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.
156 NATIONAL [\001-\006\021-\027\031\036]
157 TEX {AA}|-|{PUNCT}|{NATIONAL}|{SPECIAL_CHAR}
158 WORD {A}([-_]{A}|{A})*
165 REAL ({INT}\.{N}*)|(-?\.{N}+)
167 HORIZONTALWHITE [ \t]
170 LYRICS ({AA}|{TEX})[^0-9 \t\n\r\f]*
174 BOM_UTF8 \357\273\277
180 // swallow and ignore carriage returns
183 <extratoken>{ANY_CHAR} {
184 /* Generate a token without swallowing anything */
186 /* First unswallow the eaten character */
187 add_lexed_char (-YYLeng ());
190 /* produce requested token */
191 int type = scm_to_int (scm_caar (extra_tokens_));
192 yylval.scm = scm_cdar (extra_tokens_);
193 extra_tokens_ = scm_cdr (extra_tokens_);
194 if (scm_is_null (extra_tokens_))
200 <extratoken><<EOF>> {
201 /* Generate a token without swallowing anything */
203 /* produce requested token */
204 int type = scm_to_int (scm_caar (extra_tokens_));
205 yylval.scm = scm_cdar (extra_tokens_);
206 extra_tokens_ = scm_cdr (extra_tokens_);
207 if (scm_is_null (extra_tokens_))
213 /* Use the trailing context feature. Otherwise, the BOM will not be
214 found if the file starts with an identifier definition. */
215 <INITIAL,chords,lyrics,figures,notes>{BOM_UTF8}/.* {
216 if (this->lexloc_->line_number () != 1 || this->lexloc_->column_number () != 0)
218 LexerWarning (_ ("stray UTF-8 BOM encountered").c_str ());
221 debug_output (_ ("Skipping UTF-8 BOM"));
224 <INITIAL,chords,figures,incl,lyrics,markup,notes>{
226 yy_push_state (longcomment);
228 %[^{\n\r][^\n\r]*[\n\r] {
229 (void) YYText_utf8 ();
231 %[^{\n\r] { // backup rule
232 (void) YYText_utf8 ();
237 (void) YYText_utf8 ();
244 <INITIAL,notes,figures,chords,markup>{
250 <INITIAL,chords,lyrics,notes,figures>\\version{WHITE}* {
251 yy_push_state (version);
253 <INITIAL,chords,lyrics,notes,figures>\\sourcefilename{WHITE}* {
254 yy_push_state (sourcefilename);
256 <INITIAL,chords,lyrics,notes,figures>\\sourcefileline{WHITE}* {
257 yy_push_state (sourcefileline);
259 <version>\"[^"]*\" { /* got the version number */
260 string s (YYText_utf8 () + 1);
261 s = s.substr (0, s.rfind ('\"'));
265 SCM top_scope = scm_car (scm_last_pair (scopes_));
266 scm_module_define (top_scope, ly_symbol2scm ("version-seen"), SCM_BOOL_T);
268 if (!is_valid_version (s))
273 <sourcefilename>\"[^""]*\" {
274 string s (YYText_utf8 () + 1);
275 s = s.substr (0, s.rfind ('\"'));
278 this->here_input().get_source_file ()->name_ = s;
279 message (_f ("Renaming input to: `%s'", s.c_str ()));
280 progress_indication ("\n");
281 scm_module_define (scm_car (scopes_),
282 ly_symbol2scm ("input-file-name"),
287 <sourcefileline>{INT} {
289 sscanf (YYText (), "%d", &i);
292 this->here_input ().get_source_file ()->set_line (here_input ().start (), i);
295 <version>{ANY_CHAR} {
296 LexerError (_ ("quoted string expected after \\version").c_str ());
299 <sourcefilename>{ANY_CHAR} {
300 LexerError (_ ("quoted string expected after \\sourcefilename").c_str ());
303 <sourcefileline>{ANY_CHAR} {
304 LexerError (_ ("integer expected after \\sourcefileline").c_str ());
