RE2D(1)

📦

version 4.6 · 9 versions

4.6 · arch rolling, nixpkgs unstable (latest)
4.1 · opensuse-leap 16.0
4.2 · alpine 3.22
4.3.1 · alpine 3.23
4.5.1 · alpine 3.24, nixos 26.05
4.5.1 · ubuntu stonking
4.3 · ubuntu questing
4.4 · ubuntu resolute
4.1 · debian trixie

✨ NAME ✨

okay bestie, welcome to RE2D(1) — the manual page for re2d, which is here to generate fast lexical analyzers for D. That’s the whole NAME vibe. iconic. 💅

🌸 SYNOPSIS 🌸

re2d [ OPTIONS ] [ WARNINGS ] INPUT

INPUT can be either a file or - for stdin. yes bestie, - really is the stdin shortcut. 🫶

🦋 INTRODUCTION 🦋

okay so re2d works as a preprocessor. It reads the input file (which is usually a program in D, but can be anything) and looks for blocks of code enclosed in special-form start/end markers. The text outside of these blocks is copied verbatim into the output file. The contents of the blocks are processed by re2d. It translates them to code in D and outputs the generated code in place of the block. literally the main character of your build pipeline. ✨

Here is an example of a small program that checks if a given string contains a decimal number:

// re2d $INPUT -o $OUTPUT -i
module main;

private bool lex(const(char)* yycursor) {
    /*!re2c
        re2c:YYCTYPE = char;
        re2c:yyfill:enable = 0;

        [1-9][0-9]* { return true; }
        *           { return false; }
    */
}

void main() {
    assert(lex("1234"));
}

In the output re2d replaced the block in the middle with the generated code:

/* Generated by re2d */
// re2d $INPUT -o $OUTPUT -i
module main;

private bool lex(const(char)* yycursor) {

{
    char yych;
    yych = *yycursor;
    switch (yych) {
        case '1': .. case '9': goto yy2;
        default: goto yy1;
    }
yy1:
    ++yycursor;
    { return false; }
yy2:
    yych = *++yycursor;
    switch (yych) {
        case '0': .. case '9': goto yy2;
        default: goto yy3;
    }
yy3:
    { return true; }
}

}

void main() {
    assert(lex("1234"));
}

🎀 BASICS 🎀

A re2d program consists of a sequence of blocks intermixed with code in the target language. A block may contain definitions, configurations, rules, actions and directives in any order:

name = regular-expression ; — a definition binds name to regular-expression. Names may contain alphanumeric characters and underscore. The regular expressions section gives an overview of re2d syntax for regular expressions. Once defined, the name can be used in other regular expressions and in rules. Recursion in named definitions is not allowed, and each name should be defined before it is used. A block inherits named definitions from the global scope. Redefining a name that exists in the current scope is an error. 🚨
configuration = value ; — a configuration allows one to change re2d behavior and customize the generated code. For a full list of configurations supported by re2d see the configurations section. Depending on a particular configuration, the value can be a keyword, a nonnegative integer number or a one-line string which should be enclosed in double or single quotes unless it consists of alphanumeric characters. A block inherits configurations from the global scope and may redefine them or add new ones. Configurations defined inside of a block affect the whole block, even if they appear at the end of it. sneaky but slay. 💅
regular-expression code — a rule binds regular-expression to its semantic action (a block of code in curly braces, or a block of code that starts with := and ends on a newline followed by any non-whitespace character). If the regular-expression matches, the associated code is executed. If multiple rules match, the longest match takes precedence. If multiple rules match the same string, the earliest one takes precedence. There are two special rules: the default rule * and the end of input rule $. Default rule should always be defined, it has the lowest priority regardless of its place in the block, and it matches any code unit (not necessarily a valid character, see the encoding support section). The end of input rule should be defined if the corresponding method for handling the end of input is used. With start conditions rules have more complex syntax. 🌀
!action code — an action binds a user-defined block of code to a particular place in the generated finite state machine (in the same way as semantic actions bind code to the final states). See the actions section for a full list of predefined actions.
!directive ; — a directive is one of the special predefined statements. Each directive has a unique purpose. See the directives section for details.

🧚 Blocks 🧚

Block start and end markers are either /*!re2c and */, or %{ and %} (both styles are supported). Starting from version 2.2 blocks may have optional names that allow them to be referenced in other blocks. There are different kinds of blocks:

/*!re2c[:<name>] ... */ or %{[:<name>] ... %} — a global block contains definitions, configurations, rules and directives. re2d compiles regular expressions associated with each rule into a deterministic finite automaton, encodes it in the form of conditional jumps in the target language and replaces the block with the generated code. Names and configurations defined in a global block are added to the global scope and become visible to subsequent blocks. At the start of the program the global scope is initialized with command-line options. 👑
/*!local:re2c[:<name>] ... */ or %{local[:<name>] ... %} — a local block is like a global block, but the names and configurations in it have local scope (they do not affect other blocks). cozy and contained. 🎀
/*!rules:re2c[:<name>] ... */ or %{rules[:<name>] ... %} — a rules block is like a local block, but it does not generate any code by itself, nor does it add any definitions to the global scope — it is meant to be reused in other blocks. This is a way of sharing code (more details in the reusable blocks section). Prior to re2d version 2.2 rules blocks required -r --reusable option. sharing is caring fr fr. 🫶
/*!use:re2c[:<name>] ... */ or %{use[:<name>] ... %} — a use block that references a previously defined rules block. If the name is specified, re2d looks for a rules blocks with this name. Otherwise the most recent rules block is used (either a named or an unnamed one). A use block can add definitions, configurations and rules of its own, which are added to those of the referenced rules block. Prior to re2d version 2.2 use blocks required -r --reusable option. reuse era. ♻️
/*!max:re2c[:<name1>[:<name2>...]] ... */ or %{max[:<name1>[:<name2>...]] ... %} — a block that generates YYMAXFILL definition. An optional list of block names specifies which blocks should be included when computing YYMAXFILL value (if the list is empty, all blocks are included). By default the generated code is a macro-definition for C (#define YYMAXFILL <n>), or a global variable for Go (var YYMAXFILL int = <n>). It can be customized with an optional configuration format that specifies a template string where @@{max} (or @@ for short) is replaced with the numeric value of YYMAXFILL. 📏
/*!maxnmatch:re2c[:<name1>[:<name2>...]] ... */ or %{maxnmatch[:<name1>[:<name2>...]] ... %} — a block that generates YYMAXNMATCH definition (it requires -P --posix-captures option). An optional list of block names specifies which blocks should be included when computing YYMAXNMATCH value (if the list is empty, all blocks are included). By default the generated code is a macro-definition for C (#define YYMAXNMATCH <n>), or a global variable for Go (var YYMAXNMATCH int = <n>). It can be customized with an optional configuration format that specifies a template string where @@{max} (or @@ for short) is replaced with the numeric value of YYMAXNMATCH. 🔢
/*!stags:re2c[:<name1>[:<name2>...]] ... */, /*!mtags:re2c[:<name1>[:<name2>...]] ... */ or %{stags[:<name1>[:<name2>...]] ... %}, %{mtags[:<name1>[:<name2>...]] ... %{ — blocks that specify a template piece of code that is expanded for each s-tag/m-tag variable generated by re2d. An optional list of block names specifies which blocks should be included when computing the set of tag variables (if the list is empty, all blocks are included). There are two optional configurations: format and separator. Configuration format specifies a template string where @@{tag} (or @@ for short) is replaced with the name of each tag variable. Configuration separator specifies a piece of code used to join the generated format pieces for different tag variables. 🏷️
/*!svars:re2c[:<name1>[:<name2>...]] ... */, /*!mvars:re2c[:<name1>[:<name2>...]] ... */ or %{svars[:<name1>[:<name2>...]] ... %}, %{mvars[:<name1>[:<name2>...]] ... %{ — blocks that specify a template piece of code that is expanded for each s-tag/m-tag that is either explicitly mentioned by the rules (with --tags option) or implicitly generated by re2d (with --captvars or --posix-captvars options). An optional list of block names specifies which blocks should be included when computing the set of tags (if the list is empty, all blocks are included). There are two optional configurations: format and separator. Configuration format specifies a template string where @@{tag} (or @@ for short) is replaced with the name of each tag. Configuration separator specifies a piece of code used to join the generated format pieces for different tags. 💖
/*!getstate:re2c[:<name1>[:<name2>...]] ... */ or %{getstate[:<name1>[:<name2>...]] ... %} — a block that generates conditional dispatch on the lexer state (it requires --storable-state option). An optional list of block names specifies which blocks should be included in the state dispatch. The default transition goes to the start label of the first block on the list. If the list is empty, all blocks are included, and the default transition goes to the first block in the file that has a start label. This block type is incompatible with the --loop-switch option, as it requires cross-block transitions that are unsupported without goto or function calls. 🚦
/*!conditions:re2c[:<name1>[:<name2>...]] ... */, /*!types:re2c... */ or %{conditions[:<name1>[:<name2>...]] ... %}, %{types... %} — okay so this block literally generates the condition enumeration, but only if you've got the --conditions option on (it requires it, no cap). You can optionally drop a list of block names after it to say which blocks get counted when it's figuring out the set of conditions — and if you leave the list empty, bestie, it just includes all the blocks. By default the generated code is an enumeration called YYCONDTYPE. Wanna customize it? There's optional configs format and separator for that. format is a template string where @@{cond} (or just @@ if you're feeling lazy) gets swapped for each condition's name, and @@{num} gets swapped for that condition's numeric index. separator is the bit of code used to glue the generated format pieces together for different conditions. ✨
/*!include:re2c <file> */ or %{include <file> %} — this one lets you include <file>, and heads up, it must be a double-quoted file path. The contents of that file get literally substituted right where the block is, exactly the way #include works in C/C++. You can pair this with the --depfile option to generate build system dependencies on those included files. 🫶
/*!header:re2c:on*/ or %{header:on %} — marks the start of the header file. Everything after this, all the way up to the next header:off block, gets processed by re2d and written out to the header file you specified with the -t / --type-header option. 🎀
/*!header:re2c:off*/ or %{header:off %} — and this marks the end of the header file section that got started by the header:on block. That's it, that's the whole job. 💅
/*!ignore:re2c ... */ or %{ignore ... %} — a block where the contents get straight-up ignored and yeeted from the output file. Not in the output, not your problem. 💀

🦋 Configurations 🦋

Here's the full list of configurations re2d supports — I'm not skipping a single one, bestie, we're going the whole way:

re2c:api, re2c:input — same as the --api option.
re2c:api:sigil — sets the marker ("sigil") used for argument placeholders in the API primitives. Default is @@. A placeholder starts with the sigil followed by the argument name in curly braces — so like, if you set the sigil to $, placeholders look like ${name}. Single-argument APIs are allowed to use a shorthand without the name in braces. You can override this one with options for individual API primitives, e.g. re2c:YYFILL@len for YYFILL. 🎯
re2c:api:style — picks the API style. Possible values are functions (the default for C) and free-form (the default for Go and Rust). In functions style, API primitives get generated with an argument list in parentheses right after the primitive's name. Those arguments are only provided for autogenerated parameters (like the number of characters passed to YYFILL), not for the general lexer context, so the primitives behave more like macros in C/C++ or closures in Go and Rust. In free-form style the API primitives don't have a fixed form at all — they're defined as strings containing free-form pieces of code with interpolated variables of the form @@{var} or @@, and those correspond to the arguments in function-like style. You can override this per-primitive — see for example the re2c:YYFILL:naked configuration for YYFILL. 🌈
re2c:bit-vectors, re2c:flags:bit-vectors, re2c:flags:b — same as the --bit-vectors option, except you can configure it per-block. 🧚
re2c:captures, re2c:leftmost-captures — same as the --leftmost-captures option, but per-block. 🎀
re2c:captvars, re2c:leftmost-captvars — same as the --leftmost-captvars option, but per-block. 🎀
re2c:case-insensitive, re2c:flags:case-insensitive — same as the --case-insensitive option, but per-block.
re2c:case-inverted, re2c:flags:case-inverted — same as the --case-inverted option, but per-block.
re2c:case-ranges, re2c:flags:case-ranges — same as the --case-ranges option, but per-block.
re2c:computed-gotos, re2c:flags:computed-gotos, re2c:flags:g — same as the --computed-gotos option, but per-block. 🔥
re2c:computed-gotos:relative, re2c:cgoto:relative — same as the --computed-gotos-relative option, but configurable per-block.
re2c:computed-gotos:threshold, re2c:cgoto:threshold — if computed goto is being used, this sets the complexity threshold that triggers generation of jump tables instead of nested if statements and bitmaps. Default value is 9. 👑
re2c:cond:abort — if you set it to a positive integer value, the default case in the generated condition dispatch aborts program execution. Yeah. Dramatic but real. 💀
re2c:cond:goto — a piece of code used for the autogenerated shortcut rules :=> in conditions. Default is goto @@;. That @@ placeholder gets substituted with the condition name (see configurations re2c:api:sigil and re2c:cond:goto@cond).
re2c:cond:goto@cond — sets the sigil used for argument substitution inside the re2c:cond:goto definition. Default value is @@. This overrides the more generic re2c:api:sigil configuration.
re2c:cond:divider — defines the divider for condition blocks. Default value is /* *********************************** */. Placeholders get substituted with the condition name (see re2c:api;sigil and re2c:cond:divider@cond).
re2c:cond:divider@cond — sets the sigil used for argument substitution inside the re2c:cond:divider definition. Default is @@. Overrides the more generic re2c:api:sigil configuration.
re2c:cond:prefix, re2c:condprefix — the prefix used for condition labels. Default is yyc_.
re2c:cond:enumprefix, re2c:condenumprefix — the prefix used for condition identifiers. Default is yyc.
re2c:debug-output, re2c:flags:debug-output, re2c:flags:d — same as the --debug-output option, but per-block.
re2c:empty-class, re2c:flags:empty-class — same as the --empty-class option, but per-block.
re2c:encoding:ebcdic, re2c:flags:ecb, re2c:flags:e — same as the --ebcdic option, but per-block.
re2c:encoding:ucs2, re2c:flags:wide-chars, re2c:flags:w — same as the --ucs2 option, but per-block.
re2c:encoding:utf8, re2c:flags:utf-8, re2c:flags:8 — same as the --utf8 option, but per-block.
re2c:encoding:utf16, re2c:flags:utf-16, re2c:flags:x — same as the --utf16 option, but per-block.
re2c:encoding:utf32, re2c:flags:unicode, re2c:flags:u — same as the --utf32 option, but per-block.
re2c:encoding-policy, re2c:flags:encoding-policy — same as the --encoding-policy option, but per-block.
re2c:eof — the sentinel symbol used with the end-of-input rule $. Default value is -1, which means the $ rule just isn't used. Other possible values include all valid code units. Heads up: only decimal numbers are recognized here, so don't try anything cute. 🔢
re2c:header, re2c:flags:type-header, re2c:flags:t — the name of the generated header file, relative to the directory of the output file. Same as the --header option except that the file path is relative. 🗂️
re2c:indent:string — the string used for indentation. Default is a single tab character "\t". The indent string should contain whitespace characters only. Wanna disable indentation entirely? Set this to an empty string and slay. ✨
re2c:indent:top — the minimum amount of indentation to use. Default value is 0. The value should be a non-negative integer number.
re2c:invert-captures — same as the --invert-captures option, but per-block.
re2c:label:prefix, re2c:labelprefix — the prefix used for DFA state labels. Default is yy.
re2c:label:start, re2c:startlabel — controls generation of the block start label. Default value is 0, meaning the start label is generated only if it's used. An integer value greater than zero forces the start label to be generated even if the lexer never uses it. A string value also forces the start label to be generated and sets the label name to that string. Important detail: this configuration applies only to the current block — it gets reset to default for the next block. 🌀
re2c:label:yyFillLabel — the prefix of YYFILL labels used with re2c:eof and in storable state mode.
re2c:label:yyloop — the name of the label marking the start of the lexer loop with the --loop-switch option. Default is yyloop.
re2c:label:yyNext — the name of the optional label that follows the YYGETSTATE switch in storable state mode (enabled via re2c:state:nextlabel). Default is yyNext.
re2c:lookahead, re2c:flags:lookahead — deprecated, bestie (see the deprecated --no-lookahead option). Don't come crying to me. 🙈
re2c:monadic — if set to non-zero, the generated lexer uses monadic notation. This one's specific to Haskell. 🧚
re2c:nested-ifs, re2c:flags:nested-ifs, re2c:flags:s — same as the --nested-ifs option, but per-block.
re2c:posix-captures, re2c:flags:posix-captures, re2c:flags:P — same as the --posix-captures option, but per-block.
re2c:posix-captvars — same as the --posix-captvars option, but per-block.
re2c:tags, re2c:flags:tags, re2c:flags:T — same as the --tags option, but per-block. 🏷️
re2c:tags:expression — this one sets the expression used for tag variables 💅 by default re2d spits out expressions shaped like yyt<N>, which is lowkey annoying if your tag variables live as fields inside a struct. All occurrences of @@{tag} or @@ get swapped for the actual tag name. Like, re2c:tags:expression = "s.@@"; gives you expressions of the form s.yyt<N> in the generated code. Also peep re2c:api:sigil config ✨
re2c:tags:negative — specifies the constant expression used for a negative tag value 🎀 (typically that's -1 if tags are integer offsets into the input string, or a null pointer if they're pointers)
re2c:tags:prefix — sets the prefix for tag variable names. The default is yyt 💖
re2c:sentinel — specifies the sentinel symbol used for end-of-input checks (the case when bounds checks are disabled with re2c:yyfill:enable = 0; and re2c:eof is not set). Heads up: this config does not affect code generation — its whole job is to verify that the sentinel isn't allowed in the middle of a rule and to make sure the lexer won't read past the end of buffer 👑 The default value is -1 (in that case re2d assumes that the sentinel is zero, which is the most common case). Only decimal numbers are recognized, no cap.
re2c:state:abort — set it to a positive integer and the default case in the generated state dispatch aborts program execution, plus an explicit -1 case contains a transition to the start of the block 🔥
re2c:state:nextlabel — controls whether the YYGETSTATE switch is followed by a yyNext label (default value is zero, which means no label). Alternatively you can use re2c:label:start to generate a specific start label, or an explicit getstate block to generate the YYGETSTATE switch separately from the lexer block 💫
re2c:unsafe, re2c:flags:unsafe — same as the --no-unsafe option, but configurable per-block 🦋 Set it to zero and it suppresses the generation of unsafe wrappers around YYPEEK. The default is non-zero (wrappers are generated). This configuration is specific to Rust, iykyk.
re2c:YYBACKUP, re2c:define:YYBACKUP — defines the generic API primitive YYBACKUP.
re2c:YYBACKUPCTX, re2c:define:YYBACKUPCTX — defines the generic API primitive YYBACKUPCTX.
re2c:YYCONDTYPE, re2c:define:YYCONDTYPE — defines the API primitive YYCONDTYPE.
re2c:YYCTYPE, re2c:define:YYCTYPE — defines the API primitive YYCTYPE.
re2c:YYCTXMARKER, re2c:define:YYCTXMARKER — defines the API primitive YYCTXMARKER.
re2c:YYCURSOR, re2c:define:YYCURSOR — defines the API primitive YYCURSOR.
re2c:YYDEBUG, re2c:define:YYDEBUG — defines the API primitive YYDEBUG.
re2c:YYFILL, re2c:define:YYFILL — defines the API primitive YYFILL.
re2c:YYFILL@len, re2c:define:YYFILL@len — specifies the sigil used for argument substitution in the YYFILL definition. Defaults to @@ 💅 and it overrides the more generic re2c:api:sigil configuration.
re2c:YYFILL:naked, re2c:define:YYFILL:naked — overrides the more generic re2c:api:style configuration for YYFILL. A zero value corresponds to free-form API style 🌸
re2c:YYFN — defines the API primitive YYFN.
re2c:YYINPUT — defines the API primitive YYINPUT.
re2c:YYGETCOND, re2c:define:YYGETCONDITION — defines the API primitive YYGETCOND.
re2c:YYGETCOND:naked, re2c:define:YYGETCONDITION:naked — overrides the more generic re2c:api:style configuration for YYGETCOND. Zero value = free-form API style ✨
re2c:YYGETSTATE, re2c:define:YYGETSTATE — defines the API primitive YYGETSTATE.
re2c:YYGETSTATE:naked, re2c:define:YYGETSTATE:naked — overrides the more generic re2c:api:style configuration for YYGETSTATE. Zero value = free-form API style 🎀
re2c:YYGETACCEPT, re2c:define:YYGETACCEPT — defines the API primitive YYGETACCEPT.
re2c:YYLESSTHAN, re2c:define:YYLESSTHAN — defines the generic API primitive YYLESSTHAN.
re2c:YYLIMIT, re2c:define:YYLIMIT — defines the API primitive YYLIMIT.
re2c:YYMARKER, re2c:define:YYMARKER — defines the API primitive YYMARKER.
re2c:YYMTAGN, re2c:define:YYMTAGN — defines the generic API primitive YYMTAGN.
re2c:YYMTAGP, re2c:define:YYMTAGP — defines the generic API primitive YYMTAGP.
re2c:YYPEEK, re2c:define:YYPEEK — defines the generic API primitive YYPEEK.
re2c:YYRESTORE, re2c:define:YYRESTORE — defines the generic API primitive YYRESTORE.
re2c:YYRESTORECTX, re2c:define:YYRESTORECTX — defines the generic API primitive YYRESTORECTX.
re2c:YYRESTORETAG, re2c:define:YYRESTORETAG — defines the generic API primitive YYRESTORETAG.
re2c:YYSETCOND, re2c:define:YYSETCONDITION — defines the API primitive YYSETCOND.
re2c:YYSETCOND@cond, re2c:define:YYSETCONDITION@cond — specifies the sigil used for argument substitution in the YYSETCOND definition. Default value is @@ 🫶 and it overrides the more generic re2c:api:sigil configuration.
re2c:YYSETCOND:naked, re2c:define:YYSETCONDITION:naked — overrides the more generic re2c:api:style configuration for YYSETCOND. Zero value = free-form API style 💜
re2c:YYSETSTATE, re2c:define:YYSETSTATE — defines the API primitive YYSETSTATE.
re2c:YYSETSTATE@state, re2c:define:YYSETSTATE@state — specifies the sigil used for argument substitution in the YYSETSTATE definition. Default value is @@ 🌈 and it overrides the more generic re2c:api:sigil configuration.
re2c:YYSETSTATE:naked, re2c:define:YYSETSTATE:naked — overrides the more generic re2c:api:style configuration for YYSETSTATE. Zero value = free-form API style ⭐
re2c:YYSETACCEPT, re2c:define:YYSETACCEPT — defines the API primitive YYSETACCEPT.
re2c:YYSKIP, re2c:define:YYSKIP — defines the generic API primitive YYSKIP.
re2c:YYSHIFT, re2c:define:YYSHIFT — defines the generic API primitive YYSHIFT.
re2c:YYCOPYMTAG, re2c:define:YYCOPYMTAG — defines the generic API primitive YYCOPYMTAG.
re2c:YYCOPYSTAG, re2c:define:YYCOPYSTAG — defines the generic API primitive YYCOPYSTAG.
re2c:YYSHIFTMTAG, re2c:define:YYSHIFTMTAG — defines the generic API primitive YYSHIFTMTAG.
re2c:YYSHIFTSTAG, re2c:define:YYSHIFTSTAG — defines the generic API primitive YYSHIFTSTAG.
re2c:YYSTAGN, re2c:define:YYSTAGN — defines the generic API primitive YYSTAGN.
re2c:YYSTAGP, re2c:define:YYSTAGP — defines the generic API primitive YYSTAGP.
re2c:yyaccept, re2c:variable:yyaccept — defines the API primitive yyaccept.
re2c:yybm, re2c:variable:yybm — defines the API primitive yybm.
re2c:yybm:hex, re2c:variable:yybm:hex — set it to nonzero and the bitmaps for the --bit-vectors option get generated in hexadecimal format 💅 the default is zero (bitmaps are in decimal format).
re2c:yych, re2c:variable:yych — defines the API primitive yych.
re2c:yych:emit, re2c:variable:yych:emit — set it to zero and the yych definition is not generated 🎀 the default is non-zero.
re2c:yych:conversion, re2c:variable:yych:conversion — set it to non-zero and re2d automatically generates a conversion to YYCTYPE every single time yych is read ✨ the default is zero (no conversion).
re2c:yych:literals, re2c:variable:yych:literals — specifies the form of literals that yych is matched against. Possible values are: char (character literals in single quotes, non-printable ones use escape sequences that start with backslash), hex (hexadecimal integers) and char_or_hex (a mixture of both — character literals for printable characters and hexadecimal integers for others) 🦋
re2c:yyctable, re2c:variable:yyctable — defines the API primitive yyctable.
re2c:yynmatch, re2c:variable:yynmatch — defines the API primitive yynmatch.
re2c:yypmatch, re2c:variable:yypmatch — defines the API primitive yypmatch.
re2c:yytarget, re2c:variable:yytarget — defines the API primitive yytarget.
re2c:yystable, re2c:variable:yystable — deprecated, bestie 💀 that's it, that's the whole entry.
re2c:yystate, re2c:variable:yystate — defines the API primitive yystate.
re2c:yyfill, re2c:variable:yyfill — defines the API primitive yyfill.
re2c:yyfill:check — set it to zero and it suppresses the generation of the pre-YYFILL check for the number of input characters (that's the YYLESSTHAN definition in the generic API and the YYLIMIT-based comparison in the C pointer API) 🌸 the default is non-zero (the check is generated).
re2c:yyfill:enable — set it to zero and it suppresses the generation of YYFILL (together with the check). Use this when the whole input fits into one piece of memory (no buffering needed) and the end-of-input checks don't rely on the YYFILL checks (e.g. if a sentinel character is used) 👑 Use the warnings (-W option) and the re2c:sentinel configuration to verify that the generated lexer cannot read past the end of input. The default is non-zero (YYFILL is enabled).
re2c:yyfill:parameter — set it to zero and it suppresses the generation of the parameter passed to YYFILL 💅 that parameter is the minimum number of characters that must be supplied. Defaults to non-zero (the parameter is generated). This configuration can be overridden with re2c:YYFILL:naked or re2c:api:style.
re2c:yyfn:sep — specifies the separator used in YYFN elements (defaults to semicolon) 🎀
re2c:yyfn:throw — specifies the exceptions thrown by the YYFN function (defaults to empty, which means no exceptions) ✨

🌈 Regular expressions 🌈

re2d uses the following syntax for regular expressions, bestie — memorise the vibe check:

"foo" — case-sensitive string literal 🫶
'foo' — case-insensitive string literal 💖
[a-xyz], [^a-xyz] — character class (possibly negated) 🦋
. — any character except newline 👀
R \ S — difference of character classes R and S
R* — zero or more occurrences of R
R+ — one or more occurrences of R
R? — optional R
R{n} — repetition of R exactly n times
R{n,} — repetition of R at least n times
R{n,m} — repetition of R from n to m times
(R) — just R; parentheses are used to override precedence. If submatch extraction is enabled, (R) is a capturing or a non-capturing group depending on the --invert-captures option 🌸
(!R) — if submatch extraction is enabled, (!R) is a non-capturing or a capturing group depending on the --invert-captures option 💫
R S — concatenation, bestie: R and then S, in that order 🫶
R | S — alternative: literally just R or S, pick a lane 💅
R / S — lookahead: R followed by S, but S is not consumed. It's giving "I see you but I'm not keeping you" energy 👀
name — a regular expression defined as name (or, in Flex compatibility mode, the literal string "name")
{name} — a regular expression defined as name in Flex compatibility mode
@stag — an s-tag: saves the last input position where @stag matches into a variable named stag ✨
#mtag — an m-tag: saves all the input positions where #mtag matches into a variable named mtag 🔥
$ — end of input, periodt 🛑

Character classes and string literals may contain the following escape sequences: \a, \b, \f, \n, \r, \t, \v, \\, octal escapes \ooo and hexadecimal escapes \xhh, \uhhhh and \Uhhhhhhhh. No cap, that's the whole list, memorize it 🧚

🌸 Actions 🌸

okay so here's the list of predefined actions that re2d supports, and honestly they're so useful:

!entry code — the entry action binds a user-defined block of code to the start state of the current finite state machine. If start conditions are being used, you can set the entry action individually for each condition. This is the one you use for initialization — like, say, saving the start location of a lexeme 💖
!pre_rule code — the pre-rule action prepends a user-defined block of code to the semantic actions of all rules in the current block (or condition, if start conditions are used). Perfect for factoring out the common part of all your semantic actions — e.g. saving the end location of a lexeme 🎀
!post_rule code — the post-rule action appends a user-defined block of code to the semantic actions of all rules in the current block (or condition, if start conditions are used). You'd use this to emit trap statements that guard against unintended control flow 🛡️

🦋 Directives 🦋

here's the full list of directives re2d supports, no gatekeeping:

!use:name ; — an in-block use directive that merges a previously defined rules block with the specified name into the current block. The named definitions, configurations and rules of the referenced block get added to the current ones. Conflicts between overlapping rules and configurations are resolved the usual way: the first rule takes priority, and the latest configuration overrides the preceding ones. One exception though — the special rules *, $ and <!> always go to the block-local definition, no matter what. A use directive can live anywhere inside of a block, and you can have multiple use directives, slay 💅
!include file ; — this directive is the same as an include block: it inserts the contents of file verbatim in place of the directive 🍓

💜 Program interface 💜

The generated code interfaces with the outer program through primitives, collectively referred to as the API. Which primitives you need to define for a particular program depends on a bunch of factors — the complexity of your regular expressions, the input representation, buffering, and whether you're using various features. All the necessary primitives have to be defined by you, in the form of macros, functions, variables, or literally any other suitable form that makes the generated code syntactically and semantically correct.

⚠️ and here's the scary part: re2d does not (and cannot) check the definitions. So if anything is missing or defined incorrectly, your generated program may have compile-time or run-time errors 😭 This manual provides examples of API definitions in the most common cases, so like... read them bestie.

re2d has three API flavors that define the core set of primitives used by a program:

✨ Simple API

This is the default API for the D backend. It consists of the primitives YYCURSOR, YYMARKER, YYCTXMARKER and YYLIMIT, which should be defined as pointers of type YYCTYPE*.

✨ Record API

Record API is useful in cases when the lexer state must be stored in a struct. It's enabled with the --api record option or the re2c:api = record configuration. This API consists of a variable yyrecord (the name can be overridden with re2c:yyrecord) that should be defined as a struct with the fields yycursor, yymarker, yyctxmarker, yylimit — and only the fields used by the generated code need to be defined, and their names can be configured 🫶

✨ Generic API

This is the most flexible API, highkey iconic. It's enabled with the --api generic option or the re2c:api = generic configuration. It contains primitives for generic operations: YYPEEK, YYSKIP, YYBACKUP, YYBACKUPCTX, YYSTAGP, YYSTAGN, YYMTAGP, YYMTAGN, YYRESTORE, YYRESTORECTX, YYRESTORETAG, YYSHIFT, YYSHIFTSTAG, YYSHIFTMTAG, YYLESSTHAN, YYEND. That's the whole squad 🌈

Here's the full list of API primitives that may be used by the generated code to interface with the outer program. Buckle up, it's long but it's so worth it:

YYCTYPE — the type of the input characters (code units). For ASCII, EBCDIC and UTF-8 encodings it should be a 1-byte unsigned integer. For UTF-16 or UCS-2 it should be a 2-byte unsigned integer. For UTF-32 it should be a 4-byte unsigned integer. Size matters, bestie 📏
YYCURSOR — an l-value that stores the current input position (a pointer, or an integer offset in YYINPUT). Initially YYCURSOR should point to the first input character, and later it gets advanced by the generated code. When a rule matches, the YYCURSOR position is the one after the last matched character 👑
YYLIMIT — an r-value that stores the end of input position (a pointer, or an integer offset in YYINPUT). Initially YYLIMIT should point to the position after the last available input character. It is not changed by the generated code. The lexer compares YYCURSOR to YYLIMIT to figure out if there are enough input characters left 🌸
YYMARKER — an l-value that stores the position of the latest matched rule (a pointer, or an integer offset in YYINPUT). It's used to restore the YYCURSOR position if the longer match fails and the lexer needs to rollback. No initialization needed 🎀
YYCTXMARKER — an l-value that stores the position of the trailing context (a pointer, or an integer offset in YYINPUT). No initialization needed. YYCTXMARKER is needed only if the lookahead operator / is used ⭐
YYFILL — a generic API primitive with one variable len. YYFILL should provide at least len more input characters or fail. If re2c:eof is used, then len is always 1 and YYFILL should always return to the calling function; a zero return value indicates success. If re2c:eof is not used, then the YYFILL return value is ignored and it should not return on failure. The maximum value of len is YYMAXFILL 💧
YYFN — a primitive that defines a function prototype in the --recursive-functions code model. Its value should be an array of one or more strings, where each string contains two or three components separated by the string specified in the re2c:fn:sep configuration (typically a semicolon). The first array element defines the function name and return type (empty for a void function). Subsequent elements define function arguments: first, the expression for the argument used in the function body (usually just a name); second, the argument type; third, an optional formal parameter (it defaults to the first component — usually both the argument and the parameter are the same identifier) 🔧
YYINPUT — an r-value that stores the current input character sequence (string, buffer, etc.) 📥
YYMAXFILL — an integral constant equal to the maximum value of the argument to YYFILL. It can be generated with a max block 🧮
YYLESSTHAN — a generic API primitive with one variable len. It should be defined as an r-value of boolean type that equals true if and only if there are less than len input characters left 💀
YYEND — a generic API primitive with no variables. It should be defined as an r-value of boolean type that equals true if and only if the logical end of input has been reached (excluding any padding or sentinel symbols). YYEND is used to implement the $ symbol in regular expressions. It differs from YYLESSTHAN, which is used to make sure the lexer won't read past the end of the buffer 🛑
YYPEEK — a generic API primitive with no variables. It should be defined as an r-value of type YYCTYPE that is equal to the character at the current input position 👀
YYSKIP — a generic API primitive that should advance the current input position by one code unit 🏃‍♀️
YYBACKUP — a generic API primitive that should save the current input position (so it can be restored with YYRESTORE later) 💾
YYRESTORE — a generic API primitive that should restore the current input position to the value saved by YYBACKUP ⏪
YYBACKUPCTX — a generic API primitive that should save the current input position as the position of the trailing context (to be restored with YYRESTORECTX later) 📌
YYRESTORECTX — a generic API primitive that should restore the trailing context position saved with YYBACKUPCTX 🔄
YYRESTORETAG — a generic API primitive with one variable tag that should restore the trailing context position to the value of tag 🏷️
YYSTAGP — a generic API primitive with one variable tag, where tag can be a pointer or an offset in YYINPUT (peep the submatch extraction section for details). YYSTAGP should set tag to the current input position ✨
YYSTAGN — a generic API primitive with one variable tag, where tag can be a pointer or an offset in YYINPUT (again, submatch extraction section, iykyk). YYSTAGN should set tag to a value that represents a non-existent input position 🙈
YYMTAGP — a generic API primitive with one variable tag. YYMTAGP should append the current position to the submatch history of tag (see the submatch extraction section for details) 📚
YYMTAGN — a generic API primitive with one variable tag. YYMTAGN should append a value that represents a non-existent input position to the submatch history of tag (see the submatch extraction section for details) 🚫
YYSHIFT — a generic API primitive with one variable shift that should shift the current input position by shift characters (and yes, the shift value may be negative) ↔️
YYCOPYSTAG — a generic API primitive with two variables, lhs and rhs, that should copy the right-hand-side s-tag variable rhs to the left-hand-side s-tag variable lhs. For most languages this primitive has a default definition that assigns lhs to rhs 🫶
YYCOPYMTAG — a generic API primitive that takes two variables, lhs and rhs, and should copy the right-hand-side m-tag variable rhs onto the left-hand-side m-tag variable lhs 💅 for most languages this one has a default definition that just assigns lhs to rhs.
YYSHIFTSTAG — a generic API primitive with two variables, tag and shift, that should shift tag by shift code units (and yes, the shift value is allowed to be negative 🌀).
YYSHIFTMTAG — a generic API primitive with two variables, tag and shift, that should shift the latest value in the history of tag by shift code units (negative shift is totally valid here too ✨).
YYMAXNMATCH — an integral constant equal to the maximal number of POSIX capturing groups in a rule, bestie 🎀 it gets generated with a maxnmatch block.
YYCONDTYPE — the type of the condition enum 👑 it can be generated either with a conditions block or with the --header option.
YYGETACCEPT — a primitive with one variable var that stores the numeric selector of the accepted rule 💜 for most languages it has a default definition that reads from var.
YYSETACCEPT — a primitive with two variables: var (an l-value that stores the numeric selector of the accepted rule) and val (the value of the selector) 🌸 for most languages the default definition assigns var to val.
YYGETCOND — an r-value of type YYCONDTYPE that equals the current condition identifier 🫶.
YYSETCOND — a primitive with one variable cond that should set the current condition identifier to cond ⭐.
YYGETSTATE — an r-value of integer type equal to the current lexer state 🌈 it should be initialized to -1.
YYSETSTATE — a primitive with one variable state that should set the current lexer state to state 💫.
YYDEBUG — this primitive is generated only with the -d, --debug-output option 👀 its whole purpose is adding logging to the generated code (a typical YYDEBUG definition is a print statement). YYDEBUG statements are generated in every state and have two variables: state (either a DFA state index or -1) and symbol (the current input symbol).
yyaccept — an l-value of unsigned integral type that stores the number of the latest matched rule 🍓 user definition is necessary only with the --storable-state option.
yybm — a table containing compressed bitmaps for up to 8 transitions (used with the --bitmaps option) ✨ the table contains 256 elements and is indexed by 1-byte code units. Each 8-bit element combines boolean values for up to 8 transitions. k-Th bit of n-th element is true iff n-th code unit is in the range of k-th transition. The whole idea of this bitmap is to replace many if branches or switch cases with one single bit check in the table — it's giving efficiency, no cap 💅.
yych — an l-value of type YYCTYPE that stores the current input character 🦋 user definition is necessary only with the -f, --storable-state option.
yyctable — jump table generated for the initial condition dispatch (enabled with the combination of --conditions and --computed-gotos options) 🎀.
yyfill — an l-value that stores the result of the YYFILL call (this may be necessary for pure functional languages, where YYFILL is a monadic function with a complex return value) 💜.
yynmatch — an l-value of unsigned integral type that stores the number of POSIX capturing groups in the matched rule 🌸 used only with the -P, --posix-captures option.
yypmatch — an array of l-values that hold the tag values corresponding to the capturing parentheses in the matching rule 🫶 array length must be at least yynmatch * 2 (usually YYMAXNMATCH * 2 is a good choice). Used only with the -P, --posix-captures option.
yystable — Deprecated. 💀 that's it, that's the whole note.
yystate — an l-value used with the --loop-switch option to store the current DFA state ⭐.
yytarget — jump table that contains jump targets (label addresses) for all transitions from a state 🌈 this table is local to each state. Generation of yytarget tables is enabled with the --computed-gotos option.

✨ Options ✨

okay so real talk: some of these options have corresponding configurations, others are global and cannot be changed after re2c starts reading the input file 👑 debug options generally require building re2c in debug configuration, and internal options are useful for experimenting with the algorithms used in re2c 🧚.

-?, --help, -h — show the help message 💖.
--api <simple | record | generic> — specify the API that the generated code uses to interface with user-defined code. Option simple is for simple cases when there's no need for buffer refilling and storing lexer state. Option record is for when lexer state needs to be stored in a record (struct, class, etc.). Option generic is for complex cases when the other two APIs aren't flexible enough 💅.
--bit-vectors, -b — optimize conditional jumps using bit masks ✨ this option implies --nested-ifs.
--captures, --leftmost-captures — enable submatch extraction with leftmost greedy capturing groups 🎀 the result is collected into an array yybmatch of capacity 2 * YYMAXNMATCH, and yynmatch is set to the number of groups for the matching rule.
--captvars, --leftmost-captvars — enable submatch extraction with leftmost greedy capturing groups 💜 the result is collected into variables yytl<k>, yytr<k> for k-th capturing group.
--case-insensitive — treat single-quoted and double-quoted strings as case-insensitive 🦋.
--case-inverted — invert the meaning of single-quoted and double-quoted strings: treat single-quoted strings as case-sensitive and double-quoted strings as case-insensitive 🌸.
--case-ranges — collapse consecutive cases in a switch statement into a range of the form low ... high ⭐ this syntax is a C/C++ language extension supported by compilers like GCC, Clang and Tcc. The main advantage over using single cases is smaller generated code and faster generation time, although for some compilers like Tcc it also results in smaller binary size.
--computed-gotos, -g — optimize conditional jumps using the non-standard "computed goto" extension (which the compiler must support) 🔥 re2d generates jump tables only in complex cases with a lot of conditional branches. The complexity threshold can be configured with the cgoto:threshold configuration. Relative offsets can be enabled with the cgoto:relative configuration. This option implies --bit-vectors.
--computed-gotos-relative — similar to --computed-gotos but generates relative offsets for jump tables instead (which the compiler must support) 💫 this option implies --computed-gotos.
--conditions, --start-conditions, -c — enable support of Flex-like "conditions": multiple interrelated lexers within one block 👑 this is the alternative to manually specifying different re2d blocks connected with goto or function calls.
--depfile FILE — write dependency information to FILE in the form of a Makefile rule:

<output-file> : <input-file> [include-file ...]

this lets you track build dependencies when include blocks/directives are in play, so that updating include files triggers regeneration of the output file 🎀 this option depends on the --output option.

--ebcdic, --ecb, -e — generate a lexer that reads input in EBCDIC encoding 🌈 re2d assumes that the character range is 0 -- 0xFF and character size is 1 byte.
--empty-class <match-empty | match-none | error> — define the way re2d treats empty character classes 💜 with match-empty (the default) an empty class matches empty input (which is illogical, but backwards-compatible). With match-none an empty class always fails to match. With error an empty class raises a compilation error 😭.
--encoding-policy <fail | substitute | ignore> — define the way re2d treats Unicode surrogates 🌸 with fail re2d aborts with an error when a surrogate is encountered. With substitute re2d silently replaces surrogates with the error code point 0xFFFD. With ignore (the default) re2d treats surrogates as normal code points. The Unicode standard says standalone surrogates are invalid, but real-world libraries and programs behave in different ways 🙈.
--flex-syntax, -F — partial support for Flex syntax: in this mode named definitions don't need the equal sign and the terminating semicolon, and when used they must be surrounded with curly braces 🫶 names without curly braces are treated as double-quoted strings.
--goto-label — use the "goto/label" code model: encode DFA in the form of labeled code blocks connected with goto transitions across blocks ⭐ this is only supported for languages that have a goto statement.
--header, --type-header, -t HEADER — generate a HEADER file 💅 the contents of the file can be specified using special blocks header:on and header:off. If conditions are used, the generated header will have a condition enum automatically appended to it (unless there's an explicit conditions block).
-I PATH — add PATH to the list of locations used when searching for include files 🎀 this option is useful in combination with an include block or directive. re2d looks for FILE in the directory of the parent file and in the include locations specified with the -I option.
--input <default | custom> — deprecated alias for --api 💀 option default corresponds to simple (it is indeed the default for most backends, but not for all). Option custom corresponds to generic.
--input-encoding <ascii | utf8> — specify the way re2d parses regular expressions ✨ with ascii (the default) re2d handles input as ASCII-encoded: any sequence of code units is a sequence of standalone 1-byte characters. With utf8 re2d handles input as UTF8-encoded and recognizes multibyte characters 🦋.
--invert-captures — flips the whole capturing vibe 🔄. Normally (...) captures and (! ...) is like "no thanks I'm just here for the group photo." With this turned on, it's the other way around: (! ...) starts capturing and (...) is suddenly camera-shy. Chaotic but valid 💅.
--lang <none | c | d | go | haskell | java | js | ocaml | python | rust | swift | v | zig> — tells re2d what language you want your shiny lexer in ✨. The supported languages are C, D, Go, Haskell, Java, JS, OCaml, Python, Rust, Swift, V, and Zig (and yes bestie, more languages can be added via user-defined syntax files — peep the --syntax option 👀). If you pass none, it disables the default syntax configs so the target language is literally undefined — main character with no backstory, iykyk.
--location-format <gnu | msvc> — controls how locations show up in messages 📍. With gnu, locations print as 'filename:line:column: ...'. With msvc, they print as 'filename(line,column) ...'. The default is gnu, no cap.
--loop-switch — uses the "loop/switch" code model, which encodes the DFA as a loop over a switch statement where each individual state is a switch case 🎀. The state lives in a variable called yystate. When you transition between states, yystate gets updated to the case label of the destination state and execution continues to the head of the loop. It's giving very "we do a little looping" energy.
--computed-continue — this one requires the --loop-switch code model, so no mixing and matching bestie 🚫. It uses computed continue statements to pick the next switch case directly, which unlocks jump-table optimization similar to what --computed-gotos does in the goto/label code model. Fancy stuff ✨.
--nested-ifs, -s — uses nested if statements instead of switch statements in conditional jumps. This usually results in more efficient code when you're working with non-optimizing compilers 💪. Lowkey a glow-up for your build.
--no-debug-info, -i — doesn't output line directives. This is super useful when the generated code lives in a version control system, because otherwise you get huge autogenerated diffs on even the tiniest change 😭. Nobody wants that chaos in their git history.
--no-generation-date — suppresses the date output in the generated file 📅. Clean and minimal, we love to see it.
--no-version — suppresses the version output in the generated file. Simple as that 🫶.
--no-unsafe — doesn't generate the unsafe wrapper over YYPEEK (this one is Rust-specific btw 🦀). For performance reasons YYPEEK should avoid bounds-checking, since the lexer already does end-of-input checks in a more efficient way. The user can choose to provide a safe YYPEEK definition, or one that's unsafe only in release builds — in that case --no-unsafe helps avoid warnings about redundant unsafe blocks. Very thoughtful, very demure.
--output, -o OUTPUT — specifies the OUTPUT file 📁. That's the one, that's the whole thing.
--posix-captures, -P — enables submatch extraction with POSIX-style capturing groups. The result gets collected into an array called yybmatch with capacity 2 * YYMAXNMATCH, and yynmatch is set to the number of groups for the matching rule 🎯. Iconic behavior honestly.
--posix-captvars — also enables submatch extraction with POSIX-style capturing groups, but this time the result is collected into variables yytl<k>, yytr<k> for the k-th capturing group 💖. Different vibe, same goal.
--recursive-functions — uses a code model based on co-recursive functions, where each DFA state is its own separate function that may call other state-functions or even itself 🌀. It's giving recursive main character energy, fr fr.
--reusable, -r — deprecated since version 2.2 (reusable blocks are allowed by default now). So like… she's retired bestie 💀, but she's still in the manual for historical reasons.
--skeleton, -S — ignores user-defined interface code and generates a self-contained "skeleton" program 🦴. It also generates input files with strings derived from the regular grammar plus compressed match results, which are used to verify "skeleton" behavior on all inputs. This option is useful for finding bugs in optimizations and code generation. Important: this option is supported only for C ⚠️.
--storable-state, -f — generates a lexer that can store its inner state 💾. This is useful in push-model lexers that get stopped by an outer program when there isn't enough input, then resumed when more input becomes available. In this mode, users should additionally define YYGETSTATE and YYSETSTATE primitives, and the variables yych, yyaccept, and state should be part of the stored lexer state. Memory queen behavior 👑.
--syntax FILE — loads configurations from the specified FILE and applies them on top of the default syntax file 📜. Note that FILE can define just a few configurations (if it's used to amend the default syntax file), or it can define a whole new language backend — and in that latter case it's recommended to use the --lang none option. Flexible and powerful, we stan.
--tags, -T — enables submatch extraction with tags 🏷️. Short flag, big energy.
--ucs2, --wide-chars, -w — generates a lexer that reads UCS2-encoded input. re2d assumes the character range is 0 – 0xFFFF and character size is 2 bytes. This option implies --nested-ifs ⚠️. Wide chars, wide possibilities 💅.
--utf8, --utf-8, -8 — generates a lexer that reads input in UTF-8 encoding. re2d assumes the character range is 0 – 0x10FFFF and character size is 1 byte 🌈. The classic, the icon, the default-adjacent choice.
--utf16, --utf-16, -x — generates a lexer that reads UTF16-encoded input. re2d assumes the character range is 0 – 0x10FFFF and character size is 2 bytes. This option implies --nested-ifs ⚠️. Sixteen bits of pure vibes.
--utf32, --unicode, -u — generates a lexer that reads UTF32-encoded input. re2d assumes the character range is 0 – 0x10FFFF and character size is 4 bytes. This option implies --nested-ifs ⚠️. Maximum character budget, slay 💖.
--verbose — outputs a short message in case of success ✅. A little "you did it bestie" moment in the terminal.
--vernum, -V — shows version information in MMmmpp format (major, minor, patch) 🔢. Cute and numeric.
--version, -v — shows version information. That's it, that's the tweet 💅.
--single-pass, -1 — deprecated. Does nothing (single pass is the default now) 💀. She's literally just here for nostalgia.
--debug-output, -d — emits YYDEBUG invocations in the generated code. Super useful for tracing lexer execution 🔍. Debug bestie mode activated.
--dump-adfa — debug option: outputs the DFA after tunneling (in .dot format) 🔬. For when you wanna see the tunnels, iykyk.
--dump-cfg — debug option: outputs the control flow graph of tag variables (in .dot format) 🕸️. Flowchart realness.
--dump-closure-stats — debug option: outputs statistics on the number of states in closure 📊. Numbers bestie, numbers.
--dump-dfa-det — debug option: outputs the DFA immediately after determinization (in .dot format) ⚡. Fresh-out-the-oven DFA, no cap.
--dump-dfa-min — debug option: outputs the DFA after minimization (in .dot format) 📉. Smaller, cleaner, still iconic.
--dump-dfa-tagopt — debug option: outputs the DFA after tag optimizations (in .dot format) 🏷️✨. Tag-optimized and thriving.
--dump-dfa-tree — debug option: outputs the DFA under construction with states represented as tag history trees (in .dot format) 🌳. Tree structure, tree drama, tree everything.
--dump-dfa-raw — debug option: outputs the DFA under construction with expanded state-sets (in .dot format) 🎨. Raw, unfiltered, expanded.
--dump-interf — debug option: outputs the interference table produced by liveness analysis of tag variables 📋. Table-core.
--dump-nfa — debug option: outputs the NFA (in .dot format) 🌸. The NFA before the DFA glow-up, we love an origin story.
--emit-dot, -D — instead of normal output, generates the lexer graph in .dot format 🗺️. The output can be converted to an image with the help of Graphviz — something like dot -Tpng -odfa.png dfa.dot. Art direction, honestly 🎨.
--dfa-minimization <moore | table> — internal option: the DFA minimization algorithm used by re2d 🔧. The moore option is the Moore algorithm (and it's the default). The table option is the "table filling" algorithm. Both algorithms should produce the same DFA up to states relabeling; table filling is simpler and much slower and serves as a reference implementation 🐢. Slow but trustworthy, we respect her.
--eager-skip — internal option: makes the generated lexer advance the input position eagerly — immediately after reading the input symbol ⏩. This changes the default behavior, where the input position is advanced lazily — after transition to the next state. Eager vs lazy, the eternal mood war.
--no-lookahead — internal option, deprecated 💀. It used to enable the TDFA(0) algorithm. Unlike TDFA(1), the TDFA(0) algorithm doesn't use one-symbol lookahead. It applies register operations to the incoming transitions rather than the outgoing ones. Benchmarks showed that TDFA(0) is less efficient than TDFA(1). So yeah, she's retired for a reason bestie.
--no-optimize-tags — internal option: suppresses optimization of tag variables (useful for debugging) 🐛. Sometimes you gotta turn off the optimization to see what's really going on.
--posix-closure <gor1 | gtop> — internal option: specifies the shortest-path algorithm used for constructing epsilon-closure with POSIX disambiguation semantics 🔀. gor1 (the default) stands for the Goldberg-Radzik algorithm, and gtop stands for "global topological order." Two algorithms, one goal. Pick your fighter.
--posix-prectable <complex | naive> — internal option: specifies the algorithm used to compute the POSIX precedence table 📐. The complex algorithm computes the precedence table in one traversal of the tag history tree and has quadratic complexity in the number of TNFA states — and it's the default. The naive algorithm has worst-case cubic complexity in the number of TNFA states, but it's much simpler than complex and may be slightly faster in non-pathological cases. Complex is default, naive is the underdog, both valid 💖.
--stadfa — internal option, deprecated 💀. It used to enable the staDFA algorithm, which differs from TDFA in that register operations are placed in states rather than on transitions. Benchmarks showed staDFA is less efficient than TDFA. Another retired queen, we remember her.
--fixed-tags <none | toplevel | all> — internal option: specifies whether the fixed-tag optimization should be applied to all tags (all), none of them (none), or only those in toplevel concatenation (toplevel). The default is all ✨. "Fixed" tags are the ones located within a fixed distance to some other tag (called "base"). In those cases, only the base tag needs to be tracked, and the value of the fixed tag can be computed as the base tag's value plus a static offset 🎯. For tags under alternative or repetition, you also need to check if the base tag has a no-match value — if so, the fixed tag should also be set to no-match, disregarding the offset. For tags in top-level concatenation the check isn't needed, because they always match. Big brain optimization, no cap 🧠.

