X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=lily%2Flexer.ll;h=89c8695987af4ef8b71e2a18d74dc964e70738b0;hb=47db9a3883d726ca53e2133a3b2298f78dd6a32e;hp=9e25ed788735e0af6191c715188ce597fb966a66;hpb=32689b75bec62313992c1b452f624d65f45e8789;p=lilypond.git diff --git a/lily/lexer.ll b/lily/lexer.ll index 9e25ed7887..89c8695987 100644 --- a/lily/lexer.ll +++ b/lily/lexer.ll @@ -2,7 +2,7 @@ /* This file is part of LilyPond, the GNU music typesetter. - Copyright (C) 1996--2012 Han-Wen Nienhuys + Copyright (C) 1996--2015 Han-Wen Nienhuys Jan Nieuwenhuizen LilyPond is free software: you can redistribute it and/or modify @@ -69,7 +69,6 @@ using namespace std; #include "pitch.hh" #include "source-file.hh" #include "std-string.hh" -#include "string-convert.hh" #include "version.hh" #include "warn.hh" @@ -91,6 +90,15 @@ bool is_valid_version (string s); yylval = SCM_EOL; \ } while (0) +/* + The inside of \"violin1" is marked by commandquote mode +*/ + +#define start_command_quote() do { \ + yy_push_state (commandquote); \ + yylval = SCM_EOL; \ + } while (0) + #define yylval (*lexval_) #define yylloc (*lexloc_) @@ -114,7 +122,6 @@ SCM (* scm_parse_error_handler) (void *); %option never-interactive %option warn -%x extratoken %x chords %x figures %x incl @@ -124,6 +131,7 @@ SCM (* scm_parse_error_handler) (void *); %x markup %x notes %x quote +%x commandquote %x sourcefileline %x sourcefilename %x version @@ -150,7 +158,10 @@ 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}+ @@ -173,36 +184,6 @@ 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}/.* { @@ -348,12 +329,11 @@ BOM_UTF8 \357\273\277 } } (\$|#) { // scm for the filename - int n = 0; Input hi = here_input(); hi.step_forward (); - SCM sval = ly_parse_scm (hi.start (), &n, hi, - be_safe_global && is_main_input_, parser_); - sval = eval_scm (sval); + SCM sval = ly_parse_scm (hi, be_safe_global && is_main_input_, parser_); + sval = eval_scm (sval, hi); + int n = hi.end () - hi.start (); for (int i = 0; i < n; i++) { @@ -403,15 +383,14 @@ BOM_UTF8 \357\273\277 return MULTI_MEASURE_REST; } # { //embedded scm - int n = 0; Input hi = here_input(); hi.step_forward (); - SCM sval = ly_parse_scm (hi.start (), &n, hi, - be_safe_global && is_main_input_, parser_); + SCM sval = ly_parse_scm (hi, be_safe_global && is_main_input_, parser_); if (sval == SCM_UNDEFINED) error_level_ = 1; + int n = hi.end () - hi.start (); for (int i = 0; i < n; i++) { yyinput (); @@ -423,11 +402,11 @@ BOM_UTF8 \357\273\277 } \$ { //immediate scm - int n = 0; Input hi = here_input(); hi.step_forward (); - SCM sval = ly_parse_scm (hi.start (), &n, hi, - be_safe_global && is_main_input_, parser_); + SCM sval = ly_parse_scm (hi, be_safe_global && is_main_input_, parser_); + + int n = hi.end () - hi.start (); for (int i = 0; i < n; i++) { @@ -435,14 +414,14 @@ BOM_UTF8 \357\273\277 } char_count_stack_.back () += n; - sval = eval_scm (sval, '$'); + sval = eval_scm (sval, hi, '$'); int token = scan_scm_id (sval); if (!scm_is_eq (yylval, SCM_UNSPECIFIED)) return token; } -{ +{ \<\< { yylval = SCM_UNSPECIFIED; return DOUBLE_ANGLE_OPEN; @@ -453,7 +432,7 @@ BOM_UTF8 \357\273\277 } } -{ +{ \< { yylval = SCM_UNSPECIFIED; return ANGLE_OPEN; @@ -477,6 +456,22 @@ 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]; + } } { @@ -484,7 +479,9 @@ 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); @@ -508,7 +505,7 @@ BOM_UTF8 \357\273\277 } } -{ +{ \\{ESCAPED} { char c = escaped_char (YYText ()[1]); yylval = scm_cons (scm_from_locale_stringn (&c, 1), @@ -520,14 +517,19 @@ 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); + if (get_state () == commandquote) { + yy_pop_state (); + return scan_escaped_word (ly_scm2string (yylval)); + } + + yy_pop_state (); + return STRING; } \\ { @@ -553,16 +555,23 @@ BOM_UTF8 \357\273\277 yylval = scm_c_read_string (YYText ()); return UNSIGNED; } + \\\" { + start_command_quote (); + } {COMMAND}/[-_] | // backup rule {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]; } - [^$#{}\"\\ \t\n\r\f0-9]+ { + [^|*.=$#{}\"\\ \t\n\r\f0-9][^$#{}\"\\ \t\n\r\f0-9]* { /* ugr. This sux. */ string s (YYText_utf8 ()); yylval = SCM_UNSPECIFIED; @@ -576,9 +585,9 @@ BOM_UTF8 \357\273\277 return STRING; } /* This should really just cover {} */ - . { + [{}] { yylval = SCM_UNSPECIFIED; - return YYText ()[0]; // above catches all multibytes. + return YYText ()[0]; } } { @@ -586,6 +595,9 @@ 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); @@ -619,10 +631,6 @@ BOM_UTF8 \357\273\277 yylval = SCM_UNSPECIFIED; return CHORD_CARET; } - . { - yylval = SCM_UNSPECIFIED; - return YYText ()[0]; // WORD catches all multibyte. - } } @@ -631,6 +639,13 @@ 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); @@ -668,17 +683,17 @@ 