309 (void) YYText_utf8 ();
312 (void) YYText_utf8 ();
320 <INITIAL,chords,lyrics,notes,figures>\\maininput {
324 is_main_input_ = true;
327 error (_ ("\\maininput not allowed outside init files"));
330 <INITIAL,chords,lyrics,figures,notes>\\include {
331 yy_push_state (incl);
333 <incl>\"[^""]*\" { /* got the include file name */
334 string s (YYText_utf8 ()+1);
335 s = s.substr (0, s.rfind ('"'));
337 new_input (s, sources_);
340 <incl>\\{BLACK}*{WHITE}? { /* got the include identifier */
341 string s = YYText_utf8 () + 1;
342 strip_trailing_white (s);
343 if (s.length () && (s[s.length () - 1] == ';'))
344 s = s.substr (0, s.length () - 1);
346 SCM sid = lookup_identifier (s);
347 if (scm_is_string (sid)) {
348 new_input (ly_scm2string (sid), sources_);
351 string msg (_f ("wrong or undefined identifier: `%s'", s ));
353 LexerError (msg.c_str ());
354 SCM err = scm_current_error_port ();
355 scm_puts ("This value was found in the table: ", err);
356 scm_display (sid, err);
359 <incl>(\$|#) { // scm for the filename
361 Input hi = here_input();
363 SCM sval = ly_parse_scm (hi.start (), &n, hi,
364 be_safe_global && is_main_input_, parser_);
365 sval = eval_scm (sval);
367 for (int i = 0; i < n; i++)
371 char_count_stack_.back () += n;
373 if (scm_is_string (sval)) {
374 new_input (ly_scm2string (sval), sources_);
377 LexerError (_ ("string expected after \\include").c_str ());
378 if (sval != SCM_UNDEFINED) {
379 SCM err = scm_current_error_port ();
380 scm_puts ("This value was found instead: ", err);
381 scm_display (sval, err);
386 <incl,version,sourcefilename>\"[^""]* { // backup rule
387 error (_ ("end quote missing"));
391 /* Flex picks the longest matching pattern including trailing
392 * contexts. Without the backup pattern, r-. does not trigger the
393 * {RESTNAME} rule but rather the {WORD}/[-_] rule coming later,
394 * needed for avoiding backup states.
397 <chords,notes,figures>{RESTNAME}/[-_] | // pseudo backup rule
398 <chords,notes,figures>{RESTNAME} {
399 char const *s = YYText ();
400 yylval.scm = scm_from_locale_string (s);
403 <chords,notes,figures>q/[-_] | // pseudo backup rule
404 <chords,notes,figures>q {
405 return CHORD_REPETITION;
408 <chords,notes,figures>R/[-_] | // pseudo backup rule
409 <chords,notes,figures>R {
410 return MULTI_MEASURE_REST;
412 <INITIAL,chords,figures,lyrics,markup,notes># { //embedded scm
414 Input hi = here_input();
416 SCM sval = ly_parse_scm (hi.start (), &n, hi,
417 be_safe_global && is_main_input_, parser_);
419 if (sval == SCM_UNDEFINED)
422 for (int i = 0; i < n; i++)
426 char_count_stack_.back () += n;
432 <INITIAL,chords,figures,lyrics,markup,notes>\$ { //immediate scm
434 Input hi = here_input();
436 SCM sval = ly_parse_scm (hi.start (), &n, hi,
437 be_safe_global && is_main_input_, parser_);
439 for (int i = 0; i < n; i++)
443 char_count_stack_.back () += n;
445 sval = eval_scm (sval, '$');
447 int token = scan_scm_id (sval);
448 if (!scm_is_eq (yylval.scm, SCM_UNSPECIFIED))
452 <INITIAL,notes,lyrics>{
