MarpaX::ESLIF::BNF - MarpaX::ESLIF's BNF
version 6.0.35.1
MarpaX::ESLIF is a Scanless Interface expressed in a BNF format, that is using marpaWrapper, itself being a thin interface on top of libmarpa parser.
The MarpaX::ESLIF BNF is composed of unicode characters, in any encoding supported by the underlying convertor (ICU or iconv, in order of preference). Unsignificant whitespaces, Perl-like comments and C++-like comments are discarded.
They consist of bare names, or can be enclosed in angle brackets if whitespace if desired. They are case sensitive, and can be composed only of ASCII characters. There is no attempt to discard any leading, trailing, or repeated whitespace in angle brackets version, i.e. all the followings are different symbol names:
bare names
this <this > < this >
The grammar can contain multiple levels, the level syntax being:
::= # Alias for level 0 ~ # Alias for level 1 :\[[\d]+\]:= # General form
The level 0 must exist. We will use only ::= and/or ~ in the rest of this document for conveniene, though we are adressing any possible level.
::=
~
A rule consist of a left symbol, followed by the level, followed by one or more symbols, or a single quantified symbol:
leftsymbol1 ::= # Nothing leftsymbol2 ::= rightSymbol leftsymbol3 ::= quantifiedSymbol*
There are four types of explicit terminals.
Pseudo terminals
A zero-length terminal that matches only at the of the stream.
A zero-length terminal that matches only before a newline.
A zero-length terminal that matches only after a newline.
A zero-length terminal that always match the empty string.
Note that the newline is hardcoded to be any unicode newline, i.e. (*BSR_UNICODE)\R in PCRE2 terminology.
(*BSR_UNICODE)\R
The presence of :eol or :sol pseudo-terminal anywhere in the grammar enforces the newlineb flag in recognizers. On the contrary :eof and :eol work regardless of recognizer's newlineb flag.
newlineb
:eof
:eol
Strings
They can be single ('), double-quoted ("), or LEFT DOUBLE QUOTATION MARK (“) then RIGHT DOUBLE QUOTATION MARK (”) enclosed. The content is any valid unicode character, and the \ character can be used to escape the expected right-side quote character (i.e. ', ", or ”) or \ itself. The i modifier can be used to force case-insensitive match:
'
"
“
”
\
i
'string' 'string':i 'string\'s' "string\"s\\" “str'"\”ing”:i
and the modifier c to force unicode character mode. Internally a string is nothing else but a regular expression, so the exact implementation of the i and c modifiers correspond to the PCRE2 flags listed below in the Regular expression section. A string created with no modifier is faster when doing comparison (internally MarpaX::ESLIF will switch to a brutal memory comparison instead of calling the PCRE2 engine). Please note this is really a quoted string, not a string terminal. I.e. everything inside the quote is taken as-is, with no interpretation.
c
Character classes
They are always enclosed with left and right brackets []. Modifiers can start after a : character. A character class class is nothing else but a lexically restricted regular expression.
[]
:
Regular expression
They are always enclosed within slashes //, and the content must be valid as per the PCRE2 Perl Compatible Regular Expression library. Modifiers can start after the slash on the right. Regular expression patterns are by default anchored. The slash character itself must be preceeded by a backslash, i.e. \/ in the string seen by the parser (so, in practice, it is coded like this: "\\/").
//
\/
"\\/"
Regular expressions must be used with care in the two following scenarios:
If the regular expression ends with an unlimited quantifier at the end, i.e. * or +, it is very likely that the data will match partially until the whole input is read, effectively forcing ESLIF to read the entire input. This can break the streaming nature of your implementation.
*
+
If the regular expression ends with a negative lookahead, it can match when you think it should not. This is because negative lookahead does not trigger a partial match. In such a case, you should ensure that your regular expression forces a minimum number of characters in the subject string.
The PCRE2 syntax is supported in its entirety, this include any PCRE2 add-on. Character classes and regular expression share the same set of modifiers, executed in order of appearance, that are:
---------------------------------------------------------------- Modifiers Explanation ---------------------------------------------------------------- e Unset back-references in the pattern will match to empty strings i Case-insensitive j \u, \U and \x and unset back-references will act as JavaScript standard m Multi-line regex n Enable Unicode properties and extend meaning of meta-characters s A dot meta-character in the pattern matches all characters, including newlines x Enable comments. This has some limitation due MarpaX::ESLIF semantics D A dollar meta-character matches only at the end of the subject string J Allow duplicate names for sub-patterns U Inverts the "greediness" of the quantifiers a Meta-characters will be limited to their ASCII equivalent u Forces support of large codepoints b Could mean "forced binary" mode c Could mean "forced unicode character" mode A Remove the systematic anchoring ----------------------------------------------------------------
Internally this correspond to this set of options in PCRE2:
---------------------------------------------------------------- Modifiers PCRE2 flag unset PCR2 flag set ---------------------------------------------------------------- e PCRE2_MATCH_UNSET_BACKREF i PCRE2_CASELESS j PCRE2_ALT_BSUX|PCRE2_MATCH_UNSET_BACKREF m PCRE2_MULTILINE n PCRE2_UCP s PCRE2_DOTALL x PCRE2_EXTENDED D PCRE2_DOLLAR_ENDONLY J PCRE2_DUPNAMES U PCRE2_UNGREEDY a PCRE2_UTF N PCRE2_UCP u PCRE2_UTF b PCRE2_UTF PCRE2_NEVER_UTF c PCRE2_NEVER_UTF PCRE2_UTF A PCRE2_ANCHORED ----------------------------------------------------------------
Substitution modifiers are:
---------------------------------------------------------------- Modifiers Explanation ---------------------------------------------------------------- x Extended substitution pattern g Global substitution l Literal substitution ! Support of unknown set f Support of empty set ----------------------------------------------------------------
---------------------------------------------------------------- Modifiers PCRE2 flag unset PCR2 flag set ---------------------------------------------------------------- x PCRE2_SUBSTITUTE_EXTENDED g PCRE2_SUBSTITUTE_GLOBAL l PCRE2_SUBSTITUTE_LITERAL ! PCRE2_SUBSTITUTE_UNKNOWN_UNSET f PCRE2_SUBSTITUTE_UNSET_EMPTY ----------------------------------------------------------------
Lexemes are meta-symbols that does appear as the LHS symbol anywhere within the current grammar. Therefore they behave like terminals, except that their definition is not in the current grammar. By default such meta-symbol is looked up at the next level. For example:
rule ::= something something ~ [\d]
say that symbol something at grammar level 0 is a reference to something at grammar level 1. It is important to notice that:
something
0
1
A lexeme match respects the :discard rules at its corresponding grammar level. Nevertheless no event can happen, except the internal :discard[switch], :discard[on] and :discard[off] events.
:discard
:discard[switch]
:discard[on]
:discard[off]
A lexeme match is successful only if the valuation starting at this symbol is successful and unambiguous
The result of a lexeme match is always of type MARPAESLIF_VALUE_TYPE_ARRAY pointing to raw bytes that matched, these raw bytes will include discarded data.
