RE2GO(1) RE2GO(1) (NAME) re2go - Go (SYNOPSIS) re2go [ OPTIONS ] [ WARNINGS ] INPUT - stdin . (INTRODUCTION) re2go . ( Go ) / . . re2go . Go . : //go:generate re2go $INPUT -o $OUTPUT -i --api simple package main func lex(yyinput string) { var yycursor int /*!re2c re2c:YYCTYPE = byte; re2c:yyfill:enable = 0; [1-9][0-9]* { return } * { panic("error!") } */ } func main() { lex("1234\x00") } re2go : // Code generated by re2go, DO NOT EDIT. //go:generate re2go $INPUT -o $OUTPUT -i --api simple package main func lex(yyinput string) { var yycursor int { var yych byte yych = yyinput[yycursor] switch (yych) { case '1','2','3','4','5','6','7','8','9': goto yy2 default: goto yy1 } yy1: yycursor += 1 { panic("error!") } yy2: yycursor += 1 yych = yyinput[yycursor] switch (yych) { case '0','1','2','3','4','5','6','7','8','9': goto yy2 default: goto yy3 } yy3: { return } } } func main() { lex("1234\x00") } re2go . : name = regular-expression ; name regular-expression . - . regular expressions re2go . name . . . . configuration = value ; re2go . re2go configurations . value - . . . regular-expression code regular-expression ( := ). regular-expression code . . . : * $. ( encoding support ). handling the end of input . start conditions . !action code code ( ). actions . !directive ; . . directives . (Blocks) /*!re2c */ %{ %} ( ). 2.2 . : /*!re2c[:] ... */ or %{[:] ... %} . re2go . . options . /*!local:re2c[:] ... */ or %{local[:] ... %} ( ). /*!rules:re2c[:] ... */ or %{rules[:] ... %} -- . ( reusable blocks). 2.2 re2go -r --reusable . /*!use:re2c[:] ... */ or %{use[:] ... %} . re2go . ( ). use . 2.2 re2go use -r --reusable . /*!max:re2c[:[:...]] ... */ or %{max[:[:...]] ... %} YYMAXFILL . YYMAXFILL ( ). C (#define YYMAXFILL ) Go (var YYMAXFILL int = ) . format @@{max} ( @@) YYMAXFILL . /*!maxnmatch:re2c[:[:...]] ... */ or %{maxnmatch[:[:...]] ... %} YYMAXNMATCH ( -P --posix-captures ). YYMAXNMATCH ( ). C (#define YYMAXNMATCH ) Go (var YYMAXNMATCH int = ) . format @@{max} ( @@) YYMAXNMATCH . /*!stags:re2c[:[:...]] ... */, /*!mtags:re2c[:[:...]] ... */ or %{stags[:[:...]] ... %}, %{mtags[:[:...]] ... %{ s-tag/m-tag re2go . ( ). : format separator. format @@{tag} ( @@) . separator format . /*!svars:re2c[:[:...]] ... */, /*!mvars:re2c[:[:...]] ... */ or %{svars[:[:...]] ... %}, %{mvars[:[:...]] ... %{ s-tag/m-tag ( --tags) re2go ( --captvars --posix-captvars) . ( ). : format separator. format @@{tag} ( @@) . separator format . /*!getstate:re2c[:[:...]] ... */ or %{getstate[:[:...]] ... %} ( --storable-state ). . . . --loop-switch goto . /*!conditions:re2c[:[:...]] ... */, /*!types:re2c... */ or %{conditions[:[:...]] ... %}, %{types... %} ( --conditions ). ( ). YYCONDTYPE . format separator . format @@{cond} ( @@) @@{num} . separator format . /*!include:re2c */ or %{include %} . #include C/C++ . --depfile . /*!header:re2c:on*/ or %{header:on %} . header:off re2go -t --type-header . /*!header:re2c:off*/ or %{header:off %} header:on*/ . /*!ignore:re2c ... */ or %{ignore ... %} . re2go : re2c:api, re2c:input --api. re2c:api:sigil ("sigil") API . @@ . . $ ${name} . API . API re2c:YYFILL@len YYFILL. re2c:api:style API . functions ( C) free-form ( Go Rust). functions API . ( YYFILL) C/C++ 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 . 9 . re2c:cond: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 . re2go 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` ( re2go ). . re2c:state:abort (default case) (abort) -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) . 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) . re2c:YYGETSTATE, re2c:define:YYGETSTATE API YYGETSTATE . re2c:YYGETSTATE:naked, re2c:define:YYGETSTATE:naked re2c:api:style YYGETSTATE . API (free-form) . 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) . 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) . 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 (bitmaps) --bit-vectors () . ( ). re2c:yych, re2c:variable:yych API yych . re2c:yych:emit, re2c:variable:yych:emit yych . . re2c:yych:conversion, re2c:variable:yych:conversion re2go yych YYCTYPE . ( ). re2c:yych:literals, re2c:variable:yych:literals (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 ( ). re2go : "foo" . 'foo' . [a-xyz], [^a-xyz] ( ). . . 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 . --invert-captures (R) . (!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) re2go : !entry code (Entry action) code . start conditions . (lexeme). !pre_rule code (Pre-rule action) code ( start conditions) . ( ) . !post_rule code (Post-rule action) code ( start conditions) . (trap statements) . re2go : !use:name ; use name . . : . * $ . use use . !include file ; include : file . (Program interface) (primitives) API . . . re2go ( ) . API . re2go API : Simple API ( 4.0 ) API --api simple re2c:api = simple . : YYINPUT ( ) YYCURSOR YYMARKER YYCTXMARKER YYLIMIT ( YYINPUT ). Record API ( 4.0 ) Record API (lexer) struct . --api record re2c:api = record . API yyrecord ( re2c:yyrecord ) struct yyinput yycursor yymarker yyctxmarker yylimit ( ). Generic API API backend Go . : YYPEEK YYSKIP YYBACKUP YYBACKUPCTX YYSTAGP YYSTAGN