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 {
312 if (!is_main_input_ && include_stack_.size () == 1)
315 is_main_input_ = true;
318 error (_ ("\\maininput not allowed outside init files"));
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 error (_ ("end quote missing"));
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, '$');
440 int token = scan_scm_id (sval);
441 if (!scm_is_eq (yylval, SCM_UNSPECIFIED))
445 <INITIAL,notes,lyrics>{
447 yylval = SCM_UNSPECIFIED;
448 return DOUBLE_ANGLE_OPEN;
451 yylval = SCM_UNSPECIFIED;
452 return DOUBLE_ANGLE_CLOSE;
458 yylval = SCM_UNSPECIFIED;
462 yylval = SCM_UNSPECIFIED;
469 yylval = SCM_UNSPECIFIED;
473 yylval = SCM_UNSPECIFIED;
477 yylval = SCM_UNSPECIFIED;
483 {WORD}/[-_] | // backup rule
485 return scan_bare_word (YYText_utf8 ());
488 {COMMAND}/[-_] | // backup rule
490 return scan_escaped_word (YYText_utf8 () + 1);
493 yylval = scan_fraction (YYText ());
496 {UNSIGNED}/\/ | // backup rule
498 yylval = scm_c_read_string (YYText ());
502 yylval = scm_c_read_string (YYText () + 1);
509 char c = escaped_char (YYText ()[1]);
510 yylval = scm_cons (scm_from_locale_stringn (&c, 1),
514 yylval = scm_cons (scm_from_locale_string (YYText_utf8 ()),
521 /* yylval is union. Must remember STRING before setting SCM*/
523 yylval = scm_string_concatenate_reverse (yylval,
527 return is_lyric_state () ? LYRICS_STRING : STRING;
530 yylval = scm_cons (scm_from_locale_string (YYText ()),
540 yylval = scan_fraction (YYText ());
543 {UNSIGNED}/\/ | // backup rule
545 yylval = scm_c_read_string (YYText ());
548 {COMMAND}/[-_] | // backup rule
550 return scan_escaped_word (YYText_utf8 () + 1);
552 /* Characters needed to express durations, assignments, barchecks */
554 yylval = SCM_UNSPECIFIED;
557 [^$#{}\"\\ \t\n\r\f0-9]+ {
559 string s (YYText_utf8 ());
560 yylval = SCM_UNSPECIFIED;
566 yylval = ly_string2scm (s);
568 return LYRICS_STRING;
570 /* This should really just cover {} */
572 yylval = SCM_UNSPECIFIED;
573 return YYText ()[0]; // above catches all multibytes.
577 {WORD}/[-_] | // backup rule
579 return scan_bare_word (YYText_utf8 ());
581 {COMMAND}/[-_] | // backup rule
583 return scan_escaped_word (YYText_utf8 () + 1);
586 yylval = scan_fraction (YYText ());
589 {UNSIGNED}/\/ | // backup rule
591 yylval = scm_c_read_string (YYText ());
595 yylval = SCM_UNSPECIFIED;
599 yylval = SCM_UNSPECIFIED;
603 yylval = SCM_UNSPECIFIED;
607 yylval = SCM_UNSPECIFIED;
611 yylval = SCM_UNSPECIFIED;
615 yylval = SCM_UNSPECIFIED;
616 return YYText ()[0]; // WORD catches all multibyte.
623 yylval = SCM_UNSPECIFIED;
626 {COMMAND}/[-_] | // backup rule
628 string str (YYText_utf8 () + 1);
630 int token_type = MARKUP_FUNCTION;
631 SCM s = lookup_markup_command (str);
633 // lookup-markup-command returns a pair with the car
634 // being the function to call, and the cdr being the
635 // call signature specified to define-markup-command,
636 // a list of predicates.
638 if (!scm_is_pair (s)) {
639 // If lookup-markup-command was not successful, we
640 // try lookup-markup-list-command instead.
641 // If this fails as well, we just scan and return
643 s = lookup_markup_list_command (str);
645 token_type = MARKUP_LIST_FUNCTION;
647 return scan_escaped_word (str);
650 // If the list of predicates is, say,
651 // (number? number? markup?), then tokens
652 // EXPECT_MARKUP EXPECT_SCM EXPECT_SCM EXPECT_NO_MORE_ARGS
653 // will be generated. Note that we have to push them
654 // in reverse order, so the first token pushed in the
655 // loop will be EXPECT_NO_MORE_ARGS.