MARPAESLIF_VALUE_TYPE_ARRAY
In contrast a terminal is an explicit quoted string, character class or regular expression, e.g.:
the_rhs_is_a_terminal ::= '"'
Lexemes can be references:
Without any indication, a lexeme is always assumed to be at the grammar of the next level
X ::= Y@'Grammar Description'
This is working because a grammar description is unique across all sub-grammars. Note that when accessing a grammar by description, the later must have been declared before, using the :desc rule. Therefore a good practice is to forward declare all grammars at the beginning, e.g.:
:desc
:desc ::= 'Main Grammar' :desc ~ 'Sub Grammar 1' :desc :[2]:= 'Sub Grammar 2' :desc :[3]:= 'Sub Grammar 3'
X ::= Y@+1 X ::= Y@-2 X ::= Y@3
The signed integer is interpreted as a delta with current grammar level.
X ::= Y@=1
The unsigned integer is interpreted as an explicit grammar level.
Everytime expected terminals cannot be match, MarpaX::ESLIF will try to match the special rule :discard. The :discard rule also have precedence if it matches longer than the longest acceptable lexeme or terminal. and can not be ambiguous (else discard silently fail).
The important thing to remember is that parameterized rules parameters and expressions follow the lua semantics.
A parameterized LHS must be writen in the form
LHS<-(lua optional parameter list)
or
LHS<--(lua optional parameter list)
A parameterized RHS must be writen in the form
RHS->(lua optional expression list)
RHS-->(lua optional expression list)
The -- indicates if the internal processing of context passing is a precompiled lua byte chunk or not. Most of the times you want to use the -- notation, as if it you would have been writen like a lua library that is loaded once, this is also the fastest mode. Otherwise the context passing is recompiled everytime, allowing to have runtime dependencies, sometimes useful for debugging, sometimes necessary for your application.
--
The parameter and expression processings are done using an internal lua interpreter, embedded into marpaESLIF. The output of this internal interpreter is then injected into host's VM using the language bindings.
it is important to note that MarpaX::ESLIF will always consider a parameterized RHS like a lexeme, i.e. a grammar terminal bound to a sub-grammar. Therefore no interaction with the recognizer is possible until the lexeme sub-grammar is processed, with one exception: the dynamic lexeme, which is where the end-user may implement specific functionality with respect to parameters and expressions. This mean a parameterized rhs can also be writen in the form of a generation action name, e.g.:
/* Call host's recognizer action "external_action" */ LHS<-(x,y) ::= . => external_action->(x+1, y+2) /* Embedded anonymous lua function */ LHS<-(x,y) ::= . => ::lua->function(x, y) return "read below!" end -> (x+1, y+2) /* Embedded explicit lua function name "lua_action" writen inside your grammar */ LHS<-(x,y) ::= . => ::lua->lua_action->(x+1, y+2)
<luascript> function lua_action(x,y) return " </luascript>
In any of these cases, the call must be successful a return a string that is compatible with an RHS, that ESLIF execute as a lexeme. ESLIF will be try to clever to not recompile grammars everytime, though it will use a very basic check: the whole UTF-8 version of the generated RHS, together with an internally generated hook on grammar level, will be kept in memory, and only grammars with a different UTF-8 version at different grammar levels will be recompiled.
End-user must take care, because infinite recursivity is very easy with parameterized rules, and ESLIF is not protected against that.
A concrete lua example is e.g.:
local recognizerInterface = { ["init"] = function(self, input) self._input = input self._nbParameterizedRhsCalls = 0 end, ["read"] = function(self) self._data = self._input return true end, ["isEof"] = function(self) return true end, ["isCharacterStream"] = function(self) return true end, ["encoding"] = function(self) return nil end, ["data"] = function(self) return self._data end, ["isWithDisableThreshold"] = function(self) return false end, ["isWithExhaustion"] = function(self) return false end, ["isWithNewline"] = function(self) return true end, ["isWithTrack"] = function(self) return false end, ["parameterizedRhs"] = function(self, parameter) self._nbParameterizedRhsCalls = self._nbParameterizedRhsCalls + 1 local output if (self._nbParameterizedRhsCalls == 5) then output = "'5'" elseif (self._nbParameterizedRhsCalls > 5) then output = "'no match'" else parameter = parameter + 1 output = ". => parameterizedRhs->("..parameter..")" end return output end } local valueInterface = { ["isWithHighRankOnly"] = function(self) return true end, ["isWithOrderByRank"] = function(self) return true end, ["isWithAmbiguous"] = function(self) return false end, ["isWithNull"] = function(self) return false end, ["maxParses"] = function(self) return 0 end, ["getResult"] = function(self) return self._result end, ["setResult"] = function(self, result) self._result = result end, } local logger = { ["trace"] = function(self, msgs) self:tracef("%s", msgs) end, ["debug"] = function(self, msgs) self:debugf("%s", msgs) end, ["info"] = function(self, msgs) self:infof("%s", msgs) end, ["notice"] = function(self, msgs) self:noticef("%s", msgs) end, ["warning"] = function(self, msgs) self:warningf("%s", msgs) end, ["error"] = function(self, msgs) self:errorf("%s", msgs) end, ["critical"] = function(self, msgs) self:criticalf("%s", msgs) end, ["emergency"] = function(self, msgs) self:emergencyf("%s", msgs) end, -- -- Used by us -- ["tracef"] = function(self, fmts, ...) print(string.format("%-9s "..fmts, 'TRACE', ...)) end, ["debugf"] = function(self, fmts, ...) print(string.format("%-9s "..fmts, 'DEBUG', ...)) end, ["infof"] = function(self, fmts, ...) print(string.format("%-9s "..fmts, 'INFO', ...)) end, ["noticef"] = function(self, fmts, ...) print(string.format("%-9s "..fmts, 'NOTICE', ...)) end, ["warningf"] = function(self, fmts, ...) print(string.format("%-9s "..fmts, 'WARN', ...)) end, ["errorf"] = function(self, fmts, ...) print(string.format("%-9s "..fmts, 'ERROR', ...)) end, ["criticalf"] = function(self, fmts, ...) print(string.format("%-9s "..fmts, 'CRITICAL', ...)) end, ["emergencyf"] = function(self, fmts, ...) print(string.format("%-9s "..fmts, 'EMERGENCY', ...)) end } local marpaESLIFLua = require 'marpaESLIFLua' local marpaESLIFp = marpaESLIFLua.MarpaX::ESLIF::new(logger) marpaESLIFGrammarp = marpaESLIFp:marpaESLIFGrammar_new([[ :default ::= action => ::shift top ::= . => parameterizedRhs->(1) | . => parameterizedRhs->(2) | . => parameterizedRhs->(3) | . => parameterizedRhs->(4) | . => ::luac->function(x) return "'will not match'" end ->(5) | . => ::luac->function(x) print('Called with x='..x) return "'will not match'" end ->(15) | . => ::lua->grammar_ok->(10,12) | . => ::lua->grammar_ko->(10,12) | . => ::lua->action_raising_error->(10,12) | . => ::lua->unknown_action->(10,12) <luascript> function grammar_ok(x,y) return "'Y'" end function grammar_ko(x,y) return "Y" end function action_raising_error(x,y) error('Errors are trapped...') end </luascript> ]]) recognizerInterface:init('5') marpaESLIFGrammarp:parse(recognizerInterface, valueInterface) logger:noticef('... Grammar parse result: %s', valueInterface:getResult())
Output is:
Called with x=15 ERROR Looking at rules in grammar level 0 (Grammar level 0): symbol 1 (Y) must be resolved as <Y> in grammar at level 0 or 1 ERROR grammar_ko callback returned a string that cannot be converted to a grammar ERROR <luascript/>:9: Errors are trapped... ERROR action_raising_error callback failed ERROR <luascript/>:9: No such function unknown_action ERROR unknown_action callback failed NOTICE ... Grammar parse result: 5
Any RHS surrounded by (?=...) or (?!...) is interpreted to a positive or negative lookahead, respectively. Internally, it is processed like a lexeme, and will result in a zero-length lexeme in case of match. You can skip it in the rule valuation by surrounding again with (-...-), e.g.:
(?=...)