YYMTAGP YYMTAGN YYRESTORE YYRESTORECTX YYRESTORETAG YYSHIFT YYSHIFTSTAG YYSHIFTMTAG YYLESSTHAN YYEND. Generic API : Free-form (Free-form) . re2c:api:style = free-form . @@{var} @@ . . Generic API free-form cursor limit marker ctxmarker ( byte slice) data : /*!re2c re2c:YYPEEK = "data[cursor]"; re2c:YYSKIP = "cursor++"; re2c:YYBACKUP = "marker = cursor"; re2c:YYRESTORE = "cursor = marker"; re2c:YYBACKUPCTX = "ctxmarker = cursor"; re2c:YYRESTORECTX = "cursor = ctxmarker"; re2c:YYRESTORETAG = "cursor = ${tag}"; re2c:YYLESSTHAN = "limit - cursor < @@{len}"; re2c:YYEND = "limit == cursor"; re2c:YYSTAGP = "@@{tag} = cursor"; re2c:YYSTAGN = "@@{tag} = -1"; re2c:YYSHIFT = "cursor += @@{shift}"; re2c:YYSHIFTSTAG = "@@{tag} += @@{shift}"; */ Function-like (Function-like) re2c:api:style = functions . closure . Generic API function-like cursor limit marker ctxmarker ( byte slice) data : YYPEEK := func() byte { return data[cursor] } YYSKIP := func() { cursor++ } YYBACKUP := func() { marker = cursor } YYRESTORE := func() { cursor = marker } YYBACKUPCTX := func() { ctxmarker = cursor } YYRESTORECTX := func() { cursor = ctxmarker } YYLESSTHAN := func(n int) bool { return limit-cursor < n } YYEND := func(n int) bool { return limit == cursor } YYSHIFT := func(n int) { cursor += n } API . YYCTYPE ( code units). 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) . ..TP YYCTXMARKER l-value (trailing context) ( YYINPUT). . YYCTXMARKER / . YYFILL 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 API len. r-value (boolean) len true . YYEND API . r-value ( ) true . YYEND $ . YYLESSTHAN . YYPEEK API . r-value YYCTYPE . YYSKIP API . YYBACKUP API ( YYRESTORE ). YYRESTORE API YYBACKUP . YYBACKUPCTX API ( YYRESTORECTX ). YYRESTORECTX API YYBACKUPCTX . YYRESTORETAG API tag 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 (condition 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 ..TP 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 . (OPTIONS) re2c . re2c . re2c . -? --help -h . --api API . simple . record (struct class ) . generic API . --bit-vectors -b . --nested-ifs . --captures, --leftmost-captures . 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 . --computed-gotos -g "computed goto" ( ). re2go . cgoto:threshold . cgoto:relative . --bit-vectors . --computed-gotos-relative --computed-gotos ( ). --computed-gotos . --conditions --start-conditions -c "conditions" Flex: . re2go goto . --depfile FILE FILE Makefile : [include-file ...]. / include include . --output . --ebcdic --ecb -e EBCDIC . re2go 0 -- 0xFF . --empty-class re2go . match-empty () ( ). match-none . error . --encoding-policy re2go (Unicode surrogates). fail re2go . substitute re2go 0xFFFD . ignore () re2go . . --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 . re2go FILE include -I . --input --api. default simple ( ). custom generic . --input-encoding re2go. ascii () re2go ASCII : . utf8 re2go 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 ..TP --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 builds) --no-unsafe unsafe . --output -o OUTPUT OUTPUT. --posix-captures, -P POSIX. yybmatch 2 * YYMAXNMATCH yynmatch . --posix-captvars POSIX. yytl yytr k . --recursive-functions (co-recursive) DFA . --reusable -r 2.2 ( ). --skeleton -S ("skeleton") . "skeleton" . . C . --storable-state -f . push . YYGETSTATE YYSETSTATE yych yyaccept state . --syntax FILE FILE . FILE ( ) ( --lang none ). --tags -T . --ucs2 --wide-chars -w UCS2 . re2go 0 -- 0xFFFF . --nested-ifs . --utf8 --utf-8 -8 UTF-8 . re2go 0 -- 0x10FFFF . --utf16 --utf-16 -x UTF16 . re2go 0 -- 0x10FFFF . --nested-ifs . --utf32 --unicode -u UTF32 . re2go 0 -- 0x10FFFF . --nested-ifs . --verbose . --vernum -V MMmmpp ( ). --version -v . --single-pass -1 . ( ). --debug-output -d YYDEBUG . . --dump-adfa : DFA (tunneling) ( .dot). --dump-cfg : ( .dot). --dump-closure-stats : (closure). --dump-dfa-det : DFA (determinization) ( .dot). --dump-dfa-min : DFA ( .dot). --dump-dfa-tagopt : DFA ( .dot). --dump-dfa-tree : DFA ( .dot). --dump-dfa-raw : DFA ( .dot). --dump-interf : . --dump-nfa : NFA ( .dot). --emit-dot -D .dot . Graphviz ( dot -Tpng -odfa.png dfa.dot). --dfa-minimization : DFA re2go. moore (Moore) ( ). table " " . DFA . --eager-skip : (eager) -- . (lazy) . --no-lookahead . TDFA(0) . TDFA(1) TDFA(0) . . TDFA(0) TDFA(1) . --no-optimize-tags : ( ). --posix-closure : POSIX: gor1 () - (Goldberg-Radzik) gtop " " . --posix-prectable : POSIX. complex TNFA . naive TNFA complex ..TP --stadfa . staDFA TDFA . staDFA TDFA . --fixed-tags : (fixed-tag) (all) (none) (toplevel) . all . "" ( "base" ) . . (no-match) ( no-match ). . (Warnings) . -W . -Werror . . -W warning. -Wno- warning. -Werror- warning ( -W ). -Wno-error- warning . . -Wcondition-order . --header conditions . -Wempty-character-class . : . re2go . --empty-class . -Wmatch-empty-string ( ). (hang ). -Wswapped-range . . -Wundefined-control-flow ( ). . * . -Wunreachable-rules (shadowed) . -Wdeprecated-eof_rule $ . $ . -Wuseless-escape (escape) . re2go . -Wnondeterministic-tags n- n . -Wsentinel-in-midrule --- (crash) . . re2c:sentinel . -Wundefined-syntax-config --syntax . : . . re2c (syntax files) . -- . re2c re2c: : . ( include/syntax ) re2go . ( --lang none ). : (inplace) . (Feature lists) re2go : supported_apis API simple record generic. supported_api_styles API functions free-form. supported_code_models goto-label loop-switch recursive-functions. supported_targets (codegen) code dot skeleton. supported_features nested-ifs bitmaps computed-gotos case-ranges monadic unsafe tags captures captvars. (Language configurations) re2go : semicolons (semicolon) . backtick_quoted_strings (backtick) . single_quoted_strings (single quote) . indentation_sensitive (indentation) . wrap_blocks_in_braces . (Inplace configurations) re2c: . . (Code templates) . - : 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 ']' . . : re2go . ( -1 ). . ( ) ( ) . re2go . 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 } re2go . . . 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 . code:array_local () . : type ( ) name ( ) size ( ) row ( ) elem ( -- row ). : .const ( ). code:array_global code:array_local . code:array_elem (). : array ( ) index ( ). code:enum enumeration ( ). type ( ) elem ( ) init ( ). : .init ( ). code:enum_elem ( code:enum). name ( ) type ( ). code:assign . lhs ( ) rhs ( ). code:type_int (signed integer). code:type_uint (unsigned integer). code:type_yybm yybm. code:type_yytarget yytarget. code:type_yyctable yyctable. code:cmp_eq <<>> ("equals"). code:cmp_ne <<>> ("not equals"). code:cmp_lt << >> ("less than"). code:cmp_gt << >> ("greater than"). code:cmp_le << >> ("less or equal"). code:cmp_ge << >> ("greater or equal"). 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 (computed 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 ( re2go ) 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 . indent . dedent . topindent . . : .api.simple (True) API (--api simple re2c:api = simple). .api.generic (True) API (--api generic re2c:api = generic). .api.record (True) API (--api record re2c:api = record). .api_style.functions (True) API (re2c:api-style = functions). .api_style.freeform (True) API (re2c:api-style = free-form). .case_ranges (True) (--case-ranges re2c:case-ranges = 1). .cgoto.relative (True) goto (--computed-gotos-relative re2c:cgoto:relative = 1). .code_model.goto_label (True) goto/label (--goto-label). .code_model.loop_switch (True) loop/switch (--loop-switch). .code_model.recursive_functions (True) (--recursive-function). .date (True) . .loop_label (True) re2go (re2c:label:yyloop ). .monadic (True) (re2c:monadic = 1). . .start_conditions (True) (--start-conditions). .storable_state (True) (--storable-state). .unsafe (True) re2go "unsafe" (--unsafe re2c:unsafe = 1). "unsafe" . .version (True) re2go . .yyfill.enable (True) YYFILL (re2c:yyfill:enable = 1). .yyfn.throw (True) re2c:yyfn:throw . (lexer) . : o ( *) o o <<>> : . : . . re2go . . : o (Sentinel). . <<>> (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 . . ( ) . YYMAXFILL max . YYFILL . YYFILL re2c:eof -1 ( ). o (Custom checks). Generic API (override) . . Generic API --api custom re2c:api = custom; re2c:yyfill:enable = 0; re2c:yyfill:check = 0; . ..SS (Sentinel) (sentinel) . null . null : (lexeme) ( ). null * (crash) ( -Wsentinel-in-midrule re2c:sentinel ). re2c:yyfill:enable = 0; YYFILL . //go:generate re2go $INPUT -o $OUTPUT --api simple package main // Expect a null-terminated string. func lex(yyinput string) int { yycursor := 0 count := 0 for { /*!re2c re2c:yyfill:enable = 0; re2c:YYCTYPE = byte; * { return -1 } [\x00] { return count } [a-z]+ { count += 1; continue } [ ]+ { continue } */ } } func main() { assert_eq := func(x, y int) { if x != y { panic("error") } } assert_eq(lex("\000"), 0) assert_eq(lex("one two three\000"), 3) assert_eq(lex("f0ur\000"), -1) } (Sentinel with bounds checks) ( 1.2 ). (single-quoted) . null re2c:eof = 0; . sentinel null ( 'aaa\0aa'\0 'aaa\0 ). null . generic API YYLIMIT <= YYCURSOR YYLESSTHAN(1) . (YYFILL re2c:yyfill:enable = 0; YYFILL YYFILL with sentinel ). : $ ( *) -Wundefined-control-flow . //go:generate re2go $INPUT -o $OUTPUT --api simple package main // Expects a null-terminated string. func lex(yyinput string) int { var yycursor, yymarker int yylimit := len(yyinput) - 1 // lim points at the terminating null count := 0 for { /*!re2c re2c:YYCTYPE = byte; re2c:yyfill:enable = 0; re2c:eof = 0; str = ['] ([^'\\] | [\\][^])* [']; * { return -1 } $ { return count } str { count += 1; continue } [ ]+ { continue } */ } } func main() { assert_eq := func(x, y int) { if x != y { panic("error") } } assert_eq(lex("\000"), 0) assert_eq(lex("'qu\000tes' 'are' 'fine: \\'' \000"), 3) assert_eq(lex("'unterminated\\'\000"), -1) } ( ). . YYMAXFILL null YYMAXFILL max . null --- (lexeme) ( ). "stop" (null) (lexer) ( null ). . (generic API) (YYLIMIT - YYCURSOR) < n YYLESSTHAN(n) n ( n ). YYFILL(n) n . YYFILL ( ). YYFILL YYFILL with padding . //go:generate re2go $INPUT -o $OUTPUT --api simple package main import "strings" /*!max:re2c*/ // Expects YYMAXFILL-padded string. func lex(str string) int { // Pad string with YYMAXFILL zeroes at the end. yyinput := str + strings.Repeat("\000", int(YYMAXFILL)) yycursor := 0 