X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;ds=inline;f=lily%2Flexer.ll;h=50b79a848810b57e7500764ae22029b11799ddce;hb=59ee8eafd377bd11a4b0bf104db9aafc96b56d27;hp=e74dced0885534a2bee9ad22115dec42d60b4ef5;hpb=01df8ad908c92687d0c352e5ad5f067e52809423;p=lilypond.git diff --git a/lily/lexer.ll b/lily/lexer.ll index e74dced088..50b79a8488 100644 --- a/lily/lexer.ll +++ b/lily/lexer.ll @@ -120,6 +120,7 @@ SCM (* scm_parse_error_handler) (void *); %x incl %x lyrics %x longcomment +%x maininput %x markup %x notes %x quote @@ -149,12 +150,16 @@ N [0-9] ANY_CHAR (.|\n) WORD {A}([-_]{A}|{A})* COMMAND \\{WORD} - +/* SPECIAL category is for every letter that needs to get passed to + * the parser rather than being redefinable by the user */ +SPECIAL [-+*/=<>{}!?_^'',.:] +SHORTHAND (.|\\.) UNSIGNED {N}+ E_UNSIGNED \\{N}+ FRACTION {N}+\/{N}+ INT -?{UNSIGNED} REAL ({INT}\.{N}*)|(-?\.{N}+) +STRICTREAL {UNSIGNED}\.{UNSIGNED} WHITE [ \n\t\f\r] HORIZONTALWHITE [ \t] BLACK [^ \n\t\f\r] @@ -216,16 +221,10 @@ BOM_UTF8 \357\273\277 "%{" { yy_push_state (longcomment); } - %[^{\n\r][^\n\r]*[\n\r] { - (void) YYText_utf8 (); - } - %[^{\n\r] { // backup rule + %[^{\n\r][^\n\r]*[\n\r]? { (void) YYText_utf8 (); } - %[\n\r] { - } - %[^{\n\r][^\n\r]* { - (void) YYText_utf8 (); + %[\n\r]? { } {WHITE}+ { @@ -314,9 +313,12 @@ BOM_UTF8 \357\273\277 start_main_input (); main_input_level_ = include_stack_.size (); is_main_input_ = true; + int state = YYSTATE; + yy_push_state (maininput); + yy_push_state (state); } else - error (_ ("\\maininput not allowed outside init files")); + LexerError (_ ("\\maininput not allowed outside init files").c_str ()); } \\include { @@ -376,8 +378,8 @@ BOM_UTF8 \357\273\277 } \"[^""]* { // backup rule - error (_ ("end quote missing")); - exit (1); + LexerError (_ ("end quote missing").c_str ()); + yy_pop_state (); } /* Flex picks the longest matching pattern including trailing @@ -478,6 +480,22 @@ BOM_UTF8 \357\273\277 yylval = SCM_UNSPECIFIED; return FIGURE_OPEN; } + \\\+ { + yylval = SCM_UNSPECIFIED; + return E_PLUS; + } + \\! { + yylval = SCM_UNSPECIFIED; + return E_EXCLAMATION; + } + \\\\ { + yylval = SCM_UNSPECIFIED; + return E_BACKSLASH; + } + [][] { + yylval = SCM_UNSPECIFIED; + return YYText ()[0]; + } } { @@ -494,8 +512,12 @@ BOM_UTF8 \357\273\277 yylval = scan_fraction (YYText ()); return FRACTION; } - {UNSIGNED}/\/ | // backup rule - {UNSIGNED} { + {STRICTREAL} { + yylval = scm_c_read_string (YYText ()); + return REAL; + } + {UNSIGNED}/[/.] | // backup rule + {UNSIGNED} { yylval = scm_c_read_string (YYText ()); return UNSIGNED; } @@ -525,7 +547,7 @@ BOM_UTF8 \357\273\277 SCM_UNDEFINED, SCM_UNDEFINED); - return is_lyric_state () ? LYRICS_STRING : STRING; + return STRING; } \\ { yylval = scm_cons (scm_from_locale_string (YYText ()), @@ -541,7 +563,11 @@ BOM_UTF8 \357\273\277 yylval = scan_fraction (YYText ()); return FRACTION; } - {UNSIGNED}/\/ | // backup rule + {STRICTREAL} { + yylval = scm_c_read_string (YYText ()); + return REAL; + } + {UNSIGNED}/[/.] | // backup rule {UNSIGNED} { yylval = scm_c_read_string (YYText ()); return UNSIGNED; @@ -550,8 +576,12 @@ BOM_UTF8 \357\273\277 {COMMAND} { return scan_escaped_word (YYText_utf8 () + 1); } - /* Characters needed to express durations, assignments, barchecks */ - [*.=|] { + \\.|\| { + // UTF-8 already covered by COMMAND + return scan_shorthand (YYText ()); + } + /* Characters needed to express durations, assignments */ + [*.