(?!...)
(-...-)
lhs ::= (-(?= /anything/ )-)
will do a positive lookahead of regular expression /anything/ and skip it in the rule's valuation.
/anything/
By default, the first symbol of a grammar of level n is its start symbol. This can be set once with e.g.:
n
:start ::= symbolname
By default, a grammar of level n has the description Grammar level n. This can be set once with e.g.:
Grammar level n
:desc ::= 'A single-quoted string'
Any setting consist of a reserved keyword, followed by => or the Rightwards Double Arrow UTF-8 character ⇒, followed by a setting-specific value:
=>
⇒
action => myAction action ⇒ myAction
When you use the ⇒ UTF-8 character, it is recommended to say that the grammar itself is encoded in UTF-8 (c.f. marpaESLIFGrammarOption structure).
marpaESLIFGrammarOption
By default, symbol action is ::transfer and rule action is ::concat, i.e. the parse tree value of a grammar is a binary concatenation of every input representation (see the representation section below), without the eventual discard. Stack manipulation may require the trigger of a free function, and this has no default. Only expected terminals or lexemes are looked up, this is the Longest Acceptable Token Match (LATM) setting, defaulting to a true value. You should not change that. Defaults can be set once, for example like this:
::transfer
::concat
:default ::= action => defaultRuleAction latm => 1 discard-is-fallback => 0 symbol-action => defaultSymbolAction regex-action => defaultRegexAction default-encoding => UTF-8 fallback-encoding => UTF-16
Predefined actions are available for rules and symbols. Please refer the API documentation to know more about value types.
The symbol-action adverb is how a match within a sub-grammar is transfered.
symbol-action
The default-encoding adverb gives default encoding when the recognizer runs in character mode and the end-user gave no encoding.
default-encoding
The fallback-encoding adverb gives fallback encoding when the recognizer runs in character mode and the end-user gave no encoding nor default-encoding. Then MarpaX::ESLIF will try to guess the encoding, and if this fail, will fallback to this setting. This is the desired setting when you grammar accept input in different encoding, and defaults to a given encoding if the guess fail. Please note that, when guessing an encoding, MarpaX::ESLIF takes into account an eventual BOM, retreiving the later from input characters if it exists.
fallback-encoding
The discard-is-fallback adverb indicates ESLIF that :discard should not be tried whenever scan/resume is entered. The default is the safe implementation, i.e. it is always tried, and no grammar terminal will match if :discard matches an equal number of bytes or longer. Nevertheless, if you are sure that the grammar's :discard cannot conflict with any other grammar terminal, then :discard, if successful, will always hit as a fallback alternative. The discard-is-fallback enforces this mode, providing a safe performance improvement for grammars that has no conflict between :discard and any other grammar terminal.
discard-is-fallback
::undef
Creates a value of type UNDEF.
Meaningful for both rule and symbol actions.
::ascii
Creates a value of type STRING, with encoding "ASCII", from the right-hand side representation, guaranteed to be a NUL byte terminated sequence of ASCII characters, or UNDEF if representation is empty. Please refer to the representation section below.
::convert[[^]]+]
Creates a value of type STRING from the right-hand side representation encoded in the charset specified within the brackets, or UNDEF if representation is empty. iconv convention is used for the charset, i.e. a charset name, followed by eventual options like //TRANSLIT and/or //IGNORE. Any other option depend on how the tconv library is built, and may probably not be supported. Note that using the iconv notation does not mean that this is iconv running behind.
//TRANSLIT
//IGNORE
Please refer to the representation section below.
Creates a value of type ARRAY from the binary concatenatation all the RHS's representation, or UNDEF if representation is empty.
Meaningful for both rule and symbol actions, and is the default rule action.
::copy[x]
Copies the RHS number x (first RHS is at indice 0), putting UNDEF if it does not exist. This action is the only one that guarantees that the nature of the RHS value is unchanged.
x
Meaningful only for rule actions.
::shift
Alias for ::copy[0].
::copy[0]
Copies the single RHS number value. This action guarantees that the nature of the RHS value is unchanged.
Meaningful only for symbol actions, and is the default symbol action.
::true
Creates a value of type BOOL, containing a true value.
::false
Creates a value of type BOOL, containing a false value.
::json
Creates a value of type STRING in the UTF-8 encoding, containing a strict JSON string as per original JSON specification, i.e. it is using UTF-16 surrogates to describe characters above 0xFFFF.
::jsonf
Creates a value of type STRING in the UTF-8 encoding, containing a JSON string as per original JSON specification plus infinity and nan extensions. Without these extensions, infinity and nan are writen as "null".
::row
Creates a value of type ROW, that contains all RHS's values.
::table
Creates a value of type TABLE, that contains all RHS's values. The number of RHS must be odd.
::ast
Creates a value of type TABLE, where the single key is a string containing the lhs name, and the single value is a row containing all RHS's values, or UNDEF if rule is nullable.
The :discard symbol, despite belonging to a given grammar, is not accessible directly, and can only be set as a meta setting. An event can be associated upon discard completion, there can be multiple :discard statements:
:discard ::= symbolname1 event => discard_symbolname1$ :discard ::= symbolname2 event => discard_symbolname2$
Note than when an event is set, this will be triggered only on the :discard's RHS completion, therefore the RHS of the :discard must be an LHS in the same grammar when there is an event setting and when :discard refers to a symbol.
An explicit terminal can also be set directly, e.g.:
:discard ::= /[\s]+/ event => whitespace$ :discard ::= /#[^\n]*/u event => perl_comment$ :discard ::= /\/\/[^\n]*|\/\*(?:[^\*]+|\*(?!\/))*\*\//u event => cplusplus_comment$
this form is highly recommended, because ESLIF recognizes the case where :discard consist only of explicit terminals, and applies an optimization that prevent the intanciation of an internal parse.
Event names
They are composed of a restricted set of the ASCII graph characters. Special cases are:
:symbol
Transformed to the symbol name for which the event is triggered.
Hook that is disabling :discard rule for the current recognizer. Equivalent to a call to MarpaX::ESLIF::recognizer_hook_discardb(1). Not propagated. Take care, this is a permanent setting.