⚠️ Warnings ⚠️

Warnings can be individually enabled, disabled, and even turned into an error 😱. Choose your own adventure, bestie.

-W — turns on all the warnings 🚨. Every single one. Maximum awareness mode.
-Werror — turns warnings into errors 💀. Note: this option alone doesn't turn on any warnings — it only affects those warnings that have been turned on so far or will be turned on later. So it's like a multiplier, not a summoner. Very important distinction, don't get it twisted.
-W<warning> — turns on that specific warning 💡. Targeted, precise, iconic.
-Wno-<warning> — turns off that specific warning 🙈. Sometimes you just need peace and quiet, and that's valid.
-Werror-<warning> — turns that warning on and promotes it to a full-on error, bestie. This implies -W<warning> too, so you don't need both. 💅
-Wno-error-<warning> — okay so this one demotes a specific warning back down so it's not an error anymore. Careful though: it does not turn the warning itself off. It still yells at you, just less dramatically. 🙈
-Wcondition-order — screams if your generated program is out here making implicit assumptions about condition numbering. Rude. The fix: use the --header option or a conditions block to generate a mapping of condition names to numbers, then use the autogenerated condition names. Slay responsibly. ✨
-Wempty-character-class — warns you when a regular expression contains an empty character class. Like, trying to match an empty character class literally makes no sense — it should always fail. But for backwards compatibility reasons re2d permits empty character classes and treats them as empty strings. Use the --empty-class option to change the default behavior. 🫶
-Wmatch-empty-string — warns if a rule is nullable (i.e. it matches an empty string). Bestie, if the lexer runs in a loop and that empty match was not intentional, your lexer may unexpectedly hang in an infinite loop. Not the vibe. 💀
-Wswapped-range — warns if the lower bound of a range is greater than its upper bound. Default behavior? Silently swap the range bounds. Iconic chaos, honestly. 🔥
-Wundefined-control-flow — warns if some input strings cause undefined control flow in the lexer (and it reports the faulty patterns). This is a dangerous and common mistake, ngl. It's easily fixed by adding the default rule *, which has the lowest priority, matches any code unit, and always consumes a single code unit. 👑
-Wunreachable-rules — warns about rules that are shadowed by other rules and will never match. Living in someone else's shadow, tragic. 😭
-Wdeprecated-eof_rule — warns about standalone end of input rules $ that will be broken by future changes and require fixing. At the moment these rules take precedence when conflicting with other rules, but after the introduction of the generalized end of input symbol $, precedence order will change and these rules will become shadowed by other rules. So like, fix them now fr fr.
-Wuseless-escape — warns if a symbol is escaped when it shouldn't be. By default re2d silently ignores such escapes, but this may as well indicate a typo or an error in the escape sequence. 🍓
-Wnondeterministic-tags — warns if a tag has n-th degree of nondeterminism, where n is greater than 1. That's a lot of degrees, bestie. 🎀
-Wsentinel-in-midrule — warns if the sentinel symbol occurs in the middle of a rule — which may cause reads past the end of buffer, crashes or memory corruption in the generated lexer. Yikes. This warning is only applicable if the sentinel method of checking for the end of input is used. It is set to an error if re2c:sentinel configuration is used. 💀
-Wundefined-syntax-config — warns if the syntax file specified with --syntax option is missing definitions of some configurations. So helpful honestly — it helps maintain user-defined syntax files: if a new release adds configurations, an old syntax file will raise a warning and the user gets notified. And if some configurations are unused and don't need a definition, they should be explicitly set to <undefined>. 🌸

✨ Syntax files ✨

So the support for different languages in re2c is based on the idea of syntax files. A syntax file is a configuration file that defines the syntax of the target language — not the whole language, just a small part of it that's used by the generated code. Syntax files make re2c super flexible, but bestie, they should not be used as a replacement for re2c: configurations — their purpose is to define syntax of the target language, not to customize one particular lexer. 🦋

All supported languages have default syntax files that are part of the distribution (see include/syntax subdirectory); they're also embedded in the re2d binary. Users may provide a custom syntax file that overrides a few configurations for one of the supported languages, or they may choose to redefine all configurations (in that case the --lang none option should be used). Syntax files contain configurations of four different kinds: feature lists, language configurations, inplace configurations and code templates.

🌈 Feature lists

A few list configurations define various features supported by a given backend, so that re2d may give a clear error if the user tries to enable an unsupported feature:

supported_apis — a list of supported APIs with possible elements simple, record, generic.
supported_api_styles — a list of supported API styles with possible elements functions, free-form.
supported_code_models — a list of supported code models with possible elements goto-label, loop-switch, recursive-functions.
supported_targets — a list of supported codegen targets with possible elements code, dot, skeleton.
supported_features — a list of supported features with possible elements nested-ifs, bitmaps, computed-gotos, case-ranges, monadic, unsafe, tags, captures, captvars, computed-continue.

💜 Language configurations

A few boolean configurations describe features of the target language that affect the re2d parser and code generator:

semicolons — non-zero if the language uses semicolons after statements.
backtick_quoted_strings — non-zero if the language has backtick-quoted strings.
single_quoted_strings — non-zero if the language has single-quoted strings.
indentation_sensitive — non-zero if the language is indentation sensitive.
wrap_blocks_in_braces — non-zero if compound statements must be wrapped in curly braces.

🎀 Inplace configurations

Syntax files define initial values of all re2c: configurations, as they may differ for different languages. See the configurations section for a full list of all inplace configurations and their meaning.

🌟 Code templates

Code templates define the syntax of the target language. They're written in a simple domain-specific language with the following formal grammar:

code-template ::
      name '=' code-exprs ';'
    | CODE_TEMPLATE ';'
    | '<undefined>' ';'

code-exprs ::
      <EMPTY>
    | 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 ']'

A code template is a sequence of string literals, variables, optional elements and lists, or a reference to another code template, or a special value <undefined>. Variables are placeholders that are substituted during the code generation phase. List variables are special: when expanding list templates, re2d repeats expressions on the right hand side of the column a few times, each time replacing occurrences of the list variable with a value specific to this repetition. Lists have optional bounds (negative values are counted from the end, e.g. -1 means the last element). Conditional names start with a dot. Both conditionals and variables may be either local (specific to the given code template) or global (allowed in all code templates). When re2d reads a syntax file, it checks that each code template uses only the variables and conditionals that are allowed in it. ✨

For example, the following code template defines an if-then-else construct for a C-like language:

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;

Here branch is a list variable: branch{0} expands to the first branch (which is special, as there is no else part), branch{1:-1} expands to all remaining branches (if any). stmt is also a list variable: [stmt: stmt] is a nested list that expands to a list of statements in the body of the current branch. topindent, indent, dedent and nl are global variables, and .cond is a local conditional (their meaning is described below). This code template could produce the following code:

if x {
    // do something
} else if y {
    // do something else
} else {
    // don't do anything
}

Here's a list of all code templates supported by re2d with their local variables and conditionals. Note that a particular definition may, but does not have to, use local variables and conditionals. Any unused code templates should be set to <undefined>. 💅

code:var_local — declaration or definition of a local variable. Supported variables: type (the type of the variable), name (its name) and init (initial value, if any). Conditionals: .init (true if there is an initializer).
code:var_global — same as code:var_local, except that it's used in top-level.
code:const_local — definition of a local constant. Supported variables: type (the type of the constant), name (its name) and init (initial value).
code:const_global — same as code:const_local, except that it's used in top-level.
code:array_local — definition of a local array (table). Supported variables: type (the type of array elements), name (array name), size (its size), row (a list variable that does not itself produce any code, but expands the list expression as many times as there are rows in the table) and elem (a list variable that expands to all table elements in the current row — it's meant to be nested in the row list). Supported conditional: .const (true if the array is immutable).
code:array_global — same as code:array_local, except that it's used in top-level.
code:array_elem — reference to an element of an array (table). Supported variables: array (the name of the array) and index (index of the element).
code:enum — the definition of an enumeration, bestie ✨ it can be defined using a special language construct for enumerations, or simply as a few standalone constants. Supported variables are type (user-defined enumeration type or type of the constants), elem (list variable that expands to the name of each member) and init (initializer for each member). Conditionals: .init (true if there is an initializer).
code:enum_elem — one enumeration element 💅 that’s a member of a user-defined enumeration type or a name of a constant, depending on how code:enum is defined. Supported variables are name (the name of the element) and type (its type).
code:assign — assignment statement. Supported variables are lhs (left hand side) and rhs (right hand side).
code:type_int — signed integer type.
code:type_uint — unsigned integer type.
code:type_yybm — type of elements in the yybm table.
code:type_yytarget — type of elements in the yytarget table.
code:type_yyctable — type of elements in the yyctable table.
code:cmp_eq — operator "equals".
code:cmp_ne — operator "not equals".
code:cmp_lt — operator "less than".
code:cmp_gt — operator "greater than".
code:cmp_le — operator "less or equal".
code:cmp_ge — operator "greater or equal".
code:if_then_else — if-then-else statement with one or more branches 🦋 supported variables: branch (a list variable that does not itself produce any code, but expands list expression as many times as there are branches), cond (condition of the current branch) and stmt (a list variable that expands to all statements in the current branch). Conditionals: .cond (true if the current branch has a condition), .many (true if there’s more than one branch).
code:if_then_else_oneline — a specialization of code:if_then_else for the case when all branches have one-line statements. If this is <undefined>, code:if_then_else is used instead.
code:switch — a switch statement with one or more cases. Supported variables: expr (the switched-on expression) and case (a list variable that expands to all cases-groups with their code blocks).
code:switch_cases — a group of switch cases that maps to a single code block. Supported variables are case (a list variable that expands to all cases in this group) and stmt (a list variable that expands to all statements in the code block).
code:switch_cases_oneline — a specialization of code:switch_cases for the case when the code block consists of a single one-line statement. If this is <undefined>, code:switch_cases is used instead.
code:switch_case_range — a single switch case that covers a range of values (possibly consisting of a single value). Supported variable: val (a list variable that expands to all values in the range). Supported conditionals: .many (true if there’s more than one value in the range) and .char_literals (true if this is a switch on character literals -- some languages provide special syntax for this case).
code:switch_case_default — default switch case.
code:loop — a loop that runs forever (unless interrupted from the loop body) 🔁 supported variables: label (loop label), val (initial state), var (state variable), state (a list variable that expands to all state names) and stmt (a list variable that expands to all statements in the loop body or the current state).
code:continue — continue statement. Supported variables: label (label from which to continue execution), val (value passed to a labeled switch) and var (state variable).
code:goto — goto statement. Supported variables: label (label of the jump target).
code:cgoto — computed goto statement. Supported variables: array (the table containing computed goto information), index (index of the element in the table) and base (base label, only used if .cgoto.relative is true).
code:cgoto:data — initializer expression for a single element in computed goto table. Supported variables: label (the label that is used to initialize the current element), type (underlying type of the elements in the table) and base (base label - only used if .cgoto.relative is true).
code:fndecl — function declaration. Supported variables: name (function name), type (return type), throw (exceptions thrown by this function, maps to re2c:yyfn:throw configuration), arg (a list variable that does not itself produce code, but expands list expression as many times as there are function arguments), argname (name of the current argument), argtype (type of the current argument). Conditional: .type (true if this is a non-void function).
code:fndef — like code:fndecl, but used for function definitions, so it has one additional list variable stmt that expands to all statements in the function body.
code:fncall — function call statement. Supported variables: name (function name), retval (l-value where the return value is stored, if any) and arg (a list variable that expands to all function arguments). Conditionals: .args (true if the function has arguments) and .retval (true if return value needs to be saved).
code:tailcall — tail call statement. Supported variables: name (function name), and arg (a list variable that expands to all function arguments). Conditionals: .args (true if the function has arguments) and .retval (true if this is a non-void function).
code:recursive_functions — program body with --recursive-functions code model. Supported variables: fn (a list variable that does not itself produce any code, but expands list expression as many times as there are functions), fndecl (declaration of the current function) and fndef (definition of the current function).
code:fingerprint — the fingerprint at the top of the generated output file 🎀 supported variables: ver (re2d version that was used to generate this) and date (generation date).
code:line_info — the format of line directives (if this is set to <undefined>, no directives are generated). Supported variables: line (line number) and file (filename).
code:abort — a statement that aborts program execution.
code:yydebug — YYDEBUG statement, possibly specialized for different APIs. Supported variables: YYDEBUG, yyrecord, yych (map to the corresponding re2c: configurations), state (DFA state number).
code:yypeek — YYPEEK statement, possibly specialized for different APIs. Supported variables: YYPEEK, YYCTYPE, YYINPUT, YYCURSOR, yyrecord, yych (map to the corresponding re2c: configurations). Conditionals: .cast (true if re2c:yych:conversion is set to non-zero).
code:yyskip — YYSKIP statement, possibly specialized for different APIs. Supported variables: YYSKIP, YYCURSOR, yyrecord (map to the corresponding re2c: configurations).
code:yybackup — YYBACKUP statement, possibly specialized for different APIs. Supported variables: YYBACKUP, YYCURSOR, YYMARKER, yyrecord (map to the corresponding re2c: configurations).
code:yybackupctx — YYBACKUPCTX statement, possibly specialized for different APIs. Supported variables: YYBACKUPCTX, YYCURSOR, YYCTXMARKER, yyrecord (map to the corresponding re2c: configurations).
code:yyskip_yypeek — combined code:yyskip and code:yypeek statement (defaults to code:yyskip followed by code:yypeek).
code:yypeek_yyskip — combined code:yypeek and code:yyskip statement (defaults to code:yypeek followed by code:yyskip).
code:yyskip_yybackup — combined code:yyskip and code:yybackup statement (defaults to code:yyskip followed by code:yybackup).
code:yybackup_yyskip — combined code:yybackup and code:yyskip statement (defaults to code:yybackup followed by code:yyskip).
code:yybackup_yypeek — combined code:yybackup and code:yypeek statement (defaults to code:yybackup followed by code:yypeek).
code:yyskip_yybackup_yypeek — combined code:yyskip, code:yybackup and code:yypeek statement (defaults to code:yyskip followed by code:yybackup followed by code:yypeek).
code:yybackup_yypeek_yyskip — combined code:yybackup, code:yypeek and code:yyskip statement (defaults to code:yybackup followed by code:yypeek followed by code:yyskip).
code:yyrestore — YYRESTORE statement, possibly specialized for different APIs. Supported variables: YYRESTORE, YYCURSOR, YYMARKER, yyrecord (map to the corresponding re2c: configurations).
code:yyrestorectx — YYRESTORECTX statement, possibly specialized for different APIs. Supported variables: YYRESTORECTX, YYCURSOR, YYCTXMARKER, yyrecord (map to the corresponding re2c: configurations).
code:yyrestoretag — YYRESTORETAG statement, possibly specialized for different APIs. Supported variables: YYRESTORETAG, YYCURSOR, yyrecord (map to the corresponding re2c: configurations), tag (the name of tag variable used to restore position).
code:yyshift — YYSHIFT statement, possibly specialized for different APIs. Supported variables: YYSHIFT, YYCURSOR, yyrecord (map to the corresponding re2c: configurations), offset (the number of code units to shift the current position).
code:yyshiftstag — the YYSHIFTSTAG statement, and yes it can be specialized for different APIs. Supported variables: YYSHIFTSTAG, yyrecord, negative (map to the corresponding re2c: configurations), tag (the tag variable that needs to be shifted), and offset (the number of code units to shift). Conditionals: .nested (true if this is a nested tag — in this case its value may equal to re2c:tags:negative, which should not be shifted).
code:yyshiftmtag — the YYSHIFTMTAG statement, possibly specialized for different APIs. Supported variables: YYSHIFTMTAG (maps to the corresponding re2c: configuration), tag (the tag variable that needs to be shifted), and offset (the number of code units to shift).
code:yystagp — the YYSTAGP statement, possibly specialized for different APIs. Supported variables: YYSTAGP, YYCURSOR, yyrecord (map to the corresponding re2c: configurations), and tag (the tag variable that should be updated).
code:yymtagp — the YYMTAGP statement, possibly specialized for different APIs. Supported variables: YYMTAGP (maps to the corresponding re2c: configuration) and tag (the tag variable that should be updated).
code:yystagn — the YYSTAGN statement, possibly specialized for different APIs. Supported variables: YYSTAGN, negative, yyrecord (map to the corresponding re2c: configurations), and tag (the tag variable that should be updated).
code:yymtagn — the YYMTAGN statement, possibly specialized for different APIs. Supported variables: YYMTAGN (maps to the corresponding re2c: configuration) and tag (the tag variable that should be updated).
code:yycopystag — the YYCOPYSTAG statement, possibly specialized for different APIs. Supported variables: YYCOPYSTAG, yyrecord (map to the corresponding re2c: configurations), lhs, and rhs (left and right hand side tag variables of the copy operation).
code:yycopymtag — the YYCOPYMTAG statement, possibly specialized for different APIs. Supported variables: YYCOPYMTAG, yyrecord (map to the corresponding re2c: configurations), lhs, and rhs (left and right hand side tag variables of the copy operation).
code:yygetaccept — the YYGETACCEPT statement, possibly specialized for different APIs. Supported variables: YYGETACCEPT, yyrecord (map to the corresponding re2c: configurations), and var (maps to the re2c:yyaccept configuration).
code:yysetaccept — the YYSETACCEPT statement, possibly specialized for different APIs. Supported variables: YYSETACCEPT, yyrecord (map to the corresponding re2c: configurations), var (maps to the re2c:yyaccept configuration), and val (numeric value of the accepted rule).
code:yygetcond — the YYGETCOND statement, possibly specialized for different APIs. Supported variables: YYGETCOND, yyrecord (map to the corresponding re2c: configurations), and var (maps to the re2c:yycond configuration).
code:yysetcond — the YYSETCOND statement, possibly specialized for different APIs. Supported variables: YYSETCOND, yyrecord (map to the corresponding re2c: configurations), var (maps to the re2c:yycond configuration), and val (numeric condition identifier).
code:yygetstate — the YYGETSTATE statement, possibly specialized for different APIs. Supported variables: YYGETSTATE, yyrecord (map to the corresponding re2c: configurations), and var (maps to the re2c:yystate configuration).
code:yysetstate — the YYSETSTATE statement, possibly specialized for different APIs. Supported variables: YYSETSTATE, yyrecord (map to the corresponding re2c: configurations), var (maps to the re2c:yystate configuration), and val (state number).
code:yylessthan — the YYLESSTHAN statement, possibly specialized for different APIs. Supported variables: YYLESSTHAN, YYCURSOR, YYLIMIT, yyrecord (map to the corresponding re2c: configurations), and need (the number of code units to check against). Conditional: .many (true if the need is more than one).
code:yyend — the YYEND expression, possibly specialized for different APIs. Supported variables: YYEND, YYCURSOR, and YYLIMIT.
code:yybm_filter — the condition that is used to filter out yych values that are not covered by the yybm table (used with --bitmaps option). Supported variable: yych (maps to the re2c:yych configuration).
code:yybm_match — the format of the yybm table check (generated with --bitmaps option). Supported variables: yybm, yych (map to the corresponding re2c: configurations), offset (offset in the yybm table that needs to be added to yych), and mask (bit mask that should be applied to the table entry to retrieve the boolean value that needs to be checked).
code:yytarget_filter — the condition that is used to filter out yych values that are not covered by the yytarget table (used with --computed-gotos option). Supported variable: yych (maps to the re2c:yych configuration).