yylimit := len(yyinput) count := 0 for { /*!re2c re2c:YYCTYPE = byte; re2c:YYFILL = "return -1"; str = ['] ([^'\\] | [\\][^])* [']; [\x00] { // Check that it is the sentinel, not some unexpected null. if yycursor - 1 == len(str) { return count } else { return -1 } } str { count += 1; continue } [ ]+ { continue } * { return -1 } */ } } func main() { assert_eq := func(x, y int) { if x != y { panic("error") } } assert_eq(lex(""), 0) assert_eq(lex("'qu\000tes' 'are' 'fine: \\'' "), 3) assert_eq(lex("'unterminated\\'"), -1) assert_eq(lex("'unexpected \000 null\\'"), -1) } (Custom checks) API . . sentinel null . (sentinel) YYPEEK . . YYPEEK . //go:generate re2go $INPUT -o $OUTPUT package main // Returns "fake" terminating null if cursor has reached limit. func peek(str string, cur int) byte { if cur >= len(str) { return 0 // fake null } else { return str[cur] } } // Expects a string without terminating null. func lex(str string) int { var cur int count := 0 for { /*!re2c re2c:yyfill:enable = 0; re2c:YYCTYPE = byte; re2c:YYPEEK = "peek(str, cur)"; re2c:YYSKIP = "cur += 1"; * { return -1 } [\x00] { return count } [a-z]+ { count += 1; continue } [ ]+ { continue } */ } } func main() { assert_eq := func(x, y int) { if x != y { panic("error") } } assert_eq(lex(""), 0) assert_eq(lex("one two three"), 3) assert_eq(lex("f0ur"), -1) } : ( ) . . . (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 re2go ) o (context marker): (YYCTXMARKER API C YYBACKUPCTX/YYRESTORECTX API ) o (tag variables): ( stags mtags API YYSTAGP/YYSTAGN/YYMTAGP/YYMTAGN) YYFILL . token cursor token token limit . YYFILL ( ). YYFILL (EOF) : EOF (bounds-checking) (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 . //go:generate re2go $INPUT -o $OUTPUT package main import ( "os" "strings" ) const BUFSIZE uint = 4096 type Input struct { file *os.File yyinput []byte yycursor uint yymarker uint yylimit uint token uint eof bool } func fill(in *Input) int { if in.eof { return -1 } // unexpected EOF // Error: lexeme too long. In real life can reallocate a larger buffer. if in.token < 1 { return -2 } // Shift buffer contents (discard everything up to the current token). copy(in.yyinput[0:], in.yyinput[in.token:in.yylimit]) in.yycursor -= in.token in.yymarker -= in.token in.yylimit -= in.token in.token = 0 // Fill free space at the end of buffer with new data from file. n, _ := in.file.Read(in.yyinput[in.yylimit:BUFSIZE]) in.yylimit += uint(n) in.yyinput[in.yylimit] = 0 // If read less than expected, this is the end of input. in.eof = in.yylimit < BUFSIZE return 0 } func lex(yyrecord *Input) int { count := 0 for { yyrecord.token = yyrecord.yycursor /*!re2c re2c:api = record; re2c:eof = 0; re2c:YYCTYPE = byte; re2c:YYFILL = "fill(yyrecord) == 0"; str = ['] ([^'\\] | [\\][^])* [']; * { return -1 } $ { return count } str { count += 1; continue } [ ]+ { continue } */ } } func main() () { fname := "input" content := "'qu\000tes' 'are' 'fine: \\'' "; // Prepare input file: a few times the size of the buffer, containing // strings with zeroes and escaped quotes. f, _ := os.Create(fname) f.WriteString(strings.Repeat(content, int(BUFSIZE))) f.Seek(0, 0) count := 3 * int(BUFSIZE) // number of quoted strings written to file // Prepare lexer state: all offsets are at the end of buffer. in := &Input{ file: f, // Sentinel at `yylimit` offset is set to zero, which triggers YYFILL. yyinput: make([]byte, BUFSIZE+1), yycursor: BUFSIZE, yymarker: BUFSIZE, yylimit: BUFSIZE, token: BUFSIZE, eof: false, } // Run the lexer. if lex(in) != count { panic("error"); } // Cleanup: remove input file. f.Close(); os.Remove(fname); } ( EOF ) YYFILL . YYFILL (YYLIMIT - YYCURSOR) < n API C YYLESSTHAN(n) API (generic API) . YYFILL . YYFILL lexer ( ). 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 (bounds-checking) . //go:generate re2go $INPUT -o $OUTPUT package main import ( "os" "strings" ) /*!max:re2c*/ const BUFSIZE uint = 4096 type Input struct { file *os.File yyinput []byte yycursor uint yylimit uint token uint eof bool } func fill(in *Input, need uint) int { if in.eof { return -1 } // unexpected EOF // Error: lexeme too long. In real life can reallocate a larger buffer. if in.token < need { return -2 } // Shift buffer contents (discard everything up to the current token). copy(in.yyinput[0:], in.yyinput[in.token:in.yylimit]) in.yycursor -= in.token in.yylimit -= in.token in.token = 0 // Fill free space at the end of buffer with new data from file. n, _ := in.file.Read(in.yyinput[in.yylimit:BUFSIZE]) in.yylimit += uint(n) // If read less than expected, this is end of input => add zero padding // so that the lexer can access characters at the end of buffer. if in.yylimit < BUFSIZE { in.eof = true for i := uint(0); i < YYMAXFILL; i += 1 { in.yyinput[in.yylimit+i] = 0 } in.yylimit += YYMAXFILL } return 0 } func lex(yyrecord *Input) int { count := 0 for { yyrecord.token = yyrecord.yycursor /*!re2c re2c:api = record; re2c:YYCTYPE = byte; re2c:YYFILL = "if r := fill(yyrecord, @@); r != 0 { return r }"; str = ['] ([^'\\] | [\\][^])* [']; [\x00] { // Check that it is the sentinel, not some unexpected null. if yyrecord.token == yyrecord.yylimit - YYMAXFILL { return count } else { return -1 } } str { count += 1; continue } [ ]+ { continue } * { return -1 } */ } } func main() () { fname := "input" content := "'qu\000tes' 'are' 'fine: \\'' "; // Prepare input file: a few times the size of the buffer, containing // strings with zeroes and escaped quotes. f, _ := os.Create(fname) f.WriteString(strings.Repeat(content, int(BUFSIZE))) f.Seek(0, 0) count := 3 * int(BUFSIZE) // number of quoted strings written to file // Prepare lexer state: all offsets are at the end of buffer. // This immediately triggers YYFILL, as the YYLESSTHAN condition is true. in := &Input{ file: f, yyinput: make([]byte, BUFSIZE+YYMAXFILL), yycursor: BUFSIZE, yylimit: BUFSIZE, token: BUFSIZE, eof: false, } // Run the lexer. if lex(in) != count { panic("error"); } // Cleanup: remove input file. f.Close(); os.Remove(fname); } (Multiple blocks) (lexer) ( ). re2go . start conditions . . goto ( C/C++ Go) . goto . goto ( Rust) switch /switch yystate re2go . (binary) (octal) (decimal) (hexadecimal) . . . . //go:generate re2go $INPUT -o $OUTPUT -i --api simple package main import "errors" const u32Limit uint64 = 1<<32 var ( eSyntax = errors.New("syntax error") eOverflow = errors.New("overflow error") ) func parse_u32(yyinput string) (uint32, error) { var yycursor, yymarker int result := uint64(0) add := func(base uint64, offset byte) { result = result * base + uint64(yyinput[yycursor-1] - offset) if result >= u32Limit { result = u32Limit } } /*!re2c re2c:yyfill:enable = 0; re2c:YYCTYPE = byte; end = "\x00"; '0b' / [01] { goto bin } "0" { goto oct } "" / [1-9] { goto dec } '0x' / [0-9a-fA-F] { goto hex } * { goto err } */ bin: /*!re2c end { goto end } [01] { add(2, '0'); goto bin } * { goto err } */ oct: /*!re2c end { goto end } [0-7] { add(8, '0'); goto oct } * { goto err } */ dec: /*!re2c end { goto end } [0-9] { add(10, '0'); goto dec } * { goto err } */ hex: /*!re2c end { goto end } [0-9] { add(16, '0'); goto hex } [a-f] { add(16, 'a'-10); goto hex } [A-F] { add(16, 'A'-10); goto hex } * { goto err } */ end: if result < u32Limit { return uint32(result), nil } else { return 0, eOverflow } err: return 0, eSyntax } func main() { test := func(num uint32, str string, err error) { if n, e := parse_u32(str); !(n == num && e == err) { panic("error") } } test(1234567890, "1234567890\000", nil) test(13, "0b1101\000", nil) test(0x7fe, "0x007Fe\000", nil) test(0644, "0644\000", nil) test(0, "9999999999\000", eOverflow) test(0, "123??\000", eSyntax) } --start-conditions . (automata) re2go . re2go . : 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 . re2go re2go goto/label loop/switch . goto ( C/C++ Go) yyc_ cond ( re2c:cond:prefix ). goto . re2go switch YYGETSTATE . :=> switch ( re2c:cond:goto ). . goto ( Rust) re2go yystate . switch switch DFA . switch . ( YYGETCOND/YYSETCOND ). . . cond conditions . //go:generate re2go -c $INPUT -o $OUTPUT -i --api simple package main import "errors" var ( eSyntax = errors.New("syntax error") eOverflow = errors.New("overflow error") ) /*!conditions:re2c*/ const u32Limit uint64 = 1<<32 func parse_u32(yyinput string) (uint32, error) { var yycursor, yymarker int result := uint64(0) yycond := yycinit add := func(base uint64, offset byte) { result = result * base + uint64(yyinput[yycursor-1] - offset) if result >= u32Limit { result = u32Limit } } /*!re2c re2c:yyfill:enable = 0; re2c:YYCTYPE = byte; re2c:YYGETCOND = "cond"; re2c:YYSETCOND = "cond = @@"; <*> * { return 0, eSyntax } '0b' / [01] :=> bin "0" :=> oct "" / [1-9] :=> dec '0x' / [0-9a-fA-F] :=> hex "\x00" { if result < u32Limit { return uint32(result), nil } else { return 0, eOverflow } } [01] { add(2, '0'); goto yyc_bin } [0-7] { add(8, '0'); goto yyc_oct } [0-9] { add(10, '0'); goto yyc_dec } [0-9] { add(16, '0'); goto yyc_hex } [a-f] { add(16, 'a'-10); goto yyc_hex } [A-F] { add(16, 'A'-10); goto yyc_hex } */ } func main() { test := func(num uint32, str string, err error) { if n, e := parse_u32(str); !(n == num && e == err) { panic("error") } } test(1234567890, "1234567890\000", nil) test(13, "0b1101\000", nil) test(0x7fe, "0x007Fe\000", nil) test(0644, "0644\000", nil) test(0, "9999999999\000", eOverflow) test(0, "123??\000", eSyntax) } --storable-state re2go . re2go <<>> ("pull") <<>> ("pulls"). ( (parser) ). : <<>> ("push") . . . <<>> ("pull") : 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) . ( ). <<>> ("push") . . <<>> ("sending") . . //go:generate re2go -f $INPUT -o $OUTPUT package main import ( "fmt" "os" ) // Use a small buffer to cover the case when a lexeme doesn't fit. // In real world use a larger buffer. const BUFSIZE int = 10 type State struct { file *os.File yyinput []byte yycursor int yymarker int yylimit int token int yystate int } const ( lexEnd = iota lexReady lexWaitingForInput lexPacketBroken lexPacketTooBig ) func fill(st *State) int { shift := st.token used := st.yylimit - st.token free := BUFSIZE - used // Error: no space. In real life can reallocate a larger buffer. if free < 1 { return lexPacketTooBig } // Shift buffer contents (discard already processed data). copy(st.yyinput[0:], st.yyinput[shift:shift+used]) st.yycursor -= shift st.yymarker -= shift st.yylimit -= shift st.token -= shift // Fill free space at the end of buffer with new data. n, _ := st.file.Read(st.yyinput[st.yylimit:BUFSIZE]) st.yylimit += n st.yyinput[st.yylimit] = 0 // append sentinel symbol return lexReady } func lex(yyrecord *State, recv *int) int { var yych byte /*!getstate:re2c*/ loop: yyrecord.token = yyrecord.yycursor /*!re2c re2c:api = record; re2c:eof = 0; re2c:YYFILL = "return lexWaitingForInput"; packet = [a-z]+[;]; * { return lexPacketBroken } $ { return lexEnd } packet { *recv = *recv + 1; goto loop } */ } func test(expect int, packets []string) { // Create a pipe (open the same file for reading and writing). fname := "pipe" fw, _ := os.Create(fname) fr, _ := os.Open(fname) // Initialize lexer state: `state` value is -1, all offsets are at the end // of buffer. st := &State{ file: fr, // Sentinel at `yylimit` offset is set to zero, which triggers YYFILL. yyinput: make([]byte, BUFSIZE+1), yycursor: BUFSIZE, yymarker: BUFSIZE, yylimit: BUFSIZE, token: BUFSIZE, yystate: -1, } // 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. var status int send := 0 recv := 0 for { status = lex(st, &recv) if status == lexEnd { break } else if status == lexWaitingForInput { if send < len(packets) { fw.WriteString(packets[send]) send += 1 } status = fill(st) if status != lexReady { break } } else if status == lexPacketBroken { break } } // Check results. if status != expect || (status == lexEnd && recv != send) { panic(fmt.Sprintf("got %d, want %d", status, expect)) } // Cleanup: remove input file. fr.Close() fw.Close() os.Remove(fname) } func main() { test(lexEnd, []string{}) test(lexEnd, []string{"zero;", "one;", "two;", "three;", "four;"}) test(lexPacketBroken, []string{"??;"}) test(lexPacketTooBig, []string{"looooooooooooong;"}) } /*!rules:re2c[:] ... */ %{rules[:] ... %} re2go . . rules use . rules . use /*!use:re2c[:] ... */ %{use[:] ... %} . : rules ( ). use . use (lexer) : rules ( ). use !use:; re2go . . : . * $ ( /condition) . use re2go ( ). use 2.2 re2go . ( ). 2.2 -r --reusable . 1.2 . !use //go:generate re2go $INPUT -o $OUTPUT --api simple package main // 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. const ( Color = iota Fish Dunno ) /*!rules:re2c:colors * { panic("eh!") } "red" | "salmon" | "magenta" { return Color } */ /*!rules:re2c:fish * { panic("oh!") } "haddock" | "salmon" | "eel" { return Fish } */ func lex(yyinput string) int { var yycursor, yymarker int /*!re2c re2c:yyfill:enable = 0; re2c:YYCTYPE = byte; !use:fish; !use:colors; * { return Dunno } // overrides inherited '*' rules */ } func main() { assert_eq := func(x, y int) { if x != y { panic("error") } } assert_eq(lex("salmon"), Fish); assert_eq(lex("what?"), Dunno); } //go:generate re2go $INPUT -o $OUTPUT --input-encoding utf8 --api simple package main // 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 re2c:yyfill:enable = 0; "x y" { return 0; } * { return 1; } */ func lexUTF8(yyinput []uint8) int { var yycursor, yymarker int /*!use:re2c re2c:encoding:utf8 = 1; re2c:YYCTYPE = uint8; */ } func lexUTF32(yyinput []uint32) int { var yycursor, yymarker int /*!use:re2c re2c:encoding:utf32 = 1; re2c:YYCTYPE = uint32; */ } func main() { assert_eq := func(x, y int) { if x != y { panic("error") } } assert_eq(lexUTF8([]uint8{0xe2, 0x88, 0x80, 0x78, 0x20, 0xe2, 0x88, 0x83, 0x79}), 0) assert_eq(lexUTF32([]uint32{0x2200, 0x78, 0x20, 0x2203, 0x79}), 0) } (Submatch extraction) re2go . Tags tag @stag #mtag stag mtag . -T --tags re2c:tags = 1 . : ( ) . S-tag m-tag ( ). svars mvars. (leftmost greedy) . Captures (capturing groups) . --captures re2c:captures = 1 . : --leftmost-captures () --posix-captures POSIX. : (! ... ). --invert-captures re2c:invert-captures = 1 . yynmatch ( ) yypmatch . yynmatch yypmatch yypmatch [yynmatch * 2] . maxnmatch YYMAXNMATCH yynmatch . Captvars --captvars re2c:captvars = 1 . --captures : yynmatch yypmatch re2go yytl yytr k- ( svars ). : --leftmost-captvars re2c:leftmost-captvars = 1 ( ) --posix-captvars re2c:posix-captvars POSIX. (Tagged Deterministic Finite Automata with Lookahead) TDFA . TDFA (overhead) . TDFA DFA . . TDFA (tag variables) : . . stags mtags . . YYFILL . S-tag : o s-tag: t = YYCURSOR API C YYSTAGP(t) API o s-tag: t = NULL API C YYSTAGN(t) API o s-tag : t1 = t2 M-tag : 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 . //go:generate re2go $INPUT -o $OUTPUT --api simple package main import "reflect" type SemVer struct { major, minor, patch int } func s2n(s string) int { // convert pre-parsed string to a number n := 0 for _, c := range s { n = n*10 + int(c-'0') } return n } func parse(yyinput string) *SemVer { var yycursor, yymarker int // Final tag variables available in semantic action. /*!svars:re2c format = 'var @@ int;'; */ // Intermediate tag variables used by the lexer (must be autogenerated). /*!stags:re2c format = 'var @@ int;'; */ /*!re2c re2c:yyfill:enable = 0; re2c:YYCTYPE = byte; re2c:tags = 1; num = [0-9]+; @t1 num @t2 "." @t3 num @t4 ("." @t5 num)? [\x00] { major := s2n(yyinput[t1:t2]) minor := s2n(yyinput[t3:t4]) patch := 0 if