MarpaX::ESLIF::recognizer_hook_discardb(1)
Hook that is enabling :discard rule for the current recognizer. Equivalent to a call to MarpaX::ESLIF::recognizer_hook_discardb(0). Not propagated. Take care, this is a permanent setting.
MarpaX::ESLIF::recognizer_hook_discardb(0)
Hook that is switching :discard rule for the current recognizer. Equivalent to a call to MarpaX::ESLIF::recognizer_hook_discard_switchb(). Not propagated. Take care, this is a permanent setting.
MarpaX::ESLIF::recognizer_hook_discard_switchb()
Please note that the :discard[on], :discard[off] and :discard[switch] events will always happen if specified in the grammar with an true initial state and if associated to lexemes or terminals. No callback to the end user will happen. This is because these are internal events, categorized as parsing hooks.
Event initializers
By default, events are on, this is equivalent to appending =on after the event name. The =off characters are putting event off at startup.
=on
=off
Lexemes are different than non-lexeme symbols because they are treated in the grammar as terminals, others are not.
Symbols that are lexemes or terminals (i.e. they are grammar terminals in the common terminology) can have pause events, before mean that the scanning recognized them, after mean they have been consumed, e.g.:
before
after
:symbol ::= symbolname pause => before event => ^symbolname :symbol ::= symbolname pause => after event => symbolname$
The grammar's symbol-action can be overwritten by setting such a :symbol adverb, e.g.:
:symbol ::= symbolname symbol-action => ::u8"Custom String" :symbol ::= symbolname symbol-action => Custom_action :symbol ::= 'ThisTerminal' symbol-action => ::lua->lua_custom_action
if-action
Itapplies only when the symbol is discovered via the automatic scan, or if it is an external symbol: The if-action must refer to a callback in user-land's recognizer interface or in embedded lua interpreter, and must return a boolean:
:symbol ::= symbolname if-action => Custom_action1 :symbol ::= symbolname if-action => ::lua->Custom_action2
Please note that the if-action will always be called with an argument of type ARRAY (i.e. up to the implementation to convert that to a multibyte string, if any).
regex-action
A regex callout is in the form (?CX) (where X is a number or a double-quoted string). The grammar's regex-action is called with a marpaESLIFCalloutBlockp argument that is of type MARPAESLIF_VALUE_TYPE_TABLE, i.e. a hash in perl, a table in lua, and an ESLIFRegexCallout instance in java.
(?CX)
X
MARPAESLIF_VALUE_TYPE_TABLE
For any other symbol that is not a lexeme nor a terminal, completion, predicted or nulled events are supported, targetting a meta symbol name. These are grammar events.
For example:
event symbolname$ = completed symbolname event symbolname[]=off = nulled symbolname event ^symbolname=on = predicted symbolname
The grammar, in :default meta rule, can have an event action callback that will end into user-land's recognizer interface or in the embedded lua interpreter, e.g.
:default
:default event-action => Custom_action1
:default event-action => ::lua->Custom_action2 <luascript> -- In the embedded lua, globals are automatically -- created. During parsing phase, they are: -- marpaESLIF -- marpaESLIFGrammar -- marpaESLIFRecognizer function Custom_action2(events) -- "events" is a table, exactly like the output -- of marpaESLIFRecognizer.events() -- This function must return a boolean return true end </luascript>
The event callback must return a boolean, a false value indicates that the parsing failed.
Rules can be autoranked, the higest of a set of alternative having the highest rank, default is off:
autorank is on by default autorank is off by default
Inaccessible statements can generate warnings, can be ignored, or be error on demand, default is to ignore them:
inaccessible is warn by default inaccessible is ok by default inaccessible is fatal by default
A statement have a symbol name on the left-hand side (LHS) and zero or more symbol names, or terminals, on the right-hand side (RHS):
LHS ::= RHS1 RHS2 etc...
There are two exceptions:
Its semantic is a single symbol name following by another single symbol name, with - in the middle:
-
LHS ::= RHS1 - RHS2
Constraints are:
RHS1
RHS2
This mean that the LHS of an exception statement can never be nullable. You must explicitely declare so if this is wanted, i.e.:
LHS
LHS ::=
To understand how ESLIF proceeds, everytime the left-side of an exception reaches the start completion, the left-side match is checked against the right-side of the exception. The longest match that does not match the exception is the winner.
Simple case is when a exception applies to a full match, e.g. the PITarget of XML language:
<PITarget> ::= <NAME> - 'xml':i <NAME> ~ /[:A-Z_a-z\x{C0}-\x{D6}\x{D8}-\x{F6}\x{F8}-\x{2FF}\x{370}-\x{37D}\x{37F}-\x{1FFF}\x{200C}-\x{200D}\x{2070}-\x{218F}\x{2C00}-\x{2FEF}\x{3001}-\x{D7FF}\x{F900}-\x{FDCF}\x{FDF0}-\x{FFFD}\x{10000}-\x{EFFFF}][:A-Z_a-z\x{C0}-\x{D6}\x{D8}-\x{F6}\x{F8}-\x{2FF}\x{370}-\x{37D}\x{37F}-\x{1FFF}\x{200C}-\x{200D}\x{2070}-\x{218F}\x{2C00}-\x{2FEF}\x{3001}-\x{D7FF}\x{F900}-\x{FDCF}\x{FDF0}-\x{FFFD}\x{10000}-\x{EFFFF}\-.0-9\x{B7}\x{0300}-\x{036F}\x{203F}-\x{2040}]*/u
Or any number but '23':
number ::= /\\d+/ - '23'
More complex is when the exception is floating, e.g. the CData of XML language:
CData ::= CHARDATA - CHARDATAEXCEPTION CData ::= # Using CHARDATA removed CData nullable aspect CHARDATA ~ [\x{9}\x{A}\x{D}\x{20}-\x{D7FF}\x{E000}-\x{FFFD}\x{10000}-\x{10FFFF}]:u* CHARDATAEXCEPTION ~ CHARDATA ']]>' CHARDATA
In the above example note that CHARDATA is writen as a sequence rule, allowing ESLIF to detect every CHARDATA completion, and match it against CHARDATAEXCEPTION. Although very closed, the following would not work:
CData ::= CHARDATA - CHARDATAEXCEPTION CData ::= # Using CHARDATA removed CData nullable aspect CHARDATA ~ /[\x{9}\x{A}\x{D}\x{20}-\x{D7FF}\x{E000}-\x{FFFD}\x{10000}-\x{10FFFF}]*/u CHARDATAEXCEPTION ~ CHARDATA ']]>' CHARDATA
An exception can be consuming, eventually reading the whole data if it is writen without special care. Trying to minimize the number of characters needed and/or using well-thinked regular expressions often lead to the same result with better performance.
Internally, the longest RHS1 is first matched, then ESLIF rollbacks to every previous RHS1's completion as long as RHS2 matches.
