.\" Man page generated from reStructuredText.
.
.
.nr rst2man-indent-level 0
.
.de1 rstReportMargin
\\$1 \\n[an-margin]
level \\n[rst2man-indent-level]
level margin: \\n[rst2man-indent\\n[rst2man-indent-level]]
-
\\n[rst2man-indent0]
\\n[rst2man-indent1]
\\n[rst2man-indent2]
..
.de1 INDENT
.\" .rstReportMargin pre:
. RS \\$1
. nr rst2man-indent\\n[rst2man-indent-level] \\n[an-margin]
. nr rst2man-indent-level +1
.\" .rstReportMargin post:
..
.de UNINDENT
. RE
.\" indent \\n[an-margin]
.\" old: \\n[rst2man-indent\\n[rst2man-indent-level]]
.nr rst2man-indent-level -1
.\" new: \\n[rst2man-indent\\n[rst2man-indent-level]]
.in \\n[rst2man-indent\\n[rst2man-indent-level]]u
..
.TH "RE2ZIG" 1 "" "" ""
.SH NAME
re2zig \- generate fast lexical analyzers for Zig
.SH SYNOPSIS
.sp
re2zig \fB[ OPTIONS ]\fP \fB[ WARNINGS ]\fP \fBINPUT\fP
.sp
Input can be either a file or \fB\-\fP for stdin.
.SH INTRODUCTION
.sp
re2zig works as a preprocessor. It reads the input file (which is usually a
program in Zig, 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 re2zig\&. It translates them to code in Zig and outputs the generated
code in place of the block.
.sp
Here is an example of a small program that checks if a given string contains a
decimal number:
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT
const std = @import(\(dqstd\(dq);
fn lex(yyinput: [:0]const u8) bool {
var yycursor: u32 = 0;
%{
re2c:yyfill:enable = 0;
[1\-9][0\-9]* { return true; }
* { return false; }
%}
}
test {
try std.testing.expect(lex(\(dq1234\(dq));
}
.ft P
.fi
.UNINDENT
.UNINDENT
.sp
In the output re2zig replaced the block in the middle with the generated code:
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// Generated by re2zig
// re2zig $INPUT \-o $OUTPUT
const std = @import(\(dqstd\(dq);
fn lex(yyinput: [:0]const u8) bool {
var yycursor: u32 = 0;
var yych: u8 = 0;
var yystate: u32 = 0;
yyl: while (true) {
switch (yystate) {
0 => {
yych = yyinput[yycursor];
yycursor += 1;
switch (yych) {
0x31...0x39 => {
yystate = 2;
continue :yyl;
},
else => {
yystate = 1;
continue :yyl;
},
}
},
1 => { return false; },
2 => {
yych = yyinput[yycursor];
switch (yych) {
0x30...0x39 => {
yycursor += 1;
yystate = 2;
continue :yyl;
},
else => {
yystate = 3;
continue :yyl;
},
}
},
3 => { return true; },
else => { @panic(\(dqinternal lexer error\(dq); },
}
}
}
test {
try std.testing.expect(lex(\(dq1234\(dq));
}
.ft P
.fi
.UNINDENT
.UNINDENT
.SH BASICS
.sp
A re2zig program consists of a sequence of \fIblocks\fP intermixed with code in the
target language. A block may contain \fIdefinitions\fP, \fIconfigurations\fP, \fIrules\fP,
\fIactions\fP and \fIdirectives\fP in any order:
.INDENT 0.0
.TP
.B \fBname = regular\-expression ;\fP
A \fIdefinition\fP binds \fBname\fP to \fBregular\-expression\fP\&. Names may contain
alphanumeric characters and underscore. The \fI\%regular expressions\fP section
gives an overview of re2zig syntax for regular expressions. Once defined,
the \fBname\fP 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.
.TP
.B \fBconfiguration = value ;\fP
A \fIconfiguration\fP allows one to change re2zig behavior and customize the
generated code. For a full list of configurations supported by re2zig see
the \fI\%configurations\fP section. Depending on a particular configuration, the
\fBvalue\fP 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.
.TP
.B \fBregular\-expression code\fP
A \fIrule\fP binds \fBregular\-expression\fP to its semantic action (a block of
code in curly braces, or a block of code that starts with \fB:=\fP and ends on
a newline followed by any non\-whitespace character).
If the \fBregular\-expression\fP matches, the associated \fBcode\fP 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 \fB*\fP and the end of input rule \fB$\fP\&.
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 \fI\%encoding support\fP section). The end of input rule
should be defined if the corresponding method for
\fI\%handling the end of input\fP is used.
With \fI\%start conditions\fP rules have more complex syntax.
.TP
.B \fB!action code\fP
An \fIaction\fP binds a user\-defined block of \fBcode\fP 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 \fI\%actions\fP section for a full list of
predefined actions.
.TP
.B \fB!directive ;\fP
A \fIdirective\fP is one of the special predefined statements. Each directive
has a unique purpose. See the \fI\%directives\fP section for details.
.UNINDENT
.SS Blocks
.sp
Block start and end markers are either \fB/*!re2c\fP and \fB*/\fP, or \fB%{\fP and
\fB%}\fP (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:
.INDENT 0.0
.TP
.B \fB/*!re2c[:<name>] ... */\fP or \fB%{[:<name>] ... %}\fP
A \fIglobal block\fP contains definitions, configurations, rules and directives.
re2zig 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 \fI\%options\fP\&.
.TP
.B \fB/*!local:re2c[:<name>] ... */\fP or \fB%{local[:<name>] ... %}\fP
A \fIlocal block\fP is like a global block, but the names and configurations in
it have local scope (they do not affect other blocks).
.TP
.B \fB/*!rules:re2c[:<name>] ... */\fP or \fB%{rules[:<name>] ... %}\fP
A \fIrules block\fP 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 \fI\%reusable blocks\fP section). Prior to re2zig version 2.2 rules blocks
required \fB\-r \-\-reusable\fP option.
.TP
.B \fB/*!use:re2c[:<name>] ... */\fP or \fB%{use[:<name>] ... %}\fP
A use block that references a previously defined rules block. If the name is
specified, re2zig 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 re2zig version 2.2 use blocks
required \fB\-r \-\-reusable\fP option.
.TP
.B \fB/*!max:re2c[:<name1>[:<name2>...]] ... */\fP or \fB%{max[:<name1>[:<name2>...]] ... %}\fP
A block that generates \fBYYMAXFILL\fP definition. An optional list of block
names specifies which blocks should be included when computing \fBYYMAXFILL\fP
value (if the list is empty, all blocks are included).
By default the generated code is a macro\-definition for C
(\fB#define YYMAXFILL <n>\fP), or a global variable for Go
(\fBvar YYMAXFILL int = <n>\fP). It can be customized with an optional
configuration \fBformat\fP that specifies a template string where \fB@@{max}\fP
(or \fB@@\fP for short) is replaced with the numeric value of \fBYYMAXFILL\fP\&.
.TP
.B \fB/*!maxnmatch:re2c[:<name1>[:<name2>...]] ... */\fP or \fB%{maxnmatch[:<name1>[:<name2>...]] ... %}\fP
A block that generates \fBYYMAXNMATCH\fP definition (it requires
\fB\-P \-\-posix\-captures\fP option). An optional list of block names specifies
which blocks should be included when computing \fBYYMAXNMATCH\fP value (if the
list is empty, all blocks are included).
By default the generated code is a macro\-definition for C
(\fB#define YYMAXNMATCH <n>\fP), or a global variable for Go
(\fBvar YYMAXNMATCH int = <n>\fP). It can be customized with an optional
configuration \fBformat\fP that specifies a template string where \fB@@{max}\fP
(or \fB@@\fP for short) is replaced with the numeric value of \fBYYMAXNMATCH\fP\&.
.TP
.B \fB/*!stags:re2c[:<name1>[:<name2>...]] ... */\fP, \fB/*!mtags:re2c[:<name1>[:<name2>...]] ... */\fP or \fB%{stags[:<name1>[:<name2>...]] ... %}\fP, \fB%{mtags[:<name1>[:<name2>...]] ... %{\fP
Blocks that specify a template piece of code that is expanded for each
s\-tag/m\-tag variable generated by re2zig\&. 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: \fBformat\fP and \fBseparator\fP\&.
Configuration \fBformat\fP specifies a template string where \fB@@{tag}\fP (or
\fB@@\fP for short) is replaced with the name of each tag variable.
Configuration \fBseparator\fP specifies a piece of code used to join the
generated \fBformat\fP pieces for different tag variables.
.TP
.B \fB/*!svars:re2c[:<name1>[:<name2>...]] ... */\fP, \fB/*!mvars:re2c[:<name1>[:<name2>...]] ... */\fP or \fB%{svars[:<name1>[:<name2>...]] ... %}\fP, \fB%{mvars[:<name1>[:<name2>...]] ... %{\fP
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
\fB\-\-tags\fP option) or implicitly generated by re2zig (with \fB\-\-captvars\fP or
\fB\-\-posix\-captvars\fP 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: \fBformat\fP and \fBseparator\fP\&.
Configuration \fBformat\fP specifies a template string where \fB@@{tag}\fP (or
\fB@@\fP for short) is replaced with the name of each tag.
Configuration \fBseparator\fP specifies a piece of code used to join the
generated \fBformat\fP pieces for different tags.
.TP
.B \fB/*!getstate:re2c[:<name1>[:<name2>...]] ... */\fP or \fB%{getstate[:<name1>[:<name2>...]] ... %}\fP
A block that generates conditional dispatch on the lexer state (it requires
\fB\-\-storable\-state\fP 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 \fB\-\-loop\-switch\fP option, as it
requires cross\-block transitions that are unsupported without \fBgoto\fP or
function calls.
.TP
.B \fB/*!conditions:re2c[:<name1>[:<name2>...]] ... */\fP, \fB/*!types:re2c... */\fP or \fB%{conditions[:<name1>[:<name2>...]] ... %}\fP, \fB%{types... %}\fP
A block that generates condition enumeration (it requires \fB\-\-conditions\fP
option). An optional list of block names specifies which blocks should be
included when computing the set of conditions (if the list is empty, all
blocks are included).
By default the generated code is an enumeration \fBYYCONDTYPE\fP\&. It can be
customized with optional configurations \fBformat\fP and \fBseparator\fP\&.
Configuration \fBformat\fP specifies a template string where \fB@@{cond}\fP (or
\fB@@\fP for short) is replaced with the name of each condition, and
\fB@@{num}\fP is replaced with a numeric index of that condition.
Configuration \fBseparator\fP specifies a piece of code used to join the
generated \fBformat\fP pieces for different conditions.
.TP
.B \fB/*!include:re2c <file> */\fP or \fB%{include <file> %}\fP
This block allows one to include \fB<file>\fP, which must be a double\-quoted
file path. The contents of the file are literally substituted in place of
the block, in the same way as \fB#include\fP works in C/C++. This block can be
used together with the \fB\-\-depfile\fP option to generate build system
dependencies on the included files.
.TP
.B \fB/*!header:re2c:on*/\fP or \fB%{header:on %}\fP
This block marks the start of header file. Everything after it and up to the
following \fBheader:off\fP block is processed by re2zig and written to the
header file specified with \fB\-t \-\-type\-header\fP option.
.TP
.B \fB/*!header:re2c:off*/\fP or \fB%{header:off %}\fP
This block marks the end of header file started with \fBheader:on*/\fP block.
.TP
.B \fB/*!ignore:re2c ... */\fP or \fB%{ignore ... %}\fP
A block which contents are ignored and removed from the output file.
.UNINDENT
.SS Configurations
.sp
Here is a full list of configurations supported by re2zig:
.INDENT 0.0
.TP
.B \fBre2c:api\fP, \fBre2c:input\fP
Same as the \fB\-\-api\fP option.
.TP
.B \fBre2c:api:sigil\fP
Specify the marker (\(dqsigil\(dq) that is used for argument placeholders in the
API primitives. The default is \fB@@\fP\&. A placeholder starts with sigil
followed by the argument name in curly braces. For example, if sigil is set
to \fB$\fP, then placeholders will have the form \fB${name}\fP\&. Single\-argument
APIs may use shorthand notation without the name in braces. This option can
be overridden by options for individual API primitives, e.g.
\fBre2c:YYFILL@len\fP for \fBYYFILL\fP\&.
.TP
.B \fBre2c:api:style\fP
Specify API style. Possible values are \fBfunctions\fP (the default for C) and
\fBfree\-form\fP (the default for Go and Rust).
In \fBfunctions\fP style API primitives are generated with an argument list in
parentheses following the name of the primitive. The arguments are provided
only for autogenerated parameters (such as the number of characters passed
to \fBYYFILL\fP), but 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 API primitives do not have a fixed form: they should be
defined as strings containing free\-form pieces of code with interpolated
variables of the form \fB@@{var}\fP or \fB@@\fP (they correspond to arguments in
function\-like style).
This configuration may be overridden for individual API primitives, see for
example \fBre2c:YYFILL:naked\fP configuration for \fBYYFILL\fP\&.
.TP
.B \fBre2c:bit\-vectors\fP, \fBre2c:flags:bit\-vectors\fP, \fBre2c:flags:b\fP
Same as the \fB\-\-bit\-vectors\fP option, but can be configured on per\-block
basis.
.TP
.B \fBre2c:captures\fP, \fBre2c:leftmost\-captures\fP
Same as the \fB\-\-leftmost\-captures\fP option, but can be configured on
per\-block basis.
.TP
.B \fBre2c:captvars\fP, \fBre2c:leftmost\-captvars\fP
Same as the \fB\-\-leftmost\-captvars\fP option, but can be configured on
per\-block basis.
.TP
.B \fBre2c:case\-insensitive\fP, \fBre2c:flags:case\-insensitive\fP
Same as the \fB\-\-case\-insensitive\fP option, but can be configured on
per\-block basis.
.TP
.B \fBre2c:case\-inverted\fP, \fBre2c:flags:case\-inverted\fP
Same as the \fB\-\-case\-inverted\fP option, but can be configured on per\-block
basis.
.TP
.B \fBre2c:case\-ranges\fP, \fBre2c:flags:case\-ranges\fP
Same as the \fB\-\-case\-ranges\fP option, but can be configured on per\-block
basis.
.TP
.B \fBre2c:computed\-gotos\fP, \fBre2c:flags:computed\-gotos\fP, \fBre2c:flags:g\fP
Same as the \fB\-\-computed\-gotos\fP option, but can be configured on per\-block
basis.
.TP
.B \fBre2c:computed\-gotos:relative\fP, \fBre2c:cgoto:relative\fP
Same as the \fB\-\-computed\-gotos\-relative\fP option, but can be configured on
per\-block basis.
.TP
.B \fBre2c:computed\-gotos:threshold\fP, \fBre2c:cgoto:threshold\fP
If computed \fBgoto\fP is used, this configuration specifies the complexity
threshold that triggers the generation of jump tables instead of nested
\fBif\fP statements and bitmaps. The default value is \fB9\fP\&.
.TP
.B \fBre2c:cond:abort\fP
If set to a positive integer value, the default case in the generated
condition dispatch aborts program execution.
.TP
.B \fBre2c:cond:goto\fP
Specifies a piece of code used for the autogenerated shortcut rules \fB:=>\fP
in conditions. The default is \fBgoto @@;\fP\&.
The \fB@@\fP placeholder is substituted with condition name (see
configurations \fBre2c:api:sigil\fP and \fBre2c:cond:goto@cond\fP).
.TP
.B \fBre2c:cond:goto@cond\fP
Specifies the sigil used for argument substitution in \fBre2c:cond:goto\fP
definition. The default value is \fB@@\fP\&.
Overrides the more generic \fBre2c:api:sigil\fP configuration.
.TP
.B \fBre2c:cond:divider\fP
Defines the divider for condition blocks.
The default value is \fB/* *********************************** */\fP\&.
Placeholders are substituted with condition name (see \fBre2c:api;sigil\fP and
\fBre2c:cond:divider@cond\fP).
.TP
.B \fBre2c:cond:divider@cond\fP
Specifies the sigil used for argument substitution in \fBre2c:cond:divider\fP
definition. The default is \fB@@\fP\&.
Overrides the more generic \fBre2c:api:sigil\fP configuration.
.TP
.B \fBre2c:cond:prefix\fP, \fBre2c:condprefix\fP
Specifies the prefix used for condition labels.
The default is \fByyc_\fP\&.
.TP
.B \fBre2c:cond:enumprefix\fP, \fBre2c:condenumprefix\fP
Specifies the prefix used for condition identifiers.
The default is \fByyc\fP\&.
.TP
.B \fBre2c:debug\-output\fP, \fBre2c:flags:debug\-output\fP, \fBre2c:flags:d\fP
Same as the \fB\-\-debug\-output\fP option, but can be configured on per\-block
basis.
.TP
.B \fBre2c:empty\-class\fP, \fBre2c:flags:empty\-class\fP
Same as the \fB\-\-empty\-class\fP option, but can be configured on per\-block
basis.
.TP
.B \fBre2c:encoding:ebcdic\fP, \fBre2c:flags:ecb\fP, \fBre2c:flags:e\fP
Same as the \fB\-\-ebcdic\fP option, but can be configured on per\-block basis.
.TP
.B \fBre2c:encoding:ucs2\fP, \fBre2c:flags:wide\-chars\fP, \fBre2c:flags:w\fP
Same as the \fB\-\-ucs2\fP option, but can be configured on per\-block basis.
.TP
.B \fBre2c:encoding:utf8\fP, \fBre2c:flags:utf\-8\fP, \fBre2c:flags:8\fP
Same as the \fB\-\-utf8\fP option, but can be configured on per\-block basis.
.TP
.B \fBre2c:encoding:utf16\fP, \fBre2c:flags:utf\-16\fP, \fBre2c:flags:x\fP
Same as the \fB\-\-utf16\fP option, but can be configured on per\-block basis.
.TP
.B \fBre2c:encoding:utf32\fP, \fBre2c:flags:unicode\fP, \fBre2c:flags:u\fP
Same as the \fB\-\-utf32\fP option, but can be configured on per\-block basis.
.TP
.B \fBre2c:encoding\-policy\fP, \fBre2c:flags:encoding\-policy\fP
Same as the \fB\-\-encoding\-policy\fP option, but can be configured on per\-block
basis.
.TP
.B \fBre2c:eof\fP
Specifies the sentinel symbol used with the end\-of\-input rule \fB$\fP\&. The
default value is \fB\-1\fP (\fB$\fP rule is not used). Other possible values
include all valid code units. Only decimal numbers are recognized.
.TP
.B \fBre2c:header\fP, \fBre2c:flags:type\-header\fP, \fBre2c:flags:t\fP
Specifies the name of the generated header file relative to the directory of
the output file. Same as the \fB\-\-header\fP option except that the file path
is relative.
.TP
.B \fBre2c:indent:string\fP
Specifies the string used for indentation. The default is a single tab
character \fB\(dq\et\(dq\fP\&. Indent string should contain whitespace characters only.
To disable indentation entirely, set this configuration to an empty string.
.TP
.B \fBre2c:indent:top\fP
Specifies the minimum amount of indentation to use. The default value is
zero. The value should be a non\-negative integer number.
.TP
.B \fBre2c:invert\-captures\fP
Same as the \fB\-\-invert\-captures\fP option, but can be configured on per\-block
basis.
.TP
.B \fBre2c:label:prefix\fP, \fBre2c:labelprefix\fP
Specifies the prefix used for DFA state labels. The default is \fByy\fP\&.
.TP
.B \fBre2c:label:start\fP, \fBre2c:startlabel\fP
Controls the generation of a block start label. The default value is zero,
which means that the start label is generated only if it is used. An integer
value greater than zero forces the generation of start label even if it is
unused by the lexer. A string value also forces start label generation and
sets the label name to the specified string. This configuration applies only
to the current block (it is reset to default for the next block).
.TP
.B \fBre2c:label:yyFillLabel\fP
Specifies the prefix of \fBYYFILL\fP labels used with \fBre2c:eof\fP and in
storable state mode.
.TP
.B \fBre2c:label:yyloop\fP
Specifies the name of the label marking the start of the lexer loop with
\fB\-\-loop\-switch\fP option. The default is \fByyloop\fP\&.
.TP
.B \fBre2c:label:yyNext\fP
Specifies the name of the optional label that follows \fBYYGETSTATE\fP switch
in storable state mode (enabled with \fBre2c:state:nextlabel\fP). The default
is \fByyNext\fP\&.
.TP
.B \fBre2c:lookahead\fP, \fBre2c:flags:lookahead\fP
Deprecated (see the deprecated \fB\-\-no\-lookahead\fP option).
.TP
.B \fBre2c:monadic\fP
If set to non\-zero, the generated lexer will use monadic notation (this
configuration is specific to Haskell).
.TP
.B \fBre2c:nested\-ifs\fP, \fBre2c:flags:nested\-ifs\fP, \fBre2c:flags:s\fP
Same as the \fB\-\-nested\-ifs\fP option, but can be configured on per\-block
basis.