Okay bestie, here's the full list of global variables that are allowed in syntax files:

nl — a newline.
indent — a variable that does not produce any code, but has a side-effect of increasing indentation level.
dedent — a variable that does not produce any code, but has a side-effect of decreasing indentation level.
topindent — indentation string for the current statement. Indentation level is tracked and automatically updated by the code generator.

And here's the full list of global conditionals that are allowed in syntax files:

.api.simple — true if the simple API is used (--api simple or re2c:api = simple).
.api.generic — true if the generic API is used (--api generic or re2c:api = generic).
.api.record — true if the record API is used (--api record or re2c:api = record).
.api_style.functions — true if the function-like API style is used (re2c:api-style = functions).
.api_style.freeform — true if the free-form API style is used (re2c:api-style = free-form).
.case_ranges — true if the case ranges feature is enabled (--case-ranges or re2c:case-ranges = 1).
.cgoto.relative — true if the relative form of computed goto is used (--computed-gotos-relative or re2c:cgoto:relative = 1).
.computed_continue — true if the loop/switch model uses labeled-switch transitions (--computed-continue).
.code_model.goto_label — true if the code model based on goto/label is used (--goto-label).
.code_model.loop_switch — true if the code model based on loop/switch is used (--loop-switch).
.code_model.recursive_functions — true if the code model based on recursive functions is used (--recursive-function).
.date — true if the generated fingerprint should contain the generation date.
.loop_label — true if re2d generated loops must have a label (re2c:label:yyloop is set to a nonempty string).
.monadic — true if the generated code should be monadic (re2c:monadic = 1). This is only relevant for pure functional languages.
.start_conditions — true if start conditions are enabled (--start-conditions).
.storable_state — true if storable state is enabled (--storable-state).
.unsafe — true if re2d should use "unsafe" blocks in order to generate faster code (--unsafe, re2c:unsafe = 1). This is only relevant for languages that have the "unsafe" feature.
.version — true if the generated fingerprint should contain the re2d version.
.yyfill.enable — true if YYFILL is enabled (re2c:yyfill:enable = 1).
.yyfn.throw — true if the re2c:yyfn:throw configuration is defined to a nonempty string.

🌈 HANDLING THE END OF INPUT 🌈

One of the main problems for the lexer is knowing when to stop. There are a few terminating conditions:

the lexer may match some rule (including default rule *) and come to a final state
the lexer may fail to match any rule and come to a default state
the lexer may reach the end of input

The first two conditions terminate the lexer in a "natural" way: it comes to a state with no outgoing transitions, and the matching automatically stops. The third condition, end of input, is different: it may happen in any state, and the lexer should be able to handle it. Checking for the end of input interrupts the normal lexer workflow and adds conditional branches to the generated program, therefore it is necessary to minimize the number of such checks. re2d supports a few different methods for handling the end of input. Which one to use depends on the complexity of regular expressions, the need for buffering, performance considerations and other factors. Here is a list of methods:

Sentinel. This method eliminates the need for the end of input checks altogether. It is simple and efficient, but limited to the case when there is a natural "sentinel" character that can never occur in valid input. This character may still occur in invalid input, but it should not be allowed by the regular expressions, except perhaps as the last character of a rule. The sentinel is appended at the end of input and serves as a stop signal: when the lexer reads this character, it is either a syntax error or the end of input. In both cases the lexer should stop. This method is used if YYFILL is disabled with re2c:yyfill:enable = 0; and re2c:eof has the default value -1.
Sentinel with bounds checks — okay so this one is generic in the best way: it lets you handle any input with zero restrictions on your regexes 💅 The whole idea is to cut down on end-of-input checks by only doing them on certain characters. Same energy as the plain "sentinel" method — you pick one character as the "sentinel" and slap it at the end of the input — except there's no restriction on where the sentinel can show up (fr, literally any character can be your sentinel). When the lexer reads that character, it also does a bounds check. If the current position is still within bounds, the lexer just keeps matching and treats the sentinel like a normal character. If it's out of bounds, it invokes YYFILL (unless that's disabled). If more input gets supplied, the lexer rematches that last character and carries on like the sentinel was never there. Otherwise it's the real end of input and the lexer stops. This method kicks in when re2c:eof has a non-negative value (set it to the numeric value of your sentinel). YYFILL is optional. ✨
Bounds checks with padding — this one is generic too, and it can be faster than "sentinel with bounds checks", but it's also more complex (highkey a little extra 😅). The idea: partition the DFA states into strongly connected components (SCCs) and generate a single check per SCC for enough characters to cover the longest non-looping path in that SCC. Fewer checks, slay — but there's a catch with short lexemes at the end of input, because the check needs enough characters to cover the longest lexeme. Fix: pad the input with a few fake characters that don't form a valid lexeme suffix (so the lexer literally cannot match them). The padding length should be YYMAXFILL, generated with a max block. If there isn't enough input, the lexer invokes YYFILL, which should supply at least the required number of characters or not return. This method is used if YYFILL is enabled and re2c:eof is -1 (that's the default config, btw 🎀).
Custom checks — the generic API lets you override basic operations like reading a character, which means you can fold the end-of-input checks right into them. Cute, but ⚠️ this approach is error-prone and should be used with caution. To use a custom method, enable the generic API with --api custom or re2c:api = custom; and disable the default bounds checks with re2c:yyfill:enable = 0; or re2c:yyfill:check = 0;.

The following subsections contain an example of each method. Let's gooo 🦋

🌸 Sentinel

This example uses a sentinel character to handle the end of input. The program counts space-separated words in a null-terminated string. The sentinel is null: it's the last character of each input string, and none of the rules allow it in the middle of a lexeme (in particular, it's not included in character ranges where it'd be easy to overlook 👀). If a null shows up in the middle of a string, that's a syntax error and the lexer will match the default rule *, but it won't read past the end of input or crash (use the -Wsentinel-in-midrule warning and the re2c:sentinel configuration to verify this). The configuration re2c:yyfill:enable = 0; suppresses the generation of bounds checks and YYFILL invocations. 🫶

// re2d $INPUT -o $OUTPUT
module main;

// Expect a null-terminated string.
private int lex(const(char)* yycursor) {
    uint count = 0;

    for (;;) {
    /*!re2c
        re2c:YYCTYPE = char;
        re2c:yyfill:enable = 0;

        *      { return -1; }
        [\x00] { return count; }
        [a-z]+ { ++count; continue; }
        [ ]+   { continue; }
    */
    }
    assert(0); // unreachable
}

void main() {
    assert(lex("") == 0);
    assert(lex("one two three") == 3);
    assert(lex("f0ur") == -1);
}

💜 Sentinel with bounds checks

This example uses sentinel with bounds checks to handle the end of input (this method was added in version 1.2, ok history icon 📜). The program counts space-separated single-quoted strings. The sentinel character is null, specified with the re2c:eof = 0; configuration. Just like in the sentinel method, null is the last character of each input string — but here it's allowed in the middle of a rule (for example, 'aaa\0aa'\0 is valid input, but 'aaa\0 is a syntax error 😭). Bounds checks are generated in each state that matches an input character, but they're scoped to the branch that handles null. Checks look like YYLIMIT <= YYCURSOR or YYLESSTHAN(1) with the generic API. If the check condition is true, the lexer has hit the end of input and should stop (YYFILL is disabled with re2c:yyfill:enable = 0; because the input fits into one buffer — see the YYFILL with sentinel section for an example that actually uses YYFILL). Reaching the end of input opens three possibilities: if the lexer is in the initial state it'll match the end-of-input rule $, otherwise it may fallback to a previously matched rule (including default rule *) or go to a default state, which causes -Wundefined-control-flow. 💀

// re2d $INPUT -o $OUTPUT
module main;

// Expect a null-terminated string.
private int lex(immutable char[] s) {
    const(char)* yycursor = s.ptr, yylimit = s.ptr + s.length, yymarker;
    int count = 0;

    for (;;) {
    /*!re2c
        re2c:YYCTYPE = char;
        re2c:yyfill:enable = 0;
        re2c:eof = 0;

        str = ['] ([^'\\] | [\\][^])* ['];

        *    { return -1; }
        $    { return count; }
        str  { ++count; continue; }
        [ ]+ { continue; }
    */
    }
    assert(0); // unreachable
}

void main() {
    assert(lex("") == 0);
    assert(lex("'qu\0tes' 'are' 'fine: \\'' ") == 3);
    assert(lex("'unterminated\\'") == -1);
}

🎀 Bounds checks with padding

This example uses bounds checks with padding to handle the end of input (this method is enabled by default, bestie — it's the one you get for free ✨). The program counts space-separated single-quoted strings. There's a padding of YYMAXFILL null characters appended at the end of input, where the YYMAXFILL value is autogenerated with a max block. You don't have to use null for padding — any characters work as long as they don't form a valid lexeme suffix (in this example, padding should not contain single quotes, since they might get mistaken for a suffix of a single-quoted string 🙈). There's a "stop" rule that matches the first padding character (null) and terminates the lexer (note that it checks whether null is at the beginning of the padding — otherwise it's a syntax error 👑). Bounds checks are generated only in some states, determined by the strongly connected components of the underlying automaton. Checks have the form (YYLIMIT - YYCURSOR) < n or YYLESSTHAN(n) with the generic API, where n is the minimum number of characters needed for the lexer to proceed (which also means the next bounds check will happen in at most n characters). If the check condition is true, the lexer has reached the end of input and will invoke YYFILL(n), which should either supply at least n input characters or not return. In this example YYFILL always fails and terminates the lexer with an error (totally fine, since the input fits into one buffer). See the YYFILL with padding section for an example that actually refills the input buffer with YYFILL. 🔥

// re2d $INPUT -o $OUTPUT
module main;

import core.stdc.stdlib;
import core.stdc.string;

/*!max:re2c*/

private int lex(immutable char[] s) {
    // Make a copy of the string with YYMAXFILL zeroes at the end.
    char *buf = cast(char*) malloc(s.length + YYMaxFill);
    memcpy(buf, cast(const(void*)) s, s.length);
    memset(buf + s.length, 0, YYMaxFill);

    const(char)* yycursor = buf;
    const(char)* yylimit = buf + s.length + YYMaxFill;
    int count = 0;

loop:
    /*!re2c
        re2c:YYCTYPE = char;
        re2c:YYFILL = "goto fail;";

        str = ['] ([^'\\] | [\\][^])* ['];

        [\x00] {
            // Check that it is the sentinel, not some unexpected null.
            if (yycursor - 1 == buf + s.length) goto exit; else goto fail;
        }
        str  { ++count; goto loop; }
        [ ]+ { goto loop; }
        *    { goto fail; }
    */
fail:
    count = -1;
exit:
    free(buf);
    return count;
}

void main() {
    assert(lex("") == 0);
    assert(lex("'qu\0tes' 'are' 'fine: \\'' ") == 3);
    assert(lex("'unterminated\\'") == -1);
    assert(lex("'unexpected \0 null\\'") == -1);
}