This is a single symbol name following by the * or the + character:
LHS1 ::= RHS1* LHS2 ::= RHS2+
Empty rule have no RHS:
EMPTYRULE ::=
Eventual ambiguities in the grammar itself may be solved by adding the ; character at the end of a rule, or by enclosing zero or more statements within { and } characters:
;
{
}
EMPTYRULE ::= ; { LHS1 ::= RHS1 LHS2 ::= RHS2 - RHS3 }
There are two types of alternatives: the standard | meaning this is an or, or the loosen character || meaning that this is an alternative starting a prioritized group of alternatives, for example the calculator grammar is:
|
||
Expression ::= /[\d]+/ | '(' Expression ')' assoc => group || Expression '**' Expression assoc => right || Expression '*' Expression | Expression '/' Expression || Expression '+' Expression | Expression '-' Expression
which is strictly equivalent, in traditional BNF syntax to:
Expression ::= Expression0 Expression0 ::= Expression1 Expression1 ::= Expression2 Expression2 ::= Expression3 Expression3 ::= /[\d]+/ | '(' Expression0 ')' Expression2 ::= Expression3 '**' Expression2 Expression1 ::= Expression1 '*' Expression2 | Expression1 '/' Expression2 Expression0 ::= Expression0 '+' Expression1 Expression0 ::= Expression0 '-' Expression1
As you can see statements has been grouped at every occurence of || operator. Therefore the loosen operator || is a convenience operator, it is always possible to write an equivalent grammar without it, though this can become quite tedious. The assoc adverb has a meaning only in the presence of prioritized alternatives, else it has no effect.
assoc
The following is copied almost verbatim from the Marpa::R2 section on precedence:
In prioritized statements, every alternative has an arity. The arity is the number of times an operand appears on the RHS. A RHS symbol is an operand if and only if it is the same as the LHS symbol. Anything else is considered as an operator. When the arity is 0, precedence and associativy are meaningless and ignored. When the arity is 1, precedence has effect, but not left nor right associativity.
If arity is 2 or more and the alternative is left associative, the leftmost operand associates and operands after the first will have the next-tighest priority level. If arity is 2 or more and the alternative is right associative, the last operand associates and operands before the last will have the next-tighest priority level. In group associativity, all operands associate at the lowest priority.
Any rule can be followed by zero or more of these adverbs, if an adverb appears more than once, the latest is the winner:
During valuation, a specific action can be associated to a rule:
action => my_action
It is possible to set a hardcoded UTF-8 string as result, using a string literal:
action => ::u8"string literal, supporting \x{0D}, \u{0972} and \U{0001F600}"
where \x{hh} will translate to a byte having the hexadecimal value hh, and \u{uuuu} and \U{uuuuuuuu} will translate to the UTF-8 version of uuuu unicode code point, uuuuuuuu is for the very large, less common, code points.
\x{hh}
hh
\u{uuuu}
\U{uuuuuuuu}
uuuu
uuuuuuuu
Actions can be be writen in plain lua, conforming to the lua version embedded by ESLIF, e:g:
action => action => ::lua->function(x,y) return x^y end action => action => ::luac->function(x,y) return x^y end
The lua semantics apply immediately after the first opening pathenthesis, up to the end of the function body:
action => action => ::lua->function(x,y) --[=[ Lua semantics here ]=] return x^y end
The workflow for lua actions is:
Content of grammar's luascript is always loaded first
Lua actions are loaded dynamcally and can have two forms:
::lua->function
The actions starting with ::lua->function are always loaded as string, i.e. compiled and then executed
::luac->function
The actions starting with ::luac->function are compiled once if needed, and the binary chunk is re-injected as is for execution.
Internally ESLIF creates a function that returns the user function, e.g. if in your grammar you have:
action => action => ::lua->function(x,y) return true end
then ESLIF will inject this:
return function(x,y) return true end
The top of the lua stack is then executed, leaving to user's wanted anonymous function.
In a prioritized statement, associate with the left-most operand:
assoc => left
In a prioritized statement, associate with the right-most operand:
assoc => right
All operands associate at the lowest priority:
assoc => group
Sequence rules can have a separator, that can be a symbol name, a string, a character class or a regular expression.
separator => comma separator => ',' separator => [,] separator => /,/
Modifiers are allowed after string, character class or regular expressions.
Sequence rules can be proper, i.e. without trailing separator:
proper => 1 proper => true proper => 0 proper => false
Default for sequence rules actions is to always include the separator in the stack. This may be changed in the grammar using:
hide-separator => 1 hide-separator => true hide-separator => 0 hide-separator => false
During valuation, rules can have a rank to get prioritized. Rank is a signed integer and default to 0:
rank => -2
Any other value but 0 is not allowed if autoranking is set to a true value.
Nulling symbols can rank high low, the default is low.
null-ranking => 'low' null-ranking => 'high'
Lexemes and terminals can be prioritized, using a signed integer:
priority => 15
Lexemes can be verbose when they fail to parse:
verbose => 1 verbose => true verbose => 0 verbose => false
If verbose is a true value, ESLIF will printout the sub-recognizer state and stream if the lexeme valuation fails.
verbose
Scanner can be paused before a lexeme or a terminal is recognized, or just after it has been completed:
pause => before pause => after
Events can be specified, with an eventual initializer, given that default initialization is =on:
event => eventName event => eventName=on event => eventName=off
A name can be associated to a rule, in the form:
LHS ::= etc... RHS etc... name => something LHS ::= etc... RHS etc... name => 'quoted string literal' # No modifier is allowed after the string LHS ::= etc... RHS etc... name => "quoted string literal" # No modifier is allowed after the string
and to terminals in the form:
:symbol ::= /X/ name => XRegexp
A terminal that has the name adverb can be accessed later in the grammar at the same level using the name preceded by the dollar sign $, e.g;:
$
LHS ::= etc... $XRegexp ...
Note that two terminals cannot share the same name adverb.
