X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=lily%2Flexer.ll;h=8dd76671456cebaf4341527c55254fe61844b494;hb=5b4b0d6e9a197e8f9eb085b7c2ad78b8be3e5cfc;hp=0a4db9346160b2648accf5d6f10b1026b9c304c3;hpb=b37e3f652677ae0298423db9fa0e552e5fce0c92;p=lilypond.git diff --git a/lily/lexer.ll b/lily/lexer.ll index 0a4db93461..8dd7667145 100644 --- a/lily/lexer.ll +++ b/lily/lexer.ll @@ -4,7 +4,7 @@ source file of the LilyPond music typesetter - (c) 1996--2006 Han-Wen Nienhuys + (c) 1996--2008 Han-Wen Nienhuys Jan Nieuwenhuizen */ @@ -41,8 +41,9 @@ using namespace std; #include "interval.hh" #include "lily-guile.hh" #include "lily-lexer.hh" -#include "lilypond-input-version.hh" +#include "lilypond-version.hh" #include "main.hh" +#include "music.hh" #include "music-function.hh" #include "parse-scm.hh" #include "parser.hh" @@ -57,21 +58,21 @@ RH 7 fix (?) */ #define isatty HORRIBLEKLUDGE -void strip_trailing_white (std::string&); -void strip_leading_white (std::string&); -std::string lyric_fudge (std::string s); -int music_function_type (SCM); -SCM lookup_markup_command (std::string s); -bool is_valid_version (std::string s); +void strip_trailing_white (string&); +void strip_leading_white (string&); +string lyric_fudge (string s); +SCM lookup_markup_command (string s); +SCM lookup_markup_list_command (string s); +bool is_valid_version (string s); #define start_quote() \ yy_push_state (quote);\ - yylval.string = new std::string + yylval.string = new string #define start_lyric_quote() \ yy_push_state (lyric_quote);\ - yylval.string = new std::string + yylval.string = new string #define yylval \ (*(YYSTYPE*)lexval) @@ -80,14 +81,9 @@ bool is_valid_version (std::string s); (*(YYLTYPE*)lexloc) #define YY_USER_ACTION add_lexed_char (YYLeng ()); -/* - -LYRICS ({AA}|{TEX})[^0-9 \t\n\f]* -*/ - -SCM scan_fraction (std::string); +SCM scan_fraction (string); SCM (* scm_parse_error_handler) (void *); @@ -103,6 +99,7 @@ SCM (* scm_parse_error_handler) (void *); %option never-interactive %option warn +%x extratoken %x chords %x figures %x incl @@ -112,18 +109,25 @@ SCM (* scm_parse_error_handler) (void *); %x markup %x notes %x quote +%x sourcefileline %x sourcefilename %x version -A [a-zA-Z] +A [a-zA-Z\200-\377] AA {A}|_ N [0-9] AN {AA}|{N} +ANY_CHAR (.|\n) PUNCT [?!:'`] ACCENT \\[`'"^] -NATIONAL [\001-\006\021-\027\031\036\200-\377] +NATIONAL [\001-\006\021-\027\031\036] TEX {AA}|-|{PUNCT}|{ACCENT}|{NATIONAL} WORD {A}{AN}* +DASHED_WORD {A}({AN}|-)* +DASHED_KEY_WORD \\{DASHED_WORD} + + + ALPHAWORD {A}+ DIGIT {N} UNSIGNED {N}+ @@ -143,6 +147,7 @@ ESCAPED [nt\\'"] EXTENDER __ HYPHEN -- BOM_UTF8 \357\273\277 + %% @@ -150,7 +155,25 @@ BOM_UTF8 \357\273\277 // windows-suck-suck-suck } -{BOM_UTF8} { +{ANY_CHAR} { + /* Generate a token without swallowing anything */ + + /* First unswallow the eaten character */ + add_lexed_char (-YYLeng ()); + yyless (0); + + /* produce requested token */ + int type = extra_token_types_.back (); + extra_token_types_.pop_back (); + if (extra_token_types_.empty ()) + 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}/.