RE2PY(1) RE2PY(1) (NAME) re2py - Python (SYNOPSIS) re2py [ OPTIONS ] [ WARNINGS ] INPUT - stdin . (INTRODUCTION) re2py . ( Python ) / . . re2py . Python . : # re2py $INPUT -o $OUTPUT def lex(yyinput): yycursor = 0 %{ re2c:yyfill:enable = 0; re2c:indent:top = 1; [1-9][0-9]* { return True } * { return False } %} assert lex(b"1234\0") re2py : # Generated by re2py # re2py $INPUT -o $OUTPUT def lex(yyinput): yycursor = 0 yystate = 0 while True: match yystate: case 0: yych = yyinput[yycursor] yycursor += 1 if yych <= 0x30: yystate = 1 continue if yych <= 0x39: yystate = 2 continue yystate = 1 continue case 1: return False case 2: yych = yyinput[yycursor] if yych <= 0x2F: yystate = 3 continue if yych <= 0x39: yycursor += 1 yystate = 2 continue yystate = 3 continue case 3: return True case _: raise "internal lexer error" assert lex(b"1234\0") re2py . : name = regular-expression ; name regular-expression . - . re2py . name . . . . configuration = value ; re2py . re2py . value - . . . regular-expression code regular-expression ( := ). regular-expression code . . . : * $. ( ). . . !action code code ( ). . !directive ; . . . (Blocks) /*!re2c */ %{ %} ( ). 2.2 . : /*!re2c[:] ... */ %{[:] ... %} . re2py . . . /*!local:re2c[:] ... */ %{local[:] ... %} ( ). /*!rules:re2c[:] ... */ %{rules[:] ... %} -- . ( ). 2.2 re2py -r --reusable . /*!use:re2c[:] ... */ %{use[:] ... %} . re2py . ( ). use . 2.2 re2py use -r --reusable . /*!max:re2c[:[:...]] ... */ %{max[:[:...]] ... %} YYMAXFILL . YYMAXFILL ( ). C (#define YYMAXFILL ) Go (var YYMAXFILL int = ) . format @@{max} ( @@) YYMAXFILL . /*!maxnmatch:re2c[:[:...]] ... */ %{maxnmatch[:[:...]] ... %} YYMAXNMATCH ( -P --posix-captures ). YYMAXNMATCH ( ). C (#define YYMAXNMATCH ) Go (var YYMAXNMATCH int = ) . format @@{max} ( @@) YYMAXNMATCH . /*!stags:re2c[:[:...]] ... */ /*!mtags:re2c[:[:...]] ... */ %{stags[:[:...]] ... %} %{mtags[:[:...]] ... %{ s-tag/m-tag re2py . ( ). : format separator. format @@{tag} ( @@) . separator format . /*!svars:re2c[:[:...]] ... */ /*!mvars:re2c[:[:...]] ... */ %{svars[:[:...]] ... %} %{mvars[:[:...]] ... %{ s-tag/m-tag ( --tags) re2py ( --captvars --posix-captvars). ( ). : format separator. format @@{tag} ( @@) . separator format . /*!getstate:re2c[:[:...]] ... */ %{getstate[:[:...]] ... %} ( --storable-state ). . . . --loop-switch goto . /*!conditions:re2c[:[:...]] ... */ /*!types:re2c... */ %{conditions[:[:...]] ... %} %{types... %} ( --conditions ). ( ). YYCONDTYPE . format separator . format @@{cond} ( @@) @@{num} . separator format . /*!include:re2c */ %{include %} . #include C/C++ . --depfile . /*!header:re2c:on*/ %{header:on %} . header:off re2py -t --type-header . /*!header:re2c:off*/ %{header:off %} header:on*/ . /*!ignore:re2c ... */ %{ignore ... %} . re2py : re2c:api, re2c:input --api . re2c:api:sigil ("sigil") API . @@ . sigil . sigil $ ${name} . API . API re2c:YYFILL@len YYFILL. re2c:api:style API . functions ( C) free-form ( Go Rust). functions API . ( YYFILL) (lexer) C/C++ (closures) Go Rust . (free-form) API : @@{var} @@ ( ). API re2c:YYFILL:naked YYFILL . re2c:bit-vectors, re2c:flags:bit-vectors, re2c:flags:b --bit-vectors . re2c:captures, re2c:leftmost-captures --leftmost-captures . re2c:captvars, re2c:leftmost-captvars --leftmost-captvars . re2c:case-insensitive, re2c:flags:case-insensitive --case-insensitive . re2c:case-inverted, re2c:flags:case-inverted --case-inverted . re2c:case-ranges, re2c:flags:case-ranges --case-ranges . re2c:computed-gotos, re2c:flags:computed-gotos, re2c:flags:g --computed-gotos . re2c:computed-gotos:relative, re2c:cgoto:relative --computed-gotos-relative . re2c:computed-gotos:threshold, re2c:cgoto:threshold goto if (bitmaps) . 9 . re2c:cond:abort (default case) (abort) . re2c:cond:goto :=> . goto @@; . @@ ( re2c:api:sigil re2c:cond:goto@cond ). re2c:cond:goto@cond re2c:cond:goto . @@ . re2c:api:sigil . re2c:cond:divider . /* *********************************** */ . ( re2c:api;sigil re2c:cond:divider@cond ). re2c:cond:divider@cond re2c:cond:divider . @@ . re2c:api:sigil . re2c:cond:prefix, re2c:condprefix . yyc_ . re2c:cond:enumprefix, re2c:condenumprefix . yyc . re2c:debug-output, re2c:flags:debug-output, re2c:flags:d --debug-output . re2c:empty-class, re2c:flags:empty-class --empty-class . re2c:encoding:ebcdic, re2c:flags:ecb, re2c:flags:e --ebcdic . re2c:encoding:ucs2, re2c:flags:wide-chars, re2c:flags:w --ucs2 . re2c:encoding:utf8, re2c:flags:utf-8, re2c:flags:8 --utf8 . re2c:encoding:utf16, re2c:flags:utf-16, re2c:flags:x --utf16 . re2c:encoding:utf32, re2c:flags:unicode, re2c:flags:u --utf32 . re2c:encoding-policy, re2c:flags:encoding-policy --encoding-policy . re2c:eof $ . -1 ( $ ). . . re2c:header, re2c:flags:type-header, re2c:flags:t . --header . re2c:indent:string . "\t" . . . re2c:indent:top . . . re2c:invert-captures --invert-captures . re2c:label:prefix, re2c:labelprefix DFA . yy . re2c:label:start, re2c:startlabel . . . . ( ). re2c:label:yyFillLabel YYFILL re2c:eof . re2c:label:yyloop --loop-switch . yyloop . re2c:label:yyNext YYGETSTATE ( re2c:state:nextlabel) . yyNext . re2c:lookahead, re2c:flags:lookahead ( --no-lookahead ). re2c:monadic ( Haskell )..TP re2c:nested-ifs, re2c:flags:nested-ifs, re2c:flags:s --nested-ifs . re2c:posix-captures, re2c:flags:posix-captures, re2c:flags:P --posix-captures . re2c:posix-captvars --posix-captvars . re2c:tags, re2c:flags:tags, re2c:flags:T --tags . re2c:tags:expression . re2py yyt . struct . @@{tag} @@ . re2c:tags:expression = "s.