1 %{ // -*- mode: c++; c-file-style: "linux" -*-
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() \
90 yy_push_state (quote);\
91 yylval.string = new string
93 #define start_lyric_quote() \
94 yy_push_state (lyric_quote);\
95 yylval.string = new string
97 #define yylval (*lexval_)
99 #define yylloc (*lexloc_)
101 #define YY_USER_ACTION add_lexed_char (YYLeng ());
104 SCM scan_fraction (string);
105 SCM (* scm_parse_error_handler) (void *);
115 %option yyclass="Lily_lexer"
117 %option never-interactive
134 /* The strategy concerning multibyte characters is to accept them but
135 * call YYText_utf8 for patterns that might contain them, in order to
136 * get a single code path responsible for flagging non-UTF-8 input:
137 * Patterns for accepting only valid UTF-8 without backing up are
138 * really hard to do and complex, and if nice error messages are
139 * wanted, one would need patterns catching the invalid input as well.
141 * Since editors and operating environments don't necessarily behave
142 * reasonably in the presence of mixed encodings, we flag encoding
143 * errors also in identifiers, comments, and strings where it would be
144 * conceivable to just transparently work with the byte string. But
145 * the whole point of caring about UTF-8 in here at all is too avoid
146 * stranger errors later when input passes into backends or log files
147 * or console output or error messages.
157 NATIONAL [\001-\006\021-\027\031\036]
158 TEX {AA}|-|{PUNCT}|{NATIONAL}|{SPECIAL_CHAR}
159 DASHED_WORD {A}({AN}|-)*
160 DASHED_KEY_WORD \\{DASHED_WORD}
169 REAL ({INT}\.{N}*)|(-?\.{N}+)
171 HORIZONTALWHITE [ \t]
175 MARKUPCOMMAND \\({A}|[-_])+
176 LYRICS ({AA}|{TEX})[^0-9 \t\n\r\f]*
180 BOM_UTF8 \357\273\277
186 // swallow and ignore carriage returns
189 <extratoken>{ANY_CHAR} {
190 /* Generate a token without swallowing anything */
192 /* First unswallow the eaten character */
193 add_lexed_char (-YYLeng ());
196 /* produce requested token */
197 int type = scm_to_int (scm_caar (extra_tokens_));
198 yylval.scm = scm_cdar (extra_tokens_);
199 extra_tokens_ = scm_cdr (extra_tokens_);
200 if (scm_is_null (extra_tokens_))
206 <extratoken><<EOF>> {
207 /* Generate a token without swallowing anything */
209 /* produce requested token */
210 int type = scm_to_int (scm_caar (extra_tokens_));
211 yylval.scm = scm_cdar (extra_tokens_);
212 extra_tokens_ = scm_cdr (extra_tokens_);
213 if (scm_is_null (extra_tokens_))
219 /* Use the trailing context feature. Otherwise, the BOM will not be
220 found if the file starts with an identifier definition. */
221 <INITIAL,chords,lyrics,figures,notes>{BOM_UTF8}/.* {
222 if (this->lexloc_->line_number () != 1 || this->lexloc_->column_number () != 0)
224 LexerWarning (_ ("stray UTF-8 BOM encountered").c_str ());
227 debug_output (_ ("Skipping UTF-8 BOM"));
230 <INITIAL,chords,figures,incl,lyrics,markup,notes>{
232 yy_push_state (longcomment);
234 %[^{\n\r][^\n\r]*[\n\r] {
235 (void) YYText_utf8 ();
237 %[^{\n\r] { // backup rule
238 (void) YYText_utf8 ();
243 (void) YYText_utf8 ();
250 <INITIAL,notes,figures,chords,markup>{
256 <INITIAL,chords,lyrics,notes,figures>\\version{WHITE}* {
257 yy_push_state (version);
259 <INITIAL,chords,lyrics,notes,figures>\\sourcefilename{WHITE}* {