454 return DOUBLE_ANGLE_OPEN;
457 return DOUBLE_ANGLE_CLOSE;
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.scm = scan_fraction (YYText ());
496 {UNSIGNED}/\/ | // backup rule
498 yylval.scm = scm_c_read_string (YYText ());
502 yylval.i = String_convert::dec2int (string (YYText () +1));
509 *yylval.string += to_string (escaped_char (YYText ()[1]));
512 *yylval.string += YYText_utf8 ();
518 /* yylval is union. Must remember STRING before setting SCM*/
519 string *sp = yylval.string;
520 yylval.scm = ly_string2scm (*sp);
522 return is_lyric_state () ? LYRICS_STRING : STRING;
525 *yylval.string += YYText ();
531 start_lyric_quote ();
534 yylval.scm = scan_fraction (YYText ());
537 {UNSIGNED}/\/ | // backup rule
539 yylval.scm = scm_c_read_string (YYText ());
542 {COMMAND}/[-_] | // backup rule
544 return scan_escaped_word (YYText_utf8 () + 1);
548 string s (YYText_utf8 ());
550 return yylval.i = EXTENDER;
552 return yylval.i = HYPHEN;
555 char c = s[s.length () - 1];
556 if (c == '{' || c == '}') // brace open is for not confusing dumb tools.
557 here_input ().warning (
558 _ ("Brace found at end of lyric. Did you forget a space?"));
559 yylval.scm = ly_string2scm (s);
562 return LYRICS_STRING;
565 return YYText ()[0]; // LYRICS already catches all multibytes.
569 {WORD}/[-_] | // backup rule
571 return scan_bare_word (YYText_utf8 ());
573 {COMMAND}/[-_] | // backup rule
575 return scan_escaped_word (YYText_utf8 () + 1);
578 yylval.scm = scan_fraction (YYText ());
581 {UNSIGNED}/\/ | // backup rule
583 yylval.scm = scm_c_read_string (YYText ());
602 return YYText ()[0]; // WORD catches all multibyte.
611 {COMMAND}/[-_] | // backup rule
613 string str (YYText_utf8 () + 1);
615 int token_type = MARKUP_FUNCTION;
616 SCM s = lookup_markup_command (str);
618 // lookup-markup-command returns a pair with the car
619 // being the function to call, and the cdr being the
620 // call signature specified to define-markup-command,
621 // a list of predicates.
623 if (!scm_is_pair (s)) {
624 // If lookup-markup-command was not successful, we
625 // try lookup-markup-list-command instead.
626 // If this fails as well, we just scan and return
628 s = lookup_markup_list_command (str);
630 token_type = MARKUP_LIST_FUNCTION;
632 return scan_escaped_word (str);
635 // If the list of predicates is, say,
636 // (number? number? markup?), then tokens
637 // EXPECT_MARKUP EXPECT_SCM EXPECT_SCM EXPECT_NO_MORE_ARGS
638 // will be generated. Note that we have to push them
639 // in reverse order, so the first token pushed in the
640 // loop will be EXPECT_NO_MORE_ARGS.
642 yylval.scm = scm_car(s);
644 // yylval now contains the function to call as token
645 // value (for token type MARKUP_FUNCTION or
646 // MARKUP_LIST_FUNCTION).
648 push_extra_token(EXPECT_NO_MORE_ARGS);
650 for (; scm_is_pair(s); s = scm_cdr(s)) {
651 SCM predicate = scm_car(s);
653 if (predicate == ly_lily_module_constant ("markup-list?"))
654 push_extra_token(EXPECT_MARKUP_LIST);
655 else if (predicate == ly_lily_module_constant ("markup?"))