MarpaX::ESLIF BNF can be expressed in itself:
####################################################### :desc ::= 'G1' :start ::= <statements> :discard ::= /[\s]+/u :discard ::= /#[^\n]*/u :discard ::= /\/\/[^\n]*|\/\*(?:[^\*]+|\*(?!\/))*\*\//u :symbol ::= /::luac?\->function\(/ pause => after event => :discard[switch] :symbol ::= /\-\-?>\(/' pause => after event => :discard[switch] :symbol ::= <lua funcbody after lparen>@+2 pause => after event => :discard[switch] verbose => 1 :symbol ::= <lua args after lparen>@+2 pause => after event => :discard[switch] verbose => 1 event :discard[on] = nulled <discard on> event :discard[off] = nulled <discard off> <statements> ::= <statement>* <statement> ::= <start rule> | <desc rule> | <empty rule> | <null statement> | <statement group> | <priority rule> | <quantified rule> | <discard rule> | <default rule> | <lexeme rule> | <completion event declaration> | <nulled event declaration> | <prediction event declaration> | <inaccessible statement> | <exception statement> | <autorank statement> | <lua script statement> | <terminal rule> | <symbol rule> <start rule> ::= ':start' <op declare> <start symbol> <start symbol> ::= <symbol> | <start symbol> <lua functioncall> <desc rule> ::= ':desc' <op declare> <quoted string literal> <empty rule> ::= <lhs> <op declare> <adverb list> <null statement> ::= ';' <statement group> ::= '{' <statements> '}' <priority rule> ::= <lhs> <op declare> <priorities> <quantified rule> ::= <lhs> <op declare> <rhs primary> <quantifier> <adverb list> <discard rule> ::= ':discard' <op declare> <rhs primary> <adverb list> <default rule> ::= ':default' <op declare> <adverb list> <lexeme rule> ::= ':lexeme' <op declare> <rhs primary no parameter> <adverb list> /* Deprecated, use :symbol */ <terminal rule> ::= ':terminal' <op declare> <terminal> <adverb list> /* Deprecated, use :symbol */ <symbol rule> ::= ':symbol' <op declare> <rhs primary no parameter> <adverb list> <completion event declaration> ::= 'event' <event initialization> '=' 'completed' <lhs> | 'event' <event initialization> <op declare> 'completed' <lhs> <nulled event declaration> ::= 'event' <event initialization> '=' 'nulled' <lhs> | 'event' <event initialization> <op declare> 'nulled' <lhs> <prediction event declaration> ::= 'event' <event initialization> '=' 'predicted' <lhs> | 'event' <event initialization> <op declare> 'predicted' <lhs> <inaccessible statement> ::= 'inaccessible' 'is' <inaccessible treatment> 'by' 'default' <inaccessible treatment> ::= 'warn' | 'ok' | 'fatal' <exception statement> ::= <lhs> <op declare> <rhs primary> '-' <rhs primary> <adverb list> <autorank statement> ::= 'autorank' 'is' <on or off> 'by' 'default' <op declare> ::= <op declare top grammar> | <op declare lex grammar> | <op declare any grammar> <priorities> ::= <alternatives>+ separator => <op loosen> proper => 1 hide-separator => 1 <alternatives> ::= <alternative>+ separator => <op equal priority> proper => 1 hide-separator => 1 <alternative> ::= <rhs> <adverb list> <adverb list> ::= <adverb list items> <adverb list items> ::= <adverb item>* <adverb item> ::= <action> | <left association> | <right association> | <group association> | <separator specification> | <proper specification> | <rank specification> | <null ranking specification> | <priority specification> | <pause specification> | <latm specification> | <discard is fallback specification> | <naming> | <null adverb> | <symbol action> | <event specification> | <hide separator specification> | <if action> | <event action> | <default encoding> | <fallback encoding> | <regex action> | <verbose specification> <action> ::= 'action' /=>|\x{21D2}/u <action name> | 'action' /=>|\x{21D2}/u <string literal> | 'action' /=>|\x{21D2}/u <quoted string literal> <left association> ::= 'assoc' /=>|\x{21D2}/u 'left' <right association> ::= 'assoc' /=>|\x{21D2}/u 'right' <group association> ::= 'assoc' /=>|\x{21D2}/u 'group' <separator specification> ::= 'separator' /=>|\x{21D2}/u <rhs primary> <proper specification> ::= 'proper' /=>|\x{21D2}/u <false> | 'proper' /=>|\x{21D2}/u <true> <hide separator specification> ::= 'hide-separator' /=>|\x{21D2}/u <false> | 'hide-separator' /=>|\x{21D2}/u <true> <rank specification> ::= 'rank' /=>|\x{21D2}/u <signed integer> <null ranking specification> ::= 'null-ranking' /=>|\x{21D2}/u <null ranking constant> | 'null' 'rank' /=>|\x{21D2}/u <null ranking constant> <null ranking constant> ::= 'low' | 'high' <priority specification> ::= 'priority' /=>|\x{21D2}/u <signed integer> <pause specification> ::= 'pause' /=>|\x{21D2}/u 'before' | 'pause' /=>|\x{21D2}/u 'after' <event specification> ::= 'event' /=>|\x{21D2}/u <event initialization> <event initialization> ::= <event name> <event initializer> <verbose specification> ::= 'verbose' /=>|\x{21D2}/u <false> | 'verbose' /=>|\x{21D2}/u <true> <event initializer> ::= '=' <on or off> <on or off> ::= 'on' | 'off' <event initializer> ::= <latm specification> ::= 'latm' /=>|\x{21D2}/u <false> | 'latm' /=>|\x{21D2}/u <true> <discard is fallback specification> ::= 'discard-is-fallback' /=>|\x{21D2}/u <false> | 'discard-is-fallback' /=>|\x{21D2}/u <true> <naming> ::= 'name' /=>|\x{21D2}/u <alternative name> <null adverb> ::= ',' <symbol action> ::= 'symbol-action' /=>|\x{21D2}/u <symbol action name> | 'symbol-action' /=>|\x{21D2}/u <string literal> | 'symbol-action' /=>|\x{21D2}/u <quoted string literal> <if action> ::= 'if-action' /=>|\x{21D2}/u <if action name> <regex action> ::= 'regex-action' /=>|\x{21D2}/u <regex action name> <generator action> ::= '.' /=>|\x{21D2}/u <generator action name> <alternative name> ::= <standard name> | <quoted string literal> <event name> ::= <restricted ascii graph name> | ':symbol' | ':discard[on]' | ':discard[off]' | ':discard[switch]' <lhs> ::= <symbol name> | <lhs> <lua functiondecl> <rhs> ::= <rhs alternative>+ <rhs alternative> ::= <rhs primary> | '(-' <priorities> '-)' | '(' <priorities> ')' | '(-' <rhs primary> '-' <rhs primary> <adverb list> '-)' | '(' <rhs primary> '-' <rhs primary> <adverb list> ')' | '(-' <rhs primary> <quantifier> <adverb list> '-)' | '(' <rhs primary> <quantifier> <adverb list> ')' | /\(\?