260 yy_push_state (sourcefilename);
262 <INITIAL,chords,lyrics,notes,figures>\\sourcefileline{WHITE}* {
263 yy_push_state (sourcefileline);
265 <version>\"[^"]*\" { /* got the version number */
266 string s (YYText_utf8 () + 1);
267 s = s.substr (0, s.rfind ('\"'));
271 SCM top_scope = scm_car (scm_last_pair (scopes_));
272 scm_module_define (top_scope, ly_symbol2scm ("version-seen"), SCM_BOOL_T);
274 if (!is_valid_version (s))
279 <sourcefilename>\"[^""]*\" {
280 string s (YYText_utf8 () + 1);
281 s = s.substr (0, s.rfind ('\"'));
284 this->here_input().get_source_file ()->name_ = s;
285 message (_f ("Renaming input to: `%s'", s.c_str ()));
286 progress_indication ("\n");
287 scm_module_define (scm_car (scopes_),
288 ly_symbol2scm ("input-file-name"),
293 <sourcefileline>{INT} {
295 sscanf (YYText (), "%d", &i);
298 this->here_input ().get_source_file ()->set_line (here_input ().start (), i);
301 <version>{ANY_CHAR} {
302 LexerError (_ ("quoted string expected after \\version").c_str ());
305 <sourcefilename>{ANY_CHAR} {
306 LexerError (_ ("quoted string expected after \\sourcefilename").c_str ());
309 <sourcefileline>{ANY_CHAR} {
310 LexerError (_ ("integer expected after \\sourcefileline").c_str ());
315 (void) YYText_utf8 ();
318 (void) YYText_utf8 ();
326 <INITIAL,chords,lyrics,notes,figures>\\maininput {
330 is_main_input_ = true;
333 error (_ ("\\maininput not allowed outside init files"));
336 <INITIAL,chords,lyrics,figures,notes>\\include {
337 yy_push_state (incl);
339 <incl>\"[^""]*\" { /* got the include file name */
340 string s (YYText_utf8 ()+1);
341 s = s.substr (0, s.rfind ('"'));
343 new_input (s, sources_);
346 <incl>\\{BLACK}*{WHITE}? { /* got the include identifier */
347 string s = YYText_utf8 () + 1;
348 strip_trailing_white (s);
349 if (s.length () && (s[s.length () - 1] == ';'))
350 s = s.substr (0, s.length () - 1);
352 SCM sid = lookup_identifier (s);
353 if (scm_is_string (sid)) {
354 new_input (ly_scm2string (sid), sources_);
357 string msg (_f ("wrong or undefined identifier: `%s'", s ));
359 LexerError (msg.c_str ());
360 SCM err = scm_current_error_port ();
361 scm_puts ("This value was found in the table: ", err);
362 scm_display (sid, err);
365 <incl>(\$|#) { // scm for the filename
367 Input hi = here_input();
369 SCM sval = ly_parse_scm (hi.start (), &n, hi,
370 be_safe_global && is_main_input_, parser_);
371 sval = eval_scm (sval);
373 for (int i = 0; i < n; i++)
377 char_count_stack_.back () += n;
379 if (scm_is_string (sval)) {
380 new_input (ly_scm2string (sval), sources_);
383 LexerError (_ ("string expected after \\include").c_str ());
384 if (sval != SCM_UNDEFINED) {
385 SCM err = scm_current_error_port ();
386 scm_puts ("This value was found instead: ", err);
387 scm_display (sval, err);
392 <incl,version,sourcefilename>\"[^""]* { // backup rule
393 error (_ ("end quote missing"));
396 <chords,notes,figures>{RESTNAME} {
397 char const *s = YYText ();
398 yylval.scm = scm_from_locale_string (s);
401 <chords,notes,figures>q {
402 return CHORD_REPETITION;
405 <chords,notes,figures>R {
406 return MULTI_MEASURE_REST;
408 <INITIAL,chords,figures,lyrics,markup,notes># { //embedded scm
410 Input hi = here_input();
412 SCM sval = ly_parse_scm (hi.start (), &n, hi,