=] { yylval = SCM_UNSPECIFIED; return YYText ()[0]; } @@ -566,12 +596,12 @@ BOM_UTF8 \357\273\277 s = lyric_fudge (s); yylval = ly_string2scm (s); - return LYRICS_STRING; + return STRING; } /* This should really just cover {} */ - . { + [{}] { yylval = SCM_UNSPECIFIED; - return YYText ()[0]; // above catches all multibytes. + return YYText ()[0]; } } { @@ -612,10 +642,6 @@ BOM_UTF8 \357\273\277 yylval = SCM_UNSPECIFIED; return CHORD_CARET; } - . { - yylval = SCM_UNSPECIFIED; - return YYText ()[0]; // WORD catches all multibyte. - } } @@ -681,27 +707,40 @@ BOM_UTF8 \357\273\277 yylval = ly_string2scm (s); return STRING; } - . { + [{}] { yylval = SCM_UNSPECIFIED; - return YYText ()[0]; // Above is catchall for multibyte + return YYText ()[0]; } } <> { LexerError (_ ("EOF found inside a comment").c_str ()); - is_main_input_ = false; // should be safe , can't have \include in --safe. - yylval = SCM_UNSPECIFIED; - if (!close_input ()) - yyterminate (); // can't move this, since it actually rets a YY_NULL + yy_pop_state (); } +<> { + LexerError (_ ("EOF found inside string").c_str ()); + yy_pop_state (); +} + <> { yylval = SCM_UNSPECIFIED; if (is_main_input_) { is_main_input_ = include_stack_.size () > main_input_level_; if (!is_main_input_) + { main_input_level_ = 0; + pop_state (); + if (YYSTATE != maininput) + { + LexerError (_ ("Unfinished main input").c_str ()); + do { + pop_state (); + } while (YYSTATE != maininput); + } + pop_state (); + } if (!close_input () || !is_main_input_) /* Returns YY_NULL */ yyterminate (); @@ -711,6 +750,13 @@ BOM_UTF8 \357\273\277 yyterminate (); } +{ANY_CHAR} { + while (include_stack_.size () > main_input_level_ + && close_input ()) + ; + yyterminate (); +} + { {WORD}/[-_] | // backup rule {WORD} { @@ -740,51 +786,18 @@ BOM_UTF8 \357\273\277 } -[{}] { - yylval = SCM_UNSPECIFIED; - return YYText ()[0]; -} - --/\. | // backup rule -[*:=] { +-/\. { // backup rule yylval = SCM_UNSPECIFIED; return YYText ()[0]; } -. { +{SPECIAL} { yylval = SCM_UNSPECIFIED; return YYText ()[0]; } -\\. { - yylval = SCM_UNSPECIFIED; - char c = YYText ()[1]; - - switch (c) { - case '>': - return E_ANGLE_CLOSE; - case '<': - return E_ANGLE_OPEN; - case '!': - return E_EXCLAMATION; - case '(': - return E_OPEN; - case ')': - return E_CLOSE; - case '[': - return E_BRACKET_OPEN; - case '+': - return E_PLUS; - case ']': - return E_BRACKET_CLOSE; - case '~': - return E_TILDE; - case '\\': - return E_BACKSLASH; - - default: - return E_CHAR; - } +{SHORTHAND} { + return scan_shorthand (YYText_utf8 ()); // should not be utf-8 } <*>.[\200-\277]* { @@ -902,12 +915,16 @@ Lily_lexer::scan_escaped_word (string str) yylval = SCM_UNSPECIFIED; int i = lookup_keyword (str); - if (i == MARKUP && is_lyric_state ()) - return LYRIC_MARKUP; + if (i != -1) return i; SCM sid = lookup_identifier (str); + if (Music *m = unsmob_music (sid)) + { + m->set_spot (override_input (last_input_)); + } + if (sid != SCM_UNDEFINED) return scan_scm_id (sid); @@ -919,6 +936,26 @@ Lily_lexer::scan_escaped_word (string str) return STRING; } +int +Lily_lexer::scan_shorthand (string str) +{ + SCM sid = lookup_identifier (str); + if (Music *m = unsmob_music (sid)) + { + m->set_spot (override_input (last_input_)); + } + + if (sid != SCM_UNDEFINED) + return scan_scm_id (sid); + + string msg (_f ("undefined character or shorthand: %s", str)); + LexerError (msg.c_str ()); + + yylval = ly_string2scm (str); + + return STRING; +} + int Lily_lexer::scan_scm_id (SCM sid) { @@ -1067,19 +1104,26 @@ Lily_lexer::eval_scm (SCM readerdata, char extra_token) sval = scm_struct_ref (sval, SCM_INUM0); if (scm_is_pair (sval)) { - for (SCM v = scm_reverse (scm_cdr (sval)); - scm_is_pair (v); - v = scm_cdr (v)) + for (SCM p = scm_reverse (scm_cdr (sval)); + scm_is_pair (p); + p = scm_cdr (p)) { + SCM v = scm_car (p); + if (Music *m = unsmob_music (v)) + { + if (!unsmob_input (m->get_property ("origin"))) + m->set_spot (override_input (last_input_)); + } + int token; switch (extra_token) { case '$': - token = scan_scm_id (scm_car (v)); + token = scan_scm_id (v); if (!scm_is_eq (yylval, SCM_UNSPECIFIED)) push_extra_token (token, yylval); break; case '#': - push_extra_token (SCM_IDENTIFIER, scm_car (v)); + push_extra_token (SCM_IDENTIFIER, v); break; } } @@ -1088,6 +1132,12 @@ Lily_lexer::eval_scm (SCM readerdata, char extra_token) sval = SCM_UNSPECIFIED; } + if (Music *m = unsmob_music (sval)) + { + if (!unsmob_input (m->get_property ("origin"))) + m->set_spot (override_input (last_input_)); + } + return sval; }