@@"; s.yyt . re2c:api:sigil . re2c:tags:negative ( -1 (null pointer) ). re2c:tags:prefix . yyt . re2c:sentinel (sentinel) ( re2c:yyfill:enable = 0; re2c:eof ). : . -1` ( re2py ). . re2c:state:abort (default case) (state dispatch) -1 . re2c:state:nextlabel YYGETSTATE yyNext ( ). re2c:label:start getstate YYGETSTATE . re2c:unsafe, re2c:flags:unsafe --no-unsafe . unsafe YYPEEK . ( ). Rust . re2c:YYBACKUP, re2c:define:YYBACKUP API YYBACKUP . re2c:YYBACKUPCTX, re2c:define:YYBACKUPCTX API YYBACKUPCTX . re2c:YYCONDTYPE, re2c:define:YYCONDTYPE API YYCONDTYPE . re2c:YYCTYPE, re2c:define:YYCTYPE API YYCTYPE . re2c:YYCTXMARKER, re2c:define:YYCTXMARKER API YYCTXMARKER . re2c:YYCURSOR, re2c:define:YYCURSOR API YYCURSOR . re2c:YYDEBUG, re2c:define:YYDEBUG API YYDEBUG . re2c:YYFILL, re2c:define:YYFILL API YYFILL . re2c:YYFILL@len, re2c:define:YYFILL@len (sigil) YYFILL . @@ . re2c:api:sigil . re2c:YYFILL:naked, re2c:define:YYFILL:naked re2c:api:style YYFILL . API (free-form API style) . re2c:YYFN API YYFN . re2c:YYINPUT API YYINPUT . re2c:YYGETCOND, re2c:define:YYGETCONDITION API YYGETCOND . re2c:YYGETCOND:naked, re2c:define:YYGETCONDITION:naked re2c:api:style YYGETCOND . API (free-form API style) . re2c:YYGETSTATE, re2c:define:YYGETSTATE API YYGETSTATE . re2c:YYGETSTATE:naked, re2c:define:YYGETSTATE:naked re2c:api:style YYGETSTATE . API (free-form API style) . re2c:YYGETACCEPT, re2c:define:YYGETACCEPT API YYGETACCEPT . re2c:YYLESSTHAN, re2c:define:YYLESSTHAN API YYLESSTHAN . re2c:YYLIMIT, re2c:define:YYLIMIT API YYLIMIT . re2c:YYMARKER, re2c:define:YYMARKER API YYMARKER . re2c:YYMTAGN, re2c:define:YYMTAGN API YYMTAGN . re2c:YYMTAGP, re2c:define:YYMTAGP API YYMTAGP . re2c:YYPEEK, re2c:define:YYPEEK API YYPEEK . re2c:YYRESTORE, re2c:define:YYRESTORE API YYRESTORE . re2c:YYRESTORECTX, re2c:define:YYRESTORECTX API YYRESTORECTX . re2c:YYRESTORETAG, re2c:define:YYRESTORETAG API YYRESTORETAG . re2c:YYSETCOND, re2c:define:YYSETCONDITION API YYSETCOND . re2c:YYSETCOND@cond, re2c:define:YYSETCONDITION@cond (sigil) YYSETCOND . @@ . re2c:api:sigil . re2c:YYSETCOND:naked, re2c:define:YYSETCONDITION:naked re2c:api:style YYSETCOND . API (free-form API style) . re2c:YYSETSTATE, re2c:define:YYSETSTATE API YYSETSTATE . re2c:YYSETSTATE@state, re2c:define:YYSETSTATE@state (sigil) YYSETSTATE . @@ . re2c:api:sigil . re2c:YYSETSTATE:naked, re2c:define:YYSETSTATE:naked re2c:api:style YYSETSTATE . API (free-form API style) . re2c:YYSETACCEPT, re2c:define:YYSETACCEPT API YYSETACCEPT . re2c:YYSKIP, re2c:define:YYSKIP API YYSKIP . re2c:YYSHIFT, re2c:define:YYSHIFT API YYSHIFT . re2c:YYCOPYMTAG, re2c:define:YYCOPYMTAG API YYCOPYMTAG . re2c:YYCOPYSTAG, re2c:define:YYCOPYSTAG API YYCOPYSTAG ..TP re2c:YYSHIFTMTAG, re2c:define:YYSHIFTMTAG API YYSHIFTMTAG . re2c:YYSHIFTSTAG, re2c:define:YYSHIFTSTAG API YYSHIFTSTAG . re2c:YYSTAGN, re2c:define:YYSTAGN API YYSTAGN . re2c:YYSTAGP, re2c:define:YYSTAGP API YYSTAGP . re2c:yyaccept, re2c:variable:yyaccept API yyaccept . re2c:yybm, re2c:variable:yybm API yybm . re2c:yybm:hex, re2c:variable:yybm:hex --bit-vectors . ( ). re2c:yych, re2c:variable:yych API yych . re2c:yych:emit, re2c:variable:yych:emit yych . . re2c:yych:conversion, re2c:variable:yych:conversion re2py yych YYCTYPE . ( ). re2c:yych:literals, re2c:variable:yych:literals yych . : char ( ) hex ( ) char_or_hex ( ). re2c:yyctable, re2c:variable:yyctable API yyctable . re2c:yynmatch, re2c:variable:yynmatch API yynmatch . re2c:yypmatch, re2c:variable:yypmatch API yypmatch . re2c:yytarget, re2c:variable:yytarget API yytarget . re2c:yystable, re2c:variable:yystable . re2c:yystate, re2c:variable:yystate API yystate . re2c:yyfill, re2c:variable:yyfill API yyfill . re2c:yyfill:check YYFILL ( YYLESSTHAN API YYLIMIT API C) . ( ). re2c:yyfill:enable YYFILL ( ) . ( ) YYFILL ( [sentinel] ). ( -W) re2c:sentinel . (YYFILL ). re2c:yyfill:parameter YYFILL . . ( ). re2c:YYFILL:naked re2c:api:style . re2c:yyfn:sep YYFN ( ). re2c:yyfn:throw YYFN ( ). re2py : "foo" . 'foo' . [a-xyz] [^a-xyz] ( ). . (newline). R \ S R S. R* R. R+ R. R? R . R{n} R n . R{n,} R n . R{n,m} R n m . (R) R . (submatch extraction) --invert-captures (R) (capturing) (non-capturing) . (!R) --invert-captures (!R) . R S : R S. R | S : R S. R / S (Lookahead): R S S . name name ( "name" Flex). {name} name Flex. @stag s-tag: @stag stag . #mtag m-tag: #mtag mtag . $ . : \a \b \f \n \r \t \v \\ \ooo \xhh \uhhhh \Uhhhhhhhh. (Actions) re2py : !entry code code . start conditions . (lexeme) . !pre_rule code code ( start conditions) . ( ) . !post_rule code code ( start conditions) . (trap statements) . re2py : !use:name ; use name . . : . * $ . use use . !include file ; include : file . (Program interface) (primitives) API . . . re2py ( ) . API . re2py API : Simple API API . : YYINPUT ( ) YYCURSOR YYMARKER YYCTXMARKER YYLIMIT ( YYINPUT ). Record API Record API . --api record re2c:api = record . API yyrecord ( re2c:yyrecord ) yyinput yycursor yymarker yyctxmarker yylimit ( ). Generic API API . --api generic re2c:api = generic . API : YYPEEK YYSKIP YYBACKUP YYBACKUPCTX YYSTAGP YYSTAGN YYMTAGP YYMTAGN YYRESTORE YYRESTORECTX YYRESTORETAG YYSHIFT YYSHIFTSTAG YYSHIFTMTAG YYLESSTHAN YYEND. API . YYCTYPE ( ). ASCII EBCDIC UTF-8 . UTF-16 UCS-2 . UTF-32 . YYCURSOR l-value ( YYINPUT). YYCURSOR . YYCURSOR . YYLIMIT r-value ( YYINPUT). YYLIMIT . . YYCURSOR YYLIMIT . YYMARKER l-value ( YYINPUT). YYCURSOR (rollback) . . YYCTXMARKER l-value (trailing context) ( YYINPUT). . YYCTXMARKER / . YYFILL Generic API len. YYFILL len . re2c:eof len 1 YYFILL . re2c:eof YYFILL . len YYMAXFILL . YYFN --recursive-functions . re2c:fn:sep ( ) . ( void ). : ( ) ( - ). YYINPUT r-value ( ) . YYMAXFILL YYFILL. max . YYLESSTHAN Generic API len. r-value len true . YYEND Generic API . r-value ( ) true . YYEND $ . YYLESSTHAN . YYPEEK Generic API . r-value YYCTYPE . YYSKIP Generic API . YYBACKUP Generic API ( YYRESTORE ). YYRESTORE Generic API YYBACKUP . YYBACKUPCTX Generic API ( YYRESTORECTX ). YYRESTORECTX Generic API YYBACKUPCTX . YYRESTORETAG API tag (trailing context) tag . YYSTAGP API tag tag YYINPUT ( ). YYSTAGP tag . YYSTAGN API tag tag YYINPUT ( ). YYSTAGN tag . YYMTAGP API tag. YYMTAGP tag ( ). YYMTAGN API tag. YYMTAGN tag ( ). YYSHIFT API shift shift ( ). YYCOPYSTAG API lhs rhs s-tag (rhs) s-tag (lhs) . lhs rhs . YYCOPYMTAG API lhs rhs m-tag (rhs) m-tag (lhs) . lhs rhs . YYSHIFTSTAG API tag shift tag shift ( ). YYSHIFTMTAG API tag shift tag shift ( ). YYMAXNMATCH POSIX . maxnmatch . YYCONDTYPE (enum) . conditions --header . YYGETACCEPT var . var . YYSETACCEPT : var ( l-value ) val ( ). var val . YYGETCOND r-value YYCONDTYPE . YYSETCOND cond cond . YYGETSTATE r-value . -1 . YYSETSTATE state state . YYDEBUG -d --debug-output . ( YYDEBUG ). YYDEBUG : state ( DFA -1) symbol ( ). yyaccept l-value . --storable-state . yybm ( --bitmaps ). - . - . k- n- (true) n- k- . if switch . yych l-value YYCTYPE . -f --storable-state . yyctable ( --conditions --computed-gotos). yyfill l-value YYFILL ( YYFILL ). yynmatch l-value POSIX . -P --posix-captures . yypmatch l-value . yynmatch * 2 ( YYMAXNMATCH * 2 ). -P --posix-captures . yystable . yystate l-value --loop-switch DFA . yytarget ( ) . . yytarget --computed-gotos . re2c . (debug) re2c . re2c . -? --help -h . --api API . simple (lexer) . record lexer (struct class ) . generic API . --bit-vectors -b (bit masks). --nested-ifs . --captures, --leftmost-captures (submatch) (leftmost greedy). yybmatch 2 * YYMAXNMATCH yynmatch . --captvars, --leftmost-captvars . yytl yytr k- . --case-insensitive . --case-inverted : . --case-ranges case switch low ... high. C/C++ GCC Clang Tcc . case Tcc (binary) . --computed-gotos -g "computed goto" ( ). re2py . cgoto:threshold . cgoto:relative . --bit-vectors . --computed-gotos-relative --computed-gotos ( ). --computed-gotos . --conditions --start-conditions -c "" Flex: . re2py goto . --depfile FILE FILE Makefile : [include-file ...]. / include include . --output . --ebcdic --ecb -e EBCDIC . re2py 0 -- 0xFF . --empty-class re2py . match-empty () ( ). match-none . error . --encoding-policy re2py (Unicode surrogates). fail re2py . substitute re2py 0xFFFD . ignore () re2py . . --flex-syntax -F Flex: . . --goto-label "goto/label": DFA goto . goto . --header --type-header -t HEADER HEADER. header:on header:off . enum ( conditions ). -I PATH PATH include . include . re2py FILE include -I . --input --api. default simple ( backend ). custom generic . --input-encoding re2py. ascii () re2py ASCII : . utf8 re2py UTF8 . --invert-captures . (...) (! ...) . (! ...) (...) . --lang . C D Go Haskell Java JS OCaml Python Rust Swift V Zig ( --syntax ). none . --location-format . gnu 'filename:line:column: ...' . msvc 'filename(line,column) ...' . gnu . --loop-switch "loop/switch": DFA switch case switch . yystate . yystate case . --nested-ifs -s if switch . . --no-debug-info -i . ( (diff) ). --no-generation-date . --no-version . --no-unsafe unsafe YYPEEK ( Rust ). YYPEEK (bounds-checking) . YYPEEK (release) --no-unsafe unsafe . --output -o OUTPUT OUTPUT. --posix-captures, -P POSIX. yybmatch 2 * YYMAXNMATCH yynmatch ..TP --posix-captvars POSIX. k- yytl yytr . --recursive-functions (co-recursive) DFA . --reusable -r 2.2 ( ). --skeleton -S "skeleton" (). "skeleton" . . C . --storable-state -f (lexer) . push . YYGETSTATE YYSETSTATE yych yyaccept state . --syntax FILE FILE . FILE ( ) ( --lang none ). --tags -T . --ucs2 --wide-chars -w UCS2 . re2py 0 -- 0xFFFF . --nested-ifs . --utf8 --utf-8 -8 UTF-8 . re2py 0 -- 0x10FFFF . --utf16 --utf-16 -x UTF16 . re2py 0 -- 0x10FFFF . --nested-ifs . --utf32 --unicode -u UTF32 . re2py 0 -- 0x10FFFF . --nested-ifs . --verbose . --vernum -V MMmmpp ( ). --version -v . --single-pass -1 . ( ). --debug-output -d YYDEBUG . . --dump-adfa : DFA ( .dot). --dump-cfg : (CFG) ( .dot). --dump-closure-stats : (closure). --dump-dfa-det : DFA ( .dot). --dump-dfa-min : DFA ( .dot). --dump-dfa-tagopt : DFA ( .dot). --dump-dfa-tree : DFA ( .dot). --dump-dfa-raw : DFA ( .dot). --dump-interf : (liveness analysis) . --dump-nfa : NFA ( .dot). --emit-dot -D .dot . Graphviz ( dot -Tpng -odfa.png dfa.dot). --dfa-minimization : DFA re2py. moore (Moore) ( ). table "table filling" ( ) . DFA . --eager-skip : (eagerly) -- . (lazily) -- . --no-lookahead . TDFA(0) . TDFA(1) TDFA(0) . . TDFA(0) TDFA(1) . --no-optimize-tags : ( ). --posix-closure : (epsilon-closure) POSIX: gor1 () - (Goldberg-Radzik) gtop "global topological order" ( ) . --posix-prectable : POSIX. complex TNFA . naive ( ) TNFA complex . --stadfa . staDFA TDFA . staDFA TDFA . --fixed-tags : (fixed-tag) (all) (none) (toplevel) . all . <<>> ("Fixed") ( <<>> "base") . . (no-match) ( no-match ). . . -W . -Werror . . -W warning. -Wno- warning. -Werror- warning ( -W ). -Wno-error- warning . . -Wcondition-order . --header conditions . -Wempty-character-class . : . re2py . --empty-class . -Wmatch-empty-string ( ). . -Wswapped-range . . -Wundefined-control-flow ( ). . * . -Wunreachable-rules (shadowed) . -Wdeprecated-eof_rule $ . $ . -Wuseless-escape (escaped). re2py . -Wnondeterministic-tags n- n . -Wsentinel-in-midrule (sentinel) --- . . re2c:sentinel . -Wundefined-syntax-config --syntax . : . . re2c (syntax files) . -- . re2c re2c: : . ( include/syntax ) re2py . ( --lang none ). : . (Feature lists) re2py : supported_apis API simple record generic. supported_api_styles API functions free-form. supported_code_models goto-label loop-switch recursive-functions. supported_targets code dot skeleton. supported_features nested-ifs bitmaps computed-gotos case-ranges monadic unsafe tags captures captvars. (Language configurations) (parser) re2py : semicolons (semicolon) . backtick_quoted_strings . single_quoted_strings . indentation_sensitive . wrap_blocks_in_braces . (Inplace configurations) re2c: . . (Code templates) . - (DSL) : code-template :: name '=' code-exprs ';' | CODE_TEMPLATE ';' | '' ';' code-exprs :: | code-exprs code-expr code-expr :: STRING | VARIABLE | optional | list optional :: '(' CONDITIONAL '?' code-exprs ')' | '(' CONDITIONAL '?' code-exprs ':' code-exprs ')' list :: '[' VARIABLE ':' code-exprs ']' | '[' VARIABLE '{' NUMBER '}' ':' code-exprs ']' | '[' VARIABLE '{' NUMBER ',' NUMBER '}' ':' code-exprs ']' . (placeholders) . : re2py . ( -1 ). . ( ) ( ) . re2py . if-then-else C : code:if_then_else = [branch{0}: topindent "if " cond " {" nl indent [stmt: stmt] dedent] [branch{1:-1}: topindent "} else" (.cond ? " if " cond) " {" nl indent [stmt: stmt] dedent] topindent "}" nl; branch : branch{0} ( else ) branch{1:-1} ( ) . stmt : [stmt: stmt] . topindent indent dedent nl .cond ( ). : if x { // do something } else if y { // do something else } else { // don't do anything } re2py . . . code:var_local . : type ( ) name ( ) init ( ). : .init ( ). code:var_global code:var_local (top-level) . code:const_local . : type ( ) name ( ) init ( ). code:const_global code:const_local (top-level) . code:array_local () . : type ( ) name ( ) size ( ) row ( ) elem ( -- row ). : .const ( ). code:array_global code:array_local (top-level) . code:array_elem (). : array ( ) index ( ). code:enum ( ). type ( ) elem ( ) init ( ). : .init ( ). code:enum_elem ( code:enum). name ( ) type ( ). code:assign . lhs ( ) rhs ( ). code:type_int . code:type_uint . code:type_yybm yybm. code:type_yytarget yytarget. code:type_yyctable yyctable. code:cmp_eq " ". code:cmp_ne " ". code:cmp_lt " ". code:cmp_gt " ". code:cmp_le " ". code:cmp_ge " ". code:if_then_else If-then-else . : branch ( ) cond ( ) stmt ( ). : .cond ( ) .many ( ). code:if_then_else_oneline code:if_then_else . code:if_then_else . code:switch switch case. : expr ( switch) case ( case ). code:switch_cases case switch . case ( case ) stmt ( ). code:switch_cases_oneline code:switch_cases . code:switch_cases . code:switch_case_range case switch ( ). : val ( ). : .many ( ) .char_literals ( switch -- ). code:switch_case_default (default) switch. code:loop ( ). : label ( ) stmt ( ). code:continue continue. : label ( ). code:goto goto. : label ( ). code:cgoto goto . : array ( goto ) index ( ) base ( .cgoto.relative ). code:cgoto:data goto . : label ( ) type ( ) base ( - .cgoto.relative ). code:fndecl . : name ( ) type ( ) throw ( re2c:yyfn:throw ) arg ( ) argname ( ) argtype ( ). : .type ( void ). code:fndef code:fndecl stmt . code:fncall . : name ( ) retval ( l-value ) arg ( ). : .args ( ) .retval ( ). code:tailcall (tail call). : name ( ) arg ( ). : .args ( ) .retval ( void ). code:recursive_functions --recursive-functions. : fn ( ) fndecl ( ) fndef ( )..TP code:fingerprint (fingerprint) . : ver ( re2py ) date ( ). code:line_info ( ). : line ( ) file ( ). code:abort (abort) . code:yydebug YYDEBUG API . : YYDEBUG yyrecord yych ( re2c:) state ( DFA). code:yypeek YYPEEK API . : YYPEEK YYCTYPE YYINPUT YYCURSOR yyrecord yych ( re2c:). : .cast ( re2c:yych:conversion ). code:yyskip YYSKIP API . : YYSKIP YYCURSOR yyrecord ( re2c:). code:yybackup YYBACKUP API . : YYBACKUP YYCURSOR YYMARKER yyrecord ( re2c:). code:yybackupctx YYBACKUPCTX API . : YYBACKUPCTX YYCURSOR YYCTXMARKER yyrecord ( re2c:). code:yyskip_yypeek code:yyskip code:yypeek ( code:yyskip code:yypeek ). code:yypeek_yyskip code:yypeek code:yyskip ( code:yypeek code:yyskip ). code:yyskip_yybackup code:yyskip code:yybackup ( code:yyskip code:yybackup ). code:yybackup_yyskip code:yybackup code:yyskip ( code:yybackup code:yyskip ). code:yybackup_yypeek code:yybackup code:yypeek ( code:yybackup code:yypeek ). code:yyskip_yybackup_yypeek code:yyskip code:yybackup code:yypeek ( ``code:yyskip`` code:yybackup code:yypeek ). code:yybackup_yypeek_yyskip code:yybackup code:yypeek code:yyskip ( ``code:yybackup`` code:yypeek code:yyskip ). code:yyrestore YYRESTORE API . : YYRESTORE YYCURSOR YYMARKER yyrecord ( re2c:). code:yyrestorectx YYRESTORECTX API . : YYRESTORECTX YYCURSOR YYCTXMARKER yyrecord ( re2c:). code:yyrestoretag YYRESTORETAG API . : YYRESTORETAG YYCURSOR yyrecord ( re2c:) tag ( ). code:yyshift YYSHIFT API . : YYSHIFT YYCURSOR yyrecord ( re2c:) offset ( ). code:yyshiftstag YYSHIFTSTAG API . : YYSHIFTSTAG yyrecord negative ( re2c:) tag ( ) offset ( ). : .nested ( -- re2c:tags:negative ). code:yyshiftmtag YYSHIFTMTAG API . : YYSHIFTMTAG ( re2c:) tag ( ) offset ( ). code:yystagp YYSTAGP API . : YYSTAGP YYCURSOR yyrecord ( re2c:) tag ( ). code:yymtagp YYMTAGP API . : YYMTAGP ( re2c:) tag ( ). code:yystagn YYSTAGN API . : YYSTAGN negative yyrecord ( re2c:) tag ( ). code:yymtagn YYMTAGN API . : YYMTAGN ( re2c:) tag ( ). code:yycopystag YYCOPYSTAG API . : YYCOPYSTAG yyrecord ( re2c:) lhs rhs ( ). code:yycopymtag YYCOPYMTAG API . : YYCOPYMTAG yyrecord ( re2c:) lhs rhs ( ). code:yygetaccept YYGETACCEPT API . : YYGETACCEPT yyrecord ( re2c:) var ( re2c:yyaccept). code:yysetaccept YYSETACCEPT API . : YYSETACCEPT yyrecord ( re2c:) var ( re2c:yyaccept) val ( ). code:yygetcond YYGETCOND API . : YYGETCOND yyrecord ( re2c:) var ( re2c:yycond). code:yysetcond YYSETCOND API . : YYSETCOND yyrecord ( re2c:) var ( re2c:yycond) val ( ). code:yygetstate YYGETSTATE API . : YYGETSTATE yyrecord ( re2c:) var ( re2c:yystate). code:yysetstate YYSETSTATE API . : YYSETSTATE yyrecord ( re2c:) var ( re2c:yystate) val ( ). code:yylessthan YYLESSTHAN API . : YYLESSTHAN YYCURSOR YYLIMIT yyrecord ( re2c:) need ( ). : .many ( need )..TP code:yyend YYEND API . : YYEND YYCURSOR YYLIMIT. code:yybm_filter yych yybm ( --bitmaps ). : yych ( re2c:yych ). code:yybm_match yybm ( --bitmaps). : yybm yych ( re2c: ) offset ( yybm yych ) mask ( ) code:yytarget_filter yych yytarget ( --computed-gotos ). : yych ( re2c:yych ). : nl (newline). indent . dedent . topindent . . : .api.simple API (--api simple re2c:api = simple). .api.generic API (--api generic re2c:api = generic). .api.record API (--api record re2c:api = record). .api_style.functions API (re2c:api-style = functions). .api_style.freeform API (re2c:api-style = free-form). .case_ranges case (--case-ranges re2c:case-ranges = 1). .cgoto.relative goto (--computed-gotos-relative re2c:cgoto:relative = 1). .code_model.goto_label goto/label (--goto-label). .code_model.loop_switch loop/switch (--loop-switch). .code_model.recursive_functions (--recursive-function). .date (fingerprint) . .loop_label re2py (re2c:label:yyloop ). .monadic (re2c:monadic = 1). . .start_conditions (--start-conditions). .storable_state (--storable-state). .unsafe re2py "unsafe" (--unsafe re2c:unsafe = 1). "unsafe" . .version re2py . .yyfill.enable YYFILL (re2c:yyfill:enable = 1). .yyfn.throw re2c:yyfn:throw . (lexer) . : o ( *) o o "" : . : . . re2py . . : o (Sentinel). . "" . . : . . YYFILL re2c:yyfill:enable = 0; re2c:eof -1 . o (Sentinel with bounds checks). : . . "" "" . ( ). . . YYFILL ( ). . . re2c:eof ( ). YYFILL . o (Bounds checks with padding). " " . DFA (SCC) SCC SCC . (lexemes) . ( ) ( ). YYMAXFILL max . YYFILL . YYFILL re2c:eof -1 ( ). o (Custom checks). (Generic API) . . --api custom re2c:api = custom; re2c:yyfill:enable = 0; re2c:yyfill:check = 0; . ..SS (Sentinel) (sentinel) . null . null : (lexeme) ( ). null * ( -Wsentinel-in-midrule re2c:sentinel ). re2c:yyfill:enable = 0; YYFILL . # re2py $INPUT -o $OUTPUT # expect a null-terminated string def lex(yyinput): yycursor = 0 count = 0 while True: %{ re2c:yyfill:enable = 0; re2c:indent:top = 2; * { return -1 } [\x00] { return count } [ ]+ { break } [a-z]+ { count += 1 break } %} assert lex(b"\0") == 0 assert lex(b"one two three\0") == 3 assert lex(b"f0ur\0") == -1 (Sentinel with bounds checks) ( 1.2 ). . null re2c:eof = 0; . sentinel null ( 'aaa\0aa'\0 'aaa\0 ). null . YYLIMIT <= YYCURSOR API YYLESSTHAN(1) . (YYFILL re2c:yyfill:enable = 0; YYFILL YYFILL with sentinel ). : $ ( *) -Wundefined-control-flow . # re2py $INPUT -o $OUTPUT # expect a null-terminated string def lex(yyinput): yycursor = 0 yylimit = len(yyinput) - 1 # terminating null not included count = 0 while True: %{ re2c:yyfill:enable = 0; re2c:eof = 0; re2c:indent:top = 2; str = ['] ([^'\\] | [\\][^])* [']; * { return -1 } $ { return count } [ ]+ { break } str { count += 1 break } %} def test(str, count): # termunating null not included in `lim` assert count == lex(str) test(b"\0", 0); test(b"'qu\0tes' 'are' 'fine: \\'' \0", 3); test(b"'unterminated\\'\0", -1) (padding) ( ). . YYMAXFILL null YYMAXFILL max . null --- (lexeme) ( ). "stop" (null) ( null ). . API (YYLIMIT - YYCURSOR) < n YYLESSTHAN(n) n ( n ). YYFILL(n) n . YYFILL ( ). YYFILL YYFILL with padding . # re2py $INPUT -o $OUTPUT %{max %} def lex(yyinput): yycursor = 0 yylimit = len(yyinput) count = 0 while True: %{ re2c:YYFILL = "return -1"; re2c:indent:top = 2; str = ['] ([^'\\] | [\\][^])* [']; [\x00] { # check that it is the sentinel, not some unexpected null return count if yycursor == yylimit - YYMAXFILL + 1 else -1 } str { count += 1 break } [ ]+ { break } * { return -1 } %} def test(str, count): padded_str = str + (b"\0" * YYMAXFILL) assert lex(padded_str) == count test(b"", 0) test(b"'unterminated\\'", -1) test(b"'qu\x00tes' 'are' 'fine: \\'' ", 3) test(b"'unexpected \x00 null", -1) (Custom checks) generic API . . sentinel null . (sentinel) YYPEEK . . YYPEEK . # re2py $INPUT -o $OUTPUT # expect a string without terminating null def lex(str): cur = 0 lim = len(str) count = 0 while True: %{ re2c:api = generic; re2c:YYPEEK = "str[cur] if cur < lim else 0"; re2c:YYSKIP = "cur += 1"; re2c:yyfill:enable = 0; re2c:indent:top = 2; * { return -1 } [\x00] { return count } [ ]+ { break } [a-z]+ { count += 1 break } %} assert lex(b"") == 0 assert lex(b"one two three") == 3 assert lex(b"f0ur") == -1 (buffering) : ( ). (chunks) . . (lexer) : o (cursor): (YYCURSOR API C YYSKIP/YYPEEK API ). o (limit): (YYLIMIT API C YYLESSTHAN API ). o (marker): (YYMARKER API YYBACKUP/YYRESTORE API ). o (token): (lexeme) ( API re2py lexer ). o (context marker): (trailing context) (YYCTXMARKER API C YYBACKUPCTX/YYRESTORECTX API ). o (tag variables): ( stags mtags API YYSTAGP/YYSTAGN/YYMTAGP/YYMTAGN). YYFILL . token cursor token token limit . YYFILL ( lexer ). YYFILL EOF : EOF (padding) . -f --storable-state YYFILL ( ). YYFILL (YYFILL with sentinel) EOF YYFILL . YYFILL YYLIMIT <= YYCURSOR API C YYLESSTHAN() API . YYFILL . YYFILL . (Limit) limit (sentinel) re2c:eof . YYFILL ( # ). <-- shift --> >-A------------B---------C------------D#-----------E-> buffer token marker limit, cursor >-A------------B---------C------------D------------E#-> buffer, marker cursor limit token <-- shift --> >-A------------B---------C------------D#--E (EOF) buffer token marker limit, cursor >-A------------B---------C------------D---E#........ buffer, marker cursor limit token input.txt EOF . # re2py $INPUT -o $OUTPUT from enum import Enum import os BUFSIZE = 4096 class State: def __init__(self, fname): self.file = open(fname, "rb") self.yyinput = bytearray(BUFSIZE) self.yylimit = BUFSIZE - 1 # exclude terminating null self.yycursor = self.yylimit self.yymarker = self.yylimit self.token = self.yylimit self.eof = False def __del__(self): self.file.close() class Status(Enum): OK = 0 EOF = 1 LONG_LEXEME = 2 def fill(st): if st.eof: return Status.EOF # Error: lexeme too long. In real life could reallocate a larger buffer. if st.token < 1: return Status.LONG_LEXEME # Shift buffer contents (discard everything up to the current token). st.yyinput = st.yyinput[st.token:st.yylimit] st.yycursor -= st.token; st.yymarker -= st.token; st.yylimit -= st.token; st.token = 0; # Fill free space at the end of buffer with new data from file. bytes = st.file.read(BUFSIZE - st.yylimit - 1) # -1 for sentinel if not bytes: st.eof = True # end of file else: st.yylimit += len(bytes); st.yyinput += bytes st.yyinput += b'\0' # append sentinel return Status.OK def lex(yyrecord, count): while True: yyrecord.token = yyrecord.yycursor %{ re2c:api = record; re2c:define:YYFILL = "fill(yyrecord) == Status.OK"; re2c:eof = 0; re2c:indent:top = 2; str = ['] ([^'\\] | [\\][^])* [']; * { return -1 } $ { return count } [ ]+ { break } str { count += 1 break } %} def main(): fname = "input" # Prepare input file. f = open(fname, "w") for i in range(BUFSIZE): f.write("'qu\0tes' 'are' 'fine: \\'' ") f.close() # Run lexer on the prepared file. st = State(fname) assert lex(st, 0) == 3 * BUFSIZE # Cleanup. os.remove(fname) if __name__ == '__main__': main() ( EOF ) YYFILL . YYFILL (YYLIMIT - YYCURSOR) < n API C YYLESSTHAN(n) API . YYFILL . YYFILL ( ). YYFILL (limit) YYMAXFILL ( YYFILL n ). YYFILL ( YYMAXFILL # ). <-- shift --> <-- need --> >-A------------B---------C-----D-------E---F--------G-> buffer token marker cursor limit >-A------------B---------C-----D-------E---F--------G-> buffer, marker cursor limit token <-- shift --> <-- need --> >-A------------B---------C-----D-------E-F (EOF) buffer token marker cursor limit >-A------------B---------C-----D-------E-F############### buffer, marker cursor limit token <- YYMAXFILL -> input.txt (padding) . # re2py $INPUT -o $OUTPUT from enum import