413 be_safe_global && is_main_input_, parser_);
415 if (sval == SCM_UNDEFINED)
418 for (int i = 0; i < n; i++)
422 char_count_stack_.back () += n;
428 <INITIAL,chords,figures,lyrics,markup,notes>\$ { //immediate scm
430 Input hi = here_input();
432 SCM sval = ly_parse_scm (hi.start (), &n, hi,
433 be_safe_global && is_main_input_, parser_);
435 for (int i = 0; i < n; i++)
439 char_count_stack_.back () += n;
441 sval = eval_scm (sval, '$');
443 int token = scan_scm_id (sval);
444 if (!scm_is_eq (yylval.scm, SCM_UNSPECIFIED))
448 <INITIAL,notes,lyrics>{
450 return DOUBLE_ANGLE_OPEN;
453 return DOUBLE_ANGLE_CLOSE;
480 return scan_bare_word (YYText_utf8 ());
484 return scan_escaped_word (YYText_utf8 () + 1);
487 yylval.scm = scan_fraction (YYText ());
490 {UNSIGNED}/\/ | // backup rule
492 yylval.scm = scm_c_read_string (YYText ());
496 yylval.i = String_convert::dec2int (string (YYText () +1));
503 *yylval.string += to_string (escaped_char (YYText ()[1]));
506 *yylval.string += YYText_utf8 ();
512 /* yylval is union. Must remember STRING before setting SCM*/
513 string *sp = yylval.string;
514 yylval.scm = ly_string2scm (*sp);
516 return is_lyric_state () ? LYRICS_STRING : STRING;
519 *yylval.string += YYText ();
525 start_lyric_quote ();
528 yylval.scm = scan_fraction (YYText ());
531 {UNSIGNED}/\/ | // backup rule
533 yylval.scm = scm_c_read_string (YYText ());
537 return scan_escaped_word (YYText_utf8 () + 1);
541 string s (YYText_utf8 ());
543 return yylval.i = EXTENDER;
545 return yylval.i = HYPHEN;
548 char c = s[s.length () - 1];
549 if (c == '{' || c == '}') // brace open is for not confusing dumb tools.
550 here_input ().warning (
551 _ ("Brace found at end of lyric. Did you forget a space?"));
552 yylval.scm = ly_string2scm (s);
555 return LYRICS_STRING;
558 return YYText ()[0]; // LYRICS already catches all multibytes.
563 return scan_bare_word (YYText_utf8 ());
566 return scan_escaped_word (YYText_utf8 () + 1);
569 yylval.scm = scan_fraction (YYText ());
572 {UNSIGNED}/\/ | // backup rule
574 yylval.scm = scm_c_read_string (YYText ());
593 return YYText ()[0]; // ALPHAWORD catches all multibyte.
603 string str (YYText_utf8 () + 1);
605 int token_type = MARKUP_FUNCTION;
606 SCM s = lookup_markup_command (str);
608 // lookup-markup-command returns a pair with the car
609 // being the function to call, and the cdr being the
610 // call signature specified to define-markup-command,
611 // a list of predicates.
613 if (!scm_is_pair (s)) {
614 // If lookup-markup-command was not successful, we
615 // try lookup-markup-list-command instead.
616 // If this fails as well, we just scan and return
618 s = lookup_markup_list_command (str);
620 token_type = MARKUP_LIST_FUNCTION;
622 return scan_escaped_word (str);
625 // If the list of predicates is, say,
626 // (number? number? markup?), then tokens
627 // EXPECT_MARKUP EXPECT_SCM EXPECT_SCM EXPECT_NO_MORE_ARGS
628 // will be generated. Note that we have to push them
629 // in reverse order, so the first token pushed in the
630 // loop will be EXPECT_NO_MORE_ARGS.
632 yylval.scm = scm_car(s);
634 // yylval now contains the function to call as token
635 // value (for token type MARKUP_FUNCTION or
636 // MARKUP_LIST_FUNCTION).