656 push_extra_token(EXPECT_MARKUP);
658 push_extra_token(EXPECT_SCM, predicate);
665 [^$#{}\"\\ \t\n\r\f]+ {
666 string s (YYText_utf8 ());
668 char c = s[s.length () - 1];
669 /* brace open is for not confusing dumb tools. */
670 if (c == '{' || c == '}')
671 here_input ().warning (
672 _ ("Brace found at end of markup. Did you forget a space?"));
673 yylval.scm = ly_string2scm (s);
679 return YYText()[0]; // Above is catchall for multibyte
683 <longcomment><<EOF>> {
684 LexerError (_ ("EOF found inside a comment").c_str ());
685 is_main_input_ = false; // should be safe , can't have \include in --safe.
687 yyterminate (); // can't move this, since it actually rets a YY_NULL
690 <<EOF>> { if (is_main_input_)
692 /* 2 = init.ly + current file.
693 > because we're before closing, but is_main_input_ should
696 is_main_input_ = include_stack_.size () > 2;
697 if (!close_input () || !is_main_input_)
698 /* Returns YY_NULL */
701 else if (!close_input ())
702 /* Returns YY_NULL */
707 {WORD}/[-_] | // backup rule
709 return scan_bare_word (YYText_utf8 ());
711 {COMMAND}/[-_] | // backup rule
713 return scan_escaped_word (YYText_utf8 () + 1);
718 yylval.scm = scan_fraction (YYText ());
722 -{UNSIGNED} | // backup rule
724 yylval.scm = scm_c_read_string (YYText ());
728 {UNSIGNED}/\/ | // backup rule
730 yylval.scm = scm_c_read_string (YYText ());
740 -/\. | // backup rule
742 char c = YYText ()[0];
747 <INITIAL,notes,figures>. {
751 <INITIAL,lyrics,notes,figures>\\. {
752 char c = YYText ()[1];
756 return E_ANGLE_CLOSE;
760 return E_EXCLAMATION;
766 return E_BRACKET_OPEN;
770 return E_BRACKET_CLOSE;
782 string msg = _f ("invalid character: `%s'", YYText_utf8 ());
783 LexerError (msg.c_str ());
784 return '%'; // Better not return half a utf8 character.
789 /* Make the lexer generate a token of the given type as the next token.
790 TODO: make it possible to define a value for the token as well */
792 Lily_lexer::push_extra_token (int token_type, SCM scm)
794 if (scm_is_null (extra_tokens_))
796 if (YY_START != extratoken)
797 hidden_state_ = YY_START;
798 yy_push_state (extratoken);
800 extra_tokens_ = scm_acons (scm_from_int (token_type), scm, extra_tokens_);
804 Lily_lexer::push_chord_state (SCM alist)
806 SCM p = scm_assq (alist, pitchname_tab_stack_);
808 if (scm_is_false (p))
809 p = scm_cons (alist, alist_to_hashq (alist));
810 pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_);
811 yy_push_state (chords);
815 Lily_lexer::push_figuredbass_state ()
817 yy_push_state (figures);
821 Lily_lexer::push_initial_state ()
823 yy_push_state (INITIAL);
827 Lily_lexer::push_lyric_state ()
829 yy_push_state (lyrics);
833 Lily_lexer::push_markup_state ()
835 yy_push_state (markup);
839 Lily_lexer::push_note_state (SCM alist)
841 bool extra = (YYSTATE == extratoken);
843 SCM p = scm_assq (alist, pitchname_tab_stack_);
848 if (scm_is_false (p))
849 p = scm_cons (alist, alist_to_hashq (alist));
850 pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_);
851 yy_push_state (notes);
854 hidden_state_ = YYSTATE;
855 yy_push_state (extratoken);
860 Lily_lexer::pop_state ()
862 bool extra = (YYSTATE == extratoken);
867 if (YYSTATE == notes || YYSTATE == chords)
868 pitchname_tab_stack_ = scm_cdr (pitchname_tab_stack_);
873 hidden_state_ = YYSTATE;
874 yy_push_state (extratoken);
879 Lily_lexer::identifier_type (SCM sid)
881 int k = try_special_identifiers (&yylval.scm , sid);
882 return k >= 0 ? k : SCM_IDENTIFIER;
887 Lily_lexer::scan_escaped_word (string str)
889 // use more SCM for this.