* { if (this->lexloc->line_number () != 1 || this->lexloc->column_number () != 0) { LexerError (_ ("stray UTF-8 BOM encountered").c_str ()); @@ -177,26 +200,37 @@ BOM_UTF8 \357\273\277 } } +{ + \" { + start_quote (); + } +} + \\version{WHITE}* { yy_push_state (version); } \\sourcefilename{WHITE}* { yy_push_state (sourcefilename); } +\\sourcefileline{WHITE}* { + yy_push_state (sourcefileline); +} \"[^"]*\" { /* got the version number */ - std::string s (YYText () + 1); + string s (YYText () + 1); s = s.substr (0, s.rfind ('\"')); yy_pop_state (); + + SCM top_scope = scm_car (scm_last_pair (scopes_)); + scm_module_define (top_scope, ly_symbol2scm ("version-seen"), SCM_BOOL_T); + if (!is_valid_version (s)) return INVALID; - SCM top_scope = scm_car (scm_last_pair (scopes_)); - scm_module_define (top_scope, ly_symbol2scm ("version-seen?"), SCM_BOOL_T); } \"[^"]*\" { - std::string s (YYText () + 1); + string s (YYText () + 1); s = s.substr (0, s.rfind ('\"')); yy_pop_state (); @@ -205,17 +239,30 @@ BOM_UTF8 \357\273\277 progress_indication ("\n"); scm_module_define (scm_car (scopes_), ly_symbol2scm ("input-file-name"), - scm_makfrom0str (s.c_str ())); + ly_string2scm (s)); } + +{INT} { + int i; + sscanf (YYText (), "%d", &i); + + yy_pop_state (); + this->here_input ().get_source_file ()->set_line (here_input ().start (), i); +} + . { LexerError (_ ("quoted string expected after \\version").c_str ()); yy_pop_state (); } ->. { +. { LexerError (_ ("quoted string expected after \\sourcefilename").c_str ()); yy_pop_state (); } +. { + LexerError (_ ("integer expected after \\sourcefileline").c_str ()); + yy_pop_state (); +} { [^\%]* { } @@ -227,7 +274,7 @@ BOM_UTF8 \357\273\277 } <> { LexerError (_ ("EOF found inside a comment").c_str ()); - is_main_input_ = false; + is_main_input_ = false; // should be safe , can't have \include in --safe. if (! close_input ()) yyterminate (); // can't move this, since it actually rets a YY_NULL } @@ -248,14 +295,14 @@ BOM_UTF8 \357\273\277 yy_push_state (incl); } \"[^"]*\" { /* got the include file name */ - std::string s (YYText ()+1); + string s (YYText ()+1); s = s.substr (0, s.rfind ('"')); new_input (s, sources_); yy_pop_state (); } \\{BLACK}*{WHITE} { /* got the include identifier */ - std::string s = YYText () + 1; + string s = YYText () + 1; strip_trailing_white (s); if (s.length () && (s[s.length () - 1] == ';')) s = s.substr (0, s.length () - 1); @@ -265,7 +312,7 @@ BOM_UTF8 \357\273\277 new_input (ly_scm2string (sid), sources_); yy_pop_state (); } else { - std::string msg (_f ("wrong or undefined identifier: `%s'", s )); + string msg (_f ("wrong or undefined identifier: `%s'", s )); LexerError (msg.c_str ()); SCM err = scm_current_error_port (); @@ -279,7 +326,7 @@ BOM_UTF8 \357\273\277 } {RESTNAME} { char const *s = YYText (); - yylval.scm = scm_makfrom0str (s); + yylval.scm = scm_from_locale_string (s); return RESTNAME; } R { @@ -290,7 +337,7 @@ BOM_UTF8 \357\273\277 Input hi = here_input(); hi.step_forward (); SCM sval = ly_parse_scm (hi.start (), &n, hi, - be_safe_global && is_main_input_); + be_safe_global && is_main_input_, parser_); if (sval == SCM_UNDEFINED) { @@ -302,7 +349,7 @@ BOM_UTF8 \357\273\277 { yyinput (); } - char_count_stack_.top () += n; + char_count_stack_.back () += n; if (unpack_identifier (sval) != SCM_UNDEFINED) { @@ -311,16 +358,26 @@ BOM_UTF8 \357\273\277 } yylval.scm = sval; - return SCM_T; + return SCM_TOKEN; } { - \<\< { + \<\< { return DOUBLE_ANGLE_OPEN; } - \>\> { + \>\> { return DOUBLE_ANGLE_CLOSE; } } + +{ + \< { + return ANGLE_OPEN; + } + \> { + return ANGLE_CLOSE; + } +} + { _ { return FIGURE_SPACE; @@ -347,26 +404,20 @@ BOM_UTF8 \357\273\277 } {DIGIT} { - yylval.i = String_convert::dec2int (std::string (YYText ())); + yylval.i = String_convert::dec2int (string (YYText ())); return DIGIT; } {UNSIGNED} { - yylval.i = String_convert::dec2int (std::string (YYText ())); + yylval.i = String_convert::dec2int (string (YYText ())); return UNSIGNED; } {E_UNSIGNED} { - yylval.i = String_convert::dec2int (std::string (YYText () +1)); + yylval.i = String_convert::dec2int (string (YYText () +1)); return E_UNSIGNED; } - \" { - start_quote (); - } } -\" { - start_quote (); -} -{ +{ \\{ESCAPED} { *yylval.string += to_string (escaped_char (YYText ()[1])); } @@ -378,31 +429,10 @@ BOM_UTF8 \357\273\277 yy_pop_state (); /* yylval is union. Must remember STRING before setting SCM*/ - std::string *sp = yylval.string; - yylval.scm = scm_makfrom0str (sp->c_str ()); + string *sp = yylval.string; + yylval.scm = ly_string2scm (*sp); delete sp; - return STRING; - } - . { - *yylval.string += YYText (); - } -} -{ - \\{ESCAPED} { - *yylval.string += to_string (escaped_char (YYText ()[1])); - } - [^\\"]+ { - *yylval.string += YYText (); - } - \" { - - yy_pop_state (); - - /* yylval is union. Must remember STRING before setting SCM*/ - std::string *sp = yylval.string; - yylval.scm = scm_makfrom0str (sp->c_str ()); - delete sp; - return LYRICS_STRING; + return is_lyric_state () ? LYRICS_STRING : STRING; } . { *yylval.string += YYText (); @@ -418,7 +448,7 @@ BOM_UTF8 \357\273\277 return FRACTION; } {UNSIGNED} { - yylval.i = String_convert::dec2int (std::string (YYText ())); + yylval.i = String_convert::dec2int (string (YYText ())); return UNSIGNED; } {NOTECOMMAND} { @@ -426,7 +456,7 @@ BOM_UTF8 \357\273\277 } {LYRICS} { /* ugr. This sux. */ - std::string s (YYText ()); + string s (YYText ()); if (s == "__") return yylval.i = EXTENDER; if (s == "--") @@ -437,7 +467,7 @@ BOM_UTF8 \357\273\277 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.scm = scm_makfrom0str (s.c_str ()); + yylval.scm = ly_string2scm (s); return LYRICS_STRING; @@ -458,12 +488,9 @@ BOM_UTF8 \357\273\277 return FRACTION; } {UNSIGNED} { - yylval.i = String_convert::dec2int (std::string (YYText ())); + yylval.i = String_convert::dec2int (string (YYText ())); return UNSIGNED; } - \" { - start_quote (); - } - { return CHORD_MINUS; } @@ -486,16 +513,13 @@ BOM_UTF8 \357\273\277 { - \" { - start_quote (); - } \\score { return SCORE; } {MARKUPCOMMAND} { - std::string str (YYText () + 1); + string str (YYText () + 1); SCM s = lookup_markup_command (str); - + SCM s2 = lookup_markup_list_command (str); if (scm_is_pair (s) && scm_is_symbol (scm_cdr (s)) ) { yylval.scm = scm_car(s); SCM tag = scm_cdr(s); @@ -515,6 +539,10 @@ BOM_UTF8 \357\273\277 return MARKUP_HEAD_SCM0_MARKUP1; else if (tag == ly_symbol2scm ("scheme0-scheme1-markup2")) return MARKUP_HEAD_SCM0_SCM1_MARKUP2; + else if (tag == ly_symbol2scm ("scheme0-scheme1-markup2-markup3")) + return MARKUP_HEAD_SCM0_SCM1_MARKUP2_MARKUP3; + else if (tag == ly_symbol2scm ("scheme0-markup1-markup2")) + return MARKUP_HEAD_SCM0_MARKUP1_MARKUP2; else if (tag == ly_symbol2scm ("scheme0-scheme1-scheme2")) return MARKUP_HEAD_SCM0_SCM1_SCM2; else { @@ -522,6 +550,24 @@ BOM_UTF8 \357\273\277 ly_display_scm (s); assert(false); } + } else if (scm_is_pair (s2) && scm_is_symbol (scm_cdr (s2))) { + yylval.scm = scm_car(s2); + SCM tag = scm_cdr(s2); + if (tag == ly_symbol2scm("empty")) + return MARKUP_LIST_HEAD_EMPTY; + else if (tag == ly_symbol2scm ("scheme0")) + return MARKUP_LIST_HEAD_SCM0; + else if (tag == ly_symbol2scm ("markup-list0")) + return MARKUP_LIST_HEAD_LIST0; + else if (tag == ly_symbol2scm ("scheme0-markup-list1")) + return MARKUP_LIST_HEAD_SCM0_LIST1; + else if (tag == ly_symbol2scm ("scheme0-scheme1-markup-list2")) + return MARKUP_LIST_HEAD_SCM0_SCM1_LIST2; + else { + programming_error ("no parser tag defined for this markup list signature"); + ly_display_scm (s); + assert(false); + } } else return scan_escaped_word (str); } @@ -529,14 +575,14 @@ BOM_UTF8 \357\273\277 return YYText ()[0]; } [^#{}"\\ \t\n\r\f]+ { - std::string s (YYText ()); + string s (YYText ()); 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.scm = scm_makfrom0str (s.c_str ()); + yylval.scm = ly_string2scm (s); return STRING; @@ -546,10 +592,14 @@ BOM_UTF8 \357\273\277 } } -<> { +<*><> { if (is_main_input_) { - is_main_input_ = false; + /* 2 = init.ly + current file. + > because we're before closing, but is_main_input_ should + reflect after. + */ + is_main_input_ = include_stack_.size () > 2; if (!close_input ()) /* Returns YY_NULL */ yyterminate (); @@ -559,6 +609,14 @@ BOM_UTF8 \357\273\277 yyterminate (); } +{ + {DASHED_WORD} { + return scan_bare_word (YYText ()); + } + {DASHED_KEY_WORD} { + return scan_escaped_word (YYText () + 1); + } +} {WORD} { return scan_bare_word (YYText ()); @@ -577,7 +635,7 @@ BOM_UTF8 \357\273\277 } {UNSIGNED} { - yylval.i = String_convert::dec2int (std::string (YYText ())); + yylval.i = String_convert::dec2int (string (YYText ())); return UNSIGNED; } @@ -627,13 +685,27 @@ BOM_UTF8 \357\273\277 } <*>. { - std::string msg = _f ("invalid character: `%c'", YYText ()[0]); + string msg = _f ("invalid character: `%c'", YYText ()[0]); LexerError (msg.c_str ()); return YYText ()[0]; } %% +/* 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) +{ + if (extra_token_types_.empty ()) + { + if (YY_START != extratoken) + hidden_state_ = YY_START; + yy_push_state (extratoken); + } + extra_token_types_.push_back (token_type); +} + void Lily_lexer::push_chord_state (SCM tab) { @@ -677,6 +749,7 @@ Lily_lexer::pop_state () { if (YYSTATE == notes || YYSTATE == chords) pitchname_tab_stack_ = scm_cdr (pitchname_tab_stack_); + yy_pop_state (); } @@ -689,7 +762,7 @@ Lily_lexer::identifier_type (SCM sid) int -Lily_lexer::scan_escaped_word (std::string str) +Lily_lexer::scan_escaped_word (string str) { // use more SCM for this. @@ -705,7 +778,20 @@ Lily_lexer::scan_escaped_word (std::string str) if (is_music_function (sid)) { yylval.scm = get_music_function_transform (sid); - return music_function_type (yylval.scm); + + SCM s = scm_object_property (yylval.scm, ly_symbol2scm ("music-function-signature")); + push_extra_token (EXPECT_NO_MORE_ARGS); + for (; scm_is_pair (s); s = scm_cdr (s)) + { + if (scm_car (s) == ly_music_p_proc) + push_extra_token (EXPECT_MUSIC); + else if (scm_car (s) == ly_lily_module_constant ("markup?")) + push_extra_token (EXPECT_MARKUP); + else if (ly_is_procedure (scm_car (s))) + push_extra_token (EXPECT_SCM); + else programming_error ("Function parameter without type-checking predicate"); + } + return MUSIC_FUNCTION; } if (sid != SCM_UNDEFINED) @@ -714,16 +800,16 @@ Lily_lexer::scan_escaped_word (std::string str) return identifier_type (sid); } - std::string msg (_f ("unknown escaped string: `\\%s'", str)); + string msg (_f ("unknown escaped string: `\\%s'", str)); LexerError (msg.c_str ()); - yylval.scm = scm_makfrom0str (str.c_str ()); + yylval.scm = ly_string2scm (str); return STRING; } int -Lily_lexer::scan_bare_word (std::string str) +Lily_lexer::scan_bare_word (string str) { SCM sym = ly_symbol2scm (str.c_str ()); if ((YYSTATE == notes) || (YYSTATE == chords)) { @@ -738,47 +824,57 @@ Lily_lexer::scan_bare_word (std::string str) else if (scm_is_symbol (yylval.scm)) return DRUM_PITCH; } - else if ((handle = scm_hashq_get_handle (chordmodifier_tab_, sym))!= SCM_BOOL_F) + else if ((YYSTATE == chords) + && (handle = scm_hashq_get_handle (chordmodifier_tab_, sym))!= SCM_BOOL_F) { yylval.scm = scm_cdr (handle); return CHORD_MODIFIER; } } - yylval.scm = scm_makfrom0str (str.c_str ()); + yylval.scm = ly_string2scm (str); return STRING; } +int +Lily_lexer::get_state () const +{ + if (YY_START == extratoken) + return hidden_state_; + else + return YY_START; +} + bool Lily_lexer::is_note_state () const { - return YY_START == notes; + return get_state () == notes; } bool Lily_lexer::is_chord_state () const { - return YY_START == chords; + return get_state () == chords; } bool Lily_lexer::is_lyric_state () const { - return YY_START == lyrics; + return get_state () == lyrics; } bool Lily_lexer::is_figure_state () const { - return YY_START == figures; + return get_state () == figures; } /* - urg, belong to std::string (_convert) + urg, belong to string (_convert) and should be generalised */ void -strip_leading_white (std::string&s) +strip_leading_white (string&s) { ssize i = 0; for (; i < s.length (); i++) @@ -789,7 +885,7 @@ strip_leading_white (std::string&s) } void -strip_trailing_white (std::string&s) +strip_trailing_white (string&s) { ssize i = s.length (); while (i--) @@ -801,21 +897,26 @@ strip_trailing_white (std::string&s) -/* 2.1.2x something -> \property -> \set. */ -Lilypond_version oldest_version ("2.3.22"); +Lilypond_version oldest_version ("2.7.38"); bool -is_valid_version (std::string s) +is_valid_version (string s) { Lilypond_version current ( MAJOR_VERSION "." MINOR_VERSION "." PATCH_LEVEL ); Lilypond_version ver (s); - if (! ((ver >= oldest_version) && (ver <= current))) + if (int (ver) < oldest_version) { - non_fatal_error (_f ("Incorrect lilypond version: %s (%s, %s)", ver.to_string (), oldest_version.to_string (), current.to_string ())); - non_fatal_error (_ ("Consider updating the input with the convert-ly script")); + 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; - } + } + + if (ver > current) + { + non_fatal_error (_f ("program too old: %s (file requires: %s)", current.to_string (), ver.to_string ())); + return false; + } return true; } @@ -823,8 +924,8 @@ is_valid_version (std::string s) /* substitute _ and \, */ -std::string -lyric_fudge (std::string s) +string +lyric_fudge (string s) { char *chars = string_copy (s); @@ -834,7 +935,7 @@ lyric_fudge (std::string s) *p = ' '; } - s = std::string (chars); + s = string (chars); delete[] chars; ssize i = 0; @@ -851,11 +952,11 @@ lyric_fudge (std::string s) Convert "NUM/DEN" into a '(NUM . DEN) cons. */ SCM -scan_fraction (std::string frac) +scan_fraction (string frac) { ssize i = frac.find ('/'); - std::string left = frac.substr (0, i); - std::string right = frac.substr (i - 1); + 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); @@ -863,80 +964,17 @@ scan_fraction (std::string frac) } SCM -lookup_markup_command (std::string s) +lookup_markup_command (string s) { SCM proc = ly_lily_module_constant ("lookup-markup-command"); - return scm_call_1 (proc, scm_makfrom0str (s.c_str ())); + return scm_call_1 (proc, ly_string2scm (s)); } - -int -music_function_type (SCM func) +SCM +lookup_markup_list_command (string s) { - SCM type = scm_object_property (func, ly_symbol2scm ("music-function-signature-keyword")); - if (type == ly_symbol2scm ("scm")) - { - return MUSIC_FUNCTION_SCM; - } - else if (type == ly_symbol2scm ("music")) - { - return MUSIC_FUNCTION_MUSIC; - } - else if (type == ly_symbol2scm ("scm-music")) - { - return MUSIC_FUNCTION_SCM_MUSIC; - } - else if (type == ly_symbol2scm ("scm-scm")) - { - return MUSIC_FUNCTION_SCM_SCM; - } - else if (type == ly_symbol2scm ("music-music")) - { - return MUSIC_FUNCTION_MUSIC_MUSIC; - } - else if (type == ly_symbol2scm ("scm-music-music")) - { - return MUSIC_FUNCTION_SCM_MUSIC_MUSIC; - } - else if (type == ly_symbol2scm ("scm-scm-music")) - { - return MUSIC_FUNCTION_SCM_SCM_MUSIC; - } - else if (type == ly_symbol2scm ("scm-scm-scm")) - { - return MUSIC_FUNCTION_SCM_SCM_SCM; - } - else if (type == ly_symbol2scm ("markup")) - { - return MUSIC_FUNCTION_MARKUP; - } - else if (type == ly_symbol2scm ("markup-music")) - { - return MUSIC_FUNCTION_MARKUP_MUSIC; - } - else if (type == ly_symbol2scm ("markup-markup")) - { - return MUSIC_FUNCTION_MARKUP_MARKUP; - } - else if (type == ly_symbol2scm ("markup-music-music")) - { - return MUSIC_FUNCTION_MARKUP_MUSIC_MUSIC; - } - else if (type == ly_symbol2scm ("markup-markup-music")) - { - return MUSIC_FUNCTION_MARKUP_MARKUP_MUSIC; - } - else if (type == ly_symbol2scm ("noarg")) - { - return MUSIC_FUNCTION; - } - else - { - /* TODO: print location */ - error (_ ("can't find signature for music function")); - } - - return MUSIC_FUNCTION_SCM; + SCM proc = ly_lily_module_constant ("lookup-markup-list-command"); + return scm_call_1 (proc, ly_string2scm (s)); } /* Shut up lexer warnings. */