638 push_extra_token(EXPECT_NO_MORE_ARGS);
640 for (; scm_is_pair(s); s = scm_cdr(s)) {
641 SCM predicate = scm_car(s);
643 if (predicate == ly_lily_module_constant ("markup-list?"))
644 push_extra_token(EXPECT_MARKUP_LIST);
645 else if (predicate == ly_lily_module_constant ("markup?"))
646 push_extra_token(EXPECT_MARKUP);
648 push_extra_token(EXPECT_SCM, predicate);
655 [^$#{}\"\\ \t\n\r\f]+ {
656 string s (YYText_utf8 ());
658 char c = s[s.length () - 1];
659 /* brace open is for not confusing dumb tools. */
660 if (c == '{' || c == '}')
661 here_input ().warning (
662 _ ("Brace found at end of markup. Did you forget a space?"));
663 yylval.scm = ly_string2scm (s);
669 return YYText()[0]; // Above is catchall for multibyte
673 <longcomment><<EOF>> {
674 LexerError (_ ("EOF found inside a comment").c_str ());
675 is_main_input_ = false; // should be safe , can't have \include in --safe.
677 yyterminate (); // can't move this, since it actually rets a YY_NULL
680 <<EOF>> { if (is_main_input_)
682 /* 2 = init.ly + current file.
683 > because we're before closing, but is_main_input_ should
686 is_main_input_ = include_stack_.size () > 2;
688 /* Returns YY_NULL */
691 else if (!close_input ())
692 /* Returns YY_NULL */
698 return scan_bare_word (YYText_utf8 ());
701 return scan_escaped_word (YYText_utf8 () + 1);
706 yylval.scm = scan_fraction (YYText ());
710 -{UNSIGNED} | // backup rule
712 yylval.scm = scm_c_read_string (YYText ());
716 {UNSIGNED}/\/ | // backup rule
718 yylval.scm = scm_c_read_string (YYText ());
728 -/\. | // backup rule
730 char c = YYText ()[0];
735 <INITIAL,notes,figures>. {
739 <INITIAL,lyrics,notes,figures>\\. {
740 char c = YYText ()[1];
744 return E_ANGLE_CLOSE;
748 return E_EXCLAMATION;
754 return E_BRACKET_OPEN;
758 return E_BRACKET_CLOSE;
770 string msg = _f ("invalid character: `%s'", YYText_utf8 ());
771 LexerError (msg.c_str ());
772 return '%'; // Better not return half a utf8 character.
777 /* Make the lexer generate a token of the given type as the next token.
778 TODO: make it possible to define a value for the token as well */
780 Lily_lexer::push_extra_token (int token_type, SCM scm)
782 if (scm_is_null (extra_tokens_))
784 if (YY_START != extratoken)
785 hidden_state_ = YY_START;
786 yy_push_state (extratoken);
788 extra_tokens_ = scm_acons (scm_from_int (token_type), scm, extra_tokens_);
792 Lily_lexer::push_chord_state (SCM alist)
794 SCM p = scm_assq (alist, pitchname_tab_stack_);
796 if (scm_is_false (p))
797 p = scm_cons (alist, alist_to_hashq (alist));
798 pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_);
799 yy_push_state (chords);
803 Lily_lexer::push_figuredbass_state ()
805 yy_push_state (figures);
809 Lily_lexer::push_initial_state ()
811 yy_push_state (INITIAL);
815 Lily_lexer::push_lyric_state ()
817 yy_push_state (lyrics);
821 Lily_lexer::push_markup_state ()
823 yy_push_state (markup);
827 Lily_lexer::push_note_state (SCM alist)
829 SCM p = scm_assq (alist, pitchname_tab_stack_);
831 if (scm_is_false (p))
832 p = scm_cons (alist, alist_to_hashq (alist));
833 pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_);
834 yy_push_state (notes);
838 Lily_lexer::pop_state ()
840 bool extra = (YYSTATE == extratoken);
845 if (YYSTATE == notes || YYSTATE == chords)
846 pitchname_tab_stack_ = scm_cdr (pitchname_tab_stack_);
851 hidden_state_ = YYSTATE;
852 yy_push_state (extratoken);
857 Lily_lexer::identifier_type (SCM sid)
859 int k = try_special_identifiers (&yylval.scm , sid);
860 return k >= 0 ? k : SCM_IDENTIFIER;
865 Lily_lexer::scan_escaped_word (string str)
867 // use more SCM for this.
