X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=lily%2Flexer.ll;h=93184ebe085dc2d432274a06d4eeb887e565fe01;hb=8627d767e67258394cd4256740542a36dacd7868;hp=a360de48c137515ad5fd98251921e1e8caf983be;hpb=5bf2b8b2d50667582a61abea0e6257be0d033131;p=lilypond.git diff --git a/lily/lexer.ll b/lily/lexer.ll index a360de48c1..93184ebe08 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 @@ -30,9 +30,9 @@ /* backup rules - after making a change to the lexer rules, run + after making a change to the lexer rules, run flex -b - and make sure that + and make sure that lex.backup contains no backup states, but only the reminder Compressed tables always back up. @@ -69,9 +69,9 @@ using namespace std; #include "pitch.hh" #include "source-file.hh" #include "std-string.hh" -#include "string-convert.hh" #include "version.hh" #include "warn.hh" +#include "lily-imports.hh" /* RH 7 fix (?) @@ -120,7 +120,7 @@ SCM (* scm_parse_error_handler) (void *); %option debug %option yyclass="Lily_lexer" %option stack -%option never-interactive +%option never-interactive %option warn %x chords @@ -157,8 +157,8 @@ A [a-zA-Z\200-\377] AA {A}|_ N [0-9] ANY_CHAR (.|\n) -WORD {A}([-_]{A}|{A})* -COMMAND \\{WORD} +SYMBOL {A}([-_]{A}|{A})* +COMMAND \\{SYMBOL} /* SPECIAL category is for every letter that needs to get passed to * the parser rather than being redefinable by the user */ SPECIAL [-+*/=<>{}!?_^'',.:] @@ -188,7 +188,7 @@ BOM_UTF8 \357\273\277 /* Use the trailing context feature. Otherwise, the BOM will not be found if the file starts with an identifier definition. */ {BOM_UTF8}/.* { - if (this->lexloc_->line_number () != 1 || this->lexloc_->column_number () != 0) + if (lexloc_->line_number () != 1 || lexloc_->column_number () != 0) { LexerWarning (_ ("stray UTF-8 BOM encountered").c_str ()); // exit (1); @@ -244,7 +244,7 @@ BOM_UTF8 \357\273\277 s = s.substr (0, s.rfind ('\"')); yy_pop_state (); - this->here_input().get_source_file ()->name_ = s; + here_input().get_source_file ()->name_ = s; message (_f ("Renaming input to: `%s'", s.c_str ())); progress_indication ("\n"); scm_module_define (scm_car (scopes_), @@ -258,7 +258,7 @@ BOM_UTF8 \357\273\277 sscanf (YYText (), "%d", &i); yy_pop_state (); - this->here_input ().get_source_file ()->set_line (here_input ().start (), i); + here_input ().get_source_file ()->set_line (here_input ().start (), i); } {ANY_CHAR} { @@ -330,12 +330,11 @@ 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++) { @@ -348,7 +347,7 @@ BOM_UTF8 \357\273\277 yy_pop_state (); } else { LexerError (_ ("string expected after \\include").c_str ()); - if (sval != SCM_UNDEFINED) { + if (!SCM_UNBNDP (sval)) { SCM err = scm_current_error_port (); scm_puts ("This value was found instead: ", err); scm_display (sval, err); @@ -363,14 +362,14 @@ BOM_UTF8 \357\273\277 /* Flex picks the longest matching pattern including trailing * contexts. Without the backup pattern, r-. does not trigger the - * {RESTNAME} rule but rather the {WORD}/[-_] rule coming later, + * {RESTNAME} rule but rather the {SYMBOL}/[-_] rule coming later, * needed for avoiding backup states. */ {RESTNAME}/[-_] | // pseudo backup rule {RESTNAME} { char const *s = YYText (); - yylval = scm_from_locale_string (s); + yylval = scm_from_ascii_string (s); return RESTNAME; } q/[-_] | // pseudo backup rule @@ -385,15 +384,14 @@ 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) + if (SCM_UNBNDP (sval)) error_level_ = 1; + int n = hi.end () - hi.start (); for (int i = 0; i < n; i++) { yyinput (); @@ -405,11 +403,11 @@ 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++) { @@ -417,14 +415,14 @@ 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; @@ -435,7 +433,7 @@ BOM_UTF8 \357\273\277 } } -{ +{ \< { yylval = SCM_UNSPECIFIED; return ANGLE_OPEN; @@ -478,8 +476,8 @@ BOM_UTF8 \357\273\277 } { - {WORD}/[-_] | // backup rule - {WORD} { + {SYMBOL}/[-_] | // backup rule + {SYMBOL} { return scan_bare_word (YYText_utf8 ()); } \\\" { @@ -511,11 +509,11 @@ BOM_UTF8 \357\273\277 { \\{ESCAPED} { char c = escaped_char (YYText ()[1]); - yylval = scm_cons (scm_from_locale_stringn (&c, 1), + yylval = scm_cons (scm_from_ascii_stringn (&c, 1), yylval); } [^\\""]+ { - yylval = scm_cons (scm_from_locale_string (YYText_utf8 ()), + yylval = scm_cons (scm_from_utf8_string (YYText_utf8 ()), yylval); } \" { @@ -536,7 +534,7 @@ BOM_UTF8 \357\273\277 return STRING; } \\ { - yylval = scm_cons (scm_from_locale_string (YYText ()), + yylval = scm_cons (scm_from_ascii_string (YYText ()), yylval); } } @@ -585,7 +583,7 @@ BOM_UTF8 \357\273\277 s = lyric_fudge (s); yylval = ly_string2scm (s); - return STRING; + return SYMBOL; } /* This should really just cover {} */ [{}] { @@ -594,8 +592,8 @@ BOM_UTF8 \357\273\277 } } { - {WORD}/[-_] | // backup rule - {WORD} { + {SYMBOL}/[-_] | // backup rule + {SYMBOL} { return scan_bare_word (YYText_utf8 ()); } \\\" { @@ -686,17 +684,17 @@ 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); - else if (predicate == ly_lily_module_constant ("markup?")) - push_extra_token(EXPECT_MARKUP); + if (scm_is_eq (predicate, SCM (Lily::markup_list_p))) + push_extra_token (here_input (), EXPECT_MARKUP_LIST); + else if (scm_is_eq (predicate, SCM (Lily::markup_p))) + 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; } @@ -704,7 +702,7 @@ BOM_UTF8 \357\273\277 string s (YYText_utf8 ()); yylval = ly_string2scm (s); - return STRING; + return SYMBOL; } [{}] { yylval = SCM_UNSPECIFIED; @@ -758,8 +756,8 @@ BOM_UTF8 \357\273\277 } { - {WORD}/[-_] | // backup rule - {WORD} { + {SYMBOL}/[-_] | // backup rule + {SYMBOL} { return scan_bare_word (YYText_utf8 ()); } \\\" { @@ -812,12 +810,14 @@ BOM_UTF8 \357\273\277 %% -/* Make the lexer generate a token of the given type as the next token. +/* 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_acons (scm_from_int (token_type), scm, extra_tokens_); + extra_tokens_ = scm_cons (scm_cons2 (where.smobbed_copy (), + scm_from_int (token_type), + scm), extra_tokens_); } int @@ -827,8 +827,9 @@ Lily_lexer::pop_extra_token () return -1; /* produce requested token */ - int type = scm_to_int (scm_caar (extra_tokens_)); - yylval = scm_cdar (extra_tokens_); + yylloc = *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; } @@ -913,35 +914,35 @@ Lily_lexer::scan_escaped_word (const string &str) return i; SCM sid = lookup_identifier (str); - if (Music *m = unsmob_music (sid)) + if (Music *m = unsmob (sid)) { m->set_spot (override_input (here_input ())); } - if (sid != SCM_UNDEFINED) + if (!SCM_UNBNDP (sid)) return scan_scm_id (sid); - string msg (_f ("unknown escaped string: `\\%s'", str)); + string msg (_f ("unknown escaped string: `\\%s'", str)); LexerError (msg.c_str ()); yylval = ly_string2scm (str); - return STRING; + return STRING; // SYMBOL would cause additional processing } int Lily_lexer::scan_shorthand (const string &str) { SCM sid = lookup_identifier (str); - if (Music *m = unsmob_music (sid)) + if (Music *m = unsmob (sid)) { m->set_spot (override_input (here_input ())); } - if (sid != SCM_UNDEFINED) + if (!SCM_UNBNDP (sid)) return scan_scm_id (sid); - string msg (_f ("undefined character or shorthand: %s", str)); + string msg (_f ("undefined character or shorthand: %s", str)); LexerError (msg.c_str ()); yylval = ly_string2scm (str); @@ -952,13 +953,13 @@ Lily_lexer::scan_shorthand (const string &str) int Lily_lexer::scan_scm_id (SCM sid) { - if (is_music_function (sid)) + if (Music_function *fun = 