t5 != -1 { patch = s2n(yyinput[t5:yycursor-1]) } return &SemVer{major, minor, patch} } * { return nil } */ } func main() { assert_eq := func(x, y *SemVer) { if !reflect.DeepEqual(x, y) { panic("error") } } assert_eq(parse("23.34\000"), &SemVer{23, 34, 0}) assert_eq(parse("1.2.9999\000"), &SemVer{1, 2, 9999}) assert_eq(parse("1.a\000"), nil) } s-tag YYFILL . (lexer) YYFILL . s-tag . m-tag . //go:generate re2go $INPUT -o $OUTPUT --tags package main import ( "os" "reflect" "strings" ) const BUFSIZE int = 4095 type Input struct { file *os.File yyinput []byte yycursor int yymarker int yylimit int token int // Intermediate tag variables must be part of the lexer state passed to YYFILL. // They don't correspond to tags and should be autogenerated by re2c. /*!stags:re2c format = "\t@@ int\n"; */ eof bool } type SemVer struct { major, minor, patch int } func s2n(s []byte) int { // convert pre-parsed string to a number n := 0 for _, c := range s { n = n*10 + int(c-'0') } return n } func fill(in *Input) int { if in.eof { return -1 } // unexpected EOF // Error: lexeme too long. In real life can reallocate a larger buffer. if in.token < 1 { return -2 } // Shift buffer contents (discard everything up to the current token). copy(in.yyinput[0:], in.yyinput[in.token:in.yylimit]) in.yycursor -= in.token in.yymarker -= in.token in.yylimit -= in.token // Tag variables need to be shifted like other input positions. The check // for -1 is only needed if some tags are nested inside of alternative or // repetition, so that they can have -1 value. /*!stags:re2c format = "\tif in.@@ != -1 { in.@@ -= in.token }\n"; */ in.token = 0 // Fill free space at the end of buffer with new data from file. n, _ := in.file.Read(in.yyinput[in.yylimit:BUFSIZE]) in.yylimit += n in.yyinput[in.yylimit] = 0 // If read less than expected, this is the end of input. in.eof = in.yylimit < BUFSIZE return 0 } func parse(in *Input) []SemVer { // Final tag variables available in semantic action. /*!svars:re2c format = "var @@ int;"; */ vers := make([]SemVer, 0) for { in.token = in.yycursor /*!re2c re2c:api = record; re2c:eof = 0; re2c:yyrecord = in; re2c:YYCTYPE = byte; re2c:YYFILL = "fill(in) == 0"; num = [0-9]+; num @t1 "." @t2 num @t3 ("." @t4 num)? [\n] { major := s2n(in.yyinput[in.token:t1]) minor := s2n(in.yyinput[t2:t3]) patch := 0 if t4 != -1 { patch = s2n(in.yyinput[t4:in.yycursor-1]) } vers = append(vers, SemVer{major, minor, patch}) continue } $ { return vers } * { return nil } */ } } func main() () { fname := "input" content := "1.22.333\n"; expect := make([]SemVer, 0, BUFSIZE) for i := 0; i < BUFSIZE; i += 1 { expect = append(expect, SemVer{1, 22, 333}) } // Prepare input file (make sure it exceeds buffer size). f, _ := os.Create(fname) f.WriteString(strings.Repeat(content, BUFSIZE)) f.Seek(0, 0) // Initialize lexer state: all offsets are at the end of buffer. in := &Input{ file: f, // Sentinel at `yylimit` offset is set to zero, which triggers YYFILL. yyinput: make([]byte, BUFSIZE+1), yycursor: BUFSIZE, yymarker: BUFSIZE, yylimit: BUFSIZE, token: BUFSIZE, eof: false, } // Run the lexer and check results. if !reflect.DeepEqual(parse(in), expect) { panic("error"); } // Cleanup: remove input file. f.Close(); os.Remove(fname); } (capturing groups) . //go:generate re2go $INPUT -o $OUTPUT --api simple package main import "reflect" type SemVer struct { major, minor, patch int } func s2n(s string) int { // convert pre-parsed string to a number n := 0 for _, c := range s { n = n*10 + int(c-'0') } return n } func parse(yyinput string) *SemVer { var yycursor, yymarker int // Final tag variables used in semantic action. /*!svars:re2c format = 'var @@ int;'; */ // Intermediate tag variables used by the lexer (must be autogenerated). /*!stags:re2c format = 'var @@ int;'; */ /*!re2c re2c:yyfill:enable = 0; re2c:YYCTYPE = byte; re2c:captvars = 1; num = [0-9]+; (num) "." (num) ("." num)? [\x00] { _ = yytl0; _ = yytr0; // some variables are unused major := s2n(yyinput[yytl1:yytr1]) minor := s2n(yyinput[yytl2:yytr2]) patch := 0 if yytl3 != -1 { patch = s2n(yyinput[yytl3+1:yytr3]) } return &SemVer{major, minor, patch} } * { return nil } */ } func main() { assert_eq := func(x, y *SemVer) { if !reflect.DeepEqual(x, y) { panic("error") } } assert_eq(parse("23.34\000"), &SemVer{23, 34, 0}) assert_eq(parse("1.2.9999\000"), &SemVer{1, 2, 9999}) assert_eq(parse("1.a\000"), nil) } m-tag . trie . //go:generate re2go $INPUT -o $OUTPUT --api simple package main import "reflect" const ( mtagRoot int = -1 tagNone int = -1 ) // An m-tag tree is a way to store histories with an O(1) copy operation. // Histories naturally form a tree, as they have common start and fork at some // point. The tree is stored as an array of pairs (tag value, link to parent). // An m-tag is represented with a single link in the tree (array index). type mtagElem struct { elem int pred int } type mtagTrie = []mtagElem type Ver = []int // unbounded number of version components func s2n(s string) int { // convert pre-parsed string to a number n := 0 for _, c := range s { n = n*10 + int(c-'0') } return n } // Append a single value to an m-tag history. func add_mtag(trie *mtagTrie, mtag int, value int) int { *trie = append(*trie, mtagElem{value, mtag}) return len(*trie) - 1 } // Recursively unwind tag histories and collect version components. func unwind(trie mtagTrie, x int, y int, str string) Ver { // Reached the root of the m-tag tree, stop recursion. if x == mtagRoot && y == mtagRoot { return []int{} } // Unwind history further. ver := unwind(trie, trie[x].pred, trie[y].pred, str) // Get tag values. Tag histories must have equal length. if x == mtagRoot || y == mtagRoot { panic("tag histories have different length") } ex := trie[x].elem ey := trie[y].elem if ex != tagNone && ey != tagNone { // Both tags are valid string indices, extract component. ver = append(ver, s2n(str[ex:ey])) } else if !