869 // SCM sym = ly_symbol2scm (str.c_str ());
871 int i = lookup_keyword (str);
872 if (i == MARKUP && is_lyric_state ())
877 SCM sid = lookup_identifier (str);
878 if (sid != SCM_UNDEFINED)
879 return scan_scm_id (sid);
881 string msg (_f ("unknown escaped string: `\\%s'", str));
882 LexerError (msg.c_str ());
884 yylval.scm = ly_string2scm (str);
890 Lily_lexer::scan_scm_id (SCM sid)
892 if (is_music_function (sid))
894 int funtype = SCM_FUNCTION;
898 SCM s = get_music_function_signature (sid);
899 SCM cs = scm_car (s);
901 if (scm_is_pair (cs))
906 if (scm_is_eq (cs, ly_lily_module_constant ("ly:music?")))
907 funtype = MUSIC_FUNCTION;
908 else if (scm_is_eq (cs, ly_lily_module_constant ("ly:event?")))
909 funtype = EVENT_FUNCTION;
910 else if (ly_is_procedure (cs))
911 funtype = SCM_FUNCTION;
912 else programming_error ("Bad syntax function predicate");
914 push_extra_token (EXPECT_NO_MORE_ARGS);
915 for (s = scm_cdr (s); scm_is_pair (s); s = scm_cdr (s))
917 SCM optional = SCM_UNDEFINED;
920 if (scm_is_pair (cs))
922 optional = SCM_CDR (cs);
926 if (cs == Pitch_type_p_proc)
927 push_extra_token (EXPECT_PITCH);
928 else if (cs == Duration_type_p_proc)
929 push_extra_token (EXPECT_DURATION);
930 else if (ly_is_procedure (cs))
931 push_extra_token (EXPECT_SCM, cs);
934 programming_error ("Function parameter without type-checking predicate");
937 if (!scm_is_eq (optional, SCM_UNDEFINED))
938 push_extra_token (EXPECT_OPTIONAL, optional);
943 return identifier_type (sid);
947 Lily_lexer::scan_bare_word (string str)
949 SCM sym = ly_symbol2scm (str.c_str ());
950 if ((YYSTATE == notes) || (YYSTATE == chords)) {
951 SCM handle = SCM_BOOL_F;
952 if (scm_is_pair (pitchname_tab_stack_))
953 handle = scm_hashq_get_handle (scm_cdar (pitchname_tab_stack_), sym);
955 if (scm_is_pair (handle)) {
956 yylval.scm = scm_cdr (handle);
957 if (unsmob_pitch (yylval.scm))
958 return (YYSTATE == notes) ? NOTENAME_PITCH : TONICNAME_PITCH;
959 else if (scm_is_symbol (yylval.scm))
962 else if ((YYSTATE == chords)
963 && (handle = scm_hashq_get_handle (chordmodifier_tab_, sym))!= SCM_BOOL_F)
965 yylval.scm = scm_cdr (handle);
966 return CHORD_MODIFIER;
969 yylval.scm = ly_string2scm (str);
974 Lily_lexer::get_state () const
976 if (YY_START == extratoken)
977 return hidden_state_;
983 Lily_lexer::is_note_state () const
985 return get_state () == notes;
989 Lily_lexer::is_chord_state () const
991 return get_state () == chords;
995 Lily_lexer::is_lyric_state () const
997 return get_state () == lyrics;
1001 Lily_lexer::is_figure_state () const
1003 return get_state () == figures;
1006 // The extra_token parameter specifies how to convert multiple values
1007 // into additional tokens. For '#', additional values get pushed as
1008 // SCM_IDENTIFIER. For '$', they get checked for their type and get
1009 // pushed as a corresponding *_IDENTIFIER token. Since the latter
1010 // tampers with yylval, it can only be done from the lexer itself, so
1011 // this function is private.