🦋 Custom checks

This example uses a custom end-of-input handling method based on the generic API. The program counts space-separated single-quoted strings. It's the same as the sentinel example, except the input is not null-terminated. To cover for the missing sentinel character at the end of input, YYPEEK is redefined to do a bounds check before it reads the next input character. Ngl this is inefficient, because those checks happen very often 😭. If the check condition fails, YYPEEK returns the real character; otherwise it returns a fake sentinel character. 🧚

// re2d $INPUT -o $OUTPUT
module main;

import core.stdc.stdlib;
import core.stdc.string;

private int lex(immutable char[] s) {
    // For the sake of example create a string without terminating null.
    char *buf = cast(char*) malloc(s.length);
    memcpy(buf, cast(const(void*)) s, s.length);

    const(char) *cur = buf, lim = buf + s.length;
    int count = 0;

    for (;;) {
    /*!re2c
        re2c:api = generic;
        re2c:yyfill:enable = 0;
        re2c:YYCTYPE = char;
        re2c:YYPEEK = "cur < lim ? *cur : 0";  // fake null
        re2c:YYSKIP = "++cur;";

        *      { count = -1; break; }
        [\x00] { break;{} }
        [a-z]+ { ++count; continue;{} }
        [ ]+   { continue; }
    */
    }

    free(buf);
    return count;
}

void main() {
    assert(lex("") == 0);
    assert(lex("one two three ") == 3);
    assert(lex("f0ur") == -1);
}

💾 BUFFER REFILLING 💾

okay so bestie, buffering is the whole thing when the input can't just be mapped in memory all at once — either it's too large, or it comes in a streaming fashion (like reading from a socket, iykyk). The usual technique in these cases is to allocate a fixed-sized memory buffer and process input in chunks that fit into the buffer. When the current chunk is processed, it gets moved out and new data gets moved in. In practice it's somewhat more complex, because lexer state isn't just one single input position — it's a whole little friend group of interrelated positions:

cursor — the next input character to be read (YYCURSOR in the C pointer API or YYSKIP/YYPEEK in the generic API).
limit — the position after the last available input character (YYLIMIT in the C pointer API, implicitly handled by YYLESSTHAN in the generic API).
marker — the position of the most recent match, if any (YYMARKER in the default API or YYBACKUP/YYRESTORE in the generic API).
token — the start of the current lexeme (implicit in the re2d API, since it's not needed for the normal lexer operation and can be defined and updated by the user).
context marker — the position of the trailing context (YYCTXMARKER in the C pointer API or YYBACKUPCTX/YYRESTORECTX in the generic API).
tag variables — submatch positions (defined with stags and mtags blocks and the generic API primitives YYSTAGP/YYSTAGN/YYMTAGP/YYMTAGN).

Not all of these are used in every case, but if they are used, they must be updated by YYFILL. All active positions are contained in the segment between token and cursor, so everything between buffer start and token can be discarded, the segment from token up to limit should be moved to the beginning of the buffer, and the free space at the end of the buffer should be filled with new data. In order to avoid frequent YYFILL calls it's best to fill in as many input characters as possible (even though fewer characters might suffice to resume the lexer). The details of YYFILL implementation are slightly different depending on which EOF handling method is used: the case of the EOF rule is somewhat simpler than the case of bounds-checking with padding. Also note that if the -f --storable-state option is used, YYFILL has slightly different semantics (described in the section about storable state).

✨ YYFILL with sentinel ✨

If the EOF rule is used, YYFILL is a function-like primitive that accepts no arguments and returns a value which is checked against zero. YYFILL invocation is triggered by the condition YYLIMIT <= YYCURSOR in the C pointer API and YYLESSTHAN() in the generic API. A non-zero return value means that YYFILL has failed. A successful YYFILL call must supply at least one character and adjust the input positions accordingly. limit must always be set to one after the last input position in the buffer, and the character at the limit position must be the sentinel symbol specified by the re2c:eof configuration. The pictures below show the relative locations of input positions in the buffer before and after the YYFILL call (the sentinel symbol is marked with #, and the second picture shows the case when there isn't enough input to fill the whole buffer).

                         <-- 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

Here's an example of a program that reads the input file input.txt in chunks of 4096 bytes and uses the EOF rule.

          // re2d $INPUT -o $OUTPUT
          module main;

          import core.stdc.string;
          import core.stdc.stdio;

          enum BUFSIZE = 4095;

          struct Input {
              FILE* file;
              char[BUFSIZE + 1] buffer;// +1 for sentinel
              char* yylimit, yycursor, yymarker, token;
              bool eof;
          };

          private int fill(ref Input it) {
              if (it.eof) return 1;

              const size_t shift = it.token - it.buffer.ptr;
              const size_t used = it.yylimit - it.token;

              // Error: lexeme too long. In real life could reallocate a larger buffer.
              if (shift < 1) return 2;

              // Shift buffer contents (discard everything up to the current token).
              memmove(cast(void*)it.buffer.ptr, it.token, used);
              it.yylimit -= shift;
              it.yycursor -= shift;
              it.yymarker -= shift;
              it.token -= shift;

              // Fill free space at the end of buffer with new data from file.
              it.yylimit += fread(it.yylimit, 1, BUFSIZE - used, it.file);
              it.yylimit[0] = 0;
              it.eof = it.yylimit < (it.buffer.ptr + BUFSIZE);
              return 0;
          }

          private int lex(ref Input yyrecord) {
              int count = 0;
              for (;;) {
                  yyrecord.token = yyrecord.yycursor;
              /*!re2c
                  re2c:api = record;
                  re2c:YYCTYPE = "char";
                  re2c:YYFILL = "fill(yyrecord) == 0";
                  re2c:eof = 0;

                  str = ['] ([^'\\] | [\\][^])* ['`;`? Wait I need to be exact. Let me correct: the original line is `str = ['] ([^'\\] | [\\][^])* ['`;`? No, I must use the exact line. Let me paste it correctly:
              */
              }
              assert(0);
          }

          void main() {
              const char[] fname = "input";
              const char[] content = "'qu\0tes' 'are' 'fine: \\'' ";

              // Prepare input file: a few times the size of the buffer, containing
              // strings with zeroes and escaped quotes.
              FILE* f = fopen(fname.ptr, "w");
              for (int i = 0; i < BUFSIZE; ++i) {
                  fwrite(cast(const(void*)) content.ptr, 1, content.length - 1, f);
              }
              fclose(f);
              int count = 3 * BUFSIZE; // number of quoted strings written to file

              // Initialize lexer state: all pointers are at the end of buffer.
              Input it;
              it.file = fopen(fname.ptr, "r");
              it.yycursor = it.yymarker = it.token = it.yylimit = it.buffer.ptr + BUFSIZE;
              it.eof = 0;
              // Sentinel (at YYLIMIT pointer) is set to zero, which triggers YYFILL.
              it.yylimit[0] = 0;

              // Run the lexer.
              assert(lex(it) == count);

              // Cleanup: remove input file.
              fclose(it.file);
              remove(fname.ptr);
          }

🧚 YYFILL with padding 🧚

In the default case (when the EOF rule is not used), YYFILL is a function-like primitive that accepts a single argument and does not return any value. YYFILL invocation is triggered by the condition (YYLIMIT - YYCURSOR) < n in the C pointer API and YYLESSTHAN(n) in the generic API. The argument passed to YYFILL is the minimal number of characters that must be supplied. If it fails to do so, YYFILL must not return to the lexer (for that reason it's best implemented as a macro that returns from the calling function on failure). In case of a successful YYFILL invocation, the limit position must be set either to one after the last input position in the buffer, or to the end of the YYMAXFILL padding (in case YYFILL has successfully read at least n characters, but not enough to fill the entire buffer). The pictures below show the relative locations of input positions in the buffer before and after the YYFILL invocation (YYMAXFILL padding on the second picture is marked with # symbols).

                         <-- 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 ->

Here's an example of a program that reads the input file input.txt in chunks of 4096 bytes and uses bounds-checking with padding.

          // re2d $INPUT -o $OUTPUT
          module main;

          import core.stdc.string;
          import core.stdc.stdio;

          /*!max:re2c*/
          enum BufSize = (4096 - YYMaxFill);

          struct Input {
              FILE* file;
              char[BufSize + YYMaxFill] buffer;
              char* yylimit, yycursor, token;
              bool eof;
          };

          private int fill(ref Input it, size_t need) {
              if (it.eof) return 1;

              const size_t shift = it.token - it.buffer.ptr;
              const size_t used = it.yylimit - it.token;

              // Error: lexeme too long. In real life could reallocate a larger buffer.
              if (shift < need) return 2;

              // Shift buffer contents (discard everything up to the current token).
              memmove(it.buffer.ptr, it.token, used);
              it.yylimit -= shift;
              it.yycursor -= shift;
              it.token -= shift;

              // Fill free space at the end of buffer with new data from file.
              it.yylimit += fread(it.yylimit, 1, BufSize - used, it.file);

              // 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 (it.yylimit < it.buffer.ptr + BufSize) {
                  it.eof = true;
                  memset(it.yylimit, 0, YYMaxFill);
                  it.yylimit += YYMaxFill;
              }

              return 0;
          }

          private int lex(ref Input yyrecord) {
              int count = 0;
              for (;;) {
                  yyrecord.token = yyrecord.yycursor;
              /*!re2c
                  re2c:api = record;
                  re2c:YYCTYPE = "char";
                  re2c:YYFILL = "if (fill(yyrecord, @@) != 0) return -1;";

                  str = ['] ([^'\\] | [\\][^])* ['`;`? Again, exact line should be here. Let me write it properly:
              */
              }
              assert(0);
          }

          void main() {
              const char[] fname = "input";
              const char[] content = "'qu\0tes' 'are' 'fine: \\'' ";

              // Prepare input file: a few times the size of the buffer, containing
              // strings with zeroes and escaped quotes.
              FILE* f = fopen(fname.ptr, "w");
              for (int i = 0; i < BufSize; ++i) {
                  fwrite(content.ptr, 1, content.length - 1, f);
              }
              fclose(f);
              int count = 3 * BufSize; // number of quoted strings written to file

              // Initialize lexer state: all pointers are at the end of buffer.
              // This immediately triggers YYFILL, as the check `it.yycursor < it.yylimit` fails.
              Input it;
              it.file = fopen(fname.ptr, "r");
              it.yycursor = it.token = it.yylimit = it.buffer.ptr + BufSize;
              it.eof = 0;

              // Run the lexer.
              assert(lex(it) == count);

              // Cleanup: remove input file.
              fclose(it.file);
              remove(fname.ptr);
          }

🌈 FEATURES 🌈

🧚 Multiple blocks

okay so sometimes you literally need several lexers that all know about each other — like when you've got a high-level state machine bouncing between lexer modes, bestie. that's a whole vibe. 💫 you can pull it off with multiple connected re2d blocks. another option? start conditions (more on those in a sec 👀).

how you wire the blocks together totally depends on your target language. no cap:

in languages that have a goto statement (like C/C++ and Go 🫶), you can just put all the blocks in one function, each one prefixed with its own label, and hop between them with a plain goto. slay, easy.
in languages that don't have goto (looking at you, Rust 💀), you gotta use a loop with a switch on a state variable — same energy as the yystate loop/switch that re2d itself generates — or else wrap each block in its own function and use function calls.

the example below uses multiple blocks to parse binary, octal, decimal and hexadecimal numbers. every base gets its own block, the initial block figures out the base and dispatches to the others, and the common configs live in a separate block at the top of the program so the other blocks inherit them. iconic ✨

// re2d $INPUT -o $OUTPUT -i
module main;

enum ERROR = ulong.max;

private void add(ulong BASE)(ref ulong u, int d) {
    u = u * BASE + d;
    if (u > uint.max) { u = ERROR; }
}

private ulong parse_u32(const(char)* s) {
    const(char)* yycursor = s, yymarker;
    ulong u = 0;

    /*!re2c
        re2c:yyfill:enable = 0;
        re2c:YYCTYPE = char;

        end = "\x00";

        '0b' / [01]        { goto bin; }
        "0"                { goto oct; }
        "" / [1-9]         { goto dec; }
        '0x' / [0-9a-fA-F] { goto hex; }
        *                  { return ERROR; }
    */
bin:
    /*!re2c
        end   { return u; }
        [01]  { add!(2)(u, yycursor[-1] - '0'); goto bin; }
        *     { return ERROR; }
    */
oct:
    /*!re2c
        end   { return u; }
        [0-7] { add!(8)(u, yycursor[-1] - '0'); goto oct; }
        *     { return ERROR; }
    */
dec:
    /*!re2c
        end   { return u; }
        [0-9] { add!(10)(u, yycursor[-1] - '0'); goto dec; }
        *     { return ERROR; }
    */
hex:
    /*!re2c
        end   { return u; }
        [0-9] { add!(16)(u, yycursor[-1] - '0');      goto hex; }
        [a-f] { add!(16)(u, yycursor[-1] - 'a' + 10); goto hex; }
        [A-F] { add!(16)(u, yycursor[-1] - 'A' + 10); goto hex; }
        *     { return ERROR; }
    */
}

void main() {
    assert(parse_u32("") == ERROR);
    assert(parse_u32("1234567890") == 1234567890);
    assert(parse_u32("0b1101") == 13);
    assert(parse_u32("0x7Fe") == 2046);
    assert(parse_u32("0644") == 420);
    assert(parse_u32("9999999999") == ERROR);
}

🎀 Start conditions

start conditions get switched on with the --start-conditions option, and honestly? they're the moment. 💅 they give you a way to encode multiple interrelated automata inside one single re2d block. one block, many moods.

each condition is basically its own little automaton. it has a unique name that you specify, plus a unique internal number that re2d makes up for you. those numbers are what you use to switch between conditions — the generated code calls on the YYGETCOND and YYSETCOND primitives to read the current condition or set it to a given number. wanna generate numeric condition identifiers? use the conditions block, the --header option, or the re2c:header configuration. and the re2c:cond:enumprefix configuration sets the prefix on those generated identifiers. 🏷️

when you're in condition mode, every single rule has to be prefixed with a list of comma-separated condition names wrapped in angle brackets — or the wildcard <*> if you mean all conditions. the rule syntax gets extended like this:

< condition-list > regular-expression code — a rule that gets merged into every condition on the condition-list. it matches regular-expression and runs the associated code.
< condition-list > regular-expression => condition code — a rule merged into every condition on the condition-list. it matches regular-expression, sets the current condition to condition, and runs the associated code.
< condition-list > regular-expression :=> condition — a rule merged into every condition on the condition-list. it matches regular-expression and immediately transitions to condition (there's no semantic action, it just vibes on over).
< condition-list > !action code — a rule that binds code to the place defined by action in every condition on the condition-list (peep the actions section for the various types of actions 👀).
<! condition-list > code — a rule that prepends code to the semantic actions of all rules, for every condition on the condition-list. ⚠️ this syntax is deprecated — use the !pre_rule action instead (does exactly the same thing, bestie).
< > code — a rule that creates a special entry condition with number zero and name "0" which runs code before jumping to other conditions. ⚠️ this syntax is deprecated too — use the !entry action instead (it's way more fine-grained: you can specify the code per-condition, and you can jump straight to a condition's start without going through condition dispatch). glow-up fr.
< > => condition code — same as the previous rule, except it sets the next condition.
< > :=> condition — same as the previous rule, except it has no associated code and immediately jumps to condition.

the code re2d generates for conditions depends on whether it's using the goto/label approach or the loop/switch approach to encode the automata. two completely different fits, both cute. 💖

in languages with a goto statement (C/C++, Go — hiii 🫶), conditions naturally become blocks of code prefixed with labels of the form yyc_<cond>, where cond is a condition name. (you can change the label prefix with re2c:cond:prefix.) transitions between conditions are handled with goto and the condition labels. before all the conditions, re2d generates an initial switch on YYGETSTATE that jumps to the start state of the current condition. the shortcut rules :=> bypass that initial switch and jump directly to the specified condition — and re2c:cond:goto can change that default behavior. the rules with semantic actions do not automatically jump to the next condition; you have to do that yourself in your user-defined action code. no cap, that part's on you.

in languages without goto (Rust, ilysm but 😭), re2d reuses the yystate variable to store condition numbers instead. every condition gets a numeric identifier equal to the number of its start state, so switching between conditions is literally no different than switching between DFA states of a single condition. no separate initial condition switch needed. (and since that same approach is used to implement storable states, YYGETCOND/YYSETCOND are redundant if you're using both storable states and conditions. free real estate ✨)

the program below uses start conditions to parse binary, octal, decimal and hexadecimal numbers. it's a single block where each base has its own condition, and the initial condition is connected to all of them. the user-defined variable cond stores the current condition number, and it's initialized to the number of the initial condition generated with the conditions block. chef's kiss. 👑