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)) @@ -966,15 +967,15 @@ Lily_lexer::scan_scm_id (SCM sid) cs = SCM_CAR (cs); } - if (scm_is_eq (cs, ly_lily_module_constant ("ly:music?"))) + if (scm_is_eq (cs, SCM (Lily::ly_music_p))) funtype = MUSIC_FUNCTION; - else if (scm_is_eq (cs, ly_lily_module_constant ("ly:event?"))) + else if (scm_is_eq (cs, SCM (Lily::ly_event_p))) funtype = EVENT_FUNCTION; else if (ly_is_procedure (cs)) 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; @@ -985,16 +986,16 @@ Lily_lexer::scan_scm_id (SCM sid) optional = SCM_CDR (cs); cs = SCM_CAR (cs); } - + if (ly_is_procedure (cs)) - push_extra_token (EXPECT_SCM, 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; } @@ -1003,30 +1004,41 @@ Lily_lexer::scan_scm_id (SCM sid) } int -Lily_lexer::scan_bare_word (const string &str) +Lily_lexer::scan_word (SCM & output, SCM sym) { - SCM sym = ly_symbol2scm (str.c_str ()); if ((YYSTATE == notes) || (YYSTATE == chords)) { SCM handle = SCM_BOOL_F; if (scm_is_pair (pitchname_tab_stack_)) handle = scm_hashq_get_handle (scm_cdar (pitchname_tab_stack_), sym); - + if (scm_is_pair (handle)) { - yylval = scm_cdr (handle); - if (unsmob_pitch (yylval)) + output = scm_cdr (handle); + if (unsmob (yylval)) return (YYSTATE == notes) ? NOTENAME_PITCH : TONICNAME_PITCH; else if (scm_is_symbol (yylval)) return DRUM_PITCH; } else if ((YYSTATE == chords) - && (handle = scm_hashq_get_handle (chordmodifier_tab_, sym))!= SCM_BOOL_F) + && scm_is_true (handle = scm_hashq_get_handle (chordmodifier_tab_, sym))) { - yylval = scm_cdr (handle); + output = scm_cdr (handle); return CHORD_MODIFIER; } } + output = SCM_UNDEFINED; + return -1; +} + +int +Lily_lexer::scan_bare_word (const string &str) +{ + int state = scan_word (yylval, ly_symbol2scm (str.c_str ())); + if (state >= 0) + { + return state; + } yylval = ly_string2scm (str); - return STRING; + return SYMBOL; } int @@ -1067,14 +1079,14 @@ 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_); } @@ -1095,21 +1107,23 @@ 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 = unsmob (v)) { - if (!unsmob_input (m->get_property ("origin"))) + if (!unsmob (m->get_property ("origin"))) m->set_spot (override_input (here_input ())); } - + int token; switch (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; } } @@ -1118,9 +1132,9 @@ Lily_lexer::eval_scm (SCM readerdata, char extra_token) sval = SCM_UNSPECIFIED; } - if (Music *m = unsmob_music (sval)) + if (Music *m = unsmob (sval)) { - if (!unsmob_input (m->get_property ("origin"))) + if (!unsmob (m->get_property ("origin"))) m->set_spot (override_input (here_input ())); } @@ -1238,7 +1252,7 @@ Lily_lexer::YYText_utf8 () /* urg, belong to string (_convert) - and should be generalised + and should be generalised */ void strip_leading_white (string&s) @@ -1254,8 +1268,8 @@ strip_leading_white (string&s) void strip_trailing_white (string&s) { - ssize i = s.length (); - while (i--) + ssize i = s.length (); + while (i--) if (!isspace (s[i])) break; @@ -1272,8 +1286,13 @@ 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")); return false; @@ -1286,7 +1305,7 @@ is_valid_version (string s) } return true; } - + /* substitute _ @@ -1313,23 +1332,20 @@ 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 lookup_markup_command (string s) { - SCM proc = ly_lily_module_constant ("lookup-markup-command"); - return scm_call_1 (proc, ly_string2scm (s)); + return Lily::lookup_markup_command (ly_string2scm (s)); } SCM lookup_markup_list_command (string s) { - SCM proc = ly_lily_module_constant ("lookup-markup-list-command"); - return scm_call_1 (proc, ly_string2scm (s)); + return Lily::lookup_markup_list_command (ly_string2scm (s)); } /* Shut up lexer warnings. */