1014 Lily_lexer::eval_scm (SCM readerdata, char extra_token)
1016 SCM sval = SCM_UNDEFINED;
1018 if (!SCM_UNBNDP (readerdata))
1020 sval = ly_eval_scm (scm_car (readerdata),
1021 *unsmob_input (scm_cdr (readerdata)),
1022 be_safe_global && is_main_input_,
1026 if (SCM_UNBNDP (sval))
1029 return SCM_UNSPECIFIED;
1032 if (extra_token && SCM_VALUESP (sval))
1034 sval = scm_struct_ref (sval, SCM_INUM0);
1036 if (scm_is_pair (sval)) {
1037 for (SCM v = scm_reverse (scm_cdr (sval));
1042 switch (extra_token) {
1044 token = scan_scm_id (scm_car (v));
1045 if (!scm_is_eq (yylval.scm, SCM_UNSPECIFIED))
1046 push_extra_token (token, yylval.scm);
1049 push_extra_token (SCM_IDENTIFIER, scm_car (v));
1053 sval = scm_car (sval);
1055 sval = SCM_UNSPECIFIED;
1061 /* Check for valid UTF-8 that has no overlong or surrogate codes and
1062 is in the range 0-0x10ffff */
1065 Lily_lexer::YYText_utf8 ()
1067 const char * const p = YYText ();
1068 for (int i=0; p[i];) {
1069 if ((p[i] & 0xff) < 0x80) {
1073 int oldi = i; // start of character
1074 int more = 0; // # of followup bytes, 0 if bad
1075 switch (p[i++] & 0xff) {
1076 // 0xc0 and 0xc1 are overlong prefixes for
1077 // 0x00-0x3f and 0x40-0x7f respectively, bad.
1078 case 0xc2: // 0x80-0xbf
1079 case 0xc3: // 0xc0-0xff
1080 case 0xc4: // 0x100-0x13f
1081 case 0xc5: // 0x140-0x17f
1082 case 0xc6: // 0x180-0x1bf
1083 case 0xc7: // 0x1c0-0x1ff
1084 case 0xc8: // 0x200-0x23f
1085 case 0xc9: // 0x240-0x27f
1086 case 0xca: // 0x280-0x2bf
1087 case 0xcb: // 0x2c0-0x2ff
1088 case 0xcc: // 0x300-0x33f
1089 case 0xcd: // 0x340-0x37f
1090 case 0xce: // 0x380-0x3bf
1091 case 0xcf: // 0x3c0-0x3ff
1092 case 0xd0: // 0x400-0x43f
1093 case 0xd1: // 0x440-0x47f
1094 case 0xd2: // 0x480-0x4bf
1095 case 0xd3: // 0x4c0-0x4ff
1096 case 0xd4: // 0x500-0x53f
1097 case 0xd5: // 0x540-0x57f
1098 case 0xd6: // 0x580-0x5bf
1099 case 0xd7: // 0x5c0-0x5ff
1100 case 0xd8: // 0x600-0x63f
1101 case 0xd9: // 0x640-0x67f
1102 case 0xda: // 0x680-0x6bf
1103 case 0xdb: // 0x6c0-0x6ff
1104 case 0xdc: // 0x700-0x73f
1105 case 0xdd: // 0x740-0x77f
1106 case 0xde: // 0x780-0x7bf
1107 case 0xdf: // 0x7c0-0x7ff
1108 more = 1; // 2-byte sequences, 0x80-0x7ff
1111 // don't allow overlong sequences for 0-0x7ff
1112 if ((p[i] & 0xff) < 0xa0)
1114 case 0xe1: // 0x1000-0x1fff
1115 case 0xe2: // 0x2000-0x2fff
1116 case 0xe3: // 0x3000-0x3fff
1117 case 0xe4: // 0x4000-0x4fff
1118 case 0xe5: // 0x5000-0x5fff
1119 case 0xe6: // 0x6000-0x6fff
1120 case 0xe7: // 0x7000-0x7fff
1121 case 0xe8: // 0x8000-0x8fff
1122 case 0xe9: // 0x9000-0x9fff
1123 case 0xea: // 0xa000-0xafff
1124 case 0xeb: // 0xb000-0xbfff
1125 case 0xec: // 0xc000-0xcfff
1126 more = 2; // 3-byte sequences, 0x7ff-0xcfff