659 // yylval now contains the function to call as token
660 // value (for token type MARKUP_FUNCTION or
661 // MARKUP_LIST_FUNCTION).
663 push_extra_token(EXPECT_NO_MORE_ARGS);
665 for (; scm_is_pair(s); s = scm_cdr(s)) {
666 SCM predicate = scm_car(s);
668 if (predicate == ly_lily_module_constant ("markup-list?"))
669 push_extra_token(EXPECT_MARKUP_LIST);
670 else if (predicate == ly_lily_module_constant ("markup?"))
671 push_extra_token(EXPECT_MARKUP);
673 push_extra_token(EXPECT_SCM, predicate);
677 [^$#{}\"\\ \t\n\r\f]+ {
678 string s (YYText_utf8 ());
680 yylval = ly_string2scm (s);
684 yylval = SCM_UNSPECIFIED;
685 return YYText ()[0]; // Above is catchall for multibyte
689 <longcomment><<EOF>> {
690 LexerError (_ ("EOF found inside a comment").c_str ());
691 is_main_input_ = false; // should be safe , can't have \include in --safe.
692 yylval = SCM_UNSPECIFIED;
694 yyterminate (); // can't move this, since it actually rets a YY_NULL
698 yylval = SCM_UNSPECIFIED;
701 /* 2 = init.ly + current file.
702 > because we're before closing, but is_main_input_ should
705 is_main_input_ = include_stack_.size () > 2;
706 if (!close_input () || !is_main_input_)
707 /* Returns YY_NULL */
710 else if (!close_input ())
711 /* Returns YY_NULL */
716 {WORD}/[-_] | // backup rule
718 return scan_bare_word (YYText_utf8 ());
720 {COMMAND}/[-_] | // backup rule
722 return scan_escaped_word (YYText_utf8 () + 1);
727 yylval = scan_fraction (YYText ());
731 -{UNSIGNED} | // backup rule
733 yylval = scm_c_read_string (YYText ());
737 {UNSIGNED}/\/ | // backup rule
739 yylval = scm_c_read_string (YYText ());
745 yylval = SCM_UNSPECIFIED;
749 -/\. | // backup rule
751 yylval = SCM_UNSPECIFIED;
755 <INITIAL,notes,figures>. {
756 yylval = SCM_UNSPECIFIED;
760 <INITIAL,lyrics,notes,figures>\\. {
761 yylval = SCM_UNSPECIFIED;
762 char c = YYText ()[1];
766 return E_ANGLE_CLOSE;
770 return E_EXCLAMATION;
776 return E_BRACKET_OPEN;
780 return E_BRACKET_CLOSE;
792 string msg = _f ("invalid character: `%s'", YYText_utf8 ());
793 LexerError (msg.c_str ());
794 yylval = SCM_UNSPECIFIED;
795 return '%'; // Better not return half a utf8 character.
800 /* Make the lexer generate a token of the given type as the next token.
801 TODO: make it possible to define a value for the token as well */
803 Lily_lexer::push_extra_token (int token_type, SCM scm)
805 if (scm_is_null (extra_tokens_))
807 if (YY_START != extratoken)
808 hidden_state_ = YY_START;
809 yy_push_state (extratoken);
811 extra_tokens_ = scm_acons (scm_from_int (token_type), scm, extra_tokens_);
815 Lily_lexer::push_chord_state (SCM alist)
817 SCM p = scm_assq (alist, pitchname_tab_stack_);
819 if (scm_is_false (p))
820 p = scm_cons (alist, alist_to_hashq (alist));
821 pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_);
822 yy_push_state (chords);
826 Lily_lexer::push_figuredbass_state ()
828 yy_push_state (figures);
832 Lily_lexer::push_initial_state ()
834 yy_push_state (INITIAL);
838 Lily_lexer::push_lyric_state ()
840 yy_push_state (lyrics);
844 Lily_lexer::push_markup_state ()
846 yy_push_state (markup);
850 Lily_lexer::push_note_state (SCM alist)
852 bool extra = (YYSTATE == extratoken);
854 SCM p = scm_assq (alist, pitchname_tab_stack_);
859 if (scm_is_false (p))
860 p = scm_cons (alist, alist_to_hashq (alist));
861 pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_);
862 yy_push_state (notes);
865 hidden_state_ = YYSTATE;
866 yy_push_state (extratoken);
871 Lily_lexer::pop_state ()
873 bool extra = (YYSTATE == extratoken);
878 if (YYSTATE == notes || YYSTATE == chords)
879 pitchname_tab_stack_ = scm_cdr (pitchname_tab_stack_);
884 hidden_state_ = YYSTATE;
885 yy_push_state (extratoken);
890 Lily_lexer::identifier_type (SCM sid)
892 int k = try_special_identifiers (&yylval , sid);
893 return k >= 0 ? k : SCM_IDENTIFIER;
898 Lily_lexer::scan_escaped_word (string str)
900 // use more SCM for this.