.TP
.B \fBre2c:posix\-captures\fP, \fBre2c:flags:posix\-captures\fP, \fBre2c:flags:P\fP
Same as the \fB\-\-posix\-captures\fP option, but can be configured on per\-block
basis.
.TP
.B \fBre2c:posix\-captvars\fP
Same as the \fB\-\-posix\-captvars\fP option, but can be configured on per\-block
basis.
.TP
.B \fBre2c:tags\fP, \fBre2c:flags:tags\fP, \fBre2c:flags:T\fP
Same as the \fB\-\-tags\fP option, but can be configured on per\-block basis.
.TP
.B \fBre2c:tags:expression\fP
Specifies the expression used for tag variables.
By default re2zig generates expressions of the form \fByyt<N>\fP\&. This might
be inconvenient, for example if tag variables are defined as fields in a
struct. All occurrences of \fB@@{tag}\fP or \fB@@\fP are replaced with the
actual tag name. For example, \fBre2c:tags:expression = \(dqs.@@\(dq;\fP results
in expressions of the form \fBs.yyt<N>\fP in the generated code.
See also \fBre2c:api:sigil\fP configuration.
.TP
.B \fBre2c:tags:negative\fP
Specifies the constant expression that is used for negative tag value
(typically this would be \fB\-1\fP if tags are integer offsets in the input
string, or null pointer if they are pointers).
.TP
.B \fBre2c:tags:prefix\fP
Specifies the prefix for tag variable names. The default is \fByyt\fP\&.
.TP
.B \fBre2c:sentinel\fP
Specifies the sentinel symbol used for the end\-of\-input checks (when bounds
checks are disabled with \fBre2c:yyfill:enable = 0;\fP and \fBre2c:eof\fP is not
set). This configuration does not affect code generation: its purpose is to
verify that the sentinel is not allowed in the middle of a rule, and ensure
that the lexer won\(aqt read past the end of buffer. The default value is
\fI\-1\(ga\fP (in that case re2zig assumes that the sentinel is zero, which is the
most common case). Only decimal numbers are recognized.
.TP
.B \fBre2c:state:abort\fP
If set to a positive integer value, the default case in the generated
state dispatch aborts program execution, and an explicit \fB\-1\fP case
contains transition to the start of the block.
.TP
.B \fBre2c:state:nextlabel\fP
Controls if the \fBYYGETSTATE\fP switch is followed by an \fByyNext\fP label
(the default value is zero, which corresponds to no label).
Alternatively one can use \fBre2c:label:start\fP to generate a specific start
label, or an explicit \fBgetstate\fP block to generate the \fBYYGETSTATE\fP
switch separately from the lexer block.
.TP
.B \fBre2c:unsafe\fP, \fBre2c:flags:unsafe\fP
Same as the \fB\-\-no\-unsafe\fP option, but can be configured on per\-block
basis.
If set to zero, it suppresses the generation of \fBunsafe\fP wrappers around
\fBYYPEEK\fP\&. The default is non\-zero (wrappers are generated).
This configuration is specific to Rust.
.TP
.B \fBre2c:YYBACKUP\fP, \fBre2c:define:YYBACKUP\fP
Defines generic API primitive \fBYYBACKUP\fP\&.
.TP
.B \fBre2c:YYBACKUPCTX\fP, \fBre2c:define:YYBACKUPCTX\fP
Defines generic API primitive \fBYYBACKUPCTX\fP\&.
.TP
.B \fBre2c:YYCONDTYPE\fP, \fBre2c:define:YYCONDTYPE\fP
Defines API primitive \fBYYCONDTYPE\fP\&.
.TP
.B \fBre2c:YYCTYPE\fP, \fBre2c:define:YYCTYPE\fP
Defines API primitive \fBYYCTYPE\fP\&.
.TP
.B \fBre2c:YYCTXMARKER\fP, \fBre2c:define:YYCTXMARKER\fP
Defines API primitive \fBYYCTXMARKER\fP\&.
.TP
.B \fBre2c:YYCURSOR\fP, \fBre2c:define:YYCURSOR\fP
Defines API primitive \fBYYCURSOR\fP\&.
.TP
.B \fBre2c:YYDEBUG\fP, \fBre2c:define:YYDEBUG\fP
Defines API primitive \fBYYDEBUG\fP\&.
.TP
.B \fBre2c:YYFILL\fP, \fBre2c:define:YYFILL\fP
Defines API primitive \fBYYFILL\fP\&.
.TP
.B \fBre2c:YYFILL@len\fP, \fBre2c:define:YYFILL@len\fP
Specifies the sigil used for argument substitution in \fBYYFILL\fP
definition. Defaults to \fB@@\fP\&.
Overrides the more generic \fBre2c:api:sigil\fP configuration.
.TP
.B \fBre2c:YYFILL:naked\fP, \fBre2c:define:YYFILL:naked\fP
Overrides the more generic \fBre2c:api:style\fP configuration for \fBYYFILL\fP\&.
Zero value corresponds to free\-form API style.
.TP
.B \fBre2c:YYFN\fP
Defines API primitive \fBYYFN\fP\&.
.TP
.B \fBre2c:YYINPUT\fP
Defines API primitive \fBYYINPUT\fP\&.
.TP
.B \fBre2c:YYGETCOND\fP, \fBre2c:define:YYGETCONDITION\fP
Defines API primitive \fBYYGETCOND\fP\&.
.TP
.B \fBre2c:YYGETCOND:naked\fP, \fBre2c:define:YYGETCONDITION:naked\fP
Overrides the more generic \fBre2c:api:style\fP configuration for
\fBYYGETCOND\fP\&. Zero value corresponds to free\-form API style.
.TP
.B \fBre2c:YYGETSTATE\fP, \fBre2c:define:YYGETSTATE\fP
Defines API primitive \fBYYGETSTATE\fP\&.
.TP
.B \fBre2c:YYGETSTATE:naked\fP, \fBre2c:define:YYGETSTATE:naked\fP
Overrides the more generic \fBre2c:api:style\fP configuration for
\fBYYGETSTATE\fP\&. Zero value corresponds to free\-form API style.
.TP
.B \fBre2c:YYGETACCEPT\fP, \fBre2c:define:YYGETACCEPT\fP
Defines API primitive \fBYYGETACCEPT\fP\&.
.TP
.B \fBre2c:YYLESSTHAN\fP, \fBre2c:define:YYLESSTHAN\fP
Defines generic API primitive \fBYYLESSTHAN\fP\&.
.TP
.B \fBre2c:YYLIMIT\fP, \fBre2c:define:YYLIMIT\fP
Defines API primitive \fBYYLIMIT\fP\&.
.TP
.B \fBre2c:YYMARKER\fP, \fBre2c:define:YYMARKER\fP
Defines API primitive \fBYYMARKER\fP\&.
.TP
.B \fBre2c:YYMTAGN\fP, \fBre2c:define:YYMTAGN\fP
Defines generic API primitive \fBYYMTAGN\fP\&.
.TP
.B \fBre2c:YYMTAGP\fP, \fBre2c:define:YYMTAGP\fP
Defines generic API primitive \fBYYMTAGP\fP\&.
.TP
.B \fBre2c:YYPEEK\fP, \fBre2c:define:YYPEEK\fP
Defines generic API primitive \fBYYPEEK\fP\&.
.TP
.B \fBre2c:YYRESTORE\fP, \fBre2c:define:YYRESTORE\fP
Defines generic API primitive \fBYYRESTORE\fP\&.
.TP
.B \fBre2c:YYRESTORECTX\fP, \fBre2c:define:YYRESTORECTX\fP
Defines generic API primitive \fBYYRESTORECTX\fP\&.
.TP
.B \fBre2c:YYRESTORETAG\fP, \fBre2c:define:YYRESTORETAG\fP
Defines generic API primitive \fBYYRESTORETAG\fP\&.
.TP
.B \fBre2c:YYSETCOND\fP, \fBre2c:define:YYSETCONDITION\fP
Defines API primitive \fBYYSETCOND\fP\&.
.TP
.B \fBre2c:YYSETCOND@cond\fP, \fBre2c:define:YYSETCONDITION@cond\fP
Specifies the sigil used for argument substitution in \fBYYSETCOND\fP
definition. The default value is \fB@@\fP\&.
Overrides the more generic \fBre2c:api:sigil\fP configuration.
.TP
.B \fBre2c:YYSETCOND:naked\fP, \fBre2c:define:YYSETCONDITION:naked\fP
Overrides the more generic \fBre2c:api:style\fP configuration for
\fBYYSETCOND\fP\&. Zero value corresponds to free\-form API style.
.TP
.B \fBre2c:YYSETSTATE\fP, \fBre2c:define:YYSETSTATE\fP
Defines API primitive \fBYYSETSTATE\fP\&.
.TP
.B \fBre2c:YYSETSTATE@state\fP, \fBre2c:define:YYSETSTATE@state\fP
Specifies the sigil used for argument substitution in \fBYYSETSTATE\fP
definition. The default value is \fB@@\fP\&.
Overrides the more generic \fBre2c:api:sigil\fP configuration.
.TP
.B \fBre2c:YYSETSTATE:naked\fP, \fBre2c:define:YYSETSTATE:naked\fP
Overrides the more generic \fBre2c:api:style\fP configuration for
\fBYYSETSTATE\fP\&. Zero value corresponds to free\-form API style.
.TP
.B \fBre2c:YYSETACCEPT\fP, \fBre2c:define:YYSETACCEPT\fP
Defines API primitive \fBYYSETACCEPT\fP\&.
.TP
.B \fBre2c:YYSKIP\fP, \fBre2c:define:YYSKIP\fP
Defines generic API primitive \fBYYSKIP\fP\&.
.TP
.B \fBre2c:YYSHIFT\fP, \fBre2c:define:YYSHIFT\fP
Defines generic API primitive \fBYYSHIFT\fP\&.
.TP
.B \fBre2c:YYCOPYMTAG\fP, \fBre2c:define:YYCOPYMTAG\fP
Defines generic API primitive \fBYYCOPYMTAG\fP\&.
.TP
.B \fBre2c:YYCOPYSTAG\fP, \fBre2c:define:YYCOPYSTAG\fP
Defines generic API primitive \fBYYCOPYSTAG\fP\&.
.TP
.B \fBre2c:YYSHIFTMTAG\fP, \fBre2c:define:YYSHIFTMTAG\fP
Defines generic API primitive \fBYYSHIFTMTAG\fP\&.
.TP
.B \fBre2c:YYSHIFTSTAG\fP, \fBre2c:define:YYSHIFTSTAG\fP
Defines generic API primitive \fBYYSHIFTSTAG\fP\&.
.TP
.B \fBre2c:YYSTAGN\fP, \fBre2c:define:YYSTAGN\fP
Defines generic API primitive \fBYYSTAGN\fP\&.
.TP
.B \fBre2c:YYSTAGP\fP, \fBre2c:define:YYSTAGP\fP
Defines generic API primitive \fBYYSTAGP\fP\&.
.TP
.B \fBre2c:yyaccept\fP, \fBre2c:variable:yyaccept\fP
Defines API primitive \fByyaccept\fP\&.
.TP
.B \fBre2c:yybm\fP, \fBre2c:variable:yybm\fP
Defines API primitive \fByybm\fP\&.
.TP
.B \fBre2c:yybm:hex\fP, \fBre2c:variable:yybm:hex\fP
If set to nonzero, bitmaps for the \fB\-\-bit\-vectors\fP option are generated
in hexadecimal format. The default is zero (bitmaps are in decimal format).
.TP
.B \fBre2c:yych\fP, \fBre2c:variable:yych\fP
Defines API primitive \fByych\fP\&.
.TP
.B \fBre2c:yych:emit\fP, \fBre2c:variable:yych:emit\fP
If set to zero, \fByych\fP definition is not generated.
The default is non\-zero.
.TP
.B \fBre2c:yych:conversion\fP, \fBre2c:variable:yych:conversion\fP
If set to non\-zero, re2zig automatically generates a conversion to \fBYYCTYPE\fP
every time \fByych\fP is read. The default is to zero (no conversion).
.TP
.B \fBre2c:yych:literals\fP, \fBre2c:variable:yych:literals\fP
Specifies the form of literals that \fByych\fP is matched against. Possible
values are: \fBchar\fP (character literals in single quotes, non\-printable
ones use escape sequences that start with backslash), \fBhex\fP (hexadecimal
integers) and \fBchar_or_hex\fP (a mixture of both, character literals for
printable characters and hexadecimal integers for others).
.TP
.B \fBre2c:yyctable\fP, \fBre2c:variable:yyctable\fP
Defines API primitive \fByyctable\fP\&.
.TP
.B \fBre2c:yynmatch\fP, \fBre2c:variable:yynmatch\fP
Defines API primitive \fByynmatch\fP\&.
.TP
.B \fBre2c:yypmatch\fP, \fBre2c:variable:yypmatch\fP
Defines API primitive \fByypmatch\fP\&.
.TP
.B \fBre2c:yytarget\fP, \fBre2c:variable:yytarget\fP
Defines API primitive \fByytarget\fP\&.
.TP
.B \fBre2c:yystable\fP, \fBre2c:variable:yystable\fP
Deprecated.
.TP
.B \fBre2c:yystate\fP, \fBre2c:variable:yystate\fP
Defines API primitive \fByystate\fP\&.
.TP
.B \fBre2c:yyfill\fP, \fBre2c:variable:yyfill\fP
Defines API primitive \fByyfill\fP\&.
.TP
.B \fBre2c:yyfill:check\fP
If set to zero, suppresses the generation of pre\-\fBYYFILL\fP check for the
number of input characters (the \fBYYLESSTHAN\fP definition in generic API and
the \fBYYLIMIT\fP\-based comparison in C pointer API). The default is non\-zero
(generate the check).
.TP
.B \fBre2c:yyfill:enable\fP
If set to zero, suppresses the generation of \fBYYFILL\fP (together
with the check). This should be used when the whole input fits into one piece
of memory (there is no need for buffering) and the end\-of\-input checks do not
rely on the \fBYYFILL\fP checks (e.g. if a sentinel character is used).
Use warnings (\fB\-W\fP option) and \fBre2c:sentinel\fP configuration to verify
that the generated lexer cannot read past the end of input.
The default is non\-zero (\fBYYFILL\fP is enabled).
.TP
.B \fBre2c:yyfill:parameter\fP
If set to zero, suppresses the generation of parameter passed to \fBYYFILL\fP\&.
The 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 \fBre2c:YYFILL:naked\fP or
\fBre2c:api:style\fP\&.
.TP
.B \fBre2c:yyfn:sep\fP
Specifies separator used in \fBYYFN\fP elements (defaults to semicolon).
.TP
.B \fBre2c:yyfn:throw\fP
Specifies exceptions thrown by \fBYYFN\fP function (defaults to empty, which
means no exceptions).
.UNINDENT
.SS Regular expressions
.sp
re2zig uses the following syntax for regular expressions:
.INDENT 0.0
.TP
.B \fB\(dqfoo\(dq\fP
Case\-sensitive string literal.
.TP
.B \fB\(aqfoo\(aq\fP
Case\-insensitive string literal.
.TP
.B \fB[a\-xyz]\fP, \fB[^a\-xyz]\fP
Character class (possibly negated).
.TP
.B \fB\&.\fP
Any character except newline.
.TP
.B \fBR \e S\fP
Difference of character classes \fBR\fP and \fBS\fP\&.
.TP
.B \fBR*\fP
Zero or more occurrences of \fBR\fP\&.
.TP
.B \fBR+\fP
One or more occurrences of \fBR\fP\&.
.TP
.B \fBR?\fP
Optional \fBR\fP\&.
.TP
.B \fBR{n}\fP
Repetition of \fBR\fP exactly \fBn\fP times.
.TP
.B \fBR{n,}\fP
Repetition of \fBR\fP at least \fBn\fP times.
.TP
.B \fBR{n,m}\fP
Repetition of \fBR\fP from \fBn\fP to \fBm\fP times.
.TP
.B \fB(R)\fP
Just \fBR\fP; parentheses are used to override precedence. If submatch
extraction is enabled, \fB(R)\fP is a capturing or a non\-capturing group
depending on \fB\-\-invert\-captures\fP option.
.TP
.B \fB(!R)\fP
If submatch extraction is enabled, \fB(!R)\fP is a non\-capturing or a
capturing group depending on \fB\-\-invert\-captures\fP option.
.TP
.B \fBR S\fP
Concatenation: \fBR\fP followed by \fBS\fP\&.
.TP
.B \fBR | S\fP
Alternative: \fBR or S\fP\&.
.TP
.B \fBR / S\fP
Lookahead: \fBR\fP followed by \fBS\fP, but \fBS\fP is not consumed.
.TP
.B \fBname\fP
Regular expression defined as \fBname\fP (or literal string \fB\(dqname\(dq\fP in
Flex compatibility mode).
.TP
.B \fB{name}\fP
Regular expression defined as \fBname\fP in Flex compatibility mode.
.TP
.B \fB@stag\fP
An \fIs\-tag\fP: saves the last input position at which \fB@stag\fP matches in a
variable named \fBstag\fP\&.
.TP
.B \fB#mtag\fP
An \fIm\-tag\fP: saves all input positions at which \fB#mtag\fP matches in a
variable named \fBmtag\fP\&.
.UNINDENT
.sp
Character classes and string literals may contain the following escape
sequences: \fB\ea\fP, \fB\eb\fP, \fB\ef\fP, \fB\en\fP, \fB\er\fP, \fB\et\fP, \fB\ev\fP, \fB\e\e\fP,
octal escapes \fB\eooo\fP and hexadecimal escapes \fB\exhh\fP, \fB\euhhhh\fP and
\fB\eUhhhhhhhh\fP\&.
.SS Actions
.sp
Here is a list of predefined actions supported by re2zig:
.INDENT 0.0
.TP
.B \fB!entry code\fP
Entry action binds a user\-defined block of \fBcode\fP to the start state of
the current finite state machine. If \fI\%start conditions\fP are used, the entry
action can be set individually for each condition. This action may be used
to perform initialization, e.g. to save start location of a lexeme.
.TP
.B \fB!pre_rule code\fP
Pre\-rule action prepends a user\-defined block of \fBcode\fP to semantic actions
of all rules in the current block (or condition, if \fI\%start conditions\fP are
used). This action may be used to factor out the common part of all semantic
actions (e.g. saving the end location of a lexeme).
.TP
.B \fB!post_rule code\fP
Post\-rule action appends a user\-defined block of \fBcode\fP to semantic actions
of all rules in the current block (or condition, if \fI\%start conditions\fP are
used). This action may be used to emit trap statements that guard against
unintended control flow.
.UNINDENT
.SS Directives
.sp
Here is a full list of directives supported by re2zig:
.INDENT 0.0
.TP
.B \fB!use:name ;\fP
An in\-block use directive that merges a previously defined rules block with
the specified \fBname\fP into the current block. Named definitions, configurations
and rules of the referenced block are added to the current ones. Conflicts
between overlapping rules and configurations are resolved in the usual way:
the first rule takes priority, and the latest configuration overrides the
preceding ones. One exception is the special rules \fB*\fP, \fB$\fP and \fB<!>\fP
for which a block\-local definition always takes priority. A use directive
can be placed anywhere inside of a block, and multiple use directives are
allowed.
.TP
.B \fB!include file ;\fP
This directive is the same as \fBinclude\fP block: it inserts \fBfile\fP
contents verbatim in place of the directive.
.UNINDENT
.SS Program interface
.sp
The generated code interfaces with the outer program with the help of
\fIprimitives\fP, collectively referred to as the \fIAPI\fP\&.
Which primitives should be defined for a particular program depends on multiple
factors, including the complexity of regular expressions, input representation,
buffering and the use of various features. All the necessary primitives should
be defined by the user in the form of macros, functions, variables or any other
suitable form that makes the generated code syntactically and semantically
correct. re2zig does not (and cannot) check the definitions, so if anything is
missing or defined incorrectly, the generated program may have compile\-time or
run\-time errors.
This manual provides examples of API definitions in the most common cases.
.sp
re2zig has three API flavors that define the core set of primitives used by a
program:
.INDENT 0.0
.TP
.B \fBSimple API\fP
This is the default API for the Zig backend. It consists of the following
primitives: \fBYYINPUT\fP (which should be defined as a sequence of code
units, e.g. a string) and \fBYYCURSOR\fP, \fBYYMARKER\fP, \fBYYCTXMARKER\fP,
\fBYYLIMIT\fP (which should be defined as indices in \fBYYINPUT\fP).
.nf
.fi
.sp
.TP
.B \fBRecord API\fP
Record API is useful in cases when lexer state must be stored in a struct.
It is enabled with \fB\-\-api record\fP option or \fBre2c:api = record\fP
configuration. This API consists of a variable \fByyrecord\fP (the
name can be overridden with \fBre2c:yyrecord\fP) that should be
defined as a struct with fields \fByyinput\fP, \fByycursor\fP, \fByymarker\fP,
\fByyctxmarker\fP, \fByylimit\fP (only the fields used by the generated code
need to be defined, and their names can be configured).