Enum import os BUFSIZE = 4096 %{max %} class State: def __init__(self, fname): self.file = open(fname, "rb") self.yyinput = bytearray(BUFSIZE) self.yylimit = BUFSIZE - YYMAXFILL self.yycursor = self.yylimit self.yymarker = self.yylimit self.token = self.yylimit self.eof = False def __del__(self): self.file.close() class Status(Enum): OK = 0 EOF = 1 LONG_LEXEME = 2 def fill(st, need): if st.eof: return Status.EOF # Error: lexeme too long. In real life could reallocate a larger buffer. if st.token < need: return Status.LONG_LEXEME # Shift buffer contents (discard everything up to the current token). st.yyinput = st.yyinput[st.token:st.yylimit] st.yycursor -= st.token; st.yymarker -= st.token; st.yylimit -= st.token; st.token = 0; # Fill free space at the end of buffer with new data from file. bytes = st.file.read(BUFSIZE - st.yylimit - 1) # -1 for sentinel if not bytes: st.eof = True # end of file st.yylimit += YYMAXFILL st.yyinput += b"\0" * YYMAXFILL else: st.yylimit += len(bytes); st.yyinput += bytes return Status.OK def lex(yyrecord): count = 0 while True: yyrecord.token = yyrecord.yycursor %{ re2c:api = record; re2c:YYFILL = "if fill(yyrecord, @@) != Status.OK: return -1"; re2c:indent:top = 2; str = ['] ([^'\\] | [\\][^])* [']; [\x00] { # Check that it is the sentinel, not some unexpected null. return count if yyrecord.token == yyrecord.yylimit - YYMAXFILL else -1 } str { count += 1 break } [ ]+ { break } * { return -1 } %} def main(): fname = "input" # Prepare input file. f = open(fname, "w") for i in range(BUFSIZE): f.write("'qu\0tes' 'are' 'fine: \\'' ") f.close() # Run lexer on the prepared file. st = State(fname) assert lex(st) == 3 * BUFSIZE # Cleanup. os.remove(fname) if __name__ == '__main__': main() (Multiple blocks) (lexer) ( ). re2py . (start conditions) . . goto ( C/C++ Go) (label) . goto . goto ( Rust) switch (state variable) /switch yystate re2py . ( ) ( ) ( ) ( ) . . . . # re2py $INPUT -o $OUTPUT class State: def __init__(self, str): self.yyinput = str self.yycursor = 0 self.yymarker = 0 # Common re2c definitions shared between all functions. %{ re2c:api = record; re2c:yyrecord = st; re2c:yyfill:enable = 0; re2c:indent:top = 2; %} def parse_u32(str): st = State(str) %{local re2c:indent:top = 1; '0b' / [01] { return parse_bin(st) } "0" { return parse_oct(st) } "" / [1-9] { return parse_dec(st) } '0x' / [0-9a-fA-F] { return parse_hex(st) } * { return None } %} def parse_bin(st): n = 0 while True: %{ [01] { n = n * 2 + (st.yyinput[st.yycursor - 1] - 48) break } * { return n } %} def parse_oct(st): n = 0 while True: %{ [0-7] { n = n * 8 + (st.yyinput[st.yycursor - 1] - 48) break } * { return n } %} def parse_dec(st): n = 0 while True: %{ [0-9] { n = n * 10 + (st.yyinput[st.yycursor - 1] - 48) break } * { return n } %} def parse_hex(st): n = 0 while True: %{ [0-9] { n = n * 16 + (st.yyinput[st.yycursor - 1] - 48) break } [a-f] { n = n * 16 + (st.yyinput[st.yycursor - 1] - 87) break } [A-F] { n = n * 16 + (st.yyinput[st.yycursor - 1] - 55) break } * { return n } %} assert parse_u32(b"\0") == None assert parse_u32(b"1234567890\0") == 1234567890 assert parse_u32(b"0b1101\0") == 13 assert parse_u32(b"0x7Fe\0") == 2046 assert parse_u32(b"0644\0") == 420 assert parse_u32(b"9999999999\0") == 9999999999 --start-conditions . re2py . re2py . : YYGETCOND YYSETCOND . conditions --header re2c:header . re2c:cond:enumprefix . <*> . : < condition-list > regular-expression code condition-list . regular-expression code . < condition-list > regular-expression => condition code condition-list . regular-expression condition code . < condition-list > regular-expression :=> condition condition-list . regular-expression condition ( ). < condition-list > !action code code action condition-list ( actions ). code code condition-list . !pre_rule ( ). < > code "0" code . !entry ( ). < > => condition code condition . < > :=> condition condition . re2py re2py goto/label loop/switch. goto ( C/C++ Go) yyc_ cond ( re2c:cond:prefix ). goto . re2py switch YYGETSTATE . :=> switch ( re2c:cond:goto ). . goto ( Rust) re2py yystate . switch switch DFA . switch . ( YYGETCOND/YYSETCOND ). () () () () . . cond conditions . # re2py $INPUT -o $OUTPUT -c %{conditions %} def parse_u32(yyinput): yycursor = 0 yycond = YYC_INIT num = 0 while True: %{ re2c:yyfill:enable = 0; re2c:indent:top = 2; '0b' / [01] :=> BIN "0" :=> OCT "" / [1-9] :=> DEC '0x' / [0-9a-fA-F] :=> HEX * { return None } [01] { num = num * 2 + (yyinput[yycursor - 1] - 48) break } [0-7] { num = num * 8 + (yyinput[yycursor - 1] - 48) break } [0-9] { num = num * 10 + (yyinput[yycursor - 1] - 48) break } [0-9] { num = num * 16 + (yyinput[yycursor - 1] - 48) break } [a-f] { num = num * 16 + (yyinput[yycursor - 1] - 87) break } [A-F] { num = num * 16 + (yyinput[yycursor - 1] - 55) break } * { return num } %} assert parse_u32(b"\0") == None assert parse_u32(b"1234567890\0") == 1234567890 assert parse_u32(b"0b1101\0") == 13 assert parse_u32(b"0x7Fe\0") == 2046 assert parse_u32(b"0644\0") == 420 assert parse_u32(b"9999999999\0") == 9999999999 --storable-state re2py (lexer) . re2py <<>> ("pull") . ( (parser) ). : <<>> ("push") . . . <<>> : o YYSETSTATE() YYGETSTATE(state) . o yych yyaccept ( ) state . state -1 . o YYFILL . . o . o getstate YYGETSTATE . goto ( --loop-switch). o re2c:eof sentinel with bounds checks ( ) . (padding) : (greedy lexeme) . ( ). <<>> . . <<>> . . # re2py $INPUT -o $OUTPUT -f from enum import Enum import os # Use a small buffer to cover the case when a lexeme doesn't fit. # In real world use a larger buffer. BUFSIZE = 10 DEBUG = False class State: def __init__(self, file): self.file = file self.yyinput = bytearray(BUFSIZE) self.yylimit = BUFSIZE - 1 # exclude terminating null self.yycursor = self.yylimit self.yymarker = self.yylimit self.token = self.yylimit self.yystate = -1 class Status(Enum): END = 0 READY = 1 WAITING = 2 BIG_PACKET = 3 BAD_PACKET = 4 def fill(st): # Error: lexeme too long. In real life could reallocate a larger buffer. if st.token < 1: return Status.BIG_PACKET # Shift buffer contents (discard