(ex == tagNone && ey == tagNone) { panic("both tags should be tagNone") } return ver } func parse(yyinput string) []int { var yycursor, yymarker int trie := make([]mtagElem, 0) // Final tag variables available in semantic action. /*!svars:re2c format = 'var @@ int;'; */ /*!mvars:re2c format = "var @@ int;"; */ // Intermediate tag variables used by the lexer (must be autogenerated). /*!stags:re2c format = 'var @@ int'; separator = "\n\t"; */ /*!mtags:re2c format = "\t@@ := mtagRoot\n"; */ /*!re2c re2c:tags = 1; re2c:yyfill:enable = 0; re2c:YYCTYPE = byte; re2c:YYMTAGP = "@@ = add_mtag(&trie, @@, yycursor)"; re2c:YYMTAGN = "@@ = add_mtag(&trie, @@, tagNone)"; num = [0-9]+; @t1 num @t2 ("." #t3 num #t4)* [\x00] { ver := make([]int, 0) ver = append(ver, s2n(yyinput[t1:t2])) ver = append(ver, unwind(trie, t3, t4, yyinput)...) return ver } * { return nil } */ } func main() { assert_eq := func(x, y []int) { if !reflect.DeepEqual(x, y) { panic("error") } } assert_eq(parse("1\000"), []int{1}) assert_eq(parse("1.2.3.4.5.6.7\000"), []int{1, 2, 3, 4, 5, 6, 7}) assert_eq(parse("1.\000"), nil) } (code points) (code units) . . . . (fixed-length) . (variable-length) . << >> "any" [^] ( * ). : yych YYCTYPE . . re2go . : . : 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). . . . o UTF16 ( --utf16 re2c:encoding:utf16). . . . o UTF32 ( --utf32 re2c:encoding:utf32). [0-0x10FFFF] . include/unicode_blocks.re include/unicode_categories.re include/unicode_properties.re re2go (Unicode) . --input-encoding . --input-encoding utf8 re2go UTF8 ( --utf8 ). --encoding-policy ( [0xD800-0xDFFF]) re2go . UTF8 . //go:generate re2go $INPUT -o $OUTPUT -8si --api simple package main /*!include:re2c "unicode_categories.re" */ func lex(yyinput string) int { var yycursor, yymarker int /*!re2c re2c:yyfill:enable = 0; re2c:YYCTYPE = byte; // Simplified "Unicode Identifier and Pattern Syntax" // (see https://unicode.org/reports/tr31) id_start = L | Nl | [$_]; id_continue = id_start | Mn | Mc | Nd | Pc | [\u200D\u05F3]; identifier = id_start id_continue*; identifier { return 0 } * { return 1 } */ } func main() { if lex("_\000") != 0 { panic("error") } } re2go /*!include:re2c FILE */ %{include FILE %} !include FILE ; FILE . re2go include include -I . / include re2go #include C/C++ : FILE / . include include . --depfile include . re2go include include/ . ( Unicode) re2go . include . Include (definitions.go) const ( ResultOk = iota ResultFail ) /*!re2c number = [1-9][0-9]*; */ Include (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 ResultOk } (Input file) //go:generate re2go $INPUT -o $OUTPUT -i --api simple package main /*!include:re2c "definitions.go" */ func lex(yyinput string) int { var yycursor, yymarker int /*!re2c re2c:YYCTYPE = byte; re2c:yyfill:enable = 0; * { return ResultFail } number { return ResultOk } !include "extra_rules.re.inc"; */ } func main() { assert_eq := func(x, y int) { if x != y { panic("error") } } assert_eq(lex("123\000"), ResultOk) assert_eq(lex("123.4567\000"), ResultOk) } re2go --header re2c:header /*!header:re2c:on*/ /*!header:re2c:off*/ %{header:on%} %{header:off%} .re . . re2go --header re2c:header ( stdout ). re2go (translation units) . (lexer state) ( ). (Input file) //go:generate re2go $INPUT -o $OUTPUT -i --header lexer/state.go package main import "./lexer" // the package is generated by re2c /*!header:re2c:on*/ package lexer type State struct { Data string Cur /*!stags:re2c format=", @@"; */ int } /*!header:re2c:off*/ func lex(yyrecord *lexer.State) int { var t int /*!re2c re2c:header = "lexer/state.go"; re2c:api = record; re2c:YYCTYPE = byte; re2c:YYINPUT = "yyrecord.Data"; re2c:YYCURSOR = "yyrecord.Cur"; re2c:yyfill:enable = 0; re2c:tags = 1; re2c:tags:prefix = "Tag"; [a]* @t [b]* { return t } */ } func main() { st := &lexer.State{Data:"ab\x00",} if lex(st) != 1 { panic("error") } } (Header file) // Code generated by re2c, DO NOT EDIT. package lexer type State struct { Data string Cur, Mar, Tag1 int } -S, --skeleton re2go re2go C . . DFA ( ) re2go : .input DFA .keys . .input . : o re2go ( DFA ). o (benchmarking) . o (minimized) ( ). ( ). re2go DFA . : DFA . ( ASCII UTF-8 EBCDIC) DFA . DFA : re2go DFA . . re2go ( ) . (re2go ). (Visualization and debug) -D, --emit-dot re2go 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) re2go 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. RE2GO(1)