.nf
.fi
.sp
.TP
.B \fBGeneric API\fP
This is the most flexible API. It is enabled with \fB\-\-api generic\fP option
or \fBre2c:api = generic\fP configuration.
It contains primitives for generic operations:
\fBYYPEEK\fP,
\fBYYSKIP\fP,
\fBYYBACKUP\fP,
\fBYYBACKUPCTX\fP,
\fBYYSTAGP\fP,
\fBYYSTAGN\fP,
\fBYYMTAGP\fP,
\fBYYMTAGN\fP,
\fBYYRESTORE\fP,
\fBYYRESTORECTX\fP,
\fBYYRESTORETAG\fP,
\fBYYSHIFT\fP,
\fBYYSHIFTSTAG\fP,
\fBYYSHIFTMTAG\fP,
\fBYYLESSTHAN\fP\&.
.UNINDENT
.sp
Here is a full list of API primitives that may be used by the generated code in
order to interface with the outer program.
.INDENT 0.0
.TP
.B \fBYYCTYPE\fP
The type of the input characters (code units).
For ASCII, EBCDIC and UTF\-8 encodings it should be 1\-byte unsigned integer.
For UTF\-16 or UCS\-2 it should be 2\-byte unsigned integer. For UTF\-32 it
should be 4\-byte unsigned integer.
.TP
.B \fBYYCURSOR\fP
An l\-value that stores the current input position (a pointer or an integer
offset in \fBYYINPUT\fP). Initially \fBYYCURSOR\fP should point to the first
input character, and later it is advanced by the generated code. When a rule
matches, \fBYYCURSOR\fP position is the one after the last matched character.
.TP
.B \fBYYLIMIT\fP
An r\-value that stores the end of input position (a pointer or an integer
offset in \fBYYINPUT\fP). Initially \fBYYLIMIT\fP should point to the position
after the last available input character. It is not changed by the
generated code. The lexer compares \fBYYCURSOR\fP to \fBYYLIMIT\fP
in order to determine if there are enough input characters left.
.TP
.B \fBYYMARKER\fP
An l\-value that stores the position of the latest matched rule (a pointer or
an integer offset in \fBYYINPUT\fP). It is used to restore the \fBYYCURSOR\fP
position if the longer match fails and the lexer needs to rollback.
Initialization is not needed.
.TP
.B \fBYYCTXMARKER\fP
An l\-value that stores the position of the trailing context (a pointer or an
integer offset in \fBYYINPUT\fP). No initialization is needed. \fBYYCTXMARKER\fP
is needed only if the lookahead operator \fB/\fP is used.
.TP
.B \fBYYFILL\fP
A generic API primitive with one variable \fBlen\fP\&.
\fBYYFILL\fP should provide at least \fBlen\fP more input characters or fail.
If \fBre2c:eof\fP is used, then \fBlen\fP is always \fB1\fP and \fBYYFILL\fP should
always return to the calling function; zero return value indicates success.
If \fBre2c:eof\fP is not used, then \fBYYFILL\fP return value is ignored and it
should not return on failure. The maximum value of \fBlen\fP is \fBYYMAXFILL\fP\&.
.TP
.B \fBYYFN\fP
A primitive that defines function prototype in \fB\-\-recursive\-functions\fP
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
\fBre2c:fn:sep\fP configuration (typically a semicolon). The first array
element defines function name and return type (empty for a void function).
Subsequent elements define function arguments: first, the expression for the
argument used in function body (usually just a name); second, argument type;
third, an optional formal parameter (it defaults to the first component \-
usually both the argument and the parameter are the same identifier).
.TP
.B \fBYYINPUT\fP
An r\-value that stores the current input character sequence (string, buffer,
etc.).
.TP
.B \fBYYMAXFILL\fP
An integral constant equal to the maximum value of the argument to
\fBYYFILL\fP\&. It can be generated with a \fBmax\fP block.
.TP
.B \fBYYLESSTHAN\fP
A generic API primitive with one variable \fBlen\fP\&.
It should be defined as an r\-value of boolean type that equals \fBtrue\fP if
and only if there are less than \fBlen\fP input characters left.
.TP
.B \fBYYPEEK\fP
A generic API primitive with no variables.
It should be defined as an r\-value of type \fBYYCTYPE\fP that is equal to the
character at the current input position.
.TP
.B \fBYYSKIP\fP
A generic API primitive that should advance the current input position by
one code unit.
.TP
.B \fBYYBACKUP\fP
A generic API primitive that should save the current input position (to be
restored with \fBYYRESTORE\fP later).
.TP
.B \fBYYRESTORE\fP
A generic API primitive that should restore the current input position to
the value saved by \fBYYBACKUP\fP\&.
.TP
.B \fBYYBACKUPCTX\fP
A generic API primitive that should save the current input position as the
position of the trailing context (to be restored with \fBYYRESTORECTX\fP
later).
.TP
.B \fBYYRESTORECTX\fP
A generic API primitive that should restore the trailing context position
saved with \fBYYBACKUPCTX\fP\&.
.TP
.B \fBYYRESTORETAG\fP
A generic API primitive with one variable \fBtag\fP that should restore the
trailing context position to the value of \fBtag\fP\&.
.TP
.B \fBYYSTAGP\fP
A generic API primitive with one variable \fBtag\fP, where \fBtag\fP can be a
pointer or an offset in \fBYYINPUT\fP (see submatch extraction section for
details). \fBYYSTAGP\fP should set \fBtag\fP to the current input position.
.TP
.B \fBYYSTAGN\fP
A generic API primitive with one variable \fBtag\fP, where \fBtag\fP can be a
pointer or an offset in \fBYYINPUT\fP (see submatch extraction section for
details). \fBYYSTAGN\fP should to set \fBtag\fP to a value that represents
non\-existent input position.
.TP
.B \fBYYMTAGP\fP
A generic API primitive with one variable \fBtag\fP\&.
\fBYYMTAGP\fP should append the current position to the submatch history of
\fBtag\fP (see the submatch extraction section for details.)
.TP
.B \fBYYMTAGN\fP
A generic API primitive with one variable \fBtag\fP\&.
\fBYYMTAGN\fP should append a value that represents non\-existent input
position position to the submatch history of \fBtag\fP (see the submatch
extraction section for details.)
.TP
.B \fBYYSHIFT\fP
A generic API primitive with one variable \fBshift\fP that should shift the
current input position by \fBshift\fP characters (the shift value may be
negative).
.TP
.B \fBYYCOPYSTAG\fP
A generic API primitive with two variables, \fBlhs\fP and \fBrhs\fP that should
copy right\-hand\-side s\-tag variable \fBrhs\fP to the left\-hand\-side s\-tag
variable \fBlhs\fP\&. For most languages this primitive has a default definition
that assigns \fBlhs\fP to \fBrhs\fP\&.
.TP
.B \fBYYCOPYMTAG\fP
A generic API primitive with two variables, \fBlhs\fP and \fBrhs\fP that should
copy right\-hand\-side m\-tag variable \fBrhs\fP to the left\-hand\-side m\-tag
variable \fBlhs\fP\&. For most languages this primitive has a default definition
that assigns \fBlhs\fP to \fBrhs\fP\&.
.TP
.B \fBYYSHIFTSTAG\fP
A generic API primitive with two variables, \fBtag\fP and \fBshift\fP that
should shift \fBtag\fP by \fBshift\fP code units (the shift value may be
negative).
.TP
.B \fBYYSHIFTMTAG\fP
A generic API primitive with two variables, \fBtag\fP and \fBshift\fP that
should shift the latest value in the history of \fBtag\fP by \fBshift\fP code
units (the shift value may be negative).
.TP
.B \fBYYMAXNMATCH\fP
An integral constant equal to the maximal number of POSIX capturing groups
in a rule. It is generated with a \fBmaxnmatch\fP block.
.TP
.B \fBYYCONDTYPE\fP
The type of the condition enum.
It can be generated either with \fBconditions\fP block or \fB\-\-header\fP option.
.TP
.B \fBYYGETACCEPT\fP
A primitive with one variable \fBvar\fP that stores numeric selector of the
accepted rule. For most languages this primitive has a default definition
that reads from \fBvar\fP\&.
.TP
.B \fBYYSETACCEPT\fP
A primitive with two variables: \fBvar\fP (an l\-value that stores numeric
selector of the accepted rule), and \fBval\fP (the value of selector). For
most languages this primitive has a default definition that assigns \fBvar\fP
to \fBval\fP\&.
.TP
.B \fBYYGETCOND\fP
An r\-value of type \fBYYCONDTYPE\fP that is equal to the current condition
identifier.
.TP
.B \fBYYSETCOND\fP
A primitive with one variable \fBcond\fP that should set the current
condition identifier to \fBcond\fP\&.
.TP
.B \fBYYGETSTATE\fP
An r\-value of integer type that is equal to the current lexer state. It
should be initialized to \fB\-1\fP\&.
.TP
.B \fBYYSETSTATE\fP
A primitive with one variable \fBstate\fP that should set the current lexer
state to \fBstate\fP\&.
.TP
.B \fBYYDEBUG\fP
This primitive is generated only with \fB\-d\fP, \fB\-\-debug\-output\fP option.
Its purpose is to add logging to the generated code (typical \fBYYDEBUG\fP
definition is a print statement). \fBYYDEBUG\fP statements are generated in
every state and have two variables: \fBstate\fP (either a DFA state index or
\fB\-1\fP) and \fBsymbol\fP (the current input symbol).
.TP
.B \fByyaccept\fP
An l\-value of unsigned integral type that stores the number of the latest
matched rule. User definition is necessary only with \fB\-\-storable\-state\fP
option.
.TP
.B \fByybm\fP
A table containing compressed bitmaps for up to 8 transitions (used with
the \fB\-\-bitmaps\fP 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 idea of this bitmap is to replace many \fIif\fP
branches or \fIswitch\fP cases with one check of a single bit in the table.
.TP
.B \fByych\fP
An l\-value of type \fBYYCTYPE\fP that stores the current input character.
User definition is necessary only with \fB\-f\fP \fB\-\-storable\-state\fP option.
.TP
.B \fByyctable\fP
Jump table generated for the initial condition dispatch (enabled with the
combination of \fB\-\-conditions\fP and \fB\-\-computed\-gotos\fP options).
.TP
.B \fByyfill\fP
An l\-value that stores the result of \fBYYFILL\fP call (this may be necessary
for pure functional languages, where \fBYYFILL\fP is a monadic function with
complex return value).
.TP
.B \fByynmatch\fP
An l\-value of unsigned integral type that stores the number of POSIX
capturing groups in the matched rule.
Used only with \fB\-P\fP \fB\-\-posix\-captures\fP option.
.TP
.B \fByypmatch\fP
An array of l\-values that are used to hold the tag values corresponding
to the capturing parentheses in the matching rule. Array length must be
at least \fByynmatch * 2\fP (usually \fBYYMAXNMATCH * 2\fP is a good choice).
Used only with \fB\-P\fP \fB\-\-posix\-captures\fP option.
.TP
.B \fByystable\fP
Deprecated.
.TP
.B \fByystate\fP
An l\-value used with the \fB\-\-loop\-switch\fP option to store the current DFA
state.
.TP
.B \fByytarget\fP
Jump table that contains jump targets (label addresses) for all transitions
from a state. This table is local to each state. Generation of \fByytarget\fP
tables is enabled with \fB\-\-computed\-gotos\fP option.
.UNINDENT
.SS Options
.sp
Some of the options have corresponding \fI\%configurations\fP,
others are global and cannot be changed after re2c starts reading the input file.
Debug options generally require building re2c in debug configuration.
Internal options are useful for experimenting with the algorithms used in re2c.
.INDENT 0.0
.TP
.B \fB\-? \-\-help \-h\fP
Show help message.
.TP
.B \fB\-\-api <simple | record | generic>\fP
Specify the API used by the generated code to interface with used\-defined
code. Option \fBsimple\fP shold be used in simple cases when there\(aqs no need
for buffer refilling and storing lexer state. Option \fBrecord\fP should be
used when lexer state needs to be stored in a record (struct, class, etc.).
Option \fBgeneric\fP should be used in complex cases when the other two APIs
are not flexible enough.
.TP
.B \fB\-\-bit\-vectors \-b\fP
Optimize conditional jumps using bit masks.
This option implies \fB\-\-nested\-ifs\fP\&.
.TP
.B \fB\-\-captures\fP, \fB\-\-leftmost\-captures\fP
Enable submatch extraction with leftmost greedy capturing groups. The result
is collected into an array \fByybmatch\fP of capacity \fB2 * YYMAXNMATCH\fP, and
\fByynmatch\fP is set to the number of groups for the matching rule.
.TP
.B \fB\-\-captvars\fP, \fB\-\-leftmost\-captvars\fP
Enable submatch extraction with leftmost greedy capturing groups. The result
is collected into variables \fByytl<k>\fP, \fByytr<k>\fP for \fBk\fP\-th capturing
group.
.TP
.B \fB\-\-case\-insensitive\fP
Treat single\-quoted and double\-quoted strings as case\-insensitive.
.TP
.B \fB\-\-case\-inverted\fP
Invert the meaning of single\-quoted and double\-quoted strings:
treat single\-quoted strings as case\-sensitive and double\-quoted strings
as case\-insensitive.
.TP
.B \fB\-\-case\-ranges\fP
Collapse consecutive cases in a switch statements into a range of the form
\fBlow ... high\fP\&. This syntax is a C/C++ language extension that is
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.
.TP
.B \fB\-\-computed\-gotos \-g\fP
Optimize conditional jumps using non\-standard \(dqcomputed goto\(dq extension
(which must be supported by the compiler). re2zig generates jump tables
only in complex cases with a lot of conditional branches. Complexity
threshold can be configured with \fBcgoto:threshold\fP configuration.
Relative offsets can be enabled with \fBcgoto:relative\fP configuration. This
option implies \fB\-\-bit\-vectors\fP\&.
.TP
.B \fB\-\-computed\-gotos\-relative\fP
Similar to \-\-computed\-gotos but generate relative offsets for jump tables
instead (which must be supported by the compiler). This option implies
\fB\-\-computed\-gotos\fP\&.
.TP
.B \fB\-\-conditions \-\-start\-conditions \-c\fP
Enable support of Flex\-like \(dqconditions\(dq: multiple interrelated lexers
within one block. This is an alternative to manually specifying different
re2zig blocks connected with \fBgoto\fP or function calls.
.TP
.B \fB\-\-depfile FILE\fP
Write dependency information to \fBFILE\fP in the form of a Makefile rule
\fB<output\-file> : <input\-file> [include\-file ...]\fP\&. This allows one to
track build dependencies in the presence of \fBinclude\fP blocks/directives,
so that updating include files triggers regeneration of the output file.
This option depends on the \fB\-\-output\fP option.
.TP
.B \fB\-\-ebcdic \-\-ecb \-e\fP
Generate a lexer that reads input in EBCDIC encoding. re2zig assumes that
the character range is 0 \-\- 0xFF and character size is 1 byte.
.TP
.B \fB\-\-empty\-class <match\-empty | match\-none | error>\fP
Define the way re2zig treats empty character classes. With \fBmatch\-empty\fP
(the default) empty class matches empty input (which is illogical, but
backwards\-compatible). With \fBmatch\-none\fP empty class always fails to match.
With \fBerror\fP empty class raises a compilation error.
.TP
.B \fB\-\-encoding\-policy <fail | substitute | ignore>\fP
Define the way re2zig treats Unicode surrogates.
With \fBfail\fP re2zig aborts with an error when a surrogate is encountered.
With \fBsubstitute\fP re2zig silently replaces surrogates with the error code
point 0xFFFD. With \fBignore\fP (the default) re2zig treats surrogates as
normal code points. The Unicode standard says that standalone surrogates
are invalid, but real\-world libraries and programs behave in different ways.
.TP
.B \fB\-\-flex\-syntax \-F\fP
Partial support for Flex syntax: in this mode named definitions don\(aqt 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.
.TP
.B \fB\-\-goto\-label\fP
Use \(dqgoto/label\(dq code model: encode DFA in form of labeled code blocks
connected with \fBgoto\fP transitions across blocks. This is only supported
for languages that have a \fBgoto\fP statement.
.TP
.B \fB\-\-header \-\-type\-header \-t HEADER\fP
Generate a \fBHEADER\fP file. The contents of the file can be specified using
special blocks \fBheader:on\fP and \fBheader:off\fP\&. If conditions are used, the
generated header will have a condition enum automatically appended to it
(unless there is an explicit \fBconditions\fP block).
.TP
.B \fB\-I PATH\fP
Add \fBPATH\fP to the list of locations which are used when searching for
include files. This option is useful in combination with \fBinclude\fP block
or directive. re2zig looks for \fBFILE\fP in the directory of the parent file
and in the include locations specified with \fB\-I\fP option.
.TP
.B \fB\-\-input <default | custom>\fP
Deprecated alias for \fB\-\-api\fP\&. Option \fBdefault\fP corresponds to \fBsimple\fP
(it is indeed the default for most backends, but not for all). Option
\fBcustom\fP corresponds to \fBgeneric\fP\&.
.TP
.B \fB\-\-input\-encoding <ascii | utf8>\fP
Specify the way re2zig parses regular expressions.
With \fBascii\fP (the default) re2zig handles input as ASCII\-encoded: any
sequence of code units is a sequence of standalone 1\-byte characters.
With \fButf8\fP re2zig handles input as UTF8\-encoded and recognizes multibyte
characters.
.TP
.B \fB\-\-invert\-captures\fP
Invert the meaning of capturing and non\-capturing groups. By default
\fB(...)\fP is capturing and \fB(! ...)\fP is non\-capturing. With this option
\fB(! ...)\fP is capturing and \fB(...)\fP is non\-capturing.
.TP
.B \fB\-\-lang <none | c | d | go | haskell | java | js | ocaml | python | rust | swift | v | zig>\fP
Specify the target language. Supported languages are C, D, Go, Haskell,
Java, JS, OCaml, Python, Rust, Swift, V, Zig (more languages can be added via
user\-defined syntax files, see the \fB\-\-syntax\fP option). Option \fBnone\fP
disables default suntax configs, so that the target language is undefined.
.TP
.B \fB\-\-location\-format <gnu | msvc>\fP
Specify location format in messages.
With \fBgnu\fP locations are printed as \(aqfilename:line:column: ...\(aq.
With \fBmsvc\fP locations are printed as \(aqfilename(line,column) ...\(aq.
The default is \fBgnu\fP\&.
.TP
.B \fB\-\-loop\-switch\fP
Use \(dqloop/switch\(dq code model: encode DFA in form of a loop over a switch
statement, where individual states are switch cases. State is stored in a
variable \fByystate\fP\&. Transitions between states update \fByystate\fP to the
case label of the destination state and continue execution to the head of
the loop.
.TP
.B \fB\-\-nested\-ifs \-s\fP
Use nested \fBif\fP statements instead of \fBswitch\fP statements in conditional
jumps. This usually results in more efficient code with non\-optimizing
compilers.
.TP
.B \fB\-\-no\-debug\-info \-i\fP
Do not output line directives. This may be useful when the generated code is
stored in a version control system (to avoid huge autogenerated diffs on
small changes).
.TP
.B \fB\-\-no\-generation\-date\fP
Suppress date output in the generated file.
.TP
.B \fB\-\-no\-version\fP
Suppress version output in the generated file.
.TP
.B \fB\-\-no\-unsafe\fP
Do not generate \fBunsafe\fP wrapper over \fBYYPEEK\fP (this option is specific
to Rust). For performance reasons \fBYYPEEK\fP should avoid bounds\-checking,
as the lexer already performs end\-of\-input checks in a more efficient way.
The user may choose to provide a safe \fBYYPEEK\fP definition, or a definition
that is unsafe only in release builds, in which case the \fB\-\-no\-unsafe\fP
option helps to avoid warnings about redundant \fBunsafe\fP blocks.
.TP
.B \fB\-\-output \-o OUTPUT\fP
Specify the \fBOUTPUT\fP file.
.TP
.B \fB\-\-posix\-captures\fP, \fB\-P\fP
Enable submatch extraction with POSIX\-style capturing groups. The result
is collected into an array \fByybmatch\fP of capacity \fB2 * YYMAXNMATCH\fP, and
\fByynmatch\fP is set to the number of groups for the matching rule.
.TP
.B \fB\-\-posix\-captvars\fP
Enable submatch extraction with POSIX\-style capturing groups. The result
is collected into variables \fByytl<k>\fP, \fByytr<k>\fP for \fBk\fP\-th capturing
group.
.TP
.B \fB\-\-recursive\-functions\fP
Use code model based on co\-recursive functions, where each DFA state is a
separate function that may call other state\-functions or itself.
.TP
.B \fB\-\-reusable \-r\fP
Deprecated since version 2.2 (reusable blocks are allowed by default now).
.TP
.B \fB\-\-skeleton \-S\fP
Ignore user\-defined interface code and generate a self\-contained \(dqskeleton\(dq
program. Additionally, generate input files with strings derived from the
regular grammar and compressed match results that are used to verify
\(dqskeleton\(dq behavior on all inputs. This option is useful for finding bugs
in optimizations and code generation. This option is supported only for C.
