-XdZddlZddlZddlmZddlmZddlm Z m Z ddl m Z m Z mZmZmZmZmZmZmZmZmZmZmZmZmZmZmZedZGdd eZ d e!d e!fd Z"Gd dZ#eddZ$Gddee$Z%GddeZ&ede%e dZ'Gddee'e&Z(y)zT This module contains base classes and helper functions for defining Pratt parsers. N)ABCMeta)name)DecimalDecimalException)AnycastoverloadCallableDictGenericListOptionalUnionTupleTypePatternMatchMutableMappingMutableSequenceIteratorTypeVar) (start)(end)(string)(float) (decimal) (integer)(name) (invalid) (unknown)ceZdZdZy) ParseErrorz.An error when parsing source with TDOP parser.N)__name__ __module__ __qualname____doc__5/usr/lib/python3.12/site-packages/elementpath/tdop.pyr"r"1s8r(r"symbolreturncdtdtfd|jr|jS|tvr|ddjSt d|DrXdj d|D}|jj dd j d d j d d S|j d d }|jr |jj d d Sd j fd |D}|j dd j d d j d d S) zP Converts a symbol string to an identifier (only alphanumeric and '_'). cr+cz|js|dk(r|Sdtt|jzS)N_z%s_)isalnum unicode_namestrtitle)r-s r) get_id_namez)_symbol_to_classname..get_id_name9s3 99;!s(H<A/5577 7r(c3$K|]}|dv yw)z-_Nr'.0r-s r) z'_symbol_to_classname..Cs '1Q$Ys c32K|]}t|ywN)r1r8s r)r:z'_symbol_to_classname..Ds9&Qa&s-r/c3.K|] }|ywr=r')r9r-r4s r)r:z'_symbol_to_classname..Ks3FqKNFs)r2r0r3SPECIAL_SYMBOLSalljoinreplace isidentifier)r*valuer4s @r)_symbol_to_classnamerG5s8s8s8 ~~||~ ? "a|!!## ' ' '9&99{{}$$S"-55c2>FFsBOO NN3 $E {{}$$S"-- GG3F3 3E ==b ! ) )#r 2 : :3 CCr(ceZdZdZdeddfdZdedefdZdedefdZ defd Z defd Z de fd Z d edefd ZdedefdZdedefdZy) MultiLabelaV Helper class for defining multi-value label for tokens. Useful when a symbol has more roles. A label of this type has equivalence with each of its values. Example: label = MultiLabel('function', 'operator') label == 'symbol' # False label == 'function' # True label == 'operator' # True valuesr+Nc||_yr=)rJ)selfrJs r)__init__zMultiLabel.__init__Zs  r(otherc@tfd|jDS)Nc3(K|] }|k( ywr=r'r9vrNs r)r:z$MultiLabel.__eq__..^3{!5A:{anyrJrLrNs `r)__eq__zMultiLabel.__eq__]3t{{333r(c@tfd|jDS)Nc3(K|] }|k7 ywr=r'rQs r)r:z$MultiLabel.__ne__..arSrT)rBrJrWs `r)__ne__zMultiLabel.__ne__`rYr(cJ|jj|jSr=) __class__r#rJrLs r)__repr__zMultiLabel.__repr__cs00$++>>r(cXdj|jjddS)N__r;r/)rCrJrDr_s r)__str__zMultiLabel.__str__fs"yy%--c377r(c,t|jSr=)hashrJr_s r)__hash__zMultiLabel.__hash__isDKK  r(itemc@tfd|jDS)Nc3&K|]}|v ywr=r')r9rRrgs r)r:z*MultiLabel.__contains__..ms2k419ksrU)rLrgs `r) __contains__zMultiLabel.__contains__ls2dkk222r(sc@tfd|jDS)Nc3@K|]}|jywr=) startswithr9rRrks r)r:z(MultiLabel.startswith..ps8Kq1<<?KrUrLrks `r)rnzMultiLabel.startswithos8DKK888r(c@tfd|jDS)Nc3@K|]}|jywr=)endswithros r)r:z&MultiLabel.endswith..ss6+Q1::a=+rprUrqs `r)rtzMultiLabel.endswithrs6$++666r()r#r$r%r&r2rMobjectboolrXr\r`rcintrfrjrnrtr'r(r)rIrIOs 4F4t44F4t4?#?88!