X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=lily%2Flexer.ll;h=51a33fbb4b2a57b446ef954c4af9c1b962f27003;hb=8848da24e00c75d7de626b2ecc409fb45e87ce4a;hp=f0fed79c7f9b8b936233034f435acb43c09bbb32;hpb=2b5668b7afafe5939eb047e351adddf44bc40bb5;p=lilypond.git diff --git a/lily/lexer.ll b/lily/lexer.ll index f0fed79c7f..51a33fbb4b 100644 --- a/lily/lexer.ll +++ b/lily/lexer.ll @@ -1,4 +1,4 @@ -%{ // -*- mode: c++; c-file-style: "linux" -*- +%{ // -*- mode: c++; c-file-style: "linux"; indent-tabs-mode: t -*- /* This file is part of LilyPond, the GNU music typesetter. @@ -91,8 +91,12 @@ bool is_valid_version (string s); yylval = SCM_EOL; \ } while (0) -#define start_lyric_quote() do { \ - yy_push_state (lyric_quote); \ +/* + The inside of \"violin1" is marked by commandquote mode +*/ + +#define start_command_quote() do { \ + yy_push_state (commandquote); \ yylval = SCM_EOL; \ } while (0) @@ -119,16 +123,16 @@ SCM (* scm_parse_error_handler) (void *); %option never-interactive %option warn -%x extratoken %x chords %x figures %x incl %x lyrics -%x lyric_quote %x longcomment +%x maininput %x markup %x notes %x quote +%x commandquote %x sourcefileline %x sourcefilename %x version @@ -153,24 +157,23 @@ A [a-zA-Z\200-\377] AA {A}|_ N [0-9] ANY_CHAR (.|\n) -PUNCT [][()?!:'`] -SPECIAL_CHAR [&@] -NATIONAL [\001-\006\021-\027\031\036] -TEX {AA}|-|{PUNCT}|{NATIONAL}|{SPECIAL_CHAR} 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] RESTNAME [rs] -LYRICS ({AA}|{TEX})[^0-9 \t\n\r\f]* -ESCAPED [nt\\'"] +ESCAPED [nt\\''""] EXTENDER __ HYPHEN -- BOM_UTF8 \357\273\277 @@ -182,36 +185,6 @@ BOM_UTF8 \357\273\277 // swallow and ignore carriage returns } -{ANY_CHAR} { - /* Generate a token without swallowing anything */ - - /* First unswallow the eaten character */ - add_lexed_char (-YYLeng ()); - yyless (0); - - /* produce requested token */ - int type = scm_to_int (scm_caar (extra_tokens_)); - yylval = scm_cdar (extra_tokens_); - extra_tokens_ = scm_cdr (extra_tokens_); - if (scm_is_null (extra_tokens_)) - yy_pop_state (); - - return type; -} - -<> { - /* Generate a token without swallowing anything */ - - /* produce requested token */ - int type = scm_to_int (scm_caar (extra_tokens_)); - yylval = scm_cdar (extra_tokens_); - extra_tokens_ = scm_cdr (extra_tokens_); - if (scm_is_null (extra_tokens_)) - yy_pop_state (); - - return type; -} - /* Use the trailing context feature. Otherwise, the BOM will not be found if the file starts with an identifier definition. */ {BOM_UTF8}/.* { @@ -227,16 +200,10 @@ BOM_UTF8 \357\273\277 "%{" { yy_push_state (longcomment); } - %[^{\n\r][^\n\r]*[\n\r] { + %[^{\n\r][^\n\r]*[\n\r]? { (void) YYText_utf8 (); } - %[^{\n\r] { // backup rule - (void) YYText_utf8 (); - } - %[\n\r] { - } - %[^{\n\r][^\n\r]* { - (void) YYText_utf8 (); + %[\n\r]? { } {WHITE}+ { @@ -258,7 +225,7 @@ BOM_UTF8 \357\273\277 \\sourcefileline{WHITE}* { yy_push_state (sourcefileline); } -\"[^"]*\" { /* got the version number */ +\"[^""]*\" { /* got the version number */ string s (YYText_utf8 () + 1); s = s.substr (0, s.rfind ('\"')); @@ -323,10 +290,14 @@ BOM_UTF8 \357\273\277 if (!is_main_input_) { 