[=!]/ <priorities> ')' | /\(\?[=!]/ <rhs primary> '-' <rhs primary> <adverb list> ')' | /\(\?[=!]/ <rhs primary> <quantifier> <adverb list> ')' <rhs primary no parameter> ::= <single symbol> | <symbol> '@' <grammar reference> | '$' <alternative name> <rhs primary> ::= <rhs primary no parameter> | <rhs primary no parameter> <lua functioncall> | <generator action> <lua functioncall> <single symbol> ::= <symbol> | <terminal> <terminal> ::= <character class> | <regular expression> | <quoted string> | ':eof' | ':eol' | ':sol' | ':empty' | <regular expression> | <regular expression> '->' <regular substitution> <symbol> ::= <symbol name> <symbol name> ::= <bare name> | <bracketed name> <action name> ::= <restricted ascii graph name> | '::shift' | '::undef' | '::ascii' | /::convert\[[^\]]+\]/ | '::concat' | /::copy\[\d+\]/ | <lua action name> | '::true' | '::false' | '::json' | '::jsonf' | '::row' | '::table' | '::ast' | <lua function> <symbol action name> ::= <restricted ascii graph name> | '::transfer' | '::undef' | '::ascii' | /::convert\[[^\]]+\]/ | '::concat' | <lua action name> | '::true' | '::false' | '::json' | '::jsonf' | <lua function> <if action name> ::= <restricted ascii graph name> | <lua action name> | <lua function> <generator action name> ::= <restricted ascii graph name> | <lua action name> | <lua function> <regex action name> ::= <restricted ascii graph name> | <lua action name> | <lua function> <quantifier> ::= '*' | '+' <signed integer> ::= /[+-]?\d+/ <unsigned integer> ::= /\d+/ <grammar reference> ::= <quoted string> | <signed integer> | '=' <unsigned integer> <string literal> ::= <string literal unit>+ proper => 1 <string literal unit> ::= '::u8"' <discard off> <string literal inside any> '"' <discard on> <discard on> ::= <discard off> ::= <string literal inside any> ::= <string literal inside>* proper => 1 <string literal inside> ::= /[^"\\\n]/ | '\\' /["'?\\abfnrtve]/ | '\\' /x\{[a-fA-F0-9]{2}\}/ | '\\' /u\{[a-fA-F0-9]{4}\}/ | '\\' /U\{[a-fA-F0-9]{8}\}/ <lua script statement> ::= '<luascript>' <discard off> <lua script source> '</luascript>' <discard on> <lua script source> ::= /[\s\S]/* <event action> ::= 'event-action' /=>|\x{21D2}/u <event action name> <event action name> ::= <restricted ascii graph name> | <lua action name> | <lua function> <default encoding> ::= 'default-encoding' /=>|\x{21D2}/u <default encoding name> <default encoding name> ::= <graph ascii name> <fallback encoding> ::= 'fallback-encoding' /=>|\x{21D2}/u <fallback encoding name> <fallback encoding name> ::= <graph ascii name> <lua function> ::= /::luac?\->function\(/ <lua funcbody after lparen>@+2 <lua functioncall> ::= /\-\-?>\(/' <lua args after lparen>@+2 <lua functiondecl> ::= /<\-\-?\(/ <lua optional parlist after lparen>@+2 ####################################################### :desc ~ 'L0' <op declare any grammar> ~ /:\[[\d]+\]:=/ <op declare top grammar> ~ '::=' <op declare lex grammar> ~ '~' <op loosen> ~ '||' <op equal priority> ~ '|' <true> ~ '1' <false> ~ '0' <word character> ~ /[\w]/ <one or more word characters> ~ <word character>+ proper => 1 <zero or more word characters> ~ <word character>* proper => 1 <restricted ascii graph name> ~ /[!#$%&*+.\/;?\[\\\]^_`~A-Za-z0-9]+/ <lua action name> ~ /::lua->[a-zA-Z_][a-zA-Z0-9_]*/ <bare name> ~ <word character>+ proper => 1 <standard name> ~ /[a-zA-Z]/ <zero or more word characters> <bracketed name> ~ '<' <bracketed name string> '>' <bracketed name string> ~ /[\\s\\w\\[\\]]+/ <quoted string> ~ /'(?:[^\\']*(?:\\.[^\\']*)*)'|"(?:[^\\"]*(?:\\.[^\\"]*)*)"|\x{201C}(?:[^\\\x{201D}]*(?:\\.[^\\\x{201D}]*)*)\x{201D}/su | /'(?:[^\\']*(?:\\.[^\\']*)*)'|"(?:[^\\"]*(?:\\.[^\\"]*)*)"|\x{201C}(?:[^\\\x{201D}]*(?:\\.[^\\\x{201D}]*)*)\x{201D}/su ':' /ic?/ <quoted string literal> ~ /'(?:[^\\']*(?:\\.[^\\']*)*)'|"(?:[^\\"]*(?:\\.[^\\"]*)*)"|\x{201C}(?:[^\\\x{201D}]*(?:\\.[^\\\x{201D}]*)*)\x{201D}/su <character class> ~ /((?:\[(?:(?>[^\[\]]+)|(?-1))*\]))/ | /((?:\[(?:(?>[^\[\]]+)|(?-1))*\]))/ ':' /[eijmnsxDJUuaNbcA]+/ <regular expression> ~ /\/(?![*\/])(?:[^\\\/]*(?:\\.[^\\\/]*)*)\//su | /\/(?![*\/])(?:[^\\\/]*(?:\\.[^\\\/]*)*)\//su /[eijmnsxDJUuaNbcA]+/ <regular substitution> ~ /'(?:[^\\']*(?:\\.[^\\']*)*)'|"(?:[^\\"]*(?:\\.[^\\"]*)*)"|\x{201C}(?:[^\\\x{201D}]*(?:\\.[^\\\x{201D}]*)*)\x{201D}/su | /'(?:[^\\']*(?:\\.[^\\']*)*)'|"(?:[^\\"]*(?:\\.[^\\"]*)*)"|\x{201C}(?:[^\\\x{201D}]*(?:\\.[^\\\x{201D}]*)*)\x{201D}/su ':' /[xgl!f]+/ <graph ascii name> ~ /[[:graph:]]+/ ####################################################### :desc :[2]:= 'Lua 5.3' :discard :[2]:= /\s+/ :discard :[2]:= /--(?:\[(?!=*\[)|(?!\[))[^\n]*/ :discard :[2]:= /--\[(=*)\[.*?\]\1\]/s # # ----------------------------------------------------------------------- # Lua 5.3.4 grammar. Based on perl package MarpaX::Languages::Lua::Parser # ----------------------------------------------------------------------- # # # Special entries used to hook the lua grammar in ESLIF # <lua funcbody after lparen> :[2]:= <lua optional parlist> ')' <lua block> <lua keyword end> <lua args after lparen> :[2]:= <lua optional explist> ')' <lua optional parlist after lparen> :[2]:= <lua optional parlist> ')' <lua chunk> :[2]:= <lua chunk> :[2]:= <lua stat list> | <lua stat list> <lua laststat> | <lua stat list> <lua laststat> ';' | <lua laststat> ';' | <lua laststat> <lua stat list> :[2]:= <lua stat> | <lua stat> ';' | <lua stat list> <lua stat> rank => -1 | <lua stat list> <lua stat> ';' <lua block> :[2]:= <lua chunk> <lua stat> :[2]:= <lua varlist> '=' <lua explist> | <lua functioncall> rank => -1 | <lua label> | <lua keyword goto> <lua Name> | <lua keyword do> <lua block> <lua keyword end> | <lua keyword while> <lua exp> <lua keyword do> <lua block> <lua keyword end> | <lua keyword repeat> <lua block> <lua keyword until> <lua exp> | <lua keyword if> <lua exp> <lua keyword then> <lua block> <lua elseif sequence> <lua optional else block> <lua keyword end> | <lua keyword for> <lua Name> '=' <lua exp> ',' <lua exp> ',' <lua exp> <lua keyword do> <lua block> <lua keyword end> | <lua keyword for> <lua Name> '=' <lua exp> ',' <lua exp> <lua keyword do> <lua block> <lua keyword end> | <lua keyword for> <lua