 // If lookup-markup-command was not successful, we
641 // try lookup-markup-list-command instead.
642 // If this fails as well, we just scan and return
644 s = lookup_markup_list_command (str);
646 token_type = MARKUP_LIST_FUNCTION;
648 return scan_escaped_word (str);
651 // If the list of predicates is, say,
652 // (number? number? markup?), then tokens
653 // EXPECT_MARKUP EXPECT_SCM EXPECT_SCM EXPECT_NO_MORE_ARGS
654 // will be generated. Note that we have to push them
655 // in reverse order, so the first token pushed in the
656 // loop will be EXPECT_NO_MORE_ARGS.
660 // yylval now contains the function to call as token
661 // value (for token type MARKUP_FUNCTION or
662 // MARKUP_LIST_FUNCTION).
664 push_extra_token(EXPECT_NO_MORE_ARGS);
666 for (; scm_is_pair(s); s = scm_cdr(s)) {
667 SCM predicate = scm_car(s);
669 if (predicate == ly_lily_module_constant ("markup-list?"))
670 push_extra_token(EXPECT_MARKUP_LIST);
671 else if (predicate == ly_lily_module_constant ("markup?"))
672 push_extra_token(EXPECT_MARKUP);
674 push_extra_token(EXPECT_SCM, predicate);
678 [^$#{}\"\\ \t\n\r\f]+ {
679 string s (YYText_utf8 ());
681 yylval = ly_string2scm (s);
685 yylval = SCM_UNSPECIFIED;
686 return YYText ()[0]; // Above is catchall for multibyte
690 <longcomment><<EOF>> {
691 LexerError (_ ("EOF found inside a comment").c_str ());
692 is_main_input_ = false; // should be safe , can't have \include in --safe.
693 yylval = SCM_UNSPECIFIED;
695 yyterminate (); // can't move this, since it actually rets a YY_NULL
699 yylval = SCM_UNSPECIFIED;
702 is_main_input_ = include_stack_.size () > main_input_level_;
704 main_input_level_ = 0;
705 if (!close_input () || !is_main_input_)
706 /* Returns YY_NULL */
709 else if (!close_input ())
710 /* Returns YY_NULL */
715 {WORD}/[-_] | // backup rule
717 return scan_bare_word (YYText_utf8 ());
719 {COMMAND}/[-_] | // backup rule
721 return scan_escaped_word (YYText_utf8 () + 1);
726 yylval = scan_fraction (YYText ());
730 -{UNSIGNED} | // backup rule
732 yylval = scm_c_read_string (YYText ());
736 {UNSIGNED}/\/ | // backup rule
738 yylval = scm_c_read_string (YYText ());
744 yylval = SCM_UNSPECIFIED;
748 -/\. | // backup rule
750 yylval = SCM_UNSPECIFIED;
754 <INITIAL,notes,figures>. {
755 yylval = SCM_UNSPECIFIED;
759 <INITIAL,lyrics,notes,figures>\\. {
760 yylval = SCM_UNSPECIFIED;
761 char c = YYText ()[1];
765 return E_ANGLE_CLOSE;
769 return E_EXCLAMATION;
775 return E_BRACKET_OPEN;
779 return E_BRACKET_CLOSE;
791 string msg = _f ("invalid character: `%s'", YYText_utf8 ());
792 LexerError (msg.c_str ());
793 yylval = SCM_UNSPECIFIED;
794 return '%'; // Better not return half a utf8 character.
799 /* Make the lexer generate a token of the given type as the next token.