891 // SCM sym = ly_symbol2scm (str.c_str ());
893 int i = lookup_keyword (str);
894 if (i == MARKUP && is_lyric_state ())
899 SCM sid = lookup_identifier (str);
900 if (sid != SCM_UNDEFINED)
901 return scan_scm_id (sid);
903 string msg (_f ("unknown escaped string: `\\%s'", str));
904 LexerError (msg.c_str ());
906 yylval.scm = ly_string2scm (str);
912 Lily_lexer::scan_scm_id (SCM sid)
914 if (is_music_function (sid))
916 int funtype = SCM_FUNCTION;
920 SCM s = get_music_function_signature (sid);
921 SCM cs = scm_car (s);
923 if (scm_is_pair (cs))
928 if (scm_is_eq (cs, ly_lily_module_constant ("ly:music?")))
929 funtype = MUSIC_FUNCTION;
930 else if (scm_is_eq (cs, ly_lily_module_constant ("ly:event?")))
931 funtype = EVENT_FUNCTION;
932 else if (ly_is_procedure (cs))
933 funtype = SCM_FUNCTION;
934 else programming_error ("Bad syntax function predicate");
936 push_extra_token (EXPECT_NO_MORE_ARGS);
937 for (s = scm_cdr (s); scm_is_pair (s); s = scm_cdr (s))
939 SCM optional = SCM_UNDEFINED;
942 if (scm_is_pair (cs))
944 optional = SCM_CDR (cs);
948 if (cs == Pitch_type_p_proc)
949 push_extra_token (EXPECT_PITCH);
950 else if (cs == Duration_type_p_proc)
951 push_extra_token (EXPECT_DURATION);
952 else if (ly_is_procedure (cs))
953 push_extra_token (EXPECT_SCM, cs);
956 programming_error ("Function parameter without type-checking predicate");
959 if (!scm_is_eq (optional, SCM_UNDEFINED))
960 push_extra_token (EXPECT_OPTIONAL, optional);
965 return identifier_type (sid);
969 Lily_lexer::scan_bare_word (string str)
971 SCM sym = ly_symbol2scm (str.c_str ());
972 if ((YYSTATE == notes) || (YYSTATE == chords)) {
973 SCM handle = SCM_BOOL_F;
974 if (scm_is_pair (pitchname_tab_stack_))
975 handle = scm_hashq_get_handle (scm_cdar (pitchname_tab_stack_), sym);
977 if (scm_is_pair (handle)) {
978 yylval.scm = scm_cdr (handle);
979 if (unsmob_pitch (yylval.scm))
980 return (YYSTATE == notes) ? NOTENAME_PITCH : TONICNAME_PITCH;
981 else if (scm_is_symbol (yylval.scm))
984 else if ((YYSTATE == chords)
985 && (handle = scm_hashq_get_handle (chordmodifier_tab_, sym))!= SCM_BOOL_F)
987 yylval.scm = scm_cdr (handle);
988 return CHORD_MODIFIER;
991 yylval.scm = ly_string2scm (str);
996 Lily_lexer::get_state () const
998 if (YY_START == extratoken)
999 return hidden_state_;
1005 Lily_lexer::is_note_state () const
1007 return get_state () == notes;
1011 Lily_lexer::is_chord_state () const
1013 return get_state () == chords;
1017 Lily_lexer::is_lyric_state () const
1019 return get_state () == lyrics;
1023 Lily_lexer::is_figure_state () const
1025 return get_state () == figures;
1028 // The extra_token parameter specifies how to convert multiple values
1029 // into additional tokens. For '#', additional values get pushed as
1030 // SCM_IDENTIFIER. For '$', they get checked for their type and get
1031 // pushed as a corresponding *_IDENTIFIER token. Since the latter
1032 // tampers with yylval, it can only be done from the lexer itself, so
1033 // this function is private.