902 // SCM sym = ly_symbol2scm (str.c_str ());
904 yylval = SCM_UNSPECIFIED;
905 int i = lookup_keyword (str);
906 if (i == MARKUP && is_lyric_state ())
911 SCM sid = lookup_identifier (str);
912 if (sid != SCM_UNDEFINED)
913 return scan_scm_id (sid);
915 string msg (_f ("unknown escaped string: `\\%s'", str));
916 LexerError (msg.c_str ());
918 yylval = ly_string2scm (str);
924 Lily_lexer::scan_scm_id (SCM sid)
926 if (is_music_function (sid))
928 int funtype = SCM_FUNCTION;
932 SCM s = get_music_function_signature (sid);
933 SCM cs = scm_car (s);
935 if (scm_is_pair (cs))
940 if (scm_is_eq (cs, ly_lily_module_constant ("ly:music?")))
941 funtype = MUSIC_FUNCTION;
942 else if (scm_is_eq (cs, ly_lily_module_constant ("ly:event?")))
943 funtype = EVENT_FUNCTION;
944 else if (ly_is_procedure (cs))
945 funtype = SCM_FUNCTION;
946 else programming_error ("Bad syntax function predicate");
948 push_extra_token (EXPECT_NO_MORE_ARGS);
949 for (s = scm_cdr (s); scm_is_pair (s); s = scm_cdr (s))
951 SCM optional = SCM_UNDEFINED;
954 if (scm_is_pair (cs))
956 optional = SCM_CDR (cs);
960 if (cs == Pitch_type_p_proc)
961 push_extra_token (EXPECT_PITCH);
962 else if (cs == Duration_type_p_proc)
963 push_extra_token (EXPECT_DURATION);
964 else if (ly_is_procedure (cs))
965 push_extra_token (EXPECT_SCM, cs);
968 programming_error ("Function parameter without type-checking predicate");
971 if (!scm_is_eq (optional, SCM_UNDEFINED))
972 push_extra_token (EXPECT_OPTIONAL, optional);
977 return identifier_type (sid);
981 Lily_lexer::scan_bare_word (string str)
983 SCM sym = ly_symbol2scm (str.c_str ());
984 if ((YYSTATE == notes) || (YYSTATE == chords)) {
985 SCM handle = SCM_BOOL_F;
986 if (scm_is_pair (pitchname_tab_stack_))
987 handle = scm_hashq_get_handle (scm_cdar (pitchname_tab_stack_), sym);
989 if (scm_is_pair (handle)) {
990 yylval = scm_cdr (handle);
991 if (unsmob_pitch (yylval))
992 return (YYSTATE == notes) ? NOTENAME_PITCH : TONICNAME_PITCH;
993 else if (scm_is_symbol (yylval))
996 else if ((YYSTATE == chords)
997 && (handle = scm_hashq_get_handle (chordmodifier_tab_, sym))!= SCM_BOOL_F)
999 yylval = scm_cdr (handle);
1000 return CHORD_MODIFIER;
1003 yylval = ly_string2scm (str);
1008 Lily_lexer::get_state () const
1010 if (YY_START == extratoken)
1011 return hidden_state_;
1017 Lily_lexer::is_note_state () const
1019 return get_state () == notes;
1023 Lily_lexer::is_chord_state () const
1025 return get_state () == chords;
1029 Lily_lexer::is_lyric_state () const
1031 return get_state () == lyrics;
1035 Lily_lexer::is_figure_state () const
1037 return get_state () == figures;
1040 // The extra_token parameter specifies how to convert multiple values
1041 // into additional tokens. For '#', additional values get pushed as
1042 // SCM_IDENTIFIER. For '$', they get checked for their type and get
1043 // pushed as a corresponding *_IDENTIFIER token. Since the latter
1044 // tampers with yylval, it can only be done from the lexer itself, so
1045 // this function is private.