800 TODO: make it possible to define a value for the token as well */
802 Lily_lexer::push_extra_token (int token_type, SCM scm)
804 if (scm_is_null (extra_tokens_))
806 if (YY_START != extratoken)
807 hidden_state_ = YY_START;
808 yy_push_state (extratoken);
810 extra_tokens_ = scm_acons (scm_from_int (token_type), scm, extra_tokens_);
814 Lily_lexer::push_chord_state (SCM alist)
816 SCM p = scm_assq (alist, pitchname_tab_stack_);
818 if (scm_is_false (p))
819 p = scm_cons (alist, alist_to_hashq (alist));
820 pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_);
821 yy_push_state (chords);
825 Lily_lexer::push_figuredbass_state ()
827 yy_push_state (figures);
831 Lily_lexer::push_initial_state ()
833 yy_push_state (INITIAL);
837 Lily_lexer::push_lyric_state ()
839 yy_push_state (lyrics);
843 Lily_lexer::push_markup_state ()
845 yy_push_state (markup);
849 Lily_lexer::push_note_state (SCM alist)
851 bool extra = (YYSTATE == extratoken);
853 SCM p = scm_assq (alist, pitchname_tab_stack_);
858 if (scm_is_false (p))
859 p = scm_cons (alist, alist_to_hashq (alist));
860 pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_);
861 yy_push_state (notes);
864 hidden_state_ = YYSTATE;
865 yy_push_state (extratoken);
870 Lily_lexer::pop_state ()
872 bool extra = (YYSTATE == extratoken);
877 if (YYSTATE == notes || YYSTATE == chords)
878 pitchname_tab_stack_ = scm_cdr (pitchname_tab_stack_);
883 hidden_state_ = YYSTATE;
884 yy_push_state (extratoken);
889 Lily_lexer::identifier_type (SCM sid)
891 int k = try_special_identifiers (&yylval , sid);
892 return k >= 0 ? k : SCM_IDENTIFIER;
897 Lily_lexer::scan_escaped_word (string str)
899 // use more SCM for this.
901 // SCM sym = ly_symbol2scm (str.c_str ());
903 yylval = SCM_UNSPECIFIED;
904 int i = lookup_keyword (str);
905 if (i == MARKUP && is_lyric_state ())
910 SCM sid = lookup_identifier (str);
911 if (sid != SCM_UNDEFINED)
912 return scan_scm_id (sid);
914 string msg (_f ("unknown escaped string: `\\%s'", str));
915 LexerError (msg.c_str ());
917 yylval = ly_string2scm (str);
923 Lily_lexer::scan_scm_id (SCM sid)
925 if (is_music_function (sid))
927 int funtype = SCM_FUNCTION;
931 SCM s = get_music_function_signature (sid);
932 SCM cs = scm_car (s);
934 if (scm_is_pair (cs))
939 if (scm_is_eq (cs, ly_lily_module_constant ("ly:music?")))
940 funtype = MUSIC_FUNCTION;
941 else if (scm_is_eq (cs, ly_lily_module_constant ("ly:event?")))
942 funtype = EVENT_FUNCTION;
943 else if (ly_is_procedure (cs))
944 funtype = SCM_FUNCTION;
945 else programming_error ("Bad syntax function predicate");
947 push_extra_token (EXPECT_NO_MORE_ARGS);
948 for (s = scm_cdr (s); scm_is_pair (s); s = scm_cdr (s))
950 SCM optional = SCM_UNDEFINED;
953 if (scm_is_pair (cs))
955 optional = SCM_CDR (cs);
959 if (cs == Pitch_type_p_proc)
960 push_extra_token (EXPECT_PITCH);
961 else if (cs == Duration_type_p_proc)
962 push_extra_token (EXPECT_DURATION);
963 else if (ly_is_procedure (cs))
964 push_extra_token (EXPECT_SCM, cs);
967 programming_error ("Function parameter without type-checking predicate");
970 if (!scm_is_eq (optional, SCM_UNDEFINED))
971 push_extra_token (EXPECT_OPTIONAL, optional);
976 return identifier_type (sid);
980 Lily_lexer::scan_bare_word (string str)
982 SCM sym = ly_symbol2scm (str.c_str ());
983 if ((YYSTATE == notes) || (YYSTATE == chords)) {
984 SCM handle = SCM_BOOL_F;
985 if (scm_is_pair (pitchname_tab_stack_))
986 handle = scm_hashq_get_handle (scm_cdar (pitchname_tab_stack_), sym);
988 if (scm_is_pair (handle)) {
989 yylval = scm_cdr (handle);
990 if (unsmob_pitch (yylval))
991 return (YYSTATE == notes) ? NOTENAME_PITCH : TONICNAME_PITCH;
992 else if (scm_is_symbol (yylval))
995 else if ((YYSTATE == chords)
996 && (handle = scm_hashq_get_handle (chordmodifier_tab_, sym))!= SCM_BOOL_F)
998 yylval = scm_cdr (handle);
999 return CHORD_MODIFIER;
1002 yylval = ly_string2scm (str);
1007 Lily_lexer::get_state () const
1009 if (YY_START == extratoken)
1010 return hidden_state_;
1016 Lily_lexer::is_note_state () const
1018 return get_state () == notes;
1022 Lily_lexer::is_chord_state () const
1024 return get_state () == chords;
1028 Lily_lexer::is_lyric_state () const
1030 return get_state () == lyrics;
1034 Lily_lexer::is_figure_state () const
1036 return get_state () == figures;
1039 // The extra_token parameter specifies how to convert multiple values
1040 // into additional tokens. For '#', additional values get pushed as
1041 // SCM_IDENTIFIER. For '$', they get checked for their type and get
1042 // pushed as a corresponding *_IDENTIFIER token. Since the latter
1043 // tampers with yylval, it can only be done from the lexer itself, so
1044 // this function is private.