1128 case 0xed: // 0xd000-0xdfff
1129 // Don't allow surrogate codes 0xd800-0xdfff
1130 if ((p[i] & 0xff) >= 0xa0)
1132 case 0xee: // 0xe000-0xefff
1133 case 0xef: // 0xf000-0xffff
1134 more = 2; // 3-byte sequences,
1135 // 0xd000-0xd7ff, 0xe000-0xffff
1138 // don't allow overlong sequences for 0-0xffff
1139 if ((p[i] & 0xff) < 0x90)
1141 case 0xf1: // 0x40000-0x7ffff
1142 case 0xf2: // 0x80000-0xbffff
1143 case 0xf3: // 0xc0000-0xfffff
1144 more = 3; // 4-byte sequences, 0x10000-0xfffff
1147 // don't allow more than 0x10ffff
1148 if ((p[i] & 0xff) >= 0x90)
1150 more = 3; // 4-byte sequence, 0x100000-0x10ffff
1154 // check that all continuation bytes are valid
1156 if ((p[i++] & 0xc0) != 0x80)
1162 Input h = here_input ();
1163 h.set (h.get_source_file (), h.start () + oldi, h.start () + i);
1164 h.warning (_ ("non-UTF-8 input").c_str ());
1171 urg, belong to string (_convert)
1172 and should be generalised
1175 strip_leading_white (string&s)
1178 for (; i < s.length (); i++)
1179 if (!isspace (s[i]))
1186 strip_trailing_white (string&s)
1188 ssize i = s.length ();
1190 if (!isspace (s[i]))
1193 s = s.substr (0, i + 1);
1198 Lilypond_version oldest_version ("2.7.38");
1202 is_valid_version (string s)
1204 Lilypond_version current ( MAJOR_VERSION "." MINOR_VERSION "." PATCH_LEVEL );
1205 Lilypond_version ver (s);
1206 if (int (ver) < oldest_version)
1208 non_fatal_error (_f ("file too old: %s (oldest supported: %s)", ver.to_string (), oldest_version.to_string ()));
1209 non_fatal_error (_ ("consider updating the input with the convert-ly script"));
1215 non_fatal_error (_f ("program too old: %s (file requires: %s)", current.to_string (), ver.to_string ()));
1226 lyric_fudge (string s)
1230 while ((i = s.find ('_', i)) != string::npos)
1238 Convert "NUM/DEN" into a '(NUM . DEN) cons.
1241 scan_fraction (string frac)
1243 ssize i = frac.find ('/');
1244 string left = frac.substr (0, i);
1245 string right = frac.substr (i + 1, (frac.length () - i + 1));
1247 int n = String_convert::dec2int (left);
1248 int d = String_convert::dec2int (right);
1249 return scm_cons (scm_from_int (n), scm_from_int (d));
1253 lookup_markup_command (string s)
1255 SCM proc = ly_lily_module_constant ("lookup-markup-command");
1256 return scm_call_1 (proc, ly_string2scm (s));
1260 lookup_markup_list_command (string s)
1262 SCM proc = ly_lily_module_constant ("lookup-markup-list-command");
1263 return scm_call_1 (proc, ly_string2scm (s));
1266 /* Shut up lexer warnings. */
1286 silence_lexer_warnings ()
1288 (void) yy_start_stack_ptr;
1289 (void) yy_start_stack_depth;
1290 (void) yy_start_stack;
1291 (void) yy_push_state;
1292 (void) yy_pop_state;
1293 (void) yy_top_state;
1294 (void) silence_lexer_warnings;