1048 Lily_lexer::eval_scm (SCM readerdata, char extra_token)
1050 SCM sval = SCM_UNDEFINED;
1052 if (!SCM_UNBNDP (readerdata))
1054 sval = ly_eval_scm (scm_car (readerdata),
1055 *unsmob_input (scm_cdr (readerdata)),
1056 be_safe_global && is_main_input_,
1060 if (SCM_UNBNDP (sval))
1063 return SCM_UNSPECIFIED;
1066 if (extra_token && SCM_VALUESP (sval))
1068 sval = scm_struct_ref (sval, SCM_INUM0);
1070 if (scm_is_pair (sval)) {
1071 for (SCM v = scm_reverse (scm_cdr (sval));
1076 switch (extra_token) {
1078 token = scan_scm_id (scm_car (v));
1079 if (!scm_is_eq (yylval, SCM_UNSPECIFIED))
1080 push_extra_token (token, yylval);
1083 push_extra_token (SCM_IDENTIFIER, scm_car (v));
1087 sval = scm_car (sval);
1089 sval = SCM_UNSPECIFIED;
1095 /* Check for valid UTF-8 that has no overlong or surrogate codes and
1096 is in the range 0-0x10ffff */
1099 Lily_lexer::YYText_utf8 ()
1101 const char * const p = YYText ();
1102 for (int i=0; p[i];) {
1103 if ((p[i] & 0xff) < 0x80) {
1107 int oldi = i; // start of character
1108 int more = 0; // # of followup bytes, 0 if bad
1109 switch (p[i++] & 0xff) {
1110 // 0xc0 and 0xc1 are overlong prefixes for
1111 // 0x00-0x3f and 0x40-0x7f respectively, bad.
1112 case 0xc2: // 0x80-0xbf
1113 case 0xc3: // 0xc0-0xff
1114 case 0xc4: // 0x100-0x13f
1115 case 0xc5: // 0x140-0x17f
1116 case 0xc6: // 0x180-0x1bf
1117 case 0xc7: // 0x1c0-0x1ff
1118 case 0xc8: // 0x200-0x23f
1119 case 0xc9: // 0x240-0x27f
1120 case 0xca: // 0x280-0x2bf
1121 case 0xcb: // 0x2c0-0x2ff
1122 case 0xcc: // 0x300-0x33f
1123 case 0xcd: // 0x340-0x37f
1124 case 0xce: // 0x380-0x3bf
1125 case 0xcf: // 0x3c0-0x3ff
1126 case 0xd0: // 0x400-0x43f
1127 case 0xd1: // 0x440-0x47f
1128 case 0xd2: // 0x480-0x4bf
1129 case 0xd3: // 0x4c0-0x4ff
1130 case 0xd4: // 0x500-0x53f
1131 case 0xd5: // 0x540-0x57f
1132 case 0xd6: // 0x580-0x5bf
1133 case 0xd7: // 0x5c0-0x5ff
1134 case 0xd8: // 0x600-0x63f
1135 case 0xd9: // 0x640-0x67f
1136 case 0xda: // 0x680-0x6bf
1137 case 0xdb: // 0x6c0-0x6ff
1138 case 0xdc: // 0x700-0x73f
1139 case 0xdd: // 0x740-0x77f
1140 case 0xde: // 0x780-0x7bf
1141 case 0xdf: // 0x7c0-0x7ff
1142 more = 1; // 2-byte sequences, 0x80-0x7ff
1145 // don't allow overlong sequences for 0-0x7ff
1146 if ((p[i] & 0xff) < 0xa0)
1148 case 0xe1: // 0x1000-0x1fff
1149 case 0xe2: // 0x2000-0x2fff
1150 case 0xe3: // 0x3000-0x3fff
1151 case 0xe4: // 0x4000-0x4fff
1152 case 0xe5: // 0x5000-0x5fff
1153 case 0xe6: // 0x6000-0x6fff
1154 case 0xe7: // 0x7000-0x7fff
1155 case 0xe8: // 0x8000-0x8fff
1156 case 0xe9: // 0x9000-0x9fff
1157 case 0xea: // 0xa000-0xafff
1158 case 0xeb: // 0xb000-0xbfff
1159 case 0xec: // 0xc000-0xcfff
1160 more = 2; // 3-byte sequences, 0x7ff-0xcfff