.TP
.B \fB\-\-storable\-state \-f\fP
Generate a lexer which can store its inner state.
This is useful in push\-model lexers which are stopped by an outer program
when there is not enough input, and then resumed when more input becomes
available. In this mode users should additionally define \fBYYGETSTATE\fP
and \fBYYSETSTATE\fP primitives, and variables \fByych\fP, \fByyaccept\fP and
\fBstate\fP should be part of the stored lexer state.
.TP
.B \fB\-\-syntax FILE\fP
Load configurations from the specified \fBFILE\fP and apply them on top of the
default syntax file. Note that \fBFILE\fP can define only a few configurations
(if it\(aqs used to amend the default syntax file), or it can define a whole
new language backend (in the latter case it is recommended to use
\fB\-\-lang none\fP option).
.TP
.B \fB\-\-tags \-T\fP
Enable submatch extraction with tags.
.TP
.B \fB\-\-ucs2 \-\-wide\-chars \-w\fP
Generate a lexer that reads UCS2\-encoded input. re2zig assumes that the
character range is 0 \-\- 0xFFFF and character size is 2 bytes.
This option implies \fB\-\-nested\-ifs\fP\&.
.TP
.B \fB\-\-utf8 \-\-utf\-8 \-8\fP
Generate a lexer that reads input in UTF\-8 encoding. re2zig assumes that the
character range is 0 \-\- 0x10FFFF and character size is 1 byte.
.TP
.B \fB\-\-utf16 \-\-utf\-16 \-x\fP
Generate a lexer that reads UTF16\-encoded input. re2zig assumes that the
character range is 0 \-\- 0x10FFFF and character size is 2 bytes.
This option implies \fB\-\-nested\-ifs\fP\&.
.TP
.B \fB\-\-utf32 \-\-unicode \-u\fP
Generate a lexer that reads UTF32\-encoded input. re2zig assumes that the
character range is 0 \-\- 0x10FFFF and character size is 4 bytes.
This option implies \fB\-\-nested\-ifs\fP\&.
.TP
.B \fB\-\-verbose\fP
Output a short message in case of success.
.TP
.B \fB\-\-vernum \-V\fP
Show version information in \fBMMmmpp\fP format (major, minor, patch).
.TP
.B \fB\-\-version \-v\fP
Show version information.
.TP
.B \fB\-\-single\-pass \-1\fP
Deprecated. Does nothing (single pass is the default now).
.UNINDENT
.INDENT 0.0
.TP
.B \fB\-\-debug\-output \-d\fP
Emit \fBYYDEBUG\fP invocations in the generated code. This is useful to trace
lexer execution.
.TP
.B \fB\-\-dump\-adfa\fP
Debug option: output DFA after tunneling (in .dot format).
.TP
.B \fB\-\-dump\-cfg\fP
Debug option: output control flow graph of tag variables (in .dot format).
.TP
.B \fB\-\-dump\-closure\-stats\fP
Debug option: output statistics on the number of states in closure.
.TP
.B \fB\-\-dump\-dfa\-det\fP
Debug option: output DFA immediately after determinization (in .dot format).
.TP
.B \fB\-\-dump\-dfa\-min\fP
Debug option: output DFA after minimization (in .dot format).
.TP
.B \fB\-\-dump\-dfa\-tagopt\fP
Debug option: output DFA after tag optimizations (in .dot format).
.TP
.B \fB\-\-dump\-dfa\-tree\fP
Debug option: output DFA under construction with states represented as tag
history trees (in .dot format).
.TP
.B \fB\-\-dump\-dfa\-raw\fP
Debug option: output DFA under construction with expanded state\-sets
(in .dot format).
.TP
.B \fB\-\-dump\-interf\fP
Debug option: output interference table produced by liveness analysis of tag
variables.
.TP
.B \fB\-\-dump\-nfa\fP
Debug option: output NFA (in .dot format).
.TP
.B \fB\-\-emit\-dot \-D\fP
Instead of normal output generate lexer graph in .dot format.
The output can be converted to an image with the help of Graphviz
(e.g. something like \fBdot \-Tpng \-odfa.png dfa.dot\fP).
.UNINDENT
.INDENT 0.0
.TP
.B \fB\-\-dfa\-minimization <moore | table>\fP
Internal option: DFA minimization algorithm used by re2zig\&. The \fBmoore\fP
option is the Moore algorithm (it is the default). The \fBtable\fP option is
the \(dqtable filling\(dq 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.
.TP
.B \fB\-\-eager\-skip\fP
Internal option: make the generated lexer advance the input position
eagerly \-\- immediately after reading the input symbol. This changes the
default behavior when the input position is advanced lazily \-\- after
transition to the next state.
.TP
.B \fB\-\-no\-lookahead\fP
Internal option, deprecated.
It used to enable TDFA(0) algorithm. Unlike TDFA(1), TDFA(0) algorithm does
not use one\-symbol lookahead. It applies register operations to the incoming
transitions rather than the outgoing ones. Benchmarks showed that TDFA(0)
algorithm is less efficient than TDFA(1).
.TP
.B \fB\-\-no\-optimize\-tags\fP
Internal option: suppress optimization of tag variables (useful for
debugging).
.TP
.B \fB\-\-posix\-closure <gor1 | gtop>\fP
Internal option: specify shortest\-path algorithm used for the construction of
epsilon\-closure with POSIX disambiguation semantics: \fBgor1\fP (the default)
stands for Goldberg\-Radzik algorithm, and \fBgtop\fP stands for \(dqglobal
topological order\(dq algorithm.
.TP
.B \fB\-\-posix\-prectable <complex | naive>\fP
Internal option: specify the algorithm used to compute POSIX precedence
table. The \fBcomplex\fP algorithm computes precedence table in one traversal
of tag history tree and has quadratic complexity in the number of TNFA
states; it is the default. The \fBnaive\fP algorithm has worst\-case cubic
complexity in the number of TNFA states, but it is much simpler than
\fBcomplex\fP and may be slightly faster in non\-pathological cases.
.TP
.B \fB\-\-stadfa\fP
Internal option, deprecated.
It used to enable staDFA algorithm, which differs from TDFA in that register
operations are placed in states rather than on transitions. Benchmarks
showed that staDFA algorithm is less efficient than TDFA.
.TP
.B \fB\-\-fixed\-tags <none | toplevel | all>\fP
Internal option:
specify whether the fixed\-tag optimization should be applied to all tags
(\fBall\fP), none of them (\fBnone\fP), or only those in toplevel concatenation
(\fBtoplevel\fP). The default is \fBall\fP\&.
\(dqFixed\(dq tags are those that are located within a fixed distance to some
other tag (called \(dqbase\(dq). In such cases only the base tag needs to be
tracked, and the value of the fixed tag can be computed as the value of the
base tag plus a static offset. For tags that are under alternative or
repetition it is also necessary to check if the base tag has a no\-match
value (in that case fixed tag should also be set to no\-match, disregarding
the offset). For tags in top\-level concatenation the check is not needed,
because they always match.
.UNINDENT
.SS Warnings
.sp
Warnings can be invividually enabled, disabled and turned into an error.
.INDENT 0.0
.TP
.B \fB\-W\fP
Turn on all warnings.
.TP
.B \fB\-Werror\fP
Turn warnings into errors. Note that this option alone
doesn\(aqt turn on any warnings; it only affects those warnings that have
been turned on so far or will be turned on later.
.TP
.B \fB\-W<warning>\fP
Turn on \fBwarning\fP\&.
.TP
.B \fB\-Wno\-<warning>\fP
Turn off \fBwarning\fP\&.
.TP
.B \fB\-Werror\-<warning>\fP
Turn on \fBwarning\fP and treat it as an error (this implies \fB\-W<warning>\fP).
.TP
.B \fB\-Wno\-error\-<warning>\fP
Don\(aqt treat this particular \fBwarning\fP as an error. This doesn\(aqt turn off
the warning itself.
.UNINDENT
.INDENT 0.0
.TP
.B \fB\-Wcondition\-order\fP
Warn if the generated program makes implicit assumptions about condition
numbering. One should use either \fB\-\-header\fP option or \fBconditions\fP
block to generate a mapping of condition names to numbers and then use the
autogenerated condition names.
.TP
.B \fB\-Wempty\-character\-class\fP
Warn if a regular expression contains an empty character class. Trying to
match an empty character class makes no sense: it should always fail.
However, for backwards compatibility reasons re2zig permits empty character
classes and treats them as empty strings. Use the \fB\-\-empty\-class\fP option
to change the default behavior.
.TP
.B \fB\-Wmatch\-empty\-string\fP
Warn if a rule is nullable (matches an empty string).
If the lexer runs in a loop and the empty match is unintentional, the lexer
may unexpectedly hang in an infinite loop.
.TP
.B \fB\-Wswapped\-range\fP
Warn if the lower bound of a range is greater than its upper bound. The
default behavior is to silently swap the range bounds.
.TP
.B \fB\-Wundefined\-control\-flow\fP
Warn if some input strings cause undefined control flow in the lexer (the
faulty patterns are reported). This is a dangerous and common mistake. It
can be easily fixed by adding the default rule \fB*\fP which has the lowest
priority, matches any code unit, and always consumes a single code unit.
.TP
.B \fB\-Wunreachable\-rules\fP
Warn about rules that are shadowed by other rules and will never match.
.TP
.B \fB\-Wuseless\-escape\fP
Warn if a symbol is escaped when it shouldn\(aqt be.
By default, re2zig silently ignores such escapes, but this may as well
indicate a typo or an error in the escape sequence.
.TP
.B \fB\-Wnondeterministic\-tags\fP
Warn if a tag has \fBn\fP\-th degree of nondeterminism, where \fBn\fP is greater
than 1.
.TP
.B \fB\-Wsentinel\-in\-midrule\fP
Warn if the sentinel symbol occurs in the middle of a rule \-\-\- this may
cause reads past the end of buffer, crashes or memory corruption in the
generated lexer. 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 \fBre2c:sentinel\fP configuration is used.
.TP
.B \fB\-Wundefined\-syntax\-config\fP
Warn if the syntax file specified with \fB\-\-syntax\fP option is missing
definitions of some configurations. This helps to maintain user\-defined
syntax files: if a new release adds configurations, old syntax file will
raise a warning, and the user will be notified. If some configurations are
unused and do not need a definition, they should be explicitly set to
\fB<undefined>\fP\&.
.UNINDENT
.SS Syntax files
.sp
Support for different languages in re2c is based on the idea of \fIsyntax files\fP\&.
A syntax file is a configuration file that defines syntax of the target language
\-\- not the whole language, but a small part of it that is used by the generated
code. Syntax files make re2c very flexible, but they should not be used as a
replacement for \fBre2c:\fP 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
\fBinclude/syntax\fP subdirectory); they are also embedded in the re2zig binary.
Users may provide a custom syntax file that overrides a few configurations for
one of supported languages, or they may choose to redefine all configurations
(in that case \fB\-\-lang none\fP option should be used).
Syntax files contain configurations of four different kinds: feature lists,
language configurations, inplace configurations and code templates.
.sp
\fBFeature lists\fP
.INDENT 0.0
.INDENT 3.5
A few list configurations define various features supported by a given
backend, so that re2zig may give a clear error if the user tries to enable an
unsupported feature:
.INDENT 0.0
.TP
.B \fBsupported_apis\fP
A list of supported APIs with possible elements \fBsimple\fP, \fBrecord\fP,
\fBgeneric\fP\&.
.TP
.B \fBsupported_api_styles\fP
A list of supported API styles with possible elements \fBfunctions\fP,
\fBfree\-form\fP\&.
.TP
.B \fBsupported_code_models\fP
A list of supported code models with possible elements \fBgoto\-label\fP,
\fBloop\-switch\fP, \fBrecursive\-functions\fP\&.
.TP
.B \fBsupported_targets\fP
A list of supported codegen targets with possible elements \fBcode\fP,
\fBdot\fP, \fBskeleton\fP\&.
.TP
.B \fBsupported_features\fP
A list of supported features with possible elements \fBnested\-ifs\fP,
\fBbitmaps\fP, \fBcomputed\-gotos\fP, \fBcase\-ranges\fP, \fBmonadic\fP, \fBunsafe\fP,
\fBtags\fP, \fBcaptures\fP, \fBcaptvars\fP\&.
.UNINDENT
.UNINDENT
.UNINDENT
.sp
\fBLanguage configurations\fP
.INDENT 0.0
.INDENT 3.5
A few boolean configurations describe features of the target language that
affect re2zig parser and code generator:
.INDENT 0.0
.TP
.B \fBsemicolons\fP
Non\-zero if the language uses semicolons after statements.
.TP
.B \fBbacktick_quoted_strings\fP
Non\-zero if the language has backtick\-quoted strings.
.TP
.B \fBsingle_quoted_strings\fP
Non\-zero if the language has single\-quoted strings.
.TP
.B \fBindentation_sensitive\fP
Non\-zero if the language is indentation sensitive.
.TP
.B \fBwrap_blocks_in_braces\fP
Non\-zero if compound statements must be wrapped in curly braces.
.UNINDENT
.UNINDENT
.UNINDENT
.sp
\fBInplace configurations\fP
.INDENT 0.0
.INDENT 3.5
Syntax files define initial values of all \fBre2c:\fP configurations, as they
may differ for different languages. See configurations section for a full list
of all inplace configurations and their meaning.
.UNINDENT
.UNINDENT
.sp
\fBCode templates\fP
.INDENT 0.0
.INDENT 3.5
Code templates define syntax of the target language. They are written in a
simple domain\-specific language with the following formal grammar:
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
code\-template ::
name \(aq=\(aq code\-exprs \(aq;\(aq
| CODE_TEMPLATE \(aq;\(aq
| \(aq<undefined>\(aq \(aq;\(aq
code\-exprs ::
<EMPTY>
| code\-exprs code\-expr
code\-expr ::
STRING
| VARIABLE
| optional
| list
optional ::
\(aq(\(aq CONDITIONAL \(aq?\(aq code\-exprs \(aq)\(aq
| \(aq(\(aq CONDITIONAL \(aq?\(aq code\-exprs \(aq:\(aq code\-exprs \(aq)\(aq
list ::
\(aq[\(aq VARIABLE \(aq:\(aq code\-exprs \(aq]\(aq
| \(aq[\(aq VARIABLE \(aq{\(aq NUMBER \(aq}\(aq \(aq:\(aq code\-exprs \(aq]\(aq
| \(aq[\(aq VARIABLE \(aq{\(aq NUMBER \(aq,\(aq NUMBER \(aq}\(aq \(aq:\(aq code\-exprs \(aq]\(aq
.ft P
.fi
.UNINDENT
.UNINDENT
.sp
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
\fB<undefined>\fP\&. Variables are placeholders that are substituted during code
generation phase. List variables are special: when expanding list templates,
re2zig repeats expressions 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. \fB\-1\fP 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 re2zig reads
syntax file, it checks that each code template uses only the variables and
conditionals that are allowed in it.
.sp
For example, the following code template defines if\-then\-else construct for a
C\-like language:
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
code:if_then_else =
[branch{0}: topindent \(dqif \(dq cond \(dq {\(dq nl
indent [stmt: stmt] dedent]
[branch{1:\-1}: topindent \(dq} else\(dq (.cond ? \(dq if \(dq cond) \(dq {\(dq nl
indent [stmt: stmt] dedent]
topindent \(dq}\(dq nl;
.ft P
.fi
.UNINDENT
.UNINDENT
.sp
Here \fBbranch\fP is a list variable: \fBbranch{0}\fP expands to the first branch
(which is special, as there is no \fBelse\fP part), \fBbranch{1:\-1}\fP expands to
all remaining branches (if any). \fBstmt\fP is also a list variable:
\fB[stmt: stmt]\fP is a nested list that expands to a list of statements in the
body of the current branch. \fBtopindent\fP, \fBindent\fP, \fBdedent\fP and \fBnl\fP
are global variables, and \fB\&.cond\fP is a local conditional (their meaning is
described below). This code template could produce the following code:
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
if x {
// do something
} else if y {
// do something else
} else {
// don\(aqt do anything
}
.ft P
.fi
.UNINDENT
.UNINDENT
.sp
Here\(aqs a list of all code templates supported by re2zig 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 \fB<undefined>\fP\&.
.INDENT 0.0
.TP
.B \fBcode:var_local\fP
Declaration or definition of a local variable. Supported variables:
\fBtype\fP (the type of the variable), \fBname\fP (its name) and \fBinit\fP
(initial value, if any). Conditionals: \fB\&.init\fP (true if there is an
initializer).
.TP
.B \fBcode:var_global\fP
Same as \fBcode:var_local\fP, except that it\(aqs used in top\-level.
.TP
.B \fBcode:const_local\fP
Definition of a local constant. Supported variables: \fBtype\fP (the type
of the constant), \fBname\fP (its name) and \fBinit\fP (initial value).
.TP
.B \fBcode:const_global\fP
Same as \fBcode:const_local\fP, except that it\(aqs used in top\-level.
.TP
.B \fBcode:array_local\fP
Definition of a local array (table). Supported variables: \fBtype\fP (the
type of array elements), \fBname\fP (array name), \fBsize\fP (its size),
\fBrow\fP (a list variable that does not itself produce any code, but
expands list expression as many times as there are rows in the table)
and \fBelem\fP (a list variable that expands to all table elements in the
current row \-\- it\(aqs meant to be nested in the \fBrow\fP list).
Supported conditional: \fB\&.const\fP (true if the array is immutable).
.TP
.B \fBcode:array_global\fP
Same as \fBcode:array_local\fP, except that it\(aqs used in top\-level.
.TP
.B \fBcode:array_elem\fP
Reference to an element of an array (table). Supported variables:
\fBarray\fP (the name of the array) and \fBindex\fP (index of the element).
.TP
.B \fBcode:enum\fP
Definition of an enumeration (it may be defined using a special language
construct for enumerations, or simply as a few standalone constants).
Supported variables are \fBtype\fP (user\-defined enumeration type or type
of the constants), \fBelem\fP (list variable that expands to the name of
each member) and \fBinit\fP (initializer for each member). Conditionals:
\fB\&.init\fP (true if there is an initializer).
.TP
.B \fBcode:enum_elem\fP
Enumeration element (a member of a user\-defined enumeration type or a
name of a constant, depending on how \fBcode:enum\fP is defined).
Supported variables are \fBname\fP (the name of the element) and \fBtype\fP
(its type).
.TP
.B \fBcode:assign\fP
Assignment statement. Supported variables are \fBlhs\fP (left hand side)
and \fBrhs\fP (right hand side).
.TP
.B \fBcode:type_int\fP
Signed integer type.
.TP
.B \fBcode:type_uint\fP
Unsigned integer type.
.TP
.B \fBcode:type_yybm\fP
Type of elements in the \fByybm\fP table.
.TP
.B \fBcode:type_yytarget\fP
Type of elements in the \fByytarget\fP table.
.TP
.B \fBcode:type_yyctable\fP
Type of elements in the \fByyctable\fP table.
.TP
.B \fBcode:cmp_eq\fP
Operator \(dqequals\(dq.
.TP
.B \fBcode:cmp_ne\fP
Operator \(dqnot equals\(dq.
.TP
.B \fBcode:cmp_lt\fP
Operator \(dqless than\(dq.
.TP
.B \fBcode:cmp_gt\fP
Operator \(dqgreater than\(dq
.TP
.B \fBcode:cmp_le\fP
Operator \(dqless or equal\(dq
.TP
.B \fBcode:cmp_ge\fP
Operator \(dqgreater or equal\(dq
.TP
.B \fBcode:if_then_else\fP
If\-then\-else statement with one or more branches. Supported variables:
\fBbranch\fP (a list variable that does not itself produce any code, but
expands list expression as many times as there are branches), \fBcond\fP
(condition of the current branch) and \fBstmt\fP (a list variable that
expands to all statements in the current branch). Conditionals:
\fB\&.cond\fP (true if the current branch has a condition), \fB\&.many\fP (true
if there\(aqs more than one branch).
.TP
.B \fBcode:if_then_else_oneline\fP
A specialization of \fBcode:if_then_else\fP for the case when all branches
have one\-line statements. If this is \fB<undefined>\fP,
\fBcode:if_then_else\fP is used instead.
.TP
.B \fBcode:switch\fP
A switch statement with one or more cases. Supported variables: \fBexpr\fP
(the switched\-on expression) and \fBcase\fP (a list variable that expands
to all cases\-groups with their code blocks).
.TP
.B \fBcode:switch_cases\fP
A group of switch cases that maps to a single code block. Supported
variables are \fBcase\fP (a list variable that expands to all cases in
this group) and \fBstmt\fP (a list variable that expands to all statements
in the code block.
.TP
.B \fBcode:switch_cases_oneline\fP
A specialization of \fBcode:switch_cases\fP for the case when the code
block consists of a single one\-line statement. If this is
\fB<undefined>\fP, \fBcode:switch_cases\fP is used instead.