everything up to the current token). st.yyinput = st.yyinput[st.token:st.yylimit] st.yycursor -= st.token; st.yymarker -= st.token; st.yylimit -= st.token; st.token = 0; # Fill free space at the end of buffer with new data from file. bytes = st.file.read(BUFSIZE - st.yylimit - 1) # -1 for sentinel if bytes: st.yylimit += len(bytes); st.yyinput += bytes st.yyinput += b'\0' # append sentinel return Status.READY def lex(yyrecord, recv): while True: yyrecord.token = yyrecord.yycursor %{ re2c:api = record; re2c:YYFILL = "return Status.WAITING, recv"; re2c:eof = 0; re2c:indent:top = 2; packet = [a-z]+[;]; * { return Status.BAD_PACKET, recv } $ { return Status.END, recv } packet { recv += 1 break } %} def test(packets, expect): # Create a pipe (open the same file for reading and writing). fname = "pipe" fw = open(fname, "wb") fr = open(fname, "rb") # Initialize lexer state st = State(fr) # Main loop. The buffer contains incomplete data which appears packet by # packet. When the lexer needs more input it saves its internal state and # returns to the caller which should provide more input and resume lexing. send = 0 recv = 0 while True: status, recv = lex(st, recv) if status == Status.END: if DEBUG: print("done: got {} packets".format(recv)) break elif status == Status.WAITING: if DEBUG: print("waiting..."); if send < len(packets): if DEBUG: print("sent packet {}: {}".format(send, packets[send])) fw.write(packets[send]) fw.flush() send += 1 status = fill(st) if DEBUG: print("queue: '{}', status: {}".format(st.yyinput, status)) if status == Status.BIG_PACKET: if DEBUG: print("error: packet too big") break assert status == Status.READY else: assert status == Status.BAD_PACKET if DEBUG: print("error: ill-formed packet") break # Check results. assert status == expect if status == Status.END: assert recv == send # Cleanup: remove input file. fr.close() fw.close() os.remove(fname) def main(): test([], Status.END) test([b"zero;", b"one;", b"two;", b"three;", b"four;"], Status.END) test([b"zer0;"], Status.BAD_PACKET) test([b"goooooooooogle;"], Status.BIG_PACKET) if __name__ == '__main__': main() /*!rules:re2c[:] ... */ %{rules[:] ... %} re2py . . use . . use /*!use:re2c[:] ... */ %{use[:] ... %} . : ( ). use . use (lexer) : ( ). !use:; re2py . . : . * $ ( ) . use re2py ( ). use 2.2 re2py . ( ). 2.2 -r --reusable . 1.2 . !use # re2py $INPUT -o $OUTPUT # This example shows how to combine reusable re2c blocks: two blocks # ('colors' and 'fish') are merged into one. The 'salmon' rule occurs # in both blocks; the 'fish' block takes priority because it is used # earlier. Default rule * occurs in all three blocks; the local (not # inherited) definition takes priority. from enum import Enum class Ans(Enum): COLOR = 1 FISH = 2 DUNNO = 3 %{rules:colors * { raise "ah" } "red" | "salmon" | "magenta" { return Ans.COLOR } %} %{rules:fish * { raise "oh" } "haddock" | "salmon" | "eel" { return Ans.FISH } %} def lex(yyinput): yycursor = 0 %{ re2c:yyfill:enable = 0; re2c:indent:top = 1; !use:fish; !use:colors; * { return Ans.DUNNO } // overrides inherited '*' rules %} assert lex(b"salmon") == Ans.FISH assert lex(b"what?") == Ans.DUNNO /*!use:re2c ... */ # re2py $INPUT -o $OUTPUT --input-encoding utf8 # This example supports multiple input encodings: UTF-8 and UTF-32. # Both lexers are generated from the same rules block, and the use # blocks add only encoding-specific configurations. %{rules re2c:yyfill:enable = 0; re2c:indent:top = 1; "x y" { return yycursor } * { return None } %} def lex_utf8(yyinput): yycursor = 0 %{use re2c:encoding:utf8 = 1; %} def lex_utf32(yyinput): yycursor = 0 %{use re2c:encoding:utf32 = 1; %} s8 = [0xe2, 0x88, 0x80, 0x78, 0x20, 0xe2, 0x88, 0x83, 0x79] assert lex_utf8(s8) == len(s8) s32 = [0x2200, 0x78, 0x20, 0x2203, 0x79] assert lex_utf32(s32) == len(s32) re2py . Tags tags () @stag #mtag stag mtag . -T --tags re2c:tags = 1 . : ( ) . S-tags m-tags ( ). svars mvars. (leftmost greedy) . Captures capturing groups ( ) . --captures re2c:captures = 1 . --leftmost-captures () --posix-captures POSIX . (bang) : (! ... ). --invert-captures re2c:invert-captures = 1 . yynmatch ( ) yypmatch . yynmatch yypmatch yypmatch [yynmatch * 2] . maxnmatch YYMAXNMATCH yynmatch . Captvars --captvars re2c:captvars = 1 . --captures : yynmatch yypmatch re2py yytl yytr k- ( svars ). : --leftmost-captvars re2c:leftmost-captvars = 1 ( ) --posix-captvars re2c:posix-captvars POSIX. Tagged Deterministic Finite Automata with Lookahead . TDFA (overhead) . ( ) TDFA DFA . . TDFA tag variables ( ) : . . stags mtags . . YYFILL . S-tags : o s-tag: t = YYCURSOR API C YYSTAGP(t) API (generic) o s-tag: t = NULL API C YYSTAGN(t) API o s-tag : t1 = t2 M-tags : o m-tag: YYMTAGP(t) API o m-tag: YYMTAGN(t) API o m-tag : t1 = t2 S-tag ( ) . M-tag . m-tag ( ) m-tag (prefix-tree) (v, p) v p . s-tag (semantic versions) : major minor patch ( ). YYFILL . # re2py $INPUT -o $OUTPUT from collections import namedtuple SemVer = namedtuple('SemVer', 'major minor patch') NONE = -1 def parse(yyinput): yycursor = 0 %{ re2c:yyfill:enable = 0; re2c:tags = 1; re2c:indent:top = 1; num = [0-9]+; @t1 num @t2 "." @t3 num @t4 ("." @t5 num)? [\x00] { major = int(yyinput[t1:t2]) minor = int(yyinput[t3:t4]) patch = int(yyinput[t5:yycursor - 1]) if t5 != NONE else 0 return SemVer(major, minor, patch) } * { return None } %} assert parse(b"23.34\0") == SemVer(23, 34, 0) assert parse(b"1.2.99999\0") == SemVer(1, 2, 99999) assert parse(b"1.a\0") == None s-tag YYFILL . YYFILL . s-tag . ( m-tag ) . # re2py $INPUT -o $OUTPUT from collections import namedtuple from enum import Enum import os BUFSIZE = 4096 SemVer = namedtuple('SemVer', 'major minor patch') class State: def __init__(self, fname): self.file = open(fname, "rb") self.yyinput = bytearray(BUFSIZE) self.yylimit = BUFSIZE - 1 # exclude terminating null self.yycursor = self.yylimit self.yymarker = self.yylimit self.token = self.yylimit self.eof = False %{stags format = "\n self.