1162 case 0xed: // 0xd000-0xdfff
1163 // Don't allow surrogate codes 0xd800-0xdfff
1164 if ((p[i] & 0xff) >= 0xa0)
1166 case 0xee: // 0xe000-0xefff
1167 case 0xef: // 0xf000-0xffff
1168 more = 2; // 3-byte sequences,
1169 // 0xd000-0xd7ff, 0xe000-0xffff
1172 // don't allow overlong sequences for 0-0xffff
1173 if ((p[i] & 0xff) < 0x90)
1175 case 0xf1: // 0x40000-0x7ffff
1176 case 0xf2: // 0x80000-0xbffff
1177 case 0xf3: // 0xc0000-0xfffff
1178 more = 3; // 4-byte sequences, 0x10000-0xfffff
1181 // don't allow more than 0x10ffff
1182 if ((p[i] & 0xff) >= 0x90)
1184 more = 3; // 4-byte sequence, 0x100000-0x10ffff
1188 // check that all continuation bytes are valid
1190 if ((p[i++] & 0xc0) != 0x80)
1196 Input h = here_input ();
1197 h.set (h.get_source_file (), h.start () + oldi, h.start () + i);
1198 h.warning (_ ("non-UTF-8 input").c_str ());
1205 urg, belong to string (_convert)
1206 and should be generalised
1209 strip_leading_white (string&s)
1212 for (; i < s.length (); i++)
1213 if (!isspace (s[i]))
1220 strip_trailing_white (string&s)
1222 ssize i = s.length ();
1224 if (!isspace (s[i]))
1227 s = s.substr (0, i + 1);
1232 Lilypond_version oldest_version ("2.7.38");
1236 is_valid_version (string s)
1238 Lilypond_version current ( MAJOR_VERSION "." MINOR_VERSION "." PATCH_LEVEL );
1239 Lilypond_version ver (s);
1240 if (int (ver) < oldest_version)
1242 non_fatal_error (_f ("file too old: %s (oldest supported: %s)", ver.to_string (), oldest_version.to_string ()));
1243 non_fatal_error (_ ("consider updating the input with the convert-ly script"));
1249 non_fatal_error (_f ("program too old: %s (file requires: %s)", current.to_string (), ver.to_string ()));
1260 lyric_fudge (string s)
1264 while ((i = s.find ('_', i)) != string::npos)
1272 Convert "NUM/DEN" into a '(NUM . DEN) cons.
1275 scan_fraction (string frac)
1277 ssize i = frac.find ('/');
1278 string left = frac.substr (0, i);
1279 string right = frac.substr (i + 1, (frac.length () - i + 1));
1281 int n = String_convert::dec2int (left);
1282 int d = String_convert::dec2int (right);
1283 return scm_cons (scm_from_int (n), scm_from_int (d));
1287 lookup_markup_command (string s)
1289 SCM proc = ly_lily_module_constant ("lookup-markup-command");
1290 return scm_call_1 (proc, ly_string2scm (s));
1294 lookup_markup_list_command (string s)
1296 SCM proc = ly_lily_module_constant ("lookup-markup-list-command");
1297 return scm_call_1 (proc, ly_string2scm (s));
1300 /* Shut up lexer warnings. */
1320 silence_lexer_warnings ()
1322 (void) yy_start_stack_ptr;
1323 (void) yy_start_stack_depth;
1324 (void) yy_start_stack;
1325 (void) yy_push_state;
1326 (void) yy_pop_state;
1327 (void) yy_top_state;
1328 (void) silence_lexer_warnings;