namelist> <lua keyword in> <lua explist> <lua keyword do> <lua block> <lua keyword end> | <lua keyword function> <lua funcname> <lua funcbody> | <lua keyword local> <lua keyword function> <lua Name> <lua funcbody> | <lua keyword local> <lua namelist> <lua optional namelist initialization> | ';' <lua elseif sequence> :[2]:= <lua elseif sequence> :[2]:= <lua elseif sequence> <lua elseif block> <lua elseif block> :[2]:= <lua keyword elseif> <lua exp> <lua keyword then> <lua block> <lua optional else block> :[2]:= <lua optional else block> :[2]:= <lua keyword else> <lua block> <lua optional namelist initialization> :[2]:= <lua optional namelist initialization> :[2]:= '=' <lua explist> <lua laststat> :[2]:= <lua keyword return> <lua optional explist> | <lua keyword break> <lua optional explist> :[2]:= <lua optional explist> :[2]:= <lua explist> <lua funcname> :[2]:= <lua dotted name> <lua optional colon name element> <lua dotted name> :[2]:= <lua Name>+ separator => '.' proper => 1 <lua optional colon name element> :[2]:= <lua optional colon name element> :[2]:= ':' <lua Name> <lua varlist> :[2]:= <lua var>+ separator => ',' proper => 1 <lua var> :[2]:= <lua Name> | <lua prefixexp> '[' <lua exp> ']' | <lua prefixexp> '.' <lua Name> <lua namelist> :[2]:= <lua Name>+ separator => ',' proper => 1 <lua explist> :[2]:= <lua exp>+ separator => ',' proper => 1 <lua exp> :[2]:= <lua var> | '(' <lua exp> ')' assoc => group || <lua exp> <lua args> assoc => right || <lua exp> ':' <lua Name> <lua args> assoc => right | <lua keyword nil> | <lua keyword false> | <lua keyword true> | <lua Number> | <lua String> | '...' | <lua tableconstructor> | <lua function> || <lua exp> '^' <exponent> assoc => right || '-' <lua exp> | <lua keyword not> <lua exp> | '#' <lua exp> | '~' <lua exp> || <lua exp> '*' <lua exp> | <lua exp> '/' <lua exp> | <lua exp> '//' <lua exp> | <lua exp> '%' <lua exp> || <lua exp> '+' <lua exp> | <lua exp> '-' <lua exp> || <lua exp> '..' <lua exp> assoc => right || <lua exp> '<<' <lua exp> | <lua exp> '>>' <lua exp> || <lua exp> '&' <lua exp> || <lua exp> '~' <lua exp> || <lua exp> '|' <lua exp> || <lua exp> '<' <lua exp> | <lua exp> '<=' <lua exp> | <lua exp> '>' <lua exp> | <lua exp> '>=' <lua exp> | <lua exp> '==' <lua exp> rank => 1 | <lua exp> '~=' <lua exp> || <lua exp> <lua keyword and> <lua exp> rank => 1 || <lua exp> <lua keyword or> <lua exp> <exponent> :[2]:= <lua var> | '(' <lua exp> ')' || <exponent> <lua args> || <exponent> ':' <lua Name> <lua args> | <lua keyword nil> | <lua keyword false> | <lua keyword true> | <lua Number> | <lua String> | '...' | <lua tableconstructor> | <lua function> || <lua keyword not> <exponent> | '#' <exponent> | '-' <exponent> <lua prefixexp> :[2]:= <lua var> | <lua functioncall> | '(' <lua exp> ')' <lua functioncall> :[2]:= <lua prefixexp> <lua args> | <lua prefixexp> ':' <lua Name> <lua args> <lua args> :[2]:= '(' <lua optional explist> ')' | <lua tableconstructor> | <lua String> <lua function> :[2]:= <lua keyword function> <lua funcbody> <lua funcbody> :[2]:= '(' <lua optional parlist> ')' <lua block> <lua keyword end> <lua optional parlist> :[2]:= <lua optional parlist> :[2]:= <lua namelist> | <lua namelist> ',' '...' | '...' # A lone comma is not allowed in an empty fieldlist, # apparently. This is why I use a dedicated rule # for an empty table and a '+' sequence, # instead of a '*' sequence. <lua tableconstructor> :[2]:= '{' '}' | '{' <lua fieldlist> '}' <lua fieldlist> :[2]:= <lua field>+ separator => [,;] <lua field> :[2]:= '[' <lua exp> ']' '=' <lua exp> | <lua Name> '=' <lua exp> | <lua exp> <lua label> :[2]:= '::' <lua Name> '::' <lua Name> :[2]:= <LUA NAME> - <LUA RESERVED KEYWORDS> <lua String> :[2]:= /'(?:[^\\']*(?:\\.[^\\']*)*)'|"(?:[^\\"]*(?:\\.[^\\"]*)*)"|\[(=*)\[.*?\]\1\]/su # A lua number can start with '.' if the later is followed by at least one (hex) digit <lua Number> :[2]:= /(?:\.[0-9]+|[0-9]+(?:\.[0-9]*)?)(?:[eE][+-]?[0-9]+)?/ | /0[xX](?:\.[a-fA-F0-9]+|[a-fA-F0-9]+(?:\.[a-fA-F0-9]*)?)(?:\.[a-fA-F0-9]*)?(?:[pP][+-]?[0-9]+)?/ ####################################################### <lua keyword and> :[3]:= 'and' <lua keyword break> :[3]:= 'break' <lua keyword do> :[3]:= 'do' <lua keyword else> :[3]:= 'else' <lua keyword elseif> :[3]:= 'elseif' <lua keyword end> :[3]:= 'end' <lua keyword false> :[3]:= 'false' <lua keyword for> :[3]:= 'for' <lua keyword function> :[3]:= 'function' <lua keyword if> :[3]:= 'if' <lua keyword in> :[3]:= 'in' <lua keyword local> :[3]:= 'local' <lua keyword nil> :[3]:= 'nil' <lua keyword not> :[3]:= 'not' <lua keyword or> :[3]:= 'or' <lua keyword repeat> :[3]:= 'repeat' <lua keyword return> :[3]:= 'return' <lua keyword then> :[3]:= 'then' <lua keyword true> :[3]:= 'true' <lua keyword until> :[3]:= 'until' <lua keyword while> :[3]:= 'while' <lua keyword goto> :[3]:= 'goto' <LUA NAME> :[3]:= /[a-zA-Z_][a-zA-Z_0-9]*/ <LUA RESERVED KEYWORDS> :[3]:= 'and' | 'break' | 'do' | 'else' | 'elseif' | 'end' | 'false' | 'for' | 'function' | 'if' | 'in' | 'local' | 'nil' | 'not' | 'or' | 'repeat' | 'return' | 'then' | 'true' | 'until' | 'while' | 'goto'
Every call to embedded lua provide the following globals:
Lua object representing current MarpaX::ESLIF instance.
Lua object representing current marpaESLIFGrammar instance.
Lua object representing current marpaESLIFRecognizer instance.
Lua object representing current marpaESLIFValue instance, when doing valuation.
The predicted, completed and nulled grammar events have a cost, and when possible lexeme or terminal events should be prefered.
Using hide-separator and (-...-), although convenient, are likely to be less performant than by taking that into account in your own actions.
hide-separator
marpaESLIF, tconv, ICU
Jean-Damien Durand <jeandamiendurand@free.fr>
This software is copyright (c) 2017 by Jean-Damien Durand.
This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself.
To install MarpaX::ESLIF, copy and paste the appropriate command in to your terminal.
cpanm
cpanm MarpaX::ESLIF
CPAN shell
perl -MCPAN -e shell install MarpaX::ESLIF
For more information on module installation, please visit the detailed CPAN module installation guide.