// re2d $INPUT -o $OUTPUT -ci
module main;

enum ERROR = ulong.max;
/*!conditions:re2c*/

private void add(ulong BASE)(ref ulong u, int d) {
    u = u * BASE + d;
    if (u > uint.max) { u = ERROR; }
}

private ulong parse_u32(const(char)* s) {
    const(char)* yycursor = s, yymarker;
    YYCond yycond = YYCond.yycinit;
    ulong u = 0;

    /*!re2c
        re2c:yyfill:enable = 0;
        re2c:YYCTYPE = char;

        <*> * { return ERROR; }
        <init> '0b' / [01]        :=> bin
        <init> "0"                :=> oct
        <init> "" / [1-9]         :=> dec
        <init> '0x' / [0-9a-fA-F] :=> hex
        <bin, oct, dec, hex> "\x00" { return u; }
        <bin> [01]  { add!(2)(u,  yycursor[-1] - '0');      goto yyc_bin; }
        <oct> [0-7] { add!(8)(u,  yycursor[-1] - '0');      goto yyc_oct; }
        <dec> [0-9] { add!(10)(u, yycursor[-1] - '0');      goto yyc_dec; }
        <hex> [0-9] { add!(16)(u, yycursor[-1] - '0');      goto yyc_hex; }
        <hex> [a-f] { add!(16)(u, yycursor[-1] - 'a' + 10); goto yyc_hex; }
        <hex> [A-F] { add!(16)(u, yycursor[-1] - 'A' + 10); goto yyc_hex; }
    */
}

void main() {
    assert(parse_u32("") == ERROR);
    assert(parse_u32("1234567890") == 1234567890);
    assert(parse_u32("0b1101") == 13);
    assert(parse_u32("0x7Fe") == 2046);
    assert(parse_u32("0644") == 420);
    assert(parse_u32("9999999999") == ERROR);
}

💾 Storable state

Okay bestie, so with the --storable-state option, re2d generates a lexer that can store its current state, return to the caller, and later resume operations exactly where it left off. Iconic behavior, honestly. 💅

The default mode of operation in re2d is a "pull" model, where the lexer "pulls" more input whenever it needs it. That can be a problem when the input becomes available piece by piece — for example, if the lexer is invoked by the parser, or if the lexer program communicates via a socket protocol with some other program that must wait for a reply from the lexer before it transmits the next message. Storable state is intended exactly for those cases: it lets you generate lexers that work in a "push" model. When the lexer needs more input, it stores its state and returns to the caller. Later, when more input becomes available, the caller resumes the lexer exactly where it stopped. ✨

There are a few changes necessary compared to the "pull" model:

Define YYSETSTATE() and YYGETSTATE(state) primitives.
Define yych, yyaccept (if used) and state variables as a part of persistent lexer state. The state variable should be initialized to -1.
YYFILL should return to the outer program instead of trying to supply more input. Return code should indicate that lexer needs more input.
The outer program should recognize situations when lexer needs more input and respond appropriately.
Optionally use getstate block to generate YYGETSTATE switch detached from the main lexer. This only works for languages that have goto (not in --loop-switch mode).
Use re2c:eof and the sentinel with bounds checks method to handle the end of input. Padding-based method may not work because it is unclear when to append padding: the current end of input may not be the ultimate end of input, and appending padding too early may cut off a partially read greedy lexeme. Furthermore, due to high-level program logic getting more input may depend on processing the lexeme at the end of buffer (which already is blocked due to the end-of-input condition).

Here is an example of a "push" model lexer that simulates reading packets from a socket. The lexer loops until it encounters the end of input and returns to the calling function. The calling function provides more input by "sending" the next packet and resumes lexing. This process stops when all the packets have been sent, or when there is an error. 🫶

// re2d $INPUT -o $OUTPUT -f
module main;

import core.stdc.stdio;
import core.stdc.string;

// Use a small buffer to cover the case when a lexeme doesn't fit.
// In real world use a larger buffer.
enum BUFSIZE = 10;

struct State {
    FILE* file;
    char[BUFSIZE + 1] buffer;
    char* yylimit, yycursor, yymarker, token;
    int yystate;
};

enum Status {END, READY, WAITING, BAD_PACKET, BIG_PACKET};

private Status fill(ref State st) {
    const size_t shift = st.token - cast(char*)st.buffer;
    const size_t used = st.yylimit - st.token;
    const size_t free = BUFSIZE - used;

    // Error: no space. In real life can reallocate a larger buffer.
    if (free < 1) return Status.BIG_PACKET;

    // Shift buffer contents (discard already processed data).
    memmove(cast(void*)st.buffer, st.token, used);
    st.yylimit -= shift;
    st.yycursor -= shift;
    st.yymarker -= shift;
    st.token -= shift;

    // Fill free space at the end of buffer with new data.
    const size_t read = fread(st.yylimit, 1, free, st.file);
    st.yylimit += read;
    st.yylimit[0] = 0; // append sentinel symbol

    return Status.READY;
}

private Status lex(ref State yyrecord, uint* recv) {
    char yych;
    /*!getstate:re2c*/

    for (;;) {
        yyrecord.token = yyrecord.yycursor;
    /*!re2c
        re2c:api = record;
        re2c:YYCTYPE = char;
        re2c:YYFILL = "return Status.WAITING;";
        re2c:eof = 0;

        packet = [a-z]+[;];

        *      { return Status.BAD_PACKET; }
        $      { return Status.END; }
        packet { *recv = *recv + 1; continue; }
    */
    }
    assert(0); // unreachable
}

private void test(string[] packets, Status expect) {
    // Create a pipe (open the same file for reading and writing).
    const(char*) fname = "pipe";
    FILE* fw = fopen(fname, "w");
    FILE* fr = fopen(fname, "r");
    setvbuf(fw, null, _IONBF, 0);
    setvbuf(fr, null, _IONBF, 0);

    // Initialize lexer state: `state` value is -1, all pointers are at the end
    // of buffer.
    State st;
    st.file = fr;
    st.yycursor = st.yymarker = st.token = st.yylimit = cast(char*)st.buffer + BUFSIZE;
    // Sentinel (at YYLIMIT pointer) is set to zero, which triggers YYFILL.
    st.yylimit[0] = 0;
    st.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.
    Status status;
    uint send = 0, recv = 0;
    for (;;) {
        status = lex(st, &recv);
        if (status == Status.END) {
            debug{printf("done: got %u packets\n", recv);}
            break;
        } else if (status == Status.WAITING) {
            debug{printf("waiting...\n");}
            if (send < packets.length) {
                debug{printf("sent packet %u\n", send);}
                fprintf(fw, "%s", cast(char*)packets[send]);
                ++send;
            }
            status = fill(st);
            debug{printf("queue: '%s'\n", cast(char*)st.buffer);}
            if (status == Status.BIG_PACKET) {
                debug{printf("error: packet too big\n");}
                break;
            }
            assert(status == Status.READY);
        } else {
            assert(status == Status.BAD_PACKET);
            debug{printf("error: ill-formed packet\n");}
            break;
        }
    }

    // Check results.
    assert(status == expect);
    if (status == Status.END) assert(recv == send);

    // Cleanup: remove input file.
    fclose(fw);
    fclose(fr);
    remove(fname);
}

void main() {
    string[] packets1 = [];
    string[] packets2 = ["zero;", "one;", "two;", "three;", "four;"];
    string[] packets3 = ["zer0;"];
    string[] packets4 = ["looooooooooong;"];

    test(packets1, Status.END);
    test(packets2, Status.END);
    test(packets3, Status.BAD_PACKET);
    test(packets4, Status.BIG_PACKET);
}

🧩 Reusable blocks

Reusable blocks of the form /*!rules:re2c[:<name>] ... */ or %{rules[:<name>] ... %} can be reused any number of times and combined with other re2d blocks. The <name> is optional. A rules block can be used in a use block or directive. The code for a rules block is generated at every point of use. 🎀

Use blocks are defined with /*!use:re2c[:<name>] ... */ or %{use[:<name>] ... %}. The <name> is optional: if it's not specified, the associated rules block is the most recent one (whether named or unnamed). A use block can add named definitions, configurations and rules of its own. An important use case for use blocks is a lexer that supports multiple input encodings: the same rules block is reused multiple times with encoding-specific configurations (see the example below). 🌈

In-block use directive !use:<name>; can be used from inside of a re2d block. It merges the referenced block <name> into the current one. If some of the merged rules and configurations overlap with the previously defined ones, conflicts are resolved in the usual way: the earliest rule takes priority, and latest configuration overrides preceding ones. One exception are the special rules *, $ and (in condition mode) <!>, for which a block-local definition overrides any inherited ones. Use directive allows one to combine different re2d blocks together in one block (see the example below). 💅

Named blocks and in-block use directive were added in re2d version 2.2. Since that version reusable blocks are allowed by default (no special option is needed). Before version 2.2 reuse mode was enabled with -r --reusable option. Before version 1.2 reusable blocks could not be mixed with normal blocks. ⭐

✨ Example of a `!use` directive

// re2d $INPUT -o $OUTPUT
module 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.

enum What { COLOR, FISH, DUNNO };

/*!rules:re2c:colors
    *                            { assert(false); }
    "red" | "salmon" | "magenta" { return What.COLOR; }
*/

/*!rules:re2c:fish
    *                            { assert(false); }
    "haddock" | "salmon" | "eel" { return What.FISH; }
*/

private What lex(const(char)* s) {
    const(char)* yycursor = s, yymarker;
    /*!re2c
        re2c:YYCTYPE = "char";
        re2c:yyfill:enable = 0;

        !use:fish;
        !use:colors;
        * { return What.DUNNO; } // overrides inherited '*' rules
    */
}

void main() {
    assert(lex("salmon") == What.FISH);
    assert(lex("what?") == What.DUNNO);
}

🌸 Example of a `/!use:re2c ... /` block

// re2d $INPUT -o $OUTPUT --input-encoding utf8
module main;

import std.stdint;

// 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; }
*/

private int lex_utf8(const(uint8_t)* s) {
    const(uint8_t)* yycursor = s, yymarker;
    /*!use:re2c
        re2c:YYCTYPE = uint8_t;
        re2c:encoding:utf8 = 1;
    */
}

          private int lex_utf32(const(uint32_t)* s) {
              const(uint32_t)* yycursor = s, yymarker;
              /*!use:re2c
                  re2c:YYCTYPE = uint32_t;
                  re2c:encoding:utf32 = 1;
              */
          }

          void main() {
              immutable uint8_t[] s8 = // UTF-8
                  [ 0xe2, 0x88, 0x80, 0x78, 0x20, 0xe2, 0x88, 0x83, 0x79 ];

              immutable uint32_t[] s32 = // UTF32
                  [ 0x00002200, 0x00000078, 0x00000020, 0x00002203, 0x00000079 ];

              assert(lex_utf8(cast(const(uint8_t)*)s8) == 0);
              assert(lex_utf32(cast(const(uint32_t)*)s32) == 0);
          }

🎀 Submatch extraction 🎀

okay so this is the part everyone DMs me about — how do you actually pull out the little pieces of what you matched? re2d has two options for submatch extraction, and they are not the same vibe. let's go 💅

✨ Tags ✨

The first option is to use standalone tags of the form @stag or #mtag, where stag and mtag are arbitrary used-defined names. Tags are enabled with -T --tags option or re2c:tags = 1 configuration. Semantically tags are position markers: they can be inserted anywhere in a regular expression, and they bind to the corresponding position (or multiple positions) in the input string.

S-tags bind to the last matching position, and m-tags bind to a list of positions (they may be used in repetition subexpressions, where a single position in a regular expression corresponds to multiple positions in the input string). All tags should be defined by the user, either manually or with the help of svars and mvars blocks. If there is more than one way tags can be matched against the input, ambiguity is resolved using leftmost greedy disambiguation strategy.

✨ Captures ✨

The second option is to use capturing groups. They are enabled with --captures option or re2c:captures = 1 configuration. There are two flavours for different disambiguation policies: --leftmost-captures (the default) is for leftmost greedy policy, and --posix-captures is for POSIX longest-match policy.

In this mode all parenthesized subexpressions are considered capturing groups, and a bang can be used to mark non-capturing groups: (! ... ). With --invert-captures option or re2c:invert-captures = 1 configuration the meaning of bang is inverted. 🫶

The number of groups for the matching rule is stored in a variable yynmatch (the whole regular expression is group number zero), and submatch results are stored in yypmatch array. Both yynmatch and yypmatch should be defined by the user, and yypmatch size must be at least [yynmatch * 2]. Use maxnmatch block to define YYMAXNMATCH, a constant that equals to the maximum value of yynmatch among all rules.

✨ Captvars ✨

Another way to use capturing groups is the --captvars option or re2c:captvars = 1 configuration. The only difference with --captures is in the way the generated code stores submatch results: instead of yynmatch and yypmatch re2d generates variables yytl<k> and yytr<k> for k-th capturing group (the user should declare these using an svars block).

Captures with variables support two disambiguation policies: --leftmost-captvars or re2c:leftmost-captvars = 1 for leftmost greedy policy (the default one) and --posix-captvars or re2c:posix-captvars for POSIX longest-match policy.

Under the hood all these options translate into tags and Tagged Deterministic Finite Automata with Lookahead. The core idea of TDFA is to minimize the overhead on submatch extraction. In the extreme, if there're no tags or captures in a regular expression, TDFA is just an ordinary DFA. If the number of tags is moderate, the overhead is barely noticeable 👑

The generated TDFA uses a number of tag variables which do not map directly to tags: a single variable may be used for different tags, and a tag may require multiple variables to hold all its possible values. Eventually ambiguity is resolved, and only one final variable per tag survives. Tag variables should be defined using stags or mtags blocks. If lexer state is stored, tag variables should be part of it. They also need to be updated by YYFILL. 😭

S-tags support the following operations:

save input position to an s-tag: t = YYCURSOR with C pointer API or a user-defined operation YYSTAGP(t) with generic API
save default value to an s-tag: t = NULL with C pointer API or a user-defined operation YYSTAGN(t) with generic API
copy one s-tag to another: t1 = t2

M-tags support the following operations:

append input position to an m-tag: a user-defined operation YYMTAGP(t) with both default and generic API
append default value to an m-tag: a user-defined operation YYMTAGN(t) with both default and generic API
copy one m-tag to another: t1 = t2

S-tags can be implemented as scalar values (pointers or offsets). M-tags need a more complex representation, as they need to store a sequence of tag values. The most naive and inefficient representation of an m-tag is a list (array, vector) of tag values; a more efficient representation is to store all m-tags in a prefix-tree represented as array of nodes (v, p), where v is tag value and p is a pointer to parent node. 🌸

Here is a simple example of using s-tags to parse semantic versions consisting of three numeric components: major, minor, patch (the latter is optional). See below for a more complex example that uses YYFILL.