1047 Lily_lexer::eval_scm (SCM readerdata, char extra_token)
1049 SCM sval = SCM_UNDEFINED;
1051 if (!SCM_UNBNDP (readerdata))
1053 sval = ly_eval_scm (scm_car (readerdata),
1054 *unsmob_input (scm_cdr (readerdata)),
1055 be_safe_global && is_main_input_,
1059 if (SCM_UNBNDP (sval))
1062 return SCM_UNSPECIFIED;
1065 if (extra_token && SCM_VALUESP (sval))
1067 sval = scm_struct_ref (sval, SCM_INUM0);
1069 if (scm_is_pair (sval)) {
1070 for (SCM v = scm_reverse (scm_cdr (sval));
1075 switch (extra_token) {
1077 token = scan_scm_id (scm_car (v));
1078 if (!scm_is_eq (yylval, SCM_UNSPECIFIED))
1079 push_extra_token (token, yylval);
1082 push_extra_token (SCM_IDENTIFIER, scm_car (v));
1086 sval = scm_car (sval);
1088 sval = SCM_UNSPECIFIED;
1094 /* Check for valid UTF-8 that has no overlong or surrogate codes and
1095 is in the range 0-0x10ffff */
1098 Lily_lexer::YYText_utf8 ()
1100 const char * const p = YYText ();
1101 for (int i=0; p[i];) {
1102 if ((p[i] & 0xff) < 0x80) {
1106 int oldi = i; // start of character
1107 int more = 0; // # of followup bytes, 0 if bad
1108 switch (p[i++] & 0xff) {
1109 // 0xc0 and 0xc1 are overlong prefixes for
1110 // 0x00-0x3f and 0x40-0x7f respectively, bad.
1111 case 0xc2: // 0x80-0xbf
1112 case 0xc3: // 0xc0-0xff
1113 case 0xc4: // 0x100-0x13f
1114 case 0xc5: // 0x140-0x17f
1115 case 0xc6: // 0x180-0x1bf
1116 case 0xc7: // 0x1c0-0x1ff
1117 case 0xc8: // 0x200-0x23f
1118 case 0xc9: // 0x240-0x27f
1119 case 0xca: // 0x280-0x2bf
1120 case 0xcb: // 0x2c0-0x2ff
1121 case 0xcc: // 0x300-0x33f
1122 case 0xcd: // 0x340-0x37f
1123 case 0xce: // 0x380-0x3bf
1124 case 0xcf: // 0x3c0-0x3ff
1125 case 0xd0: // 0x400-0x43f
1126 case 0xd1: // 0x440-0x47f
1127 case 0xd2: // 0x480-0x4bf
1128 case 0xd3: // 0x4c0-0x4ff
1129 case 0xd4: // 0x500-0x53f
1130 case 0xd5: // 0x540-0x57f
1131 case 0xd6: // 0x580-0x5bf
1132 case 0xd7: // 0x5c0-0x5ff
1133 case 0xd8: // 0x600-0x63f
1134 case 0xd9: // 0x640-0x67f
1135 case 0xda: // 0x680-0x6bf
1136 case 0xdb: // 0x6c0-0x6ff
1137 case 0xdc: // 0x700-0x73f
1138 case 0xdd: // 0x740-0x77f
1139 case 0xde: // 0x780-0x7bf
1140 case 0xdf: // 0x7c0-0x7ff
1141 more = 1; // 2-byte sequences, 0x80-0x7ff
1144 // don't allow overlong sequences for 0-0x7ff
1145 if ((p[i] & 0xff) < 0xa0)
1147 case 0xe1: // 0x1000-0x1fff
1148 case 0xe2: // 0x2000-0x2fff
1149 case 0xe3: // 0x3000-0x3fff
1150 case 0xe4: // 0x4000-0x4fff
1151 case 0xe5: // 0x5000-0x5fff
1152 case 0xe6: // 0x6000-0x6fff
1153 case 0xe7: // 0x7000-0x7fff
1154 case 0xe8: // 0x8000-0x8fff
1155 case 0xe9: // 0x9000-0x9fff
1156 case 0xea: // 0xa000-0xafff
1157 case 0xeb: // 0xb000-0xbfff
1158 case 0xec: // 0xc000-0xcfff
1159 more = 2; // 3-byte sequences, 0x7ff-0xcfff