@@ = -1"; %} def __del__(self): self.file.close() class Status(Enum): OK = 0 EOF = 1 LONG_LEXEME = 2 def fill(st): if st.eof: return Status.EOF # Error: lexeme too long. In real life could reallocate a larger buffer. if st.token < 1: return Status.LONG_LEXEME # Shift buffer contents (discard everything up to the current token). st.yyinput = st.yyinput[st.token:st.yylimit] st.yycursor -= st.token; st.yymarker -= st.token; st.yylimit -= st.token; %{stags format = "\n if st.@@ != -1: st.@@ -= st.token"; %} st.token = 0; # Fill free space at the end of buffer with new data from file. bytes = st.file.read(BUFSIZE - st.yylimit - 1) # -1 for sentinel if not bytes: st.eof = True # end of file else: st.yylimit += len(bytes); st.yyinput += bytes st.yyinput += b'\0' # append sentinel return Status.OK def lex(st, count): vers = [] while True: st.token = st.yycursor %{ re2c:api = record; re2c:yyrecord = st; re2c:YYFILL = "fill(st) == Status.OK"; re2c:eof = 0; re2c:indent:top = 2; re2c:tags = 1; num = [0-9]+; num @t1 "." @t2 num @t3 ("." @t4 num)? [\n] { major = int(st.yyinput[st.token:t1]) minor = int(st.yyinput[t2:t3]) patch = int(st.yyinput[t4:st.yycursor - 1]) if t4 != -1 else 0 vers.append(SemVer(major, minor, patch)) break } $ { return vers } * { return None } %} def main(): fname = "input" verstr = b"1.22.333\n" expect = [SemVer(1, 22, 333)] * BUFSIZE # Prepare input file. f = open(fname, "wb") for i in range(BUFSIZE): f.write(verstr) f.close() # Run lexer on the prepared file. st = State(fname) assert lex(st, 0) == expect # Cleanup. os.remove(fname) if __name__ == '__main__': main() (capturing groups) (semantic versions) . # re2py $INPUT -o $OUTPUT from collections import namedtuple SemVer = namedtuple('SemVer', 'major minor patch') NONE = -1 def parse(yyinput): yycursor = 0 %{ re2c:yyfill:enable = 0; re2c:captvars = 1; re2c:indent:top = 1; num = [0-9]+; (num) "." (num) ("." num)? [\x00] { major = int(yyinput[yytl1:yytr1]) minor = int(yyinput[yytl2:yytr2]) patch = 0 if yytl3 == NONE else int(yyinput[yytl3 + 1:yytr3]) return SemVer(major, minor, patch) } * { return None } %} assert parse(b"23.34\0") == SemVer(23, 34, 0) assert parse(b"1.2.99999\0") == SemVer(1, 2, 99999) assert parse(b"1.a\0") == None m-tags . trie . # re2py $INPUT -o $OUTPUT NONE = -1 def parse(yyinput): yycursor = 0 %{mtags format = '\n @@ = []'; %} # autogenerated tag variables %{ re2c:YYMTAGP = "@@.append(yycursor)"; re2c:YYMTAGN = ""; // do nothing re2c:yyfill:enable = 0; re2c:tags = 1; re2c:indent:top = 1; num = [0-9]+; @t1 num @t2 ("." #t3 num #t4)* [\x00] { vers = [int(yyinput[t1:t2])] for i in range(len(t3)): vers.append(int(yyinput[t3[i]:t4[i]])) return vers } * { return None } %} assert parse(b"1\0") == [1] assert parse(b"1.2.3.4.5.6.7\0") == [1, 2, 3, 4, 5, 6, 7] assert parse(b"1.2.\0") == None (code points) (code units) . . . . . . "any" [^] ( * ). : yych YYCTYPE . . re2py . : . : o ASCII ( ). [0-255] - . o EBCDIC ( --ebcdic re2c:encoding:ebcdic). [0-255] - . o UCS2 ( --ucs2 re2c:encoding:ucs2). [0-0xFFFF] - . o UTF8 ( --utf8 re2c:encoding:utf8). Unicode . . 1 -- 4 . o UTF16 ( --utf16 re2c:encoding:utf16). Unicode . . 1 -- 2 . o UTF32 ( --utf32 re2c:encoding:utf32). Unicode [0-0x10FFFF] - . include/unicode_blocks.re include/unicode_categories.re include/unicode_properties.re re2py Unicode . --input-encoding Unicode . --input-encoding utf8 re2py UTF8 ( --utf8 ). --encoding-policy (Unicode surrogates [0xD800-0xDFFF]) re2py . Unicode UTF8 . # re2py $INPUT -o $OUTPUT --utf8 %{include "unicode_categories.re" %} def lex(yyinput): yycursor = 0 %{ re2c:yyfill:enable = 0; re2c:indent:top = 1; // Simplified "Unicode Identifier and Pattern Syntax" // (see https://unicode.org/reports/tr31) id_start = L | Nl | [$_]; id_continue = id_start | Mn | Mc | Nd | Pc | [\u200D\u05F3]; identifier = id_start id_continue*; identifier { return True } * { return False } %} assert lex(bytes("_\0", "utf-8")) re2py /*!include:re2c FILE */ %{include FILE %} !include FILE ; FILE . re2py include ( -I ) . / re2py #include C/C++ : FILE / . include . --depfile . re2py include/ . ( Unicode) re2py . . (definitions.py) from enum import Enum class Num(Enum): INT = 1 FLOAT = 2 NAN = 3 %{ number = [1-9][0-9]*; %} (extra_rules.re.inc) // floating-point numbers frac = [0-9]* "." [0-9]+ | [0-9]+ "."; exp = 'e' [+-]? [0-9]+; float = frac exp? | [0-9]+ exp; float { return Num.FLOAT } (Input file) # re2py $INPUT -o $OUTPUT %{include "definitions.py" %} def lex(yyinput): yycursor = 0 %{ re2c:yyfill:enable = 0; re2c:indent:top = 1; * { return Num.NAN } number { return Num.INT } !include "extra_rules.re.inc"; %} assert lex(b"123\0") == Num.INT assert lex(b"123.4567\0") == Num.FLOAT re2py --header re2c:header /*!header:re2c:on*/ /*!header:re2c:off*/ %{header:on%} %{header:off%} .re . . re2py --header re2c:header ( stdout ). re2py (translation units) . (lexer state) (tag variables) ( ). (Input file) # re2py $INPUT -o $OUTPUT --header lexer/state.py from lexer.state import State %{header:on %} class State: def __init__(self, str): self.yyinput = str self.yycursor = 0 %{stags format = "\n self.@@ = 0"; %} %{header:off %} def lex(yyrecord): %{ re2c:api = record; re2c:tags = 1; re2c:yyfill:enable = 0; re2c:indent:top = 1; re2c:header = "lexer/state.py"; [a]* @t [b]* { return t } %} assert lex(State(b"ab\0")) == 1 (Header file) # Generated by re2c class State: def __init__(self, str): self.yyinput = str self.yycursor = 0 self.yyt1 = 0 (Skeleton programs) -S, --skeleton re2py re2py C . . DFA ( ) re2py : .input DFA .keys . .input . : o re2py ( DFA ). o (benchmarking) . o (minimized test cases) ( ). ( ). re2py DFA . : DFA . ( ASCII UTF-8 EBCDIC) DFA . DFA : re2py DFA . . re2py ( ) . (re2py )..SS " " -D, --emit-dot re2py . DFA DOT . Graphviz DFA . DFA . --dump-nfa --dump-dfa-raw ( ). (SEE ALSO) re2c : http://re2c.org. flex(1) lex(1) quex(http://quex.sourceforge.net). (AUTHORS) re2py Peter Bumbulis (peter@csg.uwaterloo.ca) . Marcus Boerger Dan Nuffer . : Brian Young (bayoung@acm.org) Marcus Boerger Dan Nuffer (nuffer@users.sourceforge.net) Ulya Trofimovich (skvadrik@gmail.com) Serghei Iakovlev Sergei Trofimovich Petr Skocik ligfx raekye PolarGoose. RE2PY(1)