// re2d $INPUT -o $OUTPUT
module main;

struct SemVer {
    int major;
    int minor;
    int patch;
};

private int s2n(const(char)* s, const(char)* e) { // pre-parsed string to number
    int n = 0;
    for (; s < e; ++s) n = n * 10 + (*s - '0');
    return n;
}

private bool lex(const(char)* str, ref SemVer ver) {
    const(char)* yycursor = str, yymarker;

    // Final tag variables available in semantic action.
    /*!svars:re2c format = 'const(char)* @@;\n'; */

    // Intermediate tag variables used by the lexer (must be autogenerated).
    /*!stags:re2c format = 'const(char)* @@;\n'; */

    /*!re2c
        re2c:yyfill:enable = 0;
        re2c:tags = 1;
        re2c:YYCTYPE = "char";

        num = [0-9]+;

        @t1 num @t2 "." @t3 num @t4 ("." @t5 num)? [\x00] {
            ver.major = s2n(t1, t2);
            ver.minor = s2n(t3, t4);
            ver.patch = t5 != null ? s2n(t5, yycursor - 1) : 0;
            return true;
        }
        * { return false; }
    */
}

void main() {
    SemVer v;
    assert(lex("23.34", v) && v.major == 23 && v.minor == 34 && v.patch == 0);
    assert(lex("1.2.999", v) && v.major == 1 && v.minor == 2 && v.patch == 999);
    assert(!lex("1.a", v));
}

Here is a more complex example of using s-tags with YYFILL to parse a file with newline-separated semantic versions. Tag variables are part of the lexer state, and they are adjusted in YYFILL like other input positions. Note that it is necessary for s-tags because their values are invalidated after shifting buffer contents. It may not be necessary in a custom implementation where tag variables store offsets relative to the start of the input string rather than the buffer, which may be the case with m-tags. 🦋

// re2d $INPUT -o $OUTPUT --tags
module main;

import core.stdc.string;
import core.stdc.stdio;
import std.stdio;

enum BUFSIZE = 4095;

struct Input {
    FILE* file;
    char[BUFSIZE + 1] buffer;// +1 for sentinel
    char* yylimit, yycursor, yymarker, token;
    // 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 = 'char* @@;'; */
    bool eof;
};

struct SemVer {
    int major;
    int minor;
    int patch;
};

private int s2n(const(char)* s, const(char)* e) { // pre-parsed string to number
    int n = 0;
    for (; s < e; ++s) n = n * 10 + (*s - '0');
    return n;
}

private int fill(ref Input it) {
    if (it.eof) return 1;

    const size_t shift = it.token - it.buffer.ptr;
    const size_t used = it.yylimit - it.token;

    // Error: lexeme too long. In real life could reallocate a larger buffer.
    if (shift < 1) return 2;

    // Shift buffer contents (discard everything up to the current token).
    memmove(cast(void*)it.buffer.ptr, it.token, used);
    it.yylimit -= shift;
    it.yycursor -= shift;
    it.yymarker -= shift;
    it.token -= shift;
    // Tag variables need to be shifted like other input positions. The check
    // for non-null is only needed if some tags are nested inside of alternative
    // or repetition, so that they can have null value.
    /*!stags:re2c format = "if (it.@@) it.@@ -= shift;\n"; */

    // Fill free space at the end of buffer with new data from file.
    it.yylimit += fread(it.yylimit, 1, BUFSIZE - used, it.file);
    it.yylimit[0] = 0;
    it.eof = it.yylimit < (it.buffer.ptr + BUFSIZE);
    return 0;
}

private bool lex(ref Input yyrecord, ref SemVer[] vers) {
    // Final variables available in semantic actions.
    /*!svars:re2c format = 'char* @@;'; */
    for (;;) {
        yyrecord.token = yyrecord.yycursor;
    /*!re2c
        re2c:api = record;
        re2c:YYCTYPE = "char";
        re2c:YYFILL = "fill(yyrecord) == 0";
        re2c:eof = 0;

        num = [0-9]+;

        num @t1 "." @t2 num @t3 ("." @t4 num)? [\n] {
            int major = s2n(yyrecord.token, t1);
            int minor = s2n(t2, t3);
            int patch = t4 != null ? s2n(t4, yyrecord.yycursor - 1) : 0;
            SemVer ver = SemVer(major, minor, patch);
            vers ~= ver;
            continue;
        }
        $ { return true; }
        * { return false; }
    */
    }
    assert(0);
}

void main() {
    const char[] fname = "input";
    const char[] content = "1.22.333\n' ";

    SemVer[BUFSIZE] expect = SemVer(1, 22, 333);
    SemVer[] actual;

okay so we're still mid-example — the code keeps rolling like this:

              // Prepare input file: a few times the size of the buffer, containing
              // strings with zeroes and escaped quotes.
              FILE* f = fopen(fname.ptr, "w");
              for (int i = 0; i < BUFSIZE; ++i) {
                  fwrite(cast(const(void*)) content.ptr, 1, content.length - 2, f); // skip null-terminator
              }
              fclose(f);

              // Initialize lexer state: all pointers are at the end of buffer.
              Input it;
              it.file = fopen(fname.ptr, "r");
              it.yycursor = it.yymarker = it.token = it.yylimit = it.buffer.ptr + BUFSIZE;
              it.eof = 0;
              // Sentinel (at YYLIMIT pointer) is set to zero, which triggers YYFILL.
              it.yylimit[0] = 0;

              // Run the lexer.
              assert(lex(it, actual) && actual == expect);

              // Cleanup: remove input file.
              fclose(it.file);
              remove(fname.ptr);
          }

Next up: capturing groups doing semantic version parsing, slay 💅

          // re2d $INPUT -o $OUTPUT
          module main;

          struct SemVer { int major, minor, patch; };

          private int s2n(const(char)* s, const(char)* e) { // pre-parsed string to number
              int n = 0;
              for (; s < e; ++s) n = n * 10 + (*s - '0');
              return n;
          }

          private bool lex(const(char)* str, ref SemVer ver) {
              const(char)* yycursor = str, yymarker;

              // Final tag variables available in semantic action.
              /*!svars:re2c format = 'const(char)* @@;\n'; */

              // Intermediate tag variables used by the lexer (must be autogenerated).
              /*!stags:re2c format = 'const(char)* @@;\n'; */

              /*!re2c
                  re2c:yyfill:enable = 0;
                  re2c:captvars = 1;
                  re2c:YYCTYPE = "char";

                  num = [0-9]+;

                  (num) "." (num) ("." num)? [\x00] {
                      ver.major = s2n(yytl1, yytr1);
                      ver.minor = s2n(yytl2, yytr2);
                      ver.patch = yytl3 ? s2n(yytl3 + 1, yytr3) : 0;
                      return true;
                  }
                  * { return false; }
              */
          }

          void main() {
              SemVer v;
              assert(lex("23.34", v) && v.major == 23 && v.minor == 34 && v.patch == 0);
              assert(lex("1.2.999", v) && v.major == 1 && v.minor == 2 && v.patch == 999);
              assert(!lex("1.a", v));
          }

And here's the m-tags flex: parsing a version with a variable number of components. Tag variables are stored in a trie 🧚

          // re2d $INPUT -o $OUTPUT
          module main;

          enum MtagRoot = -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).
          struct Mtag {
              const(char)* elem; // tag value
              int pred; // index of the predecessor node or root
          };

          alias MtagTrie = Mtag[];
          alias Ver = int[];

          private int s2n(const(char)* s, const(char)* e) { // pre-parsed string to number
              int n = 0;
              for (; s < e; ++s) n = n * 10 + (*s - '0');
              return n;
          }

          // Append a single value to an m-tag history.
          private void add_mtag(ref MtagTrie trie, ref int mtag, const(char)* value) {
              Mtag m = {value, mtag};
              mtag = cast(int)trie.length;
              trie ~= [m];
          }

          // Recursively unwind tag histories and collect version components.
          private void unfold(const ref MtagTrie trie, int x, int y, ref Ver ver) {
              // Reached the root of the m-tag tree, stop recursion.
              if (x == MtagRoot && y == MtagRoot) return;

              // Unwind history further.
              unfold(trie, trie[x].pred, trie[y].pred, ver);

              // Get tag values. Tag histories must have equal length.
              assert(x != MtagRoot && y != MtagRoot);
              const(char)* ex = trie[x].elem, ey = trie[y].elem;

              if (ex != null && ey != null) {
                  // Both tags are valid pointers, extract component.
                  ver ~= [s2n(ex, ey)];
              } else {
                  // Both tags are null (this corresponds to zero repetitions).
                  assert(ex == null && ey == null);
              }
          }

          private bool parse(const(char)* str, ref Ver ver) {
              const(char)* yycursor = str, yymarker;
              MtagTrie mt;

              // Final tag variables available in semantic action.
              /*!svars:re2c format = "const(char)* @@;"; */
              /*!mvars:re2c format = "int @@;"; */

              // Intermediate tag variables used by the lexer (must be autogenerated).
              /*!stags:re2c format = "const(char)* @@ = null;"; */
              /*!mtags:re2c format = "int @@ = MtagRoot;"; */

              /*!re2c
                  re2c:yyfill:enable = 0;
                  re2c:tags = 1;
                  re2c:YYCTYPE = "char";
                  re2c:YYMTAGP = "add_mtag(mt, @@, yycursor);";
                  re2c:YYMTAGN = "add_mtag(mt, @@, null);";

                  num = [0-9]+;
                  @t1 num @t2 ("." #t3 num #t4)* [\x00] {
                      ver = [];
                      ver ~= [s2n(t1, t2)];
                      unfold(mt, t3, t4, ver);
                      return true;
                  }
                  * { return false; }
              */
          }

          void main() {
              Ver v;
              assert(parse("1", v) && v == [1]);
              assert(parse("1.2.3.4.5.6.7", v) && v == [1, 2, 3, 4, 5, 6, 7]);
              assert(!parse("1.2.", v));
          }

✨ Encoding support ✨

okay bestie, this is the part where we get into code points versus code units, and it actually matters fr fr. A code point is a numeric identifier of a symbol. A code unit is the smallest unit of storage in the encoded text. A single code point may be represented with one or more code units. In a fixed-length encoding all code points are represented with the same number of code units. In a variable-length encoding code points may be represented with a different number of code units.

🚨 lowkey important: the "any" rule [^] matches any code point, but not necessarily any code unit (the only way to match any code unit regardless of the encoding is the default rule *).

The generated lexer works with a stream of code units: yych stores a code unit, and YYCTYPE is the code unit type. Regular expressions, on the other hand, are specified in terms of code points. When re2d compiles regular expressions to automata it translates code points to code units. This is generally not a simple mapping: in variable-length encodings a single code point range may get translated to a complex code unit graph.

The following encodings are supported:

ASCII (enabled by default). It is a fixed-length encoding with code space [0-255] and 1-byte code points and code units.
EBCDIC (enabled with --ebcdic or re2c:encoding:ebcdic). It is a fixed-length encoding with code space [0-255] and 1-byte code points and code units.
UCS2 (enabled with --ucs2 or re2c:encoding:ucs2). It is a fixed-length encoding with code space [0-0xFFFF] and 2-byte code points and code units.
UTF8 (enabled with --utf8 or re2c:encoding:utf8). It is a variable-length Unicode encoding. Code unit size is 1 byte. Code points are represented with 1 -- 4 code units.
UTF16 (enabled with --utf16 or re2c:encoding:utf16). It is a variable-length Unicode encoding. Code unit size is 2 bytes. Code points are represented with 1 -- 2 code units.
UTF32 (enabled with --utf32 or re2c:encoding:utf32). It is a fixed-length Unicode encoding with code space [0-0x10FFFF] and 4-byte code points and code units.

Include files include/unicode_blocks.re, include/unicode_categories.re and include/unicode_properties.re provide re2d definitions for blocks, categories and properties described in the Unicode standard. 🦋

Option --input-encoding specifies source file encoding, which can be used to enable Unicode literals in regular expressions. For example --input-encoding utf8 tells re2d that the source file is in UTF8 (it differs from --utf8 which sets input text encoding). Option --encoding-policy specifies the way re2d handles Unicode surrogates (code points in range [0xD800-0xDFFF]).

Below is an example of a lexer for UTF8 encoded Unicode identifiers 👑

          // re2d $INPUT -o $OUTPUT -8 -i
          module main;

          /*!include:re2c "unicode_categories.re" */

          private int lex(const(char)* s) {
              const(char)* yycursor = s, yymarker;
              /*!re2c
                  re2c:YYCTYPE = "char";
                  re2c:yyfill:enable = 0;

                  // 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; }
              */
          }

          void main() {
              assert(lex("_Ыдентификатор") == 0);
              assert(lex("!!!")==1);
          }

🎀 Include files 🎀

re2d allows one to include other files using a block of the form /*!include:re2c FILE */ or %{include FILE %}, or an in-block directive !include FILE ;, where FILE is a path to the file to be included. re2d looks for include files in the directory of the including file and in include locations, which can be specified with the -I option. Include blocks/directives in re2d work in the same way as C/C++ #include: FILE contents are copy-pasted verbatim in place of the block/directive. Include files may have further includes of their own. Use --depfile option to track build dependencies of the output file on include files. re2d provides some predefined include files that can be found in the include/ subdirectory of the project. These files contain definitions that may be useful to other projects (such as Unicode categories) and form something like a standard library for re2d. Below is an example of using include files.

🗂️ Include file 1 (definitions.d)

          enum Result{ OK, FAIL };

          /*!re2c
              number = [1-9][0-9]*;
          */

🧁 Include file 2 (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 Result.OK; }

🌸 Input file

          // re2d $INPUT -o $OUTPUT -i

          /*!include:re2c "definitions.d" */

          private Result lex(const(char)* s) {
              const(char)* yycursor = s, yymarker;
              /*!re2c
                  re2c:YYCTYPE = "char";
                  re2c:yyfill:enable = 0;

                  *      { return Result.FAIL; }
                  number { return Result.OK; }
                  !include "extra_rules.re.inc";
              */
          }

          void main() {
              assert(lex("123") == Result.OK);
              assert(lex("123.4567") == Result.OK);
          }

🎀 Header files 🎀

Okay so re2d lets you generate a header file from the input .re file using the --header option or the re2c:header configuration — plus block pairs of the form /*!header:re2c:on*/ and /*!header:re2c:off*/, or %{header:on%} and %{header:off%}. The first block marks the beginning of the header file, and the second block marks the end of it. Everything between these blocks is processed by re2d, and the generated code is written to the file specified with --header or re2c:header (or stdout if neither option nor configuration is used). Autogenerated header file may be needed in cases when re2d is used to generate definitions that must be visible from other translation units.

Here’s an example of generating a header file that contains the definition of the lexer state with tag variables — and yes, the number of variables depends on the regular grammar and is unknown to the programmer. 💅

Input file

          // re2d $INPUT -o $OUTPUT -i --header lexer/state.d
          module main;

          import core.stdc.stddef;
          import lexer.state; // the module is generated by re2c

          /*!header:re2c:on*/
          module lexer.state;

          struct LexerState {
              const(char)* str, yycursor;
              /*!stags:re2c format = "const(char)* @@;"; */
          };
          /*!header:re2c:off*/

          private long lex(ref LexerState yyrecord) {
              const(char)* t;
              /*!re2c
                  re2c:api = record;
                  re2c:YYCTYPE = "char";
                  re2c:tags = 1;
                  re2c:yyfill:enable = 0;
                  re2c:header = "lexer/state.d";

                  [a]* @t [b]* { return t - yyrecord.str; }
              */
          }

          void main() {
              const(char)* s = "ab";
              LexerState st = {s, s /*!stags:re2c format = ", null"; */};
              assert(lex(st) == 1);

              const(char)* s2 = "aaabbbbbbbb";
              LexerState st2 = {s2, s2 /*!stags:re2c format = ", null"; */};
              assert(lex(st2) == 3);
          }

Header file

          /* Generated by re2c */

          module lexer.state;

          struct LexerState {
              const(char)* str, yycursor;
              const(char)* yyt1;
          };

✨ Skeleton programs ✨

With the -S, --skeleton option, re2d ignores all non-re2d code and generates a self-contained C program that can be further compiled and executed. The program consists of lexer code and input data. For each constructed DFA (block or condition), re2d generates a standalone lexer and two files: an .input file with strings derived from the DFA and a .keys file with expected match results. The program runs each lexer on the corresponding .input file and compares results with the expectations. Skeleton programs are very useful for a number of reasons:

They can check correctness of various re2d optimizations (the data is generated early in the process, before any DFA transformations have taken place). 🔥
Generating a set of input data with good coverage may be useful for both testing and benchmarking. 🧚
Generating self-contained executable programs allows one to get minimized test cases (the original code may be large or have a lot of dependencies). 💖

The difficulty with generating input data is that for all but the most trivial cases the number of possible input strings is too large (even if the string length is limited). re2d solves this difficulty by generating sufficiently many strings to cover almost all DFA transitions. It uses the following algorithm. First, it constructs a skeleton of the DFA. For encodings with 1-byte code unit size (such as ASCII, UTF-8 and EBCDIC) the skeleton is just an exact copy of the original DFA. For encodings with multibyte code units the skeleton is a copy of the DFA with certain transitions omitted: namely, re2d takes at most 256 code units for each disjoint continuous range that corresponds to a DFA transition. The chosen values are evenly distributed and include range bounds. Instead of trying to cover all possible paths in the skeleton (which is infeasible), re2d generates sufficiently many paths to cover all skeleton transitions, and thus trigger the corresponding conditional jumps in the lexer. The algorithm implementation is limited by ~1Gb of transitions and consumes constant amount of memory (re2d writes data to file as soon as it is generated). 😭

🔥 Visualization and debug 🔥

With the -D, --emit-dot option, re2d does not generate code. Instead, it dumps the generated DFA in DOT format. One can convert this dump to an image of the DFA using Graphviz or another library. Note that this option shows the final DFA after it has gone through a number of optimizations and transformations. Earlier stages can be dumped with various debug options, such as --dump-nfa, --dump-dfa-raw etc. (see the full list of options). 👑

🌈 SEE ALSO 🌈

You can find more information about re2c at the official website: http://re2c.org. Similar programs are flex(1), lex(1), quex(http://quex.sourceforge.net). 🦋

👑 AUTHORS 👑

re2d was originally written by Peter Bumbulis (peter@csg.uwaterloo.ca) in 1993. Marcus Boerger and Dan Nuffer spent several years to turn the original idea into a production ready code generator. Since then it has been maintained and developed by multiple volunteers, most notably, 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 and PolarGoose. 💜

Later, bestie — that’s the whole RE2D(1) page, source/version/date footer and all. 💖✨

📦 where this page comes from

arch rolling: re2c 4.6-1 · LicenseRef-re2c
nixpkgs unstable: re2c 4.6
💖 girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on 2026-11-04: a language model's rewrite of the original, under its terms