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 { @@ -386,8 +357,8 @@ 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 @@ -488,6 +459,22 @@ 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]; + } } { @@ -495,7 +482,9 @@ BOM_UTF8 \357\273\277 {WORD} { return scan_bare_word (YYText_utf8 ()); } - + \\\" { + start_command_quote (); + } {COMMAND}/[-_] | // backup rule {COMMAND} { return scan_escaped_word (YYText_utf8 () + 1); @@ -504,8 +493,12 @@ 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; } @@ -515,7 +508,7 @@ BOM_UTF8 \357\273\277 } } -{ +{ \\{ESCAPED} { char c = escaped_char (YYText ()[1]); yylval = scm_cons (scm_from_locale_stringn (&c, 1), @@ -527,15 +520,20 @@ BOM_UTF8 \357\273\277 } \" { - yy_pop_state (); - /* yylval is union. Must remember STRING before setting SCM*/ yylval = scm_string_concatenate_reverse (yylval, SCM_UNDEFINED, SCM_UNDEFINED); - return is_lyric_state () ? LYRICS_STRING : STRING; + if (get_state () == commandquote) { + yy_pop_state (); + return scan_escaped_word (ly_scm2string (yylval)); + } + + yy_pop_state (); + + return STRING; } \\ { yylval = scm_cons (scm_from_locale_string (YYText ()), @@ -545,22 +543,38 @@ BOM_UTF8 \357\273\277 { \" { - start_lyric_quote (); + start_quote (); } {FRACTION} { 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; } + \\\" { + start_command_quote (); + } {COMMAND}/[-_] | // backup rule {COMMAND} { return scan_escaped_word (YYText_utf8 () + 1); } - {LYRICS} { + \\.|\| { + // UTF-8 already covered by COMMAND + return scan_shorthand (YYText ()); + } + /* Characters needed to express durations, assignments */ + [*.=] { + yylval = SCM_UNSPECIFIED; + return YYText ()[0]; + } + [^|*.=$#{}\"\\ \t\n\r\f0-9][^$#{}\"\\ \t\n\r\f0-9]* { /* ugr. This sux. */ string s (YYText_utf8 ()); yylval = SCM_UNSPECIFIED; @@ -569,19 +583,14 @@ BOM_UTF8 \357\273\277 if (s == "--") return HYPHEN; s = lyric_fudge (s); - - char c = s[s.length () - 1]; - if (c == '{' || c == '}') // brace open is for not confusing dumb tools. - here_input ().warning ( - _ ("Brace found at end of lyric. Did you forget a space?")); yylval = ly_string2scm (s); - - return LYRICS_STRING; + return STRING; } - . { + /* This should really just cover {} */ + [{}] { yylval = SCM_UNSPECIFIED; - return YYText ()[0]; // LYRICS already catches all multibytes. + return YYText ()[0]; } } { @@ -589,6 +598,9 @@ BOM_UTF8 \357\273\277 {WORD} { return scan_bare_word (YYText_utf8 ()); } + \\\" { + start_command_quote (); + } {COMMAND}/[-_] | // backup rule {COMMAND} { return scan_escaped_word (YYText_utf8 () + 1); @@ -622,10 +634,6 @@ BOM_UTF8 \357\273\277 yylval = SCM_UNSPECIFIED; return CHORD_CARET; } - . { - yylval = SCM_UNSPECIFIED; - return YYText ()[0]; // WORD catches all multibyte. - } } @@ -634,6 +642,13 @@ BOM_UTF8 \357\273\277 yylval = SCM_UNSPECIFIED; return SCORE; } + \\score-lines { + yylval = SCM_UNSPECIFIED; + return SCORELINES; + } + \\\" { + start_command_quote (); + } {COMMAND}/[-_] | // backup rule {COMMAND} { string str (YYText_utf8 () + 1); @@ -671,60 +686,61 @@ BOM_UTF8 \357\273\277 // value (for token type MARKUP_FUNCTION or // MARKUP_LIST_FUNCTION). - push_extra_token(EXPECT_NO_MORE_ARGS); + push_extra_token (here_input (), EXPECT_NO_MORE_ARGS); s = scm_cdr(s); for (; scm_is_pair(s); s = scm_cdr(s)) { SCM predicate = scm_car(s); if (predicate == ly_lily_module_constant ("markup-list?")) - push_extra_token(EXPECT_MARKUP_LIST); + push_extra_token (here_input (), EXPECT_MARKUP_LIST); else if (predicate == ly_lily_module_constant ("markup?")) - push_extra_token(EXPECT_MARKUP); + push_extra_token (here_input (), EXPECT_MARKUP); else - push_extra_token(EXPECT_SCM, predicate); + push_extra_token (here_input (), EXPECT_SCM, predicate); } return token_type; } - [{}] { - yylval = SCM_UNSPECIFIED; - return YYText ()[0]; - } [^$#{}\"\\ \t\n\r\f]+ { string s (YYText_utf8 ()); - char c = s[s.length () - 1]; - /* brace open is for not confusing dumb tools. */ - if (c == '{' || c == '}') - here_input ().warning ( - _ ("Brace found at end of markup. Did you forget a space?")); 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_) { - /* 2 = init.ly + current file. - > because we're before closing, but is_main_input_ should - reflect after. - */ - is_main_input_ = include_stack_.size () > 2; + 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 { + yy_pop_state (); + } while (YYSTATE != maininput); + } + extra_tokens_ = SCM_EOL; + yy_pop_state (); + } if (!close_input () || !is_main_input_) /* Returns YY_NULL */ yyterminate (); @@ -734,11 +750,21 @@ BOM_UTF8 \357\273\277 yyterminate (); } +{ANY_CHAR} { + while (include_stack_.size () > main_input_level_ + && close_input ()) + ; + yyterminate (); +} + { {WORD}/[-_] | // backup rule {WORD} { return scan_bare_word (YYText_utf8 ()); } + \\\" { + start_command_quote (); + } {COMMAND}/[-_] | // backup rule {COMMAND} { return scan_escaped_word (YYText_utf8 () + 1); @@ -763,51 +789,18 @@ BOM_UTF8 \357\273\277 } -[{}] { +-/\. { // backup rule yylval = SCM_UNSPECIFIED; return YYText ()[0]; } --/\. | // backup rule -[*:=] { +{SPECIAL} { yylval = SCM_UNSPECIFIED; return YYText ()[0]; } -. { - 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]* { @@ -822,15 +815,25 @@ BOM_UTF8 \357\273\277 /* Make the lexer generate a token of the given type as the next token. TODO: make it possible to define a value for the token as well */ void -Lily_lexer::push_extra_token (int token_type, SCM scm) +Lily_lexer::push_extra_token (Input const &where, int token_type, SCM scm) +{ + extra_tokens_ = scm_cons (scm_cons2 (make_input (where), + scm_from_int (token_type), + scm), extra_tokens_); +} + +int +Lily_lexer::pop_extra_token () { if (scm_is_null (extra_tokens_)) - { - if (YY_START != extratoken) - hidden_state_ = YY_START; - yy_push_state (extratoken); - } - extra_tokens_ = scm_acons (scm_from_int (token_type), scm, extra_tokens_); + return -1; + + /* produce requested token */ + yylloc = *unsmob_input (scm_caar (extra_tokens_)); + int type = scm_to_int (scm_cadar (extra_tokens_)); + yylval = scm_cddar (extra_tokens_); + extra_tokens_ = scm_cdr (extra_tokens_); + return type; } void @@ -871,41 +874,24 @@ Lily_lexer::push_markup_state () void Lily_lexer::push_note_state (SCM alist) { - bool extra = (YYSTATE == extratoken); - SCM p = scm_assq (alist, pitchname_tab_stack_); - if (extra) - yy_pop_state (); - if (scm_is_false (p)) p = scm_cons (alist, alist_to_hashq (alist)); pitchname_tab_stack_ = scm_cons (p, pitchname_tab_stack_); yy_push_state (notes); - - if (extra) { - hidden_state_ = YYSTATE; - yy_push_state (extratoken); - } } void Lily_lexer::pop_state () { - bool extra = (YYSTATE == extratoken); - - if (extra) - yy_pop_state (); - if (YYSTATE == notes || YYSTATE == chords) pitchname_tab_stack_ = scm_cdr (pitchname_tab_stack_); - yy_pop_state (); + // don't cross the maininput threshold + if (YYSTATE != maininput) + yy_pop_state (); - if (extra) { - hidden_state_ = YYSTATE; - yy_push_state (extratoken); - } } int @@ -917,7 +903,7 @@ Lily_lexer::identifier_type (SCM sid) int -Lily_lexer::scan_escaped_word (string str) +Lily_lexer::scan_escaped_word (const string &str) { // use more SCM for this. @@ -925,12 +911,16 @@ 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 (here_input ())); + } + if (sid != SCM_UNDEFINED) return scan_scm_id (sid); @@ -942,6 +932,26 @@ Lily_lexer::scan_escaped_word (string str) return STRING; } +int +Lily_lexer::scan_shorthand (const string &str) +{ + SCM sid = lookup_identifier (str); + if (Music *m = unsmob_music (sid)) + { + m->set_spot (override_input (here_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) { @@ -967,7 +977,7 @@ Lily_lexer::scan_scm_id (SCM sid) funtype = SCM_FUNCTION; else programming_error ("Bad syntax function predicate"); - push_extra_token (EXPECT_NO_MORE_ARGS); + push_extra_token (here_input (), EXPECT_NO_MORE_ARGS); for (s = scm_cdr (s); scm_is_pair (s); s = scm_cdr (s)) { SCM optional = SCM_UNDEFINED; @@ -979,19 +989,15 @@ Lily_lexer::scan_scm_id (SCM sid) cs = SCM_CAR (cs); } - if (cs == Pitch_type_p_proc) - push_extra_token (EXPECT_PITCH); - else if (cs == Duration_type_p_proc) - push_extra_token (EXPECT_DURATION); - else if (ly_is_procedure (cs)) - push_extra_token (EXPECT_SCM, cs); + if (ly_is_procedure (cs)) + push_extra_token (here_input (), EXPECT_SCM, cs); else { programming_error ("Function parameter without type-checking predicate"); continue; } if (!scm_is_eq (optional, SCM_UNDEFINED)) - push_extra_token (EXPECT_OPTIONAL, optional); + push_extra_token (here_input (), EXPECT_OPTIONAL, optional); } return funtype; } @@ -1000,7 +1006,7 @@ Lily_lexer::scan_scm_id (SCM sid) } int -Lily_lexer::scan_bare_word (string str) +Lily_lexer::scan_bare_word (const string &str) { SCM sym = ly_symbol2scm (str.c_str ()); if ((YYSTATE == notes) || (YYSTATE == chords)) { @@ -1029,10 +1035,7 @@ Lily_lexer::scan_bare_word (string str) int Lily_lexer::get_state () const { - if (YY_START == extratoken) - return hidden_state_; - else - return YY_START; + return YY_START; } bool @@ -1090,19 +1093,28 @@ 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 (here_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); + push_extra_token (here_input (), + token, yylval); break; case '#': - push_extra_token (SCM_IDENTIFIER, scm_car (v)); + push_extra_token (here_input (), + SCM_IDENTIFIER, v); break; } } @@ -1111,6 +1123,12 @@ 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 (here_input ())); + } + return sval; }