1036 Lily_lexer::eval_scm (SCM readerdata, char extra_token)
1038 SCM sval = SCM_UNDEFINED;
1040 if (!SCM_UNBNDP (readerdata))
1042 sval = ly_eval_scm (scm_car (readerdata),
1043 *unsmob_input (scm_cdr (readerdata)),
1044 be_safe_global && is_main_input_,
1048 if (SCM_UNBNDP (sval))
1051 return SCM_UNSPECIFIED;
1054 if (extra_token && SCM_VALUESP (sval))
1056 sval = scm_struct_ref (sval, SCM_INUM0);
1058 if (scm_is_pair (sval)) {
1059 for (SCM v = scm_reverse (scm_cdr (sval));
1064 switch (extra_token) {
1066 token = scan_scm_id (scm_car (v));
1067 if (!scm_is_eq (yylval.scm, SCM_UNSPECIFIED))
1068 push_extra_token (token, yylval.scm);
1071 push_extra_token (SCM_IDENTIFIER, scm_car (v));
1075 sval = scm_car (sval);
1077 sval = SCM_UNSPECIFIED;
1083 /* Check for valid UTF-8 that has no overlong or surrogate codes and
1084 is in the range 0-0x10ffff */
1087 Lily_lexer::YYText_utf8 ()
1089 const char * const p = YYText ();
1090 for (int i=0; p[i];) {
1091 if ((p[i] & 0xff) < 0x80) {
1095 int oldi = i; // start of character
1096 int more = 0; // # of followup bytes, 0 if bad
1097 switch (p[i++] & 0xff) {
1098 // 0xc0 and 0xc1 are overlong prefixes for
1099 // 0x00-0x3f and 0x40-0x7f respectively, bad.
1100 case 0xc2: // 0x80-0xbf
1101 case 0xc3: // 0xc0-0xff
1102 case 0xc4: // 0x100-0x13f
1103 case 0xc5: // 0x140-0x17f
1104 case 0xc6: // 0x180-0x1bf
1105 case 0xc7: // 0x1c0-0x1ff
1106 case 0xc8: // 0x200-0x23f
1107 case 0xc9: // 0x240-0x27f
1108 case 0xca: // 0x280-0x2bf
1109 case 0xcb: // 0x2c0-0x2ff
1110 case 0xcc: // 0x300-0x33f
1111 case 0xcd: // 0x340-0x37f
1112 case 0xce: // 0x380-0x3bf
1113 case 0xcf: // 0x3c0-0x3ff
1114 case 0xd0: // 0x400-0x43f
1115 case 0xd1: // 0x440-0x47f
1116 case 0xd2: // 0x480-0x4bf
1117 case 0xd3: // 0x4c0-0x4ff
1118 case 0xd4: // 0x500-0x53f
1119 case 0xd5: // 0x540-0x57f
1120 case 0xd6: // 0x580-0x5bf
1121 case 0xd7: // 0x5c0-0x5ff
1122 case 0xd8: // 0x600-0x63f
1123 case 0xd9: // 0x640-0x67f
1124 case 0xda: // 0x680-0x6bf
1125 case 0xdb: // 0x6c0-0x6ff
1126 case 0xdc: // 0x700-0x73f
1127 case 0xdd: // 0x740-0x77f
1128 case 0xde: // 0x780-0x7bf
1129 case 0xdf: // 0x7c0-0x7ff
1130 more = 1; // 2-byte sequences, 0x80-0x7ff
1133 // don't allow overlong sequences for 0-0x7ff
1134 if ((p[i] & 0xff) < 0xa0)
1136 case 0xe1: // 0x1000-0x1fff
1137 case 0xe2: // 0x2000-0x2fff
1138 case 0xe3: // 0x3000-0x3fff
1139 case 0xe4: // 0x4000-0x4fff
1140 case 0xe5: // 0x5000-0x5fff
1141 case 0xe6: // 0x6000-0x6fff
1142 case 0xe7: // 0x7000-0x7fff
1143 case 0xe8: // 0x8000-0x8fff
1144 case 0xe9: // 0x9000-0x9fff
1145 case 0xea: // 0xa000-0xafff
1146 case 0xeb: // 0xb000-0xbfff
1147 case 0xec: // 0xc000-0xcfff
1148 more = 2; // 3-byte sequences, 0x7ff-0xcfff