.TP
.B \fBcode:switch_case_range\fP
A single switch case that covers a range of values (possibly consisting
of a single value). Supported variable: \fBval\fP (a list variable that
expands to all values in the range). Supported conditionals: \fB\&.many\fP
(true if there\(aqs more than one value in the range) and
\fB\&.char_literals\fP (true if this is a switch on character literals \-\-
some languages provide special syntax for this case).
.TP
.B \fBcode:switch_case_default\fP
Default switch case.
.TP
.B \fBcode:loop\fP
A loop that runs forever (unless interrupted from the loop body).
Supported variables: \fBlabel\fP (loop label), \fBstmt\fP (a list variable
that expands to all statements in the loop body).
.TP
.B \fBcode:continue\fP
Continue statement. Supported variables: \fBlabel\fP (label from which to
continue execution).
.TP
.B \fBcode:goto\fP
Goto statement. Supported variables: \fBlabel\fP (label of the jump
target).
.TP
.B \fBcode:cgoto\fP
Computed \fBgoto\fP statement.
Supported variables: \fBarray\fP (the table containing computed \fBgoto\fP
information), \fBindex\fP (index of the element in the table) and \fBbase\fP
(base label, only used if \fB\&.cgoto.relative\fP is true).
.TP
.B \fBcode:cgoto:data\fP
Initializer expression for a single element in computed \fBgoto\fP table.
Supported variables: \fBlabel\fP (the label that is used to initialize the
current element), \fBtype\fP (underlying type of the elements in the table)
and \fBbase\fP (base label \- only used if \fB\&.cgoto.relative\fP is true).
.TP
.B \fBcode:fndecl\fP
Function declaration. Supported variables: \fBname\fP (function name),
\fBtype\fP (return type), \fBthrow\fP (exceptions thrown by this function,
maps to \fBre2c:yyfn:throw\fP configuration), \fBarg\fP (a list variable that
does not itself produce code, but expands list expression as many times as
there are function arguments), \fBargname\fP (name of the current argument),
\fBargtype\fP (type of the current argument). Conditional: \fB\&.type\fP (true
if this is a non\-void function).
.TP
.B \fBcode:fndef\fP
Like \fBcode:fndecl\fP, but used for function definitions, so it has one
additional list variable \fBstmt\fP that expands to all statements in the
function body.
.TP
.B \fBcode:fncall\fP
Function call statement. Supported variables: \fBname\fP (function name),
\fBretval\fP (l\-value where the return value is stored, if any) and
\fBarg\fP (a list variable that expands to all function arguments).
Conditionals: \fB\&.args\fP (true if the function has arguments) and
\fB\&.retval\fP (true if return value needs to be saved).
.TP
.B \fBcode:tailcall\fP
Tail call statement. Supported variables: \fBname\fP (function name),
and \fBarg\fP (a list variable that expands to all function arguments).
Conditionals: \fB\&.args\fP (true if the function has arguments) and
\fB\&.retval\fP (true if this is a non\-void function).
.TP
.B \fBcode:recursive_functions\fP
Program body with \fB\-\-recursive\-functions\fP code model. Supported
variables: \fBfn\fP (a list variable that does not itself produce any
code, but expands list expression as many times as there are functions),
\fBfndecl\fP (declaration of the current function) and \fBfndef\fP
(definition of the current function).
.TP
.B \fBcode:fingerprint\fP
The fingerprint at the top of the generated output file. Supported
variables: \fBver\fP (re2zig version that was used to generate this) and
\fBdate\fP (generation date).
.TP
.B \fBcode:line_info\fP
The format of line directives (if this is set to \fB<undefined>\fP, no
directives are generated). Supported variables: \fBline\fP (line number)
and \fBfile\fP (filename).
.TP
.B \fBcode:abort\fP
A statement that aborts program execution.
.TP
.B \fBcode:yydebug\fP
\fBYYDEBUG\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYDEBUG\fP, \fByyrecord\fP, \fByych\fP (map to the
corresponding \fBre2c:\fP configurations), \fBstate\fP (DFA state number).
.TP
.B \fBcode:yypeek\fP
\fBYYPEEK\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYPEEK\fP, \fBYYCTYPE\fP, \fBYYINPUT\fP, \fBYYCURSOR\fP,
\fByyrecord\fP, \fByych\fP (map to the corresponding \fBre2c:\fP
configurations). Conditionals: \fB\&.cast\fP (true if
\fBre2c:yych:conversion\fP is set to non\-zero).
.TP
.B \fBcode:yyskip\fP
\fBYYSKIP\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYSKIP\fP, \fBYYCURSOR\fP, \fByyrecord\fP (map to the
corresponding \fBre2c:\fP configurations).
.TP
.B \fBcode:yybackup\fP
\fBYYBACKUP\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYBACKUP\fP, \fBYYCURSOR\fP, \fBYYMARKER\fP,
\fByyrecord\fP (map to the corresponding \fBre2c:\fP configurations).
.TP
.B \fBcode:yybackupctx\fP
\fBYYBACKUPCTX\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYBACKUPCTX\fP, \fBYYCURSOR\fP, \fBYYCTXMARKER\fP,
\fByyrecord\fP (map to the corresponding \fBre2c:\fP configurations).
.TP
.B \fBcode:yyskip_yypeek\fP
Combined \fBcode:yyskip\fP and \fBcode:yypeek\fP statement (defaults to
\fBcode:yyskip\fP followed by \fBcode:yypeek\fP).
.TP
.B \fBcode:yypeek_yyskip\fP
Combined \fBcode:yypeek\fP and \fBcode:yyskip\fP statement (defaults to
\fBcode:yypeek\fP followed by \fBcode:yyskip\fP).
.TP
.B \fBcode:yyskip_yybackup\fP
Combined \fBcode:yyskip\fP and \fBcode:yybackup\fP statement (defaults to
\fBcode:yyskip\fP followed by \fBcode:yybackup\fP).
.TP
.B \fBcode:yybackup_yyskip\fP
Combined \fBcode:yybackup\fP and \fBcode:yyskip\fP statement (defaults to
\fBcode:yybackup\fP followed by \fBcode:yyskip\fP).
.TP
.B \fBcode:yybackup_yypeek\fP
Combined \fBcode:yybackup\fP and \fBcode:yypeek\fP statement (defaults to
\fBcode:yybackup\fP followed by \fBcode:yypeek\fP).
.TP
.B \fBcode:yyskip_yybackup_yypeek\fP
Combined \fBcode:yyskip\fP, \fBcode:yybackup\fP and \fBcode:yypeek\fP
statement (defaults to\(ga\(gacode:yyskip\(ga\(ga followed by \fBcode:yybackup\fP
followed by \fBcode:yypeek\fP).
.TP
.B \fBcode:yybackup_yypeek_yyskip\fP
Combined \fBcode:yybackup\fP, \fBcode:yypeek\fP and \fBcode:yyskip\fP
statement (defaults to\(ga\(gacode:yybackup\(ga\(ga followed by \fBcode:yypeek\fP
followed by \fBcode:yyskip\fP).
.TP
.B \fBcode:yyrestore\fP
\fBYYRESTORE\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYRESTORE\fP, \fBYYCURSOR\fP, \fBYYMARKER\fP,
\fByyrecord\fP (map to the corresponding \fBre2c:\fP configurations).
.TP
.B \fBcode:yyrestorectx\fP
\fBYYRESTORECTX\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYRESTORECTX\fP, \fBYYCURSOR\fP, \fBYYCTXMARKER\fP,
\fByyrecord\fP (map to the corresponding \fBre2c:\fP configurations).
.TP
.B \fBcode:yyrestoretag\fP
\fBYYRESTORETAG\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYRESTORETAG\fP, \fBYYCURSOR\fP, \fByyrecord\fP (map
to the corresponding \fBre2c:\fP configurations), \fBtag\fP (the name of tag
variable used to restore position).
.TP
.B \fBcode:yyshift\fP
\fBYYSHIFT\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYSHIFT\fP, \fBYYCURSOR\fP, \fByyrecord\fP (map to the
corresponding \fBre2c:\fP configurations), \fBoffset\fP (the number of code
units to shift the current position).
.TP
.B \fBcode:yyshiftstag\fP
\fBYYSHIFTSTAG\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYSHIFTSTAG\fP, \fByyrecord\fP, \fBnegative\fP (map
to the corresponding \fBre2c:\fP configurations), \fBtag\fP (tag variable
which needs to be shifted), \fBoffset\fP (the number of code units to
shift). Conditionals: \fB\&.nested\fP (true if this is a nested tag \-\- in
this case its value may equal to \fBre2c:tags:negative\fP, which should
not be shifted).
.TP
.B \fBcode:yyshiftmtag\fP
\fBYYSHIFTMTAG\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYSHIFTMTAG\fP (maps to the corresponding
\fBre2c:\fP configuration), \fBtag\fP (tag variable which needs to be
shifted), \fBoffset\fP (the number of code units to shift).
.TP
.B \fBcode:yystagp\fP
\fBYYSTAGP\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYSTAGP\fP, \fBYYCURSOR\fP, \fByyrecord\fP (map to the
corresponding \fBre2c:\fP configurations), \fBtag\fP (tag variable that
should be updated).
.TP
.B \fBcode:yymtagp\fP
\fBYYMTAGP\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYMTAGP\fP (maps to the corresponding \fBre2c:\fP
configuration), \fBtag\fP (tag variable that should be updated).
.TP
.B \fBcode:yystagn\fP
\fBYYSTAGN\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYSTAGN\fP, \fBnegative\fP, \fByyrecord\fP (map to the
corresponding \fBre2c:\fP configurations), \fBtag\fP (tag variable that
should be updated).
.TP
.B \fBcode:yymtagn\fP
\fBYYMTAGN\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYMTAGN\fP (maps to the corresponding \fBre2c:\fP
configuration), \fBtag\fP (tag variable that should be updated).
.TP
.B \fBcode:yycopystag\fP
\fBYYCOPYSTAG\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYCOPYSTAG\fP, \fByyrecord\fP (map to the
corresponding \fBre2c:\fP configurations), \fBlhs\fP, \fBrhs\fP (left and
right hand side tag variables of the copy operation).
.TP
.B \fBcode:yycopymtag\fP
\fBYYCOPYMTAG\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYCOPYMTAG\fP, \fByyrecord\fP (map to the
corresponding \fBre2c:\fP configurations), \fBlhs\fP, \fBrhs\fP (left and
right hand side tag variables of the copy operation).
.TP
.B \fBcode:yygetaccept\fP
\fBYYGETACCEPT\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYGETACCEPT\fP, \fByyrecord\fP (map to the
corresponding \fBre2c:\fP configurations), \fBvar\fP (maps to
\fBre2c:yyaccept\fP configuration).
.TP
.B \fBcode:yysetaccept\fP
\fBYYSETACCEPT\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYSETACCEPT\fP, \fByyrecord\fP (map to the
corresponding \fBre2c:\fP configurations), \fBvar\fP (maps to
\fBre2c:yyaccept\fP configuration) and \fBval\fP (numeric value of the
accepted rule).
.TP
.B \fBcode:yygetcond\fP
\fBYYGETCOND\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYGETCOND\fP, \fByyrecord\fP (map to the
corresponding \fBre2c:\fP configurations), \fBvar\fP (maps to
\fBre2c:yycond\fP configuration).
.TP
.B \fBcode:yysetcond\fP
\fBYYSETCOND\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYSETCOND\fP, \fByyrecord\fP (map to the
corresponding \fBre2c:\fP configurations), \fBvar\fP (maps to
\fBre2c:yycond\fP configuration) and \fBval\fP (numeric condition
identifier).
.TP
.B \fBcode:yygetstate\fP
\fBYYGETSTATE\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYGETSTATE\fP, \fByyrecord\fP (map to the
corresponding \fBre2c:\fP configurations), \fBvar\fP (maps to
\fBre2c:yystate\fP configuration).
.TP
.B \fBcode:yysetstate\fP
\fBYYSETSTATE\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYSETSTATE\fP, \fByyrecord\fP (map to the
corresponding \fBre2c:\fP configurations), \fBvar\fP (maps to
\fBre2c:yystate\fP configuration) and \fBval\fP (state number).
.TP
.B \fBcode:yylessthan\fP
\fBYYLESSTHAN\fP statement, possibly specialized for different APIs.
Supported variables: \fBYYLESSTHAN\fP, \fBYYCURSOR\fP, \fBYYLIMIT\fP,
\fByyrecord\fP (map to the corresponding \fBre2c:\fP configurations),
\fBneed\fP (the number of code units to check against). Conditional:
\fB\&.many\fP (true if the \fBneed\fP is more than one).
.TP
.B \fBcode:yybm_filter\fP
Condition that is used to filter out \fByych\fP values that are not
covered by the \fByybm\fP table (used with \fB\-\-bitmaps\fP option).
Supported variable: \fByych\fP (maps to \fBre2c:yych\fP configuration).
.TP
.B \fBcode:yybm_match\fP
The format of \fByybm\fP table check (generated with \fB\-\-bitmaps\fP
option). Supported variables: \fByybm\fP, \fByych\fP (map to the
corresponding \fBre2c:\fP configurations), \fBoffset\fP (offset in the
\fByybm\fP table that needs to be added to \fByych\fP) and \fBmask\fP (bit
mask that should be applied to the table entry to retrieve the boolean
value that needs to be checked)
.TP
.B \fBcode:yytarget_filter\fP
Condition that is used to filter out \fByych\fP values that are not
covered by the \fByytarget\fP table (used with \fB\-\-computed\-gotos\fP option).
Supported variable: \fByych\fP (maps to \fBre2c:yych\fP configuration).
.UNINDENT
.sp
Here\(aqs a list of all global variables that are allowed in syntax files:
.INDENT 0.0
.TP
.B \fBnl\fP
A newline.
.TP
.B \fBindent\fP
A variable that does not produce any code, but has a side\-effect of
increasing indentation level.
.TP
.B \fBdedent\fP
A variable that does not produce any code, but has a side\-effect of
decreasing indentation level.
.TP
.B \fBtopindent\fP
Indentation string for the current statement. Indentation level is
tracked and automatically updated by the code generator.
.UNINDENT
.sp
Here\(aqs a list of all global conditionals that are allowed in syntax files:
.INDENT 0.0
.TP
.B \fB\&.api.simple\fP
True if simple API is used (\fB\-\-api simple\fP or \fBre2c:api = simple\fP).
.TP
.B \fB\&.api.generic\fP
True if generic API is used (\fB\-\-api generic\fP or
\fBre2c:api = generic\fP).
.TP
.B \fB\&.api.record\fP
True if record API is used (\fB\-\-api record\fP or \fBre2c:api = record\fP).
.TP
.B \fB\&.api_style.functions\fP
True if function\-like API style is used
(\fBre2c:api\-style = functions\fP).
.TP
.B \fB\&.api_style.freeform\fP
True if free\-form API style is used (\fBre2c:api\-style = free\-form\fP).
.TP
.B \fB\&.case_ranges\fP
True if case ranges feature is enabled (\fB\-\-case\-ranges\fP or
\fBre2c:case\-ranges = 1\fP).
.TP
.B \fB\&.cgoto.relative\fP
True if the relative form of computed \fBgoto\fP is used
(\fB\-\-computed\-gotos\-relative\fP or \fBre2c:cgoto:relative = 1\fP).
.TP
.B \fB\&.code_model.goto_label\fP
True if code model based on goto/label is used (\fB\-\-goto\-label\fP).
.TP
.B \fB\&.code_model.loop_switch\fP
True if code model based on loop/switch is used (\fB\-\-loop\-switch\fP).
.TP
.B \fB\&.code_model.recursive_functions\fP
True if code model based on recursive functions is used
(\fB\-\-recursive\-function\fP).
.TP
.B \fB\&.date\fP
True if the generated fingerprint should contain generation date.
.TP
.B \fB\&.loop_label\fP
True if re2zig generated loops must have a label (\fBre2c:label:yyloop\fP
is set to a nonempty string).
.TP
.B \fB\&.monadic\fP
True if the generated code should be monadic (\fBre2c:monadic = 1\fP).
This is only relevant for pure functional languages.
.TP
.B \fB\&.start_conditions\fP
True if start conditions are enabled (\fB\-\-start\-conditions\fP).
.TP
.B \fB\&.storable_state\fP
True if storable state is enabled (\fB\-\-storable\-state\fP).
.TP
.B \fB\&.unsafe\fP
True if re2zig should use \(dqunsafe\(dq blocks in order to generate faster
code (\fB\-\-unsafe\fP, \fBre2c:unsafe = 1\fP). This is only relevant for
languages that have \(dqunsafe\(dq feature.
.TP
.B \fB\&.version\fP
True if the generated fingerprint should contain re2zig version.
.TP
.B \fB\&.yyfn.throw\fP
True if \fBre2c:yyfn:throw\fP configuration is defined to a nonempty string.
.UNINDENT
.UNINDENT
.UNINDENT
.SH HANDLING THE END OF INPUT
.sp
One of the main problems for the lexer is to know when to stop.
There are a few terminating conditions:
.INDENT 0.0
.IP \(bu 2
the lexer may match some rule (including default rule \fB*\fP) and come to a
final state
.IP \(bu 2
the lexer may fail to match any rule and come to a default state
.IP \(bu 2
the lexer may reach the end of input
.UNINDENT
.sp
The first two conditions terminate the lexer in a \(dqnatural\(dq 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. re2zig 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:
.INDENT 0.0
.IP \(bu 2
\fBSentinel.\fP
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
\(dqsentinel\(dq 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 \fBYYFILL\fP is disabled
with \fBre2c:yyfill:enable = 0;\fP and \fBre2c:eof\fP has the default value
\fB\-1\fP\&.
.nf
.fi
.sp
.IP \(bu 2
\fBSentinel with bounds checks.\fP
This method is generic: it allows one to handle any input without restrictions on
the regular expressions. The idea is to reduce the number of end of input
checks by performing them only on certain characters. Similar to the
\(dqsentinel\(dq method, one of the characters is chosen as a \(dqsentinel\(dq and
appended at the end of input. However, there is no restriction on where the
sentinel may occur (in fact, any character can be chosen for a sentinel).
When the lexer reads this character, it additionally performs a bounds check.
If the current position is within bounds, the lexer resumes matching and
handles the sentinel as a regular character. Otherwise it invokes \fBYYFILL\fP
(unless it is disabled). If more input is supplied, the lexer will rematch the
last character and continue as if the sentinel wasn\(aqt there. Otherwise it must
be the real end of input, and the lexer stops. This method is used when
\fBre2c:eof\fP has non\-negative value (it should be set to the numeric value of
the sentinel). \fBYYFILL\fP is optional.
.nf
.fi
.sp
.IP \(bu 2
\fBBounds checks with padding.\fP
This method is generic, and it may be faster than the \(dqsentinel with bounds
checks\(dq method, but it is also more complex. The idea is to partition 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 this
SCC. This reduces the number of checks, but there is a problem with short
lexemes at the end of input, as the check requires enough characters to cover
the longest lexeme. This can be fixed by padding the input with a few fake
characters that do not form a valid lexeme suffix (so that the lexer cannot
match them). The length of padding should be \fBYYMAXFILL\fP, generated with
a \fBmax\fP block. If there is not enough input, the lexer invokes \fBYYFILL\fP
which should supply at least the required number of characters or not return.
This method is used if \fBYYFILL\fP is enabled and \fBre2c:eof\fP is \fB\-1\fP
(this is the default configuration).
.nf
.fi
.sp
.IP \(bu 2
\fBCustom checks.\fP
Generic API allows one to override basic operations like reading a character,
which makes it possible to include the end\-of\-input checks as part of them.
This approach is error\-prone and should be used with caution. To use a custom
method, enable generic API with \fB\-\-api custom\fP or \fBre2c:api = custom;\fP and
disable default bounds checks with \fBre2c:yyfill:enable = 0;\fP or
\fBre2c:yyfill:check = 0;\fP\&.
.UNINDENT
.sp
The following subsections contain an example of each method.
.SS Sentinel
.sp
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 is the last character of each input string, and it is not allowed in the
middle of a lexeme by any of the rules (in particular, it is not included in
character ranges where it is easy to overlook). If a null occurs in the middle
of a string, it is a syntax error and the lexer will match default rule \fB*\fP,
but it won\(aqt read past the end of input or crash (use
\fI\%\-Wsentinel\-in\-midrule\fP
warning and \fBre2c:sentinel\fP configuration to verify this). Configuration
\fBre2c:yyfill:enable = 0;\fP suppresses the generation of bounds checks and
\fBYYFILL\fP invocations.