1161 case 0xed: // 0xd000-0xdfff
1162 // Don't allow surrogate codes 0xd800-0xdfff
1163 if ((p[i] & 0xff) >= 0xa0)
1165 case 0xee: // 0xe000-0xefff
1166 case 0xef: // 0xf000-0xffff
1167 more = 2; // 3-byte sequences,
1168 // 0xd000-0xd7ff, 0xe000-0xffff
1171 // don't allow overlong sequences for 0-0xffff
1172 if ((p[i] & 0xff) < 0x90)
1174 case 0xf1: // 0x40000-0x7ffff
1175 case 0xf2: // 0x80000-0xbffff
1176 case 0xf3: // 0xc0000-0xfffff
1177 more = 3; // 4-byte sequences, 0x10000-0xfffff
1180 // don't allow more than 0x10ffff
1181 if ((p[i] & 0xff) >= 0x90)
1183 more = 3; // 4-byte sequence, 0x100000-0x10ffff
1187 // check that all continuation bytes are valid
1189 if ((p[i++] & 0xc0) != 0x80)
1195 Input h = here_input ();
1196 h.set (h.get_source_file (), h.start () + oldi, h.start () + i);
1197 h.warning (_ ("non-UTF-8 input").c_str ());
1204 urg, belong to string (_convert)
1205 and should be generalised
1208 strip_leading_white (string&s)
1211 for (; i < s.length (); i++)
1212 if (!isspace (s[i]))
1219 strip_trailing_white (string&s)
1221 ssize i = s.length ();
1223 if (!isspace (s[i]))
1226 s = s.substr (0, i + 1);
1231 Lilypond_version oldest_version ("2.7.38");
1235 is_valid_version (string s)
1237 Lilypond_version current ( MAJOR_VERSION "." MINOR_VERSION "." PATCH_LEVEL );
1238 Lilypond_version ver (s);
1239 if (int (ver) < oldest_version)
1241 non_fatal_error (_f ("file too old: %s (oldest supported: %s)", ver.to_string (), oldest_version.to_string ()));
1242 non_fatal_error (_ ("consider updating the input with the convert-ly script"));
1248 non_fatal_error (_f ("program too old: %s (file requires: %s)", current.to_string (), ver.to_string ()));
1259 lyric_fudge (string s)
1263 while ((i = s.find ('_', i)) != string::npos)
1271 Convert "NUM/DEN" into a '(NUM . DEN) cons.
1274 scan_fraction (string frac)
1276 ssize i = frac.find ('/');
1277 string left = frac.substr (0, i);
1278 string right = frac.substr (i + 1, (frac.length () - i + 1));
1280 int n = String_convert::dec2int (left);
1281 int d = String_convert::dec2int (right);
1282 return scm_cons (scm_from_int (n), scm_from_int (d));
1286 lookup_markup_command (string s)
1288 SCM proc = ly_lily_module_constant ("lookup-markup-command");
1289 return scm_call_1 (proc, ly_string2scm (s));
1293 lookup_markup_list_command (string s)
1295 SCM proc = ly_lily_module_constant ("lookup-markup-list-command");
1296 return scm_call_1 (proc, ly_string2scm (s));
1299 /* Shut up lexer warnings. */
1319 silence_lexer_warnings ()
1321 (void) yy_start_stack_ptr;
1322 (void) yy_start_stack_depth;
1323 (void) yy_start_stack;
1324 (void) yy_push_state;
1325 (void) yy_pop_state;
1326 (void) yy_top_state;
1327 (void) silence_lexer_warnings;