1150 case 0xed: // 0xd000-0xdfff
1151 // Don't allow surrogate codes 0xd800-0xdfff
1152 if ((p[i] & 0xff) >= 0xa0)
1154 case 0xee: // 0xe000-0xefff
1155 case 0xef: // 0xf000-0xffff
1156 more = 2; // 3-byte sequences,
1157 // 0xd000-0xd7ff, 0xe000-0xffff
1160 // don't allow overlong sequences for 0-0xffff
1161 if ((p[i] & 0xff) < 0x90)
1163 case 0xf1: // 0x40000-0x7ffff
1164 case 0xf2: // 0x80000-0xbffff
1165 case 0xf3: // 0xc0000-0xfffff
1166 more = 3; // 4-byte sequences, 0x10000-0xfffff
1169 // don't allow more than 0x10ffff
1170 if ((p[i] & 0xff) >= 0x90)
1172 more = 3; // 4-byte sequence, 0x100000-0x10ffff
1176 // check that all continuation bytes are valid
1178 if ((p[i++] & 0xc0) != 0x80)
1184 Input h = here_input ();
1185 h.set (h.get_source_file (), h.start () + oldi, h.start () + i);
1186 h.warning (_ ("non-UTF-8 input").c_str ());
1193 urg, belong to string (_convert)
1194 and should be generalised
1197 strip_leading_white (string&s)
1200 for (; i < s.length (); i++)
1201 if (!isspace (s[i]))
1208 strip_trailing_white (string&s)
1210 ssize i = s.length ();
1212 if (!isspace (s[i]))
1215 s = s.substr (0, i + 1);
1220 Lilypond_version oldest_version ("2.7.38");
1224 is_valid_version (string s)
1226 Lilypond_version current ( MAJOR_VERSION "." MINOR_VERSION "." PATCH_LEVEL );
1227 Lilypond_version ver (s);
1228 if (int (ver) < oldest_version)
1230 non_fatal_error (_f ("file too old: %s (oldest supported: %s)", ver.to_string (), oldest_version.to_string ()));
1231 non_fatal_error (_ ("consider updating the input with the convert-ly script"));
1237 non_fatal_error (_f ("program too old: %s (file requires: %s)", current.to_string (), ver.to_string ()));
1248 lyric_fudge (string s)
1252 while ((i = s.find ('_', i)) != string::npos)
1260 Convert "NUM/DEN" into a '(NUM . DEN) cons.
1263 scan_fraction (string frac)
1265 ssize i = frac.find ('/');
1266 string left = frac.substr (0, i);
1267 string right = frac.substr (i + 1, (frac.length () - i + 1));
1269 int n = String_convert::dec2int (left);
1270 int d = String_convert::dec2int (right);
1271 return scm_cons (scm_from_int (n), scm_from_int (d));
1275 lookup_markup_command (string s)
1277 SCM proc = ly_lily_module_constant ("lookup-markup-command");
1278 return scm_call_1 (proc, ly_string2scm (s));
1282 lookup_markup_list_command (string s)
1284 SCM proc = ly_lily_module_constant ("lookup-markup-list-command");
1285 return scm_call_1 (proc, ly_string2scm (s));
1288 /* Shut up lexer warnings. */
1308 silence_lexer_warnings ()
1310 (void) yy_start_stack_ptr;
1311 (void) yy_start_stack_depth;
1312 (void) yy_start_stack;
1313 (void) yy_push_state;
1314 (void) yy_pop_state;
1315 (void) yy_top_state;
1316 (void) silence_lexer_warnings;