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT
const std = @import(\(dqstd\(dq);
// Expects a null\-terminated string.
fn lex(yyinput: [:0]const u8) i32 {
var yycursor: u32 = 0;
var count: i32 = 0;
loop: while (true) {
%{
re2c:yyfill:enable = 0;
* { return \-1; }
[\ex00] { return count; }
[a\-z]+ { count += 1; continue :loop; }
[ ]+ { continue :loop; }
%}
}
}
test {
try std.testing.expectEqual(lex(\(dq\(dq), 0);
try std.testing.expectEqual(lex(\(dqone two three\(dq), 3);
try std.testing.expectEqual(lex(\(dqf0ur\(dq), \-1);
}
.ft P
.fi
.UNINDENT
.UNINDENT
.SS Sentinel with bounds checks
.sp
This example uses sentinel with bounds checks to handle the end of input (this
method was added in version 1.2). The program counts space\-separated
single\-quoted strings. The sentinel character is null, which is specified with
\fBre2c:eof = 0;\fP configuration. As in the \fI\%sentinel\fP method, null is the last
character of each input string, but it is allowed in the middle of a rule (for
example, \fB\(aqaaa\e0aa\(aq\e0\fP is valid input, but \fB\(aqaaa\e0\fP is a syntax error).
Bounds checks are generated in each state that matches an input character, but
they are scoped to the branch that handles null. Bounds checks are of the form
\fBYYLIMIT <= YYCURSOR\fP or \fBYYLESSTHAN(1)\fP with generic API. If the check
condition is true, lexer has reached the end of input and should stop
(\fBYYFILL\fP is disabled with \fBre2c:yyfill:enable = 0;\fP as the input fits into
one buffer, see the \fI\%YYFILL with sentinel\fP section for an example that uses
\fBYYFILL\fP). Reaching the end of input opens three possibilities: if the lexer
is in the initial state it will match the end\-of\-input rule \fB$\fP, otherwise it
may fallback to a previously matched rule (including default rule \fB*\fP) or go
to a default state, causing
\fI\%\-Wundefined\-control\-flow\fP\&.
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT
const std = @import(\(dqstd\(dq);
// Expects a null\-terminated string.
fn lex(yyinput: [:0]const u8) i32 {
var yycursor: usize = 0;
var yymarker: usize = 0;
const yylimit: usize = yyinput.len; // points at the terminating null
var count: i32 = 0;
loop: while (true) {
%{
re2c:yyfill:enable = 0;
re2c:eof = 0;
str = [\(aq] ([^\(aq\e\e] | [\e\e][^])* [\(aq];
* { return \-1; }
$ { return count; }
str { count += 1; continue :loop; }
[ ]+ { continue :loop; }
%}
}
}
test {
try std.testing.expectEqual(lex(\(dq\(dq), 0);
try std.testing.expectEqual(lex(\(dq\(aqqu\ex00tes\(aq \(aqare\(aq \(aqfine: \e\e\(aq\(aq \(dq), 3);
try std.testing.expectEqual(lex(\(dq\(aqunterminated\e\e\(aq\(dq), \-1);
}
.ft P
.fi
.UNINDENT
.UNINDENT
.SS Bounds checks with padding
.sp
This example uses bounds checks with padding to handle the end of input (this
method is enabled by default). The program counts space\-separated single\-quoted
strings. There is a padding of \fBYYMAXFILL\fP null characters appended at the end
of input, where \fBYYMAXFILL\fP value is autogenerated with a \fBmax\fP block. It
is not necessary to use null for padding \-\-\- any characters can be used as long
as they do not form a valid lexeme suffix (in this example padding should not
contain single quotes, as they may be mistaken for a suffix of a single\-quoted
string). There is a \(dqstop\(dq rule that matches the first padding character (null)
and terminates the lexer (note that it checks if null is at the beginning of
padding, otherwise it is a syntax error). Bounds checks are generated only in
some states that are determined by the strongly connected components of the
underlying automaton. Checks have the form \fB(YYLIMIT \- YYCURSOR) < n\fP or
\fBYYLESSTHAN(n)\fP with generic API, where \fBn\fP is the minimum number of
characters that are needed for the lexer to proceed (it also means that the next
bounds check will occur in at most \fBn\fP characters). If the check condition is
true, the lexer has reached the end of input and will invoke \fBYYFILL(n)\fP that
should either supply at least \fBn\fP input characters or not return. In this
example \fBYYFILL\fP always fails and terminates the lexer with an error (which is
fine because the input fits into one buffer). See the \fI\%YYFILL with padding\fP
section for an example that refills the input buffer with \fBYYFILL\fP\&.
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT
const std = @import(\(dqstd\(dq);
%{max %}
fn lex(str: []const u8) !i32 {
// Create a copy of the input string padded with yymaxfill zeroes at the end.
var yyinput = try std.testing.allocator.alloc(u8, str.len + yymaxfill);
defer std.testing.allocator.free(yyinput);
std.mem.copyForwards(u8, yyinput[0..], str);
std.mem.copyForwards(u8, yyinput[str.len..], &[_]u8{0} ** yymaxfill); // zero padding
var yycursor: usize = 0;
const yylimit: usize = yyinput.len;
var count: i32 = 0;
loop: while (true) {
%{
re2c:YYFILL = \(dqreturn \-1;\(dq;
str = [\(aq] ([^\(aq\e\e] | [\e\e][^])* [\(aq];
[\ex00] {
// Check that it is the sentinel, not some unexpected null.
return if (yycursor \- 1 == str.len) count else \-1;
}
str { count += 1; continue :loop; }
[ ]+ { continue :loop; }
* { return \-1; }
%}
}
}
test {
try std.testing.expectEqual(lex(\(dq\(dq), 0);
try std.testing.expectEqual(lex(\(dq\(aqqu\ex00tes\(aq \(aqare\(aq \(aqfine: \e\e\(aq\(aq \(dq), 3);
try std.testing.expectEqual(lex(\(dq\(aqunterminated\e\e\(aq\(dq), \-1);
try std.testing.expectEqual(lex(\(dq\(aqunexpected \ex00 null\e\e\(aq\(dq), \-1);
}
.ft P
.fi
.UNINDENT
.UNINDENT
.SS Custom checks
.sp
This example uses a custom end\-of\-input handling method based on generic API.
The program counts space\-separated single\-quoted strings. It is the same as the
\fI\%sentinel\fP example, except that the input is not null\-terminated. To cover up
for the absence of a sentinel character at the end of input, \fBYYPEEK\fP is
redefined to perform a bounds check before it reads the next input character.
This is inefficient because checks are done very often. If the check condition
fails, \fBYYPEEK\fP returns the real character, otherwise it returns a fake
sentinel character.
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT
const std = @import(\(dqstd\(dq);
// Expects a string without terminating null.
fn lex(str: []const u8) i32 {
var cur: usize = 0;
var count: i32 = 0;
loop: while (true) {
%{
re2c:api = generic;
re2c:yyfill:enable = 0;
// YYPEEK returns \(dqfake\(dq terminating null if cursor has reached limit.
re2c:YYPEEK = \(dqif (cur >= str.len) 0 else str[cur]\(dq;
re2c:YYSKIP = \(dqcur += 1;\(dq;
* { return \-1; }
[\ex00] { return count; }
[a\-z]+ { count += 1; continue :loop; }
[ ]+ { continue :loop; }
%}
}
}
test {
try std.testing.expectEqual(lex(\(dq\(dq), 0);
try std.testing.expectEqual(lex(\(dqone two three\(dq), 3);
try std.testing.expectEqual(lex(\(dqf0ur\(dq), \-1);
}
.ft P
.fi
.UNINDENT
.UNINDENT
.SH BUFFER REFILLING
.sp
The need for buffering arises when the input cannot be mapped in memory all at
once: either it is too large, or it comes in a streaming fashion (like reading
from a socket). The usual technique in such 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 is moved out and new data is moved in. In
practice it is somewhat more complex, because lexer state consists not of a
single input position, but a set of interrelated positions:
.INDENT 0.0
.IP \(bu 2
cursor: the next input character to be read (\fBYYCURSOR\fP in C pointer API or
\fBYYSKIP\fP/\fBYYPEEK\fP in generic API)
.IP \(bu 2
limit: the position after the last available input character (\fBYYLIMIT\fP in
C pointer API, implicitly handled by \fBYYLESSTHAN\fP in generic API)
.IP \(bu 2
marker: the position of the most recent match, if any (\fBYYMARKER\fP in default
API or \fBYYBACKUP\fP/\fBYYRESTORE\fP in generic API)
.IP \(bu 2
token: the start of the current lexeme (implicit in re2zig API, as it is not
needed for the normal lexer operation and can be defined and updated by the
user)
.IP \(bu 2
context marker: the position of the trailing context (\fBYYCTXMARKER\fP in
C pointer API or \fBYYBACKUPCTX\fP/\fBYYRESTORECTX\fP in generic API)
.IP \(bu 2
tag variables: submatch positions (defined with \fBstags\fP and \fBmtags\fP blocks
and generic API primitives \fBYYSTAGP\fP/\fBYYSTAGN\fP/\fBYYMTAGP\fP/\fBYYMTAGN\fP)
.UNINDENT
.sp
Not all these are used in every case, but if used, they must be updated by
\fBYYFILL\fP\&. All active positions are contained in the segment between token and
cursor, therefore everything between buffer start and token can be discarded,
the segment from token and up to limit should be moved to the beginning of
buffer, and the free space at the end of buffer should be filled with new data.
In order to avoid frequent \fBYYFILL\fP calls it is best to fill in as many input
characters as possible (even though fewer characters might suffice to resume the
lexer). The details of \fBYYFILL\fP implementation are slightly different
depending on which EOF handling method is used: the case of EOF rule is somewhat
simpler than the case of bounds\-checking with padding. Also note that if
\fB\-f \-\-storable\-state\fP option is used, \fBYYFILL\fP has slightly different
semantics (described in the section about storable state).
.SS YYFILL with sentinel
.sp
If EOF rule is used, \fBYYFILL\fP is a function\-like primitive that accepts
no arguments and returns a value which is checked against zero. \fBYYFILL\fP
invocation is triggered by condition \fBYYLIMIT <= YYCURSOR\fP in C pointer API and
\fBYYLESSTHAN()\fP in generic API. A non\-zero return value means that \fBYYFILL\fP
has failed. A successful \fBYYFILL\fP call must supply at least one character and
adjust input positions accordingly. Limit must always be set to one after the
last input position in buffer, and the character at the limit position must be
the sentinel symbol specified by \fBre2c:eof\fP configuration. The pictures below
show the relative locations of input positions in buffer before and after
\fBYYFILL\fP call (sentinel symbol is marked with \fB#\fP, and the second picture
shows the case when there is not enough input to fill the whole buffer).
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
<\-\- 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
.ft P
.fi
.UNINDENT
.UNINDENT
.sp
Here is an example of a program that reads input file \fBinput.txt\fP in chunks of
4096 bytes and uses EOF rule.
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT
const std = @import(\(dqstd\(dq);
const bufsize = 4095;
const State = struct {
yyinput: [bufsize + 1]u8,
yycursor: usize,
yymarker: usize,
yylimit: usize,
token: usize,
eof: bool
};
fn fill(st: *State, file: anytype) i32 {
if (st.eof) { return \-1; } // unexpected EOF
// Error: lexeme too long. In real life can reallocate a larger buffer.
if (st.token < 1) { return \-2; }
// Shift buffer contents (discard everything up to the current token).
std.mem.copyBackwards(
u8, st.yyinput[0..st.yylimit \- st.token], st.yyinput[st.token..st.yylimit]);
st.yycursor \-= st.token;
st.yymarker = @subWithOverflow(st.yymarker, st.token)[0];
st.yylimit \-= st.token;
st.token = 0;
// Fill free space at the end of buffer with new data from file.
st.yylimit += file.read(st.yyinput[st.yylimit..bufsize]) catch 0;
st.yyinput[st.yylimit] = 0; // append sentinel symbol
// If read less than expected, this is the end of input.
st.eof = st.yylimit < bufsize;
return 0;
}
fn lex(yyrecord: *State, file: anytype) i32 {
var count: i32 = 0;
loop: while (true) {
yyrecord.token = yyrecord.yycursor;
%{
re2c:api = record;
re2c:eof = 0;
re2c:YYFILL = \(dqfill(yyrecord, file) == 0\(dq;
str = [\(aq] ([^\(aq\e\e] | [\e\e][^])* [\(aq];
* { return \-1; }
$ { return count; }
str { count += 1; continue :loop; }
[ ]+ { continue :loop; }
%}
}
}
test {
const fname = \(dqinput\(dq;
const content = \(dq\(aqqu\ex00tes\(aq \(aqare\(aq \(aqfine: \e\e\(aq\(aq \(dq ** bufsize;
const count = 3 * bufsize; // number of quoted strings written to file
// Prepare input file: a few times the size of the buffer, containing
// strings with zeroes and escaped quotes.
var fw = try std.fs.cwd().createFile(fname, .{});
try fw.writeAll(content);
fw.close();
// Prepare lexer state: all offsets are at the end of buffer.
var fr = try std.fs.cwd().openFile(fname, .{ .mode = .read_only});
// Normally file would be part of the state struct, but BufferedReader type is unclear.
var br = std.io.bufferedReader(fr.reader());
var st = State{
.yyinput = undefined,
.yycursor = bufsize,
.yymarker = bufsize,
.yylimit = bufsize,
.token = bufsize,
.eof = false,
};
// Sentinel at \(gayylimit\(ga offset is set to zero, which triggers YYFILL.
st.yyinput[st.yylimit] = 0;
// Run the lexer.
try std.testing.expectEqual(lex(&st, &br), count);
// Cleanup: remove input file.
fr.close();
try std.fs.cwd().deleteFile(fname);
}
.ft P
.fi
.UNINDENT
.UNINDENT
.SS YYFILL with padding
.sp
In the default case (when EOF rule is not used) \fBYYFILL\fP is a function\-like
primitive that accepts a single argument and does not return any value.
\fBYYFILL\fP invocation is triggered by condition \fB(YYLIMIT \- YYCURSOR) < n\fP in
C pointer API and \fBYYLESSTHAN(n)\fP in generic API. The argument passed to
\fBYYFILL\fP is the minimal number of characters that must be supplied. If it
fails to do so, \fBYYFILL\fP must not return to the lexer (for that reason it is
best implemented as a macro that returns from the calling function on failure).
In case of a successful \fBYYFILL\fP invocation the limit position must be set
either to one after the last input position in buffer, or to the end of
\fBYYMAXFILL\fP padding (in case \fBYYFILL\fP has successfully read at least \fBn\fP
characters, but not enough to fill the entire buffer). The pictures below show
the relative locations of input positions in buffer before and after \fBYYFILL\fP
invocation (\fBYYMAXFILL\fP padding on the second picture is marked with \fB#\fP
symbols).
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
<\-\- 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 \->
.ft P
.fi
.UNINDENT
.UNINDENT
.sp
Here is an example of a program that reads input file \fBinput.txt\fP in chunks of
4096 bytes and uses bounds\-checking with padding.
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT
const std = @import(\(dqstd\(dq);
%{max %}
const bufsize = 4096;
const State = struct {
yyinput: [bufsize + yymaxfill]u8,
yycursor: usize,
yymarker: usize,
yylimit: usize,
token: usize,
eof: bool
};
fn fill(st: *State, need: usize, file: anytype) i32 {
if (st.eof) { return \-1; } // unexpected EOF
// Error: lexeme too long. In real life can reallocate a larger buffer.
if (st.token < need) { return \-2; }
// Shift buffer contents (discard everything up to the current token).
std.mem.copyBackwards(
u8, st.yyinput[0..st.yylimit \- st.token], st.yyinput[st.token..st.yylimit]);
st.yycursor \-= st.token;
st.yymarker = @subWithOverflow(st.yymarker, st.token)[0];
st.yylimit \-= st.token;
st.token = 0;
// Fill free space at the end of buffer with new data from file.
st.yylimit += file.read(st.yyinput[st.yylimit..bufsize]) catch 0;
// If read less than expected, this is the end of input.
if (st.yylimit < bufsize) {
st.eof = true;
@memset(st.yyinput[st.yylimit..st.yylimit + yymaxfill], 0);
st.yylimit += yymaxfill;
}
return 0;
}
fn lex(yyrecord: *State, file: anytype) i32 {
var count: i32 = 0;
loop: while (true) {
yyrecord.token = yyrecord.yycursor;
%{
re2c:api = record;
re2c:YYFILL = \(dq{ if (fill(yyrecord, @@, file) != 0) return \-2; }\(dq;
str = [\(aq] ([^\(aq\e\e] | [\e\e][^])* [\(aq];
[\ex00] {
// Check that it is the sentinel, not some unexpected null.
return if (yyrecord.token == yyrecord.yylimit \- yymaxfill) count else \-1;
}
str { count += 1; continue :loop; }
[ ]+ { continue :loop; }
* { return \-1; }
%}
}
}
test {
const fname = \(dqinput\(dq;
const content = \(dq\(aqqu\ex00tes\(aq \(aqare\(aq \(aqfine: \e\e\(aq\(aq \(dq ** bufsize;
const count = 3 * bufsize; // number of quoted strings written to file
// Prepare input file: a few times the size of the buffer, containing
// strings with zeroes and escaped quotes.
var fw = try std.fs.cwd().createFile(fname, .{});
try fw.writeAll(content);
fw.close();
// Prepare lexer state: all offsets are at the end of buffer.
// This immediately triggers YYFILL, as the YYLESSTHAN condition is true.
var fr = try std.fs.cwd().openFile(fname, .{ .mode = .read_only});
// Normally file would be part of the state struct, but BufferedReader type is unclear.
var br = std.io.bufferedReader(fr.reader());
var st = State{
.yyinput = undefined,
.yycursor = bufsize,
.yymarker = bufsize,
.yylimit = bufsize,
.token = bufsize,
.eof = false,
};
@memset(st.yyinput[st.yylimit..st.yylimit + yymaxfill], 0); // zero\-padding at the end
// Run the lexer.
try std.testing.expectEqual(lex(&st, &br), count);
// Cleanup: remove input file.
fr.close();
try std.fs.cwd().deleteFile(fname);
}
.ft P
.fi
.UNINDENT
.UNINDENT
.SH FEATURES
.SS Multiple blocks
.sp
Sometimes it is necessary to have multiple interrelated lexers (for example, if
there is a high\-level state machine that transitions between lexer modes). This
can be implemented using multiple connected re2zig blocks. Another option is to
use \fI\%start conditions\fP\&.
.sp
The implementation of connections between blocks depends on the target language.
In languages that have \fBgoto\fP statement (such as C/C++ and Go) one can have
all blocks in one function, each of them prefixed with a label. Transition from
one block to another is a simple \fBgoto\fP\&.
In languages that do not have \fBgoto\fP (such as Rust) it is necessary to use a
loop with a switch on a state variable, similar to the \fByystate\fP loop/switch
generated by re2zig, or else wrap each block in a function and use function calls.
.sp
The example below uses multiple blocks to parse binary, octal, decimal and
hexadecimal numbers. Each base has its own block. The initial block determines
base and dispatches to other blocks. Common configurations are defined in a
separate block at the beginning of the program; they are inherited by the other
blocks.
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT
const std = @import(\(dqstd\(dq);
// Store u32 number in u64 during parsing to simplify overflow handling.
const State = struct {
yyinput: [:0]const u8,
yycursor: usize,
yymarker: usize,
num: u64,
};
// Common re2c definitions shared between all functions.
%{
re2c:api = record;
re2c:yyrecord = st;
re2c:yyfill:enable = 0;
%}
const ERROR: u64 = @as(u64, std.math.maxInt(u32)) + 1; // overflow
// Add digit with the given base, checking for overflow.
fn add(st: *State, offs: u8, base: u64) void {
const digit = st.yyinput[st.yycursor \- 1] \- offs;
st.num = @min(st.num * base + digit, ERROR);
}
// Convert u64 to optional u32 (null meaning overflow or parse error).
fn maybeU32(num: u64) ?u32 {
return if (num < ERROR) @intCast(num) else null;
}
fn parse_u32(s: [:0]const u8) ?u32 {
var st = State {.yyinput = s, .yycursor = 0, .yymarker = 0, .num = 0};
%{
\(aq0b\(aq / [01] { return parse_bin(&st); }
\(dq0\(dq { return parse_oct(&st); }
\(dq\(dq / [1\-9] { return parse_dec(&st); }
\(aq0x\(aq / [0\-9a\-fA\-F] { return parse_hex(&st); }
* { return null; }
%}
}
fn parse_bin(st: *State) ?u32 {
bin: while (true) {%{
[01] { add(st, 48, 2); continue :bin; }
* { return maybeU32(st.num); }
%}}
}
fn parse_oct(st: *State) ?u32 {
oct: while (true) {%{
[0\-7] { add(st, 48, 8); continue :oct; }
* { return maybeU32(st.num); }
%}}
}
fn parse_dec(st: *State) ?u32 {
dec: while (true) {%{
[0\-9] { add(st, 48, 10); continue :dec; }
* { return maybeU32(st.num); }
%}}
}
fn parse_hex(st: *State) ?u32 {
hex: while (true) {%{
[0\-9] { add(st, 48, 16); continue :hex; }
[a\-f] { add(st, 87, 16); continue :hex; }
[A\-F] { add(st, 55, 16); continue :hex; }
* { return maybeU32(st.num); }
%}}
}
test {
try std.testing.expectEqual(parse_u32(\(dq\(dq), null);
try std.testing.expectEqual(parse_u32(\(dq1234567890\(dq), 1234567890);
try std.testing.expectEqual(parse_u32(\(dq0b1101\(dq), 13);
try std.testing.expectEqual(parse_u32(\(dq0x7Fe\(dq), 2046);
try std.testing.expectEqual(parse_u32(\(dq0644\(dq), 420);
try std.testing.expectEqual(parse_u32(\(dq9999999999\(dq), null);
}
.ft P
.fi
.UNINDENT
.UNINDENT
.SS Start conditions
.sp
Start conditions are enabled with \fB\-\-start\-conditions\fP option. They provide a
way to encode multiple interrelated automata within the same re2zig block.