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; } @@ -688,9 +703,9 @@ 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]; } } @@ -699,7 +714,7 @@ BOM_UTF8 \357\273\277 yy_pop_state (); } -<> { +<> { LexerError (_ ("EOF found inside string").c_str ()); yy_pop_state (); } @@ -717,10 +732,11 @@ BOM_UTF8 \357\273\277 { LexerError (_ ("Unfinished main input").c_str ()); do { - pop_state (); + yy_pop_state (); } while (YYSTATE != maininput); } - pop_state (); + extra_tokens_ = SCM_EOL; + yy_pop_state (); } if (!close_input () || !is_main_input_) /* Returns YY_NULL */ @@ -743,6 +759,9 @@ 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); @@ -767,51 +786,18 @@ 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]* { @@ -826,15 +812,25 @@ 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 (where.smobbed_copy (), + 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 = *Input::unsmob (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 @@ -875,41 +871,24 @@ 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 @@ -921,7 +900,7 @@ 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. @@ -934,9 +913,9 @@ Lily_lexer::scan_escaped_word (string str) return i; SCM sid = lookup_identifier (str); - if (Music *m = unsmob_music (sid)) + if (Music *m = Music::unsmob (sid)) { - m->set_spot (override_input (last_input_)); + m->set_spot (override_input (here_input ())); } if (sid != SCM_UNDEFINED) @@ -950,16 +929,36 @@ 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 = Music::unsmob (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) { - if (is_music_function (sid)) + if (Music_function *fun = Music_function::unsmob (sid)) { int funtype = SCM_FUNCTION; yylval = sid; - SCM s = get_music_function_signature (sid); + SCM s = fun->get_signature (); SCM cs = scm_car (s); if (scm_is_pair (cs)) @@ -975,7 +974,7 @@ 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; @@ -987,19 +986,15 @@ 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; } @@ -1008,7 +1003,7 @@ 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)) { @@ -1018,7 +1013,7 @@ Lily_lexer::scan_bare_word (string str) if (scm_is_pair (handle)) { yylval = scm_cdr (handle); - if (unsmob_pitch (yylval)) + if (Pitch::is_smob (yylval)) return (YYSTATE == notes) ? NOTENAME_PITCH : TONICNAME_PITCH; else if (scm_is_symbol (yylval)) return DRUM_PITCH; @@ -1037,10 +1032,7 @@ 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 @@ -1075,14 +1067,14 @@ Lily_lexer::is_figure_state () const // this function is private. SCM -Lily_lexer::eval_scm (SCM readerdata, char extra_token) +Lily_lexer::eval_scm (SCM readerdata, Input hi, char extra_token) { SCM sval = SCM_UNDEFINED; if (!SCM_UNBNDP (readerdata)) { - sval = ly_eval_scm (scm_car (readerdata), - *unsmob_input (scm_cdr (readerdata)), + sval = ly_eval_scm (readerdata, + hi, be_safe_global && is_main_input_, parser_); } @@ -1103,10 +1095,10 @@ Lily_lexer::eval_scm (SCM readerdata, char extra_token) p = scm_cdr (p)) { SCM v = scm_car (p); - if (Music *m = unsmob_music (v)) + if (Music *m = Music::unsmob (v)) { - if (!unsmob_input (m->get_property ("origin"))) - m->set_spot (override_input (last_input_)); + if (!Input::is_smob (m->get_property ("origin"))) + m->set_spot (override_input (here_input ())); } int token; @@ -1114,10 +1106,12 @@ Lily_lexer::eval_scm (SCM readerdata, char extra_token) case '$': 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, v); + push_extra_token (here_input (), + SCM_IDENTIFIER, v); break; } } @@ -1126,10 +1120,10 @@ Lily_lexer::eval_scm (SCM readerdata, char extra_token) sval = SCM_UNSPECIFIED; } - if (Music *m = unsmob_music (sval)) + if (Music *m = Music::unsmob (sval)) { - if (!unsmob_input (m->get_property ("origin"))) - m->set_spot (override_input (last_input_)); + if (!Input::is_smob (m->get_property ("origin"))) + m->set_spot (override_input (here_input ())); } return sval; @@ -1280,7 +1274,12 @@ is_valid_version (string s) { Lilypond_version current ( MAJOR_VERSION "." MINOR_VERSION "." PATCH_LEVEL ); Lilypond_version ver (s); - if (int (ver) < oldest_version) + if (!ver) + { + non_fatal_error (_f ("Invalid version string \"%s\"", s)); + return false; + } + if (ver < oldest_version) { non_fatal_error (_f ("file too old: %s (oldest supported: %s)", ver.to_string (), oldest_version.to_string ())); non_fatal_error (_ ("consider updating the input with the convert-ly script")); @@ -1321,9 +1320,8 @@ scan_fraction (string frac) string left = frac.substr (0, i); string right = frac.substr (i + 1, (frac.length () - i + 1)); - int n = String_convert::dec2int (left); - int d = String_convert::dec2int (right); - return scm_cons (scm_from_int (n), scm_from_int (d)); + return scm_cons (scm_c_read_string (left.c_str ()), + scm_c_read_string (right.c_str ())); } SCM