.sp
Each condition corresponds to a single automaton and has a unique name specified
by the user and a unique internal number defined by re2zig\&. The numbers are used
to switch between conditions: the generated code uses \fBYYGETCOND\fP and
\fBYYSETCOND\fP primitives to get the current condition or set it to the
given number. Use \fBconditions\fP block, \fB\-\-header\fP option or \fBre2c:header\fP
configuration to generate numeric condition identifiers. Configuration
\fBre2c:cond:enumprefix\fP specifies the generated identifier prefix.
.sp
In condition mode every rule must be prefixed with a list of comma\-separated
condition names in angle brackets, or a wildcard \fB<*>\fP to denote all
conditions. The rule syntax is extended as follows:
.INDENT 0.0
.INDENT 3.5
.INDENT 0.0
.TP
.B \fB< condition\-list > regular\-expression code\fP
A rule that is merged to every condition on the \fBcondition\-list\fP\&.
It matches \fBregular\-expression\fP and executes the associated \fBcode\fP\&.
.TP
.B \fB< condition\-list > regular\-expression => condition code\fP
A rule that is merged to every condition on the \fBcondition\-list\fP\&.
It matches \fBregular\-expression\fP, sets the current condition to
\fBcondition\fP and executes the associated \fBcode\fP\&.
.TP
.B \fB< condition\-list > regular\-expression :=> condition\fP
A rule that is merged to every condition on the \fBcondition\-list\fP\&.
It matches \fBregular\-expression\fP and immediately transitions to
\fBcondition\fP (there is no semantic action).
.TP
.B \fB< condition\-list > !action code\fP
A rule that binds \fBcode\fP to the place defined by \fBaction\fP in every
condition on the \fBcondition\-list\fP (see the \fI\%actions\fP section for
various types of actions).
.TP
.B \fB<! condition\-list > code\fP
A rule that prepends \fBcode\fP to semantic actions of all rules for every
condition on the \fBcondition\-list\fP\&. This syntax is deprecated and the
\fB!pre_rule\fP action should be used instead (it does exactly the same).
.TP
.B \fB< > code\fP
A rule that creates a special entry condition with number zero and name
\fB\(dq0\(dq\fP that executes \fBcode\fP before jumping to other conditions.
This syntax is deprecated, and the \fB!entry\fP action should be used
instead (it provides a more fine\-grained control, as the code can be
specified on a per\-condition basis, and one can jump directly to
condition start without going through condition dispatch).
.TP
.B \fB< > => condition code\fP
Same as the previous rule, except that it sets the next \fBcondition\fP\&.
.TP
.B \fB< > :=> condition\fP
Same as the previous rule, except that it has no associated code and
immediately jumps to \fBcondition\fP\&.
.UNINDENT
.UNINDENT
.UNINDENT
.sp
The code re2zig generates for conditions depends on whether re2zig uses
goto/label approach or loop/switch approach to encode the automata.
.sp
In languages that have \fBgoto\fP statement (such as C/C++ and Go) conditions are
naturally implemented as blocks of code prefixed with labels of the form
\fByyc_<cond>\fP, where \fBcond\fP is a condition name (label prefix can be changed
with \fBre2c:cond:prefix\fP). Transitions between conditions are implemented using
\fBgoto\fP and condition labels. Before all conditions re2zig generates an initial
switch on \fBYYGETSTATE\fP that jumps to the start state of the current condition.
The shortcut rules \fB:=>\fP bypass the initial switch and jump directly to the
specified condition (\fBre2c:cond:goto\fP can be used to change the default
behavior). The rules with semantic actions do not automatically jump to the next
condition; this should be done by the user\-defined action code.
.sp
In languages that do not have \fBgoto\fP (such as Rust) re2zig reuses the
\fByystate\fP variable to store condition numbers. Each condition gets a numeric
identifier equal to the number of its start state, and a switch between
conditions is no different than a switch between DFA states of a single
condition. There is no need for a separate initial condition switch.
(Since the same approach is used to implement storable states,
\fBYYGETCOND\fP/\fBYYSETCOND\fP are redundant if both storable states and
conditions are used).
.sp
The program below uses start conditions to parse binary, octal, decimal and
hexadecimal numbers. There is a single block where each base has its own
condition, and the initial condition is connected to all of them. User\-defined
variable \fBcond\fP stores the current condition number; it is initialized to the
number of the initial condition generated with \fBconditions\fP block.
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT \-c
const std = @import(\(dqstd\(dq);
%{conditions %}
const ERROR: u64 = @as(u64, std.math.maxInt(u32)) + 1; // overflow
// Add digit with the given base, checking for overflow.
fn add(num: *u64, str: [:0]const u8, cur: usize, offs: u8, base: u64) void {
num.* = @min(num.* * base + (str[cur \- 1] \- offs), ERROR);
}
fn parse_u32(yyinput: [:0]const u8) ?u32 {
var yycursor: usize = 0;
var yymarker: usize = 0;
var yycond = yycinit;
var num: u64 = 0; // Store number in u64 to simplify overflow checks.
loop: while(true) {
%{
re2c:yyfill:enable = 0;
<init> \(aq0b\(aq / [01] :=> bin
<init> \(dq0\(dq :=> oct
<init> \(dq\(dq / [1\-9] :=> dec
<init> \(aq0x\(aq / [0\-9a\-fA\-F] :=> hex
<init> * { return null; }
<bin> [01] { add(&num, yyinput, yycursor, 48, 2); continue :loop; }
<oct> [0\-7] { add(&num, yyinput, yycursor, 48, 8); continue :loop; }
<dec> [0\-9] { add(&num, yyinput, yycursor, 48, 10); continue :loop; }
<hex> [0\-9] { add(&num, yyinput, yycursor, 48, 16); continue :loop; }
<hex> [a\-f] { add(&num, yyinput, yycursor, 87, 16); continue :loop; }
<hex> [A\-F] { add(&num, yyinput, yycursor, 55, 16); continue :loop; }
<bin, oct, dec, hex> * {
return if (num < ERROR) @intCast(num) else null;
}
%}}
}
test {
try std.testing.expectEqual(parse_u32(\(dq\(dq), null);
try std.testing.expectEqual(parse_u32(\(dq1234567890\(dq), 1234567890);
try std.testing.expectEqual(parse_u32(\(dq0b1101\(dq), 13);
try std.testing.expectEqual(parse_u32(\(dq0x7Fe\(dq), 2046);
try std.testing.expectEqual(parse_u32(\(dq0644\(dq), 420);
try std.testing.expectEqual(parse_u32(\(dq9999999999\(dq), null);
}
.ft P
.fi
.UNINDENT
.UNINDENT
.SS Storable state
.sp
With \fB\-\-storable\-state\fP option re2zig generates a lexer that can store
its current state, return to the caller, and later resume operations exactly
where it left off. The default mode of operation in re2zig is a \(dqpull\(dq model,
in which the lexer \(dqpulls\(dq more input whenever it needs it. This may be
unacceptable in cases 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
feature is intended exactly for such cases: it allows one to generate lexers that
work in a \(dqpush\(dq 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 \(dqpull\(dq model:
.INDENT 0.0
.IP \(bu 2
Define \fBYYSETSTATE()\fP and \fBYYGETSTATE(state)\fP primitives.
.IP \(bu 2
Define \fByych\fP, \fByyaccept\fP (if used) and \fBstate\fP variables as a part of
persistent lexer state. The \fBstate\fP variable should be initialized to \fB\-1\fP\&.
.IP \(bu 2
\fBYYFILL\fP should return to the outer program instead of trying to supply more
input. Return code should indicate that lexer needs more input.
.IP \(bu 2
The outer program should recognize situations when lexer needs more input and
respond appropriately.
.IP \(bu 2
Optionally use \fBgetstate\fP block to generate \fBYYGETSTATE\fP switch detached
from the main lexer. This only works for languages that have \fBgoto\fP (not in
\fB\-\-loop\-switch\fP mode).
.IP \(bu 2
Use \fBre2c:eof\fP and the \fI\%sentinel with bounds checks\fP 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).
.UNINDENT
.sp
Here is an example of a \(dqpush\(dq 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 \(dqsending\(dq the next
packet and resumes lexing. This process stops when all the packets have been
sent, or when there is an error.
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT \-f
const std = @import(\(dqstd\(dq);
const Status = enum {
end,
ready,
waiting,
bad_packet,
big_packet
};
// Use a small buffer to cover the case when a lexeme doesn\(aqt fit.
// In real world use a larger buffer.
const bufsize = 10;
const State = struct {
yyinput: [bufsize + 1]u8,
yycursor: usize,
yymarker: usize,
yylimit: usize,
token: usize,
yystate: i32,
received: usize,
};
fn fill(st: *State, file: anytype) Status {
// Error: lexeme too long. In real life can reallocate a larger buffer.
if (st.token < 1) { return Status.big_packet; }
// Shift buffer contents (discard everything up to the current token).
std.mem.copyBackwards(
u8, st.yyinput[0..st.yylimit \- st.token], st.yyinput[st.token..st.yylimit]);
st.yycursor \-= st.token;
st.yymarker = @subWithOverflow(st.yymarker, st.token)[0];
st.yylimit \-= st.token;
st.token = 0;
// Fill free space at the end of buffer with new data from file.
st.yylimit += file.read(st.yyinput[st.yylimit..bufsize]) catch 0;
st.yyinput[st.yylimit] = 0; // append sentinel symbol
return Status.ready;
}
fn lex(yyrecord: *State) Status {
var yych: u8 = 0;
loop: while (true) {
yyrecord.token = yyrecord.yycursor;
%{
re2c:api = record;
re2c:eof = 0;
re2c:YYFILL = \(dqreturn Status.waiting;\(dq;
packet = [a\-z]+[;];
* { return Status.bad_packet; }
$ { return Status.end; }
packet { yyrecord.received += 1; continue :loop; }
%}
}
}
fn run(expect: Status, packets: []const []const u8) !void {
// Create a \(dqpipe\(dq (open the same file for reading and writing).
const fname = \(dqinput\(dq;
var fw = try std.fs.cwd().createFile(fname, .{});
var fr = try std.fs.cwd().openFile(fname, .{ .mode = .read_only});
// Initialize lexer state: \(gastate\(ga value is \-1, all offsets are at the end
// of buffer. Normally file would be part of the state, but BufferedReader
// type is unclear.
var br = std.io.bufferedReader(fr.reader());
var st = State{
.yyinput = undefined,
.yycursor = bufsize,
.yymarker = bufsize,
.yylimit = bufsize,
.token = bufsize,
.yystate = \-1,
.received = 0,
};
// Sentinel at \(gayylimit\(ga offset is set to zero, which triggers YYFILL.
st.yyinput[st.yylimit] = 0;
// Main loop. The buffer contains incomplete data which appears packet by
// packet. When the lexer needs more input it saves its internal state and
// returns to the caller which should provide more input and resume lexing.
var status = Status.ready;
var send: usize = 0;
while (true) {
status = lex(&st);
if (status == Status.end) {
break;
} else if (status == Status.waiting) {
if (send < packets.len) {
std.log.debug(\(dqsending packet {}\(dq, .{send});
try fw.writeAll(packets[send]);
send += 1;
}
status = fill(&st, &br);
std.log.debug(\(dqfilled buffer [{s}], status {}\(dq, .{st.yyinput, status});
if (status != Status.ready) {
break;
}
} else if (status == Status.bad_packet) {
break;
}
}
// Check results.
try std.testing.expectEqual(status, expect);
if (status == Status.end) { try std.testing.expectEqual(st.received, send); }
// Cleanup: remove input file.
fw.close();
fr.close();
try std.fs.cwd().deleteFile(fname);
}
test {
try run(Status.end, &[_][]const u8{});
try run(Status.end, &[_][]const u8{\(dqzero;\(dq, \(dqone;\(dq, \(dqtwo;\(dq, \(dqthree;\(dq, \(dqfour;\(dq});
try run(Status.bad_packet, &[_][]const u8{\(dq??;\(dq});
try run(Status.big_packet, &[_][]const u8{\(dqlooooooooooooong;\(dq});
}
.ft P
.fi
.UNINDENT
.UNINDENT
.SS Reusable blocks
.sp
Reusable blocks of the form \fB/*!rules:re2c[:<name>] ... */\fP or
\fB%{rules[:<name>] ... %}\fP can be reused any number of times and combined with
other re2zig blocks. The \fB<name>\fP is optional. A rules block can be used in a
\fBuse\fP block or directive. The code for a rules block is generated at every
point of use.
.sp
Use blocks are defined with \fB/*!use:re2c[:<name>] ... */\fP or
\fB%{use[:<name>] ... %}\fP\&. The \fB<name>\fP is optional: if it\(aqs 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).
.sp
In\-block use directive \fB!use:<name>;\fP can be used from inside of a re2zig
block. It merges the referenced block \fB<name>\fP 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 \fB*\fP, \fB$\fP and (in condition mode) \fB<!>\fP, for which a block\-local
definition overrides any inherited ones. Use directive allows one to combine
different re2zig blocks together in one block (see the example below).
.sp
Named blocks and in\-block use directive were added in re2zig 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 \fB\-r \-\-reusable\fP
option. Before version 1.2 reusable blocks could not be mixed with normal
blocks.
.SS Example of a \fB!use\fP directive
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT
// This example shows how to combine reusable re2c blocks: two blocks
// (\(aqcolors\(aq and \(aqfish\(aq) are merged into one. The \(aqsalmon\(aq rule occurs
// in both blocks; the \(aqfish\(aq block takes priority because it is used
// earlier. Default rule * occurs in all three blocks; the local (not
// inherited) definition takes priority.
const std = @import(\(dqstd\(dq);
const Ans = enum {color, fish, dunno};
%{rules:colors
* { @panic(\(dqah\(dq); }
\(dqred\(dq | \(dqsalmon\(dq | \(dqmagenta\(dq { return Ans.color; }
%}
%{rules:fish
* { @panic(\(dqoh\(dq); }
\(dqhaddock\(dq | \(dqsalmon\(dq | \(dqeel\(dq { return Ans.fish; }
%}
fn lex(yyinput: [:0]const u8) Ans {
var yycursor: usize = 0;
var yymarker: usize = 0;
%{
re2c:yyfill:enable = 0;
!use:fish;
!use:colors;
* { return Ans.dunno; } // overrides inherited \(aq*\(aq rules
%}
}
test {
try std.testing.expectEqual(lex(\(dqsalmon\(dq), Ans.fish);
try std.testing.expectEqual(lex(\(dqwhat?\(dq), Ans.dunno);
}
.ft P
.fi
.UNINDENT
.UNINDENT
.SS Example of a \fB/*!use:re2c ... */\fP block
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT \-\-input\-encoding utf8
// This example supports multiple input encodings: UTF\-8 and UTF\-32.
// Both lexers are generated from the same rules block, and the use
// blocks add only encoding\-specific configurations.
const std = @import(\(dqstd\(dq);
%{rules
re2c:yyfill:enable = 0;
\(dq∀x ∃y\(dq { return yycursor; }
* { return null; }
%}
fn lex_utf8(yyinput: []const u8) ?usize {
var yycursor: usize = 0;
var yymarker: usize = 0;
%{use
re2c:encoding:utf8 = 1;
re2c:YYCTYPE = u8; // the default
%}
}
fn lex_utf32(yyinput: []const u32) ?usize {
var yycursor: usize = 0;
var yymarker: usize = 0;
%{use
re2c:encoding:utf32 = 1;
re2c:YYCTYPE = u32;
%}
}
test {
const s8 = [_]u8{0xe2, 0x88, 0x80, 0x78, 0x20, 0xe2, 0x88, 0x83, 0x79};
try std.testing.expectEqual(lex_utf8(&s8), s8.len);
const s32 = [_]u32{0x2200, 0x78, 0x20, 0x2203, 0x79};
try std.testing.expectEqual(lex_utf32(&s32), s32.len);
}
.ft P
.fi
.UNINDENT
.UNINDENT
.SS Submatch extraction
.sp
re2zig has two options for submatch extraction.
.INDENT 0.0
.TP
.B \fBTags\fP
The first option is to use standalone \fItags\fP of the form \fB@stag\fP or
\fB#mtag\fP, where \fBstag\fP and \fBmtag\fP are arbitrary used\-defined names.
Tags are enabled with \fB\-T \-\-tags\fP option or \fBre2c:tags = 1\fP
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.
\fIS\-tags\fP bind to the last matching position, and \fIm\-tags\fP 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 \fBsvars\fP and \fBmvars\fP blocks. If there is more than one
way tags can be matched against the input, ambiguity is resolved using
leftmost greedy disambiguation strategy.
.TP
.B \fBCaptures\fP
The second option is to use \fIcapturing groups\fP\&. They are enabled with
\fB\-\-captures\fP option or \fBre2c:captures = 1\fP configuration. There are two
flavours for different disambiguation policies, \fB\-\-leftmost\-captures\fP
(the default) is for leftmost greedy policy, and, \fB\-\-posix\-captures\fP 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: \fB(! ... )\fP\&. With \fB\-\-invert\-captures\fP option or
\fBre2c:invert\-captures = 1\fP configuration the meaning of bang is inverted.
The number of groups for the matching rule is stored in a variable
\fByynmatch\fP (the whole regular expression is group number zero), and
submatch results are stored in \fByypmatch\fP array. Both \fByynmatch\fP and
\fByypmatch\fP should be defined by the user, and \fByypmatch\fP size must be at
least \fB[yynmatch * 2]\fP\&. Use \fBmaxnmatch\fP block to define \fBYYMAXNMATCH\fP,
a constant that equals to the maximum value of \fByynmatch\fP among all rules.
.TP
.B \fBCaptvars\fP
Another way to use capturing groups is the \fB\-\-captvars\fP option or
\fBre2c:captvars = 1\fP configuration. The only difference with \fB\-\-captures\fP
is in the way the generated code stores submatch results: instead of
\fByynmatch\fP and \fByypmatch\fP re2zig generates variables \fByytl<k>\fP and
\fByytr<k>\fP for \fIk\fP\-th capturing group (the user should declare these using
an \fBsvars\fP block). Captures with variables support two disambiguation
policies: \fB\-\-leftmost\-captvars\fP or \fBre2c:leftmost\-captvars = 1\fP for
leftmost greedy policy (the default one) and \fB\-\-posix\-captvars\fP or
\fBre2c:posix\-captvars\fP for POSIX longest\-match policy.
.UNINDENT
.sp
Under the hood all these options translate into tags and
\fI\%Tagged Deterministic Finite Automata with Lookahead\fP\&.
The core idea of TDFA is to minimize the overhead on submatch extraction.
In the extreme, if there\(aqre 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 \fItag variables\fP 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 \fBstags\fP or \fBmtags\fP blocks. If lexer state
is stored, tag variables should be part of it. They also need to be updated by
\fBYYFILL\fP\&.
.sp
S\-tags support the following operations:
.INDENT 0.0
.IP \(bu 2
save input position to an s\-tag: \fBt = YYCURSOR\fP with C pointer API or a
user\-defined operation \fBYYSTAGP(t)\fP with generic API
.IP \(bu 2
save default value to an s\-tag: \fBt = NULL\fP with C pointer API or a
user\-defined operation \fBYYSTAGN(t)\fP with generic API
.IP \(bu 2
copy one s\-tag to another: \fBt1 = t2\fP
.UNINDENT
.sp
M\-tags support the following operations:
.INDENT 0.0
.IP \(bu 2
append input position to an m\-tag: a user\-defined operation \fBYYMTAGP(t)\fP
with both default and generic API
.IP \(bu 2
append default value to an m\-tag: a user\-defined operation \fBYYMTAGN(t)\fP
with both default and generic API
.IP \(bu 2
copy one m\-tag to another: \fBt1 = t2\fP
.UNINDENT
.sp
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 \fB(v, p)\fP, where \fBv\fP is tag value
and \fBp\fP is a pointer to parent node.
.sp
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 \fBYYFILL\fP\&.
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT
const std = @import(\(dqstd\(dq);
const SemVer = struct {
major: u32,
minor: u32,
patch: u32,
};
const none = std.math.maxInt(usize);
fn s2n(str: []const u8) u32 { // convert a pre\-parsed string to a number
var n: u32 = 0;
for (str) |c| { n = n * 10 + (c \- 48); }
return n;
}
fn parse(yyinput: [:0]const u8) ?SemVer {
var yycursor: usize = 0;
var yymarker: usize = 0;
// Final tag variables available in semantic action.
%{svars format = \(dqvar @@: usize = none;\(dq; %}
// Intermediate tag variables used by the lexer (must be autogenerated).
%{stags format = \(dqvar @@: usize = none;\(dq; %}
%{
re2c:yyfill:enable = 0;
re2c:tags = 1;
num = [0\-9]+;
@t1 num @t2 \(dq.\(dq @t3 num @t4 (\(dq.\(dq @t5 num)? [\ex00] {
return SemVer {
.major = s2n(yyinput[t1..t2]),
.minor = s2n(yyinput[t3..t4]),
.patch = if (t5 == none) 0 else s2n(yyinput[t5..yycursor \- 1]),
};
}
* { return null; }
%}
}
test {
try std.testing.expectEqual(parse(\(dq23.34\(dq), SemVer{.major = 23, .minor = 34, .patch = 0});
try std.testing.expectEqual(parse(\(dq1.2.99999\(dq), SemVer{.major = 1, .minor = 2, .patch = 99999});
try std.testing.expectEqual(parse(\(dq1.a\(dq), null);
}
.ft P
.fi
.UNINDENT
.UNINDENT
.sp
Here is a more complex example of using s\-tags with \fBYYFILL\fP to parse a file
with newline\-separated semantic versions. Tag variables are part of the lexer
state, and they are adjusted in \fBYYFILL\fP 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.
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT
const std = @import(\(dqstd\(dq);
const bufsize = 4095;
const none = std.math.maxInt(usize);
const err = error.SyntaxError;
const SemVer = struct {
major: u32,
minor: u32,
patch: u32,
};
fn s2n(str: []const u8) u32 { // convert a pre\-parsed string to a number
var n: u32 = 0;
for (str) |c| { n = n * 10 + (c \- 48); }
return n;
}
const State = struct {
yyinput: [bufsize + 1]u8,
yycursor: usize,
yymarker: usize,
yylimit: usize,
token: usize,
// Intermediate tag variables must be part of the lexer state passed to YYFILL.
// They don\(aqt correspond to tags and should be autogenerated by re2c.
%{stags format = \(dq@@: usize,\en\(dq; %}
eof: bool
};
fn fill(st: *State, file: anytype) i32 {
if (st.eof) { return \-1; } // unexpected EOF
// Error: lexeme too long. In real life can reallocate a larger buffer.
if (st.token < 1) { return \-2; }
// Shift buffer contents (discard everything up to the current token).
std.mem.copyBackwards(
u8, st.yyinput[0..st.yylimit \- st.token], st.yyinput[st.token..st.yylimit]);
st.yycursor \-= st.token;
st.yymarker = @subWithOverflow(st.yymarker, st.token)[0];
st.yylimit \-= st.token;
// Tag variables need to be shifted like other input positions. The check
// for NONE is only needed if some tags are nested inside of alternative or
// repetition, so that they can have NONE value.
%{stags format = \(dqif (st.@@ != none) st.@@ = @subWithOverflow(st.@@, st.token)[0];\en\(dq; %}
st.token = 0;
// Fill free space at the end of buffer with new data from file.
st.yylimit += file.read(st.yyinput[st.yylimit..bufsize]) catch 0;
st.yyinput[st.yylimit] = 0; // append sentinel symbol
// If read less than expected, this is the end of input.
st.eof = st.yylimit < bufsize;
return 0;
}
fn parse(st: *State, file: anytype) !std.ArrayList(SemVer) {
var vers = std.ArrayList(SemVer).init(std.testing.allocator);
// Final tag variables available in semantic action.
%{svars format = \(dqvar @@: usize = 0;\en\(dq; %}
loop: while (true) {
st.token = st.yycursor;
%{
re2c:api = record;
re2c:eof = 0;
re2c:tags = 1;
re2c:yyrecord = st;
re2c:YYFILL = \(dqfill(st, file) == 0\(dq;
num = [0\-9]+;
num @t1 \(dq.\(dq @t2 num @t3 (\(dq.\(dq @t4 num)? [\en] {
try vers.append(SemVer {
.major = s2n(st.yyinput[st.token..t1]),
.minor = s2n(st.yyinput[t2..t3]),
.patch = if (t4 == none) 0 else s2n(st.yyinput[t4..st.yycursor \- 1]),
});
continue :loop;
}
$ { return vers; }
* { return error.SyntaxError; }
%}
}
}
test {
const fname = \(dqinput\(dq;
const content = \(dq1.22.333\en\(dq ** bufsize;
// Prepare input file: a few times the size of the buffer, containing
// strings with zeroes and escaped quotes.
var fw = try std.fs.cwd().createFile(fname, .{});
try fw.writeAll(content);
fw.close();
// Prepare lexer state: all offsets are at the end of buffer.
var fr = try std.fs.cwd().openFile(fname, .{ .mode = .read_only});
// Normally file would be part of the state struct, but BufferedReader type is unclear.
var br = std.io.bufferedReader(fr.reader());
var st = State{
.yyinput = undefined,
.yycursor = bufsize,
.yymarker = bufsize,
.yylimit = bufsize,
.token = bufsize,
%{stags format = \(dq.@@ = none,\en\(dq; %}
.eof = false,
};
// Sentinel at \(gayylimit\(ga offset is set to zero, which triggers YYFILL.
st.yyinput[st.yylimit] = 0;
// Manually construct expected result.
var expect = std.ArrayList(SemVer).init(std.testing.allocator);
for (0..bufsize) |_| try expect.append(SemVer{.major = 1, .minor = 22, .patch = 333});
// Run the lexer.
var result = try parse(&st, &br);
try std.testing.expectEqualDeep(result, expect);
// Cleanup: free memory and remove input file.
expect.deinit();
result.deinit();
fr.close();
try std.fs.cwd().deleteFile(fname);
}
.ft P
.fi
.UNINDENT
.UNINDENT
.sp
Here is an example of using capturing groups to parse semantic versions.
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT
const std = @import(\(dqstd\(dq);
const none = std.math.maxInt(usize);
const SemVer = struct {
major: u32,
minor: u32,
patch: u32,
};
fn s2n(str: []const u8) u32 { // convert pre\-parsed string to a number
var n: u32 = 0;
for (str) |c| { n = n * 10 + (c \- 48); }
return n;
}
fn parse(yyinput: [:0]const u8) ?SemVer {
var yycursor: usize = 0;
var yymarker: usize = 0;
// Final tag variables available in semantic action.
%{svars format = \(dqvar @@: usize = none;\(dq; %}
// Intermediate tag variables used by the lexer (must be autogenerated).
%{stags format = \(dqvar @@: usize = none;\(dq; %}
%{
re2c:yyfill:enable = 0;
re2c:captvars = 1;
num = [0\-9]+;
(num) \(dq.\(dq (num) (\(dq.\(dq num)? [\ex00] {
return SemVer {
.major = s2n(yyinput[yytl1..yytr1]),
.minor = s2n(yyinput[yytl2..yytr2]),
.patch = if (yytl3 == none) 0 else s2n(yyinput[yytl3 + 1..yytr3])
};
}
* { return null; }
%}
}
test {
try std.testing.expectEqual(parse(\(dq23.34\(dq), SemVer{.major = 23, .minor = 34, .patch = 0});
try std.testing.expectEqual(parse(\(dq1.2.99999\(dq), SemVer{.major = 1, .minor = 2, .patch = 99999});
try std.testing.expectEqual(parse(\(dq1.a\(dq), null);
}
.ft P
.fi
.UNINDENT
.UNINDENT
.sp
Here is an example of using m\-tags to parse a version with a variable number of
components. Tag variables are stored in a trie.
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT
const std = @import(\(dqstd\(dq);
const none = std.math.maxInt(usize);
const mtag_root = none \- 1;
const err = error.SyntaxError;
// 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).
const MtagElem = struct {
elem: usize, // tag value
pred: usize, // index of the predecessor node or root
};
// Append a single value to an m\-tag history.
fn add_mtag(trie: *std.ArrayList(MtagElem), mtag: usize, value: usize) !usize {
try trie.append(MtagElem{.elem = value, .pred = mtag});
return trie.items.len \- 1;
}
// Recursively unwind tag histories and collect version components.
fn unwind(trie: *std.ArrayList(MtagElem),
x: usize,
y: usize,
str: []const u8,
ver: *std.ArrayList(u32)) !void {
// Reached the root of the m\-tag tree, stop recursion.
if (x == mtag_root and y == mtag_root) return;
// Unwind history further.
try unwind(trie, trie.items[x].pred, trie.items[y].pred, str, ver);
// Get tag values. Tag histories must have equal length.
std.debug.assert(x != mtag_root and y != mtag_root);
const ex = trie.items[x].elem;
const ey = trie.items[y].elem;
if (ex != none and ey != none) {
// Both tags are valid string indices, extract component.
try ver.append(s2n(str[ex..ey]));
} else {
// Both tags are none (this corresponds to zero repetitions).
std.debug.assert(ex == none and ey == none);
}
}
fn s2n(str: []const u8) u32 { // convert a pre\-parsed string to a number
var n: u32 = 0;
for (str) |c| { n = n * 10 + (c \- 48); }
return n;
}
fn parse(yyinput: [:0]const u8) !std.ArrayList(u32) {
var yycursor: usize = 0;
var yymarker: usize = 0;
var mt = std.ArrayList(MtagElem).init(std.testing.allocator);
defer mt.deinit();
// Final tag variables available in semantic action.
%{svars format = \(dqvar @@: usize = none;\(dq; %}
%{mvars format = \(dqvar @@: usize = mtag_root;\(dq; %}
// Intermediate tag variables used by the lexer (must be autogenerated).
%{stags format = \(dqvar @@: usize = none;\(dq; %}
%{mtags format = \(dqvar @@: usize = mtag_root;\(dq; %}
%{
re2c:YYMTAGP = \(dq@@ = add_mtag(&mt, @@, yycursor) catch none;\(dq;
re2c:YYMTAGN = \(dq@@ = add_mtag(&mt, @@, none) catch none;\(dq;
re2c:yyfill:enable = 0;
re2c:tags = 1;
num = [0\-9]+;
@t1 num @t2 (\(dq.\(dq #t3 num #t4)* [\ex00] {
var ver = std.ArrayList(u32).init(std.testing.allocator);
try ver.append(s2n(yyinput[t1..t2]));
try unwind(&mt, t3, t4, yyinput, &ver);
return ver;
}
* { return error.SyntaxError; }
%}
}
test {
var result = try parse(\(dq1\(dq);
var expect = std.ArrayList(u32).init(std.testing.allocator);
try expect.appendSlice(&[_]u32{1});
try std.testing.expectEqualDeep(result, expect);
expect.deinit();
result.deinit();
}
test {
var result = try parse(\(dq1.2.3.4.5.6.7\(dq);
var expect = std.ArrayList(u32).init(std.testing.allocator);
try expect.appendSlice(&[_]u32{1, 2, 3, 4, 5, 6, 7});
try std.testing.expectEqualDeep(result, expect);
expect.deinit();
result.deinit();
}
test {
const result = parse(\(dq1.2.\(dq) catch null;
try std.testing.expectEqualDeep(result, null);
}
.ft P
.fi
.UNINDENT
.UNINDENT
.SS Encoding support
.sp
It is necessary to understand the difference between \fBcode points\fP and
\fBcode units\fP\&. 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.
Note that the \(dqany\(dq rule \fB[^]\fP 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 \fB*\fP).
The generated lexer works with a stream of code units: \fByych\fP stores a code
unit, and \fBYYCTYPE\fP is the code unit type. Regular expressions, on the other
hand, are specified in terms of code points. When re2zig 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:
.INDENT 0.0
.IP \(bu 2
\fBASCII\fP (enabled by default). It is a fixed\-length encoding with code space
\fB[0\-255]\fP and 1\-byte code points and code units.
.IP \(bu 2
\fBEBCDIC\fP (enabled with \fB\-\-ebcdic\fP or \fBre2c:encoding:ebcdic\fP). It is a
fixed\-length encoding with code space \fB[0\-255]\fP and 1\-byte code points and
code units.
.IP \(bu 2
\fBUCS2\fP (enabled with \fB\-\-ucs2\fP or \fBre2c:encoding:ucs2\fP). It is a
fixed\-length encoding with code space \fB[0\-0xFFFF]\fP and 2\-byte code points
and code units.
.IP \(bu 2
\fBUTF8\fP (enabled with \fB\-\-utf8\fP or \fBre2c:encoding:utf8\fP). It is a
variable\-length Unicode encoding. Code unit size is 1 byte. Code points are
represented with 1 \-\- 4 code units.
.IP \(bu 2
\fBUTF16\fP (enabled with \fB\-\-utf16\fP or \fBre2c:encoding:utf16\fP). It is a
variable\-length Unicode encoding. Code unit size is 2 bytes. Code points are
represented with 1 \-\- 2 code units.
.IP \(bu 2
\fBUTF32\fP (enabled with \fB\-\-utf32\fP or \fBre2c:encoding:utf32\fP). It is a
fixed\-length Unicode encoding with code space \fB[0\-0x10FFFF]\fP and 4\-byte code
points and code units.
.UNINDENT
.sp
Include file \fBinclude/unicode_categories.re\fP provides re2zig definitions for the
standard Unicode categories.
.sp
Option \fB\-\-input\-encoding\fP specifies source file encoding, which can be used to
enable Unicode literals in regular expressions. For example
\fB\-\-input\-encoding utf8\fP tells re2zig that the source file is in UTF8 (it differs
from \fB\-\-utf8\fP which sets input text encoding). Option \fB\-\-encoding\-policy\fP
specifies the way re2zig handles Unicode surrogates (code points in range
\fB[0xD800\-0xDFFF]\fP).
.sp
Below is an example of a lexer for UTF8 encoded Unicode identifiers.
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT \-\-utf8
const std = @import(\(dqstd\(dq);
%{include \(dqunicode_categories.re\(dq %}
fn lex(yyinput: [:0]const u8) bool {
var yycursor: u32 = 0;
var yymarker: u32 = 0;
%{
re2c:yyfill:enable = 0;
// Simplified \(dqUnicode Identifier and Pattern Syntax\(dq
// (see https://unicode.org/reports/tr31)
id_start = L | Nl | [$_];
id_continue = id_start | Mn | Mc | Nd | Pc | [\eu200D\eu05F3];
identifier = id_start id_continue*;
identifier { return true; }
* { return false; }
%}
}
test {
try std.testing.expect(lex(\(dq_Ыдентификатор\(dq));
}
.ft P
.fi
.UNINDENT
.UNINDENT
.SS Include files
.sp
re2zig allows one to include other files using a block of the form
\fB/*!include:re2c FILE */\fP or \fB%{include FILE %}\fP, or an in\-block directive
\fB!include FILE ;\fP, where \fBFILE\fP is a path to the file to be included.
re2zig looks for include files in the directory of the including file and in
include locations, which can be specified with the \fB\-I\fP option. Include
blocks/directives in re2zig work in the same way as C/C++ \fB#include\fP: \fBFILE\fP
contents are copy\-pasted verbatim in place of the block/directive. Include files
may have further includes of their own. Use \fB\-\-depfile\fP option to track build
dependencies of the output file on include files.
re2zig provides some predefined include files that can be found in the
\fBinclude/\fP 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 re2zig\&. Below is an example of using include files.
.SS Include file 1 (definitions.zig)
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
const Num = enum { integer, floating, nan };
%{
number = [1\-9][0\-9]*;
%}
.ft P
.fi
.UNINDENT
.UNINDENT
.SS Include file 2 (extra_rules.re.inc)
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// floating\-point numbers
frac = [0\-9]* \(dq.\(dq [0\-9]+ | [0\-9]+ \(dq.\(dq;
exp = \(aqe\(aq [+\-]? [0\-9]+;
float = frac exp? | [0\-9]+ exp;
float { return Num.floating; }
.ft P
.fi
.UNINDENT
.UNINDENT
.SS Input file
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT
const std = @import(\(dqstd\(dq);
%{include \(dqdefinitions.zig\(dq %}
fn lex(yyinput: [:0]const u8) Num {
var yycursor: u32 = 0;
var yymarker: u32 = 0;
%{
re2c:yyfill:enable = 0;
* { return Num.nan; }
number { return Num.integer; }
!include \(dqextra_rules.re.inc\(dq;
%}
}
test {
try std.testing.expectEqual(lex(\(dq123\(dq), Num.integer);
try std.testing.expectEqual(lex(\(dq123.4567\(dq), Num.floating);
}
.ft P
.fi
.UNINDENT
.UNINDENT
.SS Header files
.sp
re2zig allows one to generate header file from the input \fB\&.re\fP file using
\fB\-\-header\fP option or \fBre2c:header\fP configuration and block pairs of the form
\fB/*!header:re2c:on*/\fP and \fB/*!header:re2c:off*/\fP, or \fB%{header:on%}\fP and
\fB%{header:off%}\fP\&. The first block marks the beginning of header file, and the
second block marks the end of it. Everything between these blocks is processed by
re2zig, and the generated code is written to the file specified with \fB\-\-header\fP
option or \fBre2c:header\fP configuration (or \fBstdout\fP if neither option nor
configuration is used). Autogenerated header file may be needed in cases when
re2zig is used to generate definitions that must be visible from other
translation units.
.sp
Here is an example of generating a header file that contains definition of the
lexer state with tag variables (the number variables depends on the regular
grammar and is unknown to the programmer).
.SS Input file
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// re2zig $INPUT \-o $OUTPUT \-\-header lexer/state.zig
const std = @import(\(dqstd\(dq);
const state = @import(\(dqlexer/state.zig\(dq); // the module is generated by re2c
%{header:on %}
pub const State = struct {
yyinput: [:0]const u8,
yycursor: usize,
%{stags format = \(dq@@: usize,\(dq; %}
};
%{header:off %}
fn lex(yyrecord: *state.State) usize {
var t: usize = 0;
%{
re2c:header = \(dqlexer/state.zig\(dq;
re2c:api = record;
re2c:yyfill:enable = 0;
re2c:tags = 1;
[a]* @t [b]* { return t; }
%}
}
test {
var st = state.State {
.yyinput = \(dqab\(dq,
.yycursor = 0,
%{stags format = \(dq.@@ = 0,\(dq; %}
};
try std.testing.expectEqual(lex(&st), 1);
}
.ft P
.fi
.UNINDENT
.UNINDENT
.SS Header file
.INDENT 0.0
.INDENT 3.5
.sp
.nf
.ft C
// Generated by re2c
pub const State = struct {
yyinput: [:0]const u8,
yycursor: usize,
yyt1: usize,
};
.ft P
.fi
.UNINDENT
.UNINDENT
.SS Skeleton programs
.sp
With the \fB\-S, \-\-skeleton\fP option, re2zig ignores all non\-re2zig 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) re2zig generates a standalone lexer and two files: an
\fB\&.input\fP file with strings derived from the DFA and a \fB\&.keys\fP file with
expected match results. The program runs each lexer on the corresponding
\fB\&.input\fP file and compares results with the expectations.
Skeleton programs are very useful for a number of reasons:
.INDENT 0.0
.IP \(bu 2
They can check correctness of various re2zig optimizations (the data is
generated early in the process, before any DFA transformations have taken
place).
.IP \(bu 2
Generating a set of input data with good coverage may be useful for both
testing and benchmarking.
.IP \(bu 2
Generating self\-contained executable programs allows one to get minimized test
cases (the original code may be large or have a lot of dependencies).
.UNINDENT
.sp
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). re2zig 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) skeleton is just an exact copy
of the original DFA. For encodings with multibyte code units skeleton is a copy
of DFA with certain transitions omitted: namely, re2zig 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) re2zig
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 (re2zig writes data to file as soon as it is
generated).
.SS Visualization and debug
.sp
With the \fB\-D, \-\-emit\-dot\fP option, re2zig 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 \fB\-\-dump\-nfa\fP, \fB\-\-dump\-dfa\-raw\fP etc. (see the full list of options).
.SH SEE ALSO
.sp
You can find more information about re2c at the official website: \fI\%http://re2c.org\fP\&.
Similar programs are flex(1), lex(1), quex(\fI\%http://quex.sourceforge.net\fP).
.SH AUTHORS
.sp
re2zig was originally written by Peter Bumbulis (\fI\%peter@csg.uwaterloo.ca\fP) 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 (\fI\%bayoung@acm.org\fP),
\fI\%Marcus Boerger\fP,
Dan Nuffer (\fI\%nuffer@users.sourceforge.net\fP),
\fI\%Ulya Trofimovich\fP (\fI\%skvadrik@gmail.com\fP),
\fI\%Serghei Iakovlev\fP,
\fI\%Sergei Trofimovich\fP,
\fI\%Petr Skocik\fP,
\fI\%ligfx\fP
\fI\%raekye\fP
and \fI\%PolarGoose\fP\&.
.\" Generated by docutils manpage writer.
.