#!3339C9D97#7$7r(rITKz Token[Any])boundceZdZUdZdZeed<dZeed<dZe ed<dZ e ed<dZ e ed<d Z e e ed <d Zeeed <d ed<e eed<eeefed< d:dd de edd fdZededefdZededeefdZdeeefdeeeeffdZdeeefdedd fdZdeeefdd fdZdefdZdededd fdZde fdZde fdZd ede fd!Z!e"defd"Z#e"de fd#Z$e"de fd$Z%e"deeeffd%Z&d;d'Z'de fd(Z(de fd)Z)dd4e de6fd9Z7y )?Tokena Token base class for defining a parser based on Pratt's method. Each token instance is a list-like object. The number of token's items is the arity of the represented operator, where token's items are the operands. Nullary operators are used for symbols, names and literals. Tokens with items represent the other operators (unary, binary and so on). Each token class has a *symbol*, a lbp (left binding power) value and a rbp (right binding power) value, that are used in the sense described by the Pratt's method. This implementation of Pratt tokens includes two extra attributes, *pattern* and *label*, that can be used to simplify the parsing of symbols in a concrete parser. :param parser: The parser instance that creates the token instance. :param value: The token value. If not provided defaults to token symbol. :cvar symbol: the symbol of the token class. :cvar lbp: Pratt's left binding power, defaults to 0. :cvar rbp: Pratt's right binding power, defaults to 0. :cvar pattern: the regex pattern used for the token class. Defaults to the escaped symbol. Can be customized to match more detailed conditions (e.g. a function with its left round bracket), in order to simplify the related code. :cvar label: defines the typology of the token class. Its value is used in representations of the token instance and can be used to restrict code choices without more complicated analysis. The label value can be set as needed by the parser implementation (eg. 'function', 'axis', 'constructor function' are used by the XPath parsers). In the base parser class defaults to 'symbol' with 'literal' and 'operator' as possible alternatives. If set by a tuple of values the token class label is transformed to a multi-value label, that means the token class can covers multiple roles (e.g. as XPath function or axis). In those cases the definitive role is defined at parse time (nud and/or led methods) after the token instance creation. rlbprbpr>r* lookup_namelabelNpattern)_itemsparserrFspanrzParser[Token[TK]]rrFrr+cg|_||_||n |j|_|jd|_y|jj |_y)N)rr)rrr*rF next_matchr)rLrrFs r)rMzToken.__init__sI  #/UT[[ $//7F V=N=N=S=S=U r(icyr=r'rLrs r) __getitem__zToken.__getitem__s),r(rkcyr=r'rqs r)rzToken.__getitem__sr^r#idr_s r)r`zToken.__repr__ #t~~'>'>4&IIIr(rNct|tr4|j|jk(xr|j|jk(Sy)NF) isinstancer{r*rFrWs r)rXz Token.__eq__s4 eU #;;%,,.L4::3L Lr(ct|Sr=rr_s r)arityz Token.aritys 4yr(c|jdk(rd|jzS|jtvrd|jzS|jdk(r9t|dk(r|djSddj d|Dd S|sd|jzSd|jddj d |Dd S) z%Returns a tree representation string.rz(%s)z(%r)(r5rr;c34K|]}|jywr=treer9rgs r)r:zToken.tree..s;dd d)c34K|]}|jywr=rrs r)r:zToken.tree..s-IDDdiiDr)r*rFrArrrCr_s r)rz Token.trees ;;( "DJJ& & [[O +DJJ& & [[C 4yA~Aw||#sxx;d;; >DKK' 't{{m1SXX-ID-I%I$J!L Lr(c|j}|dk(rtt|jS|dk(rt|jS|tvr%t |jj ddSt|}|s|S|dk(r6d|jvr|djd|S|d|djS|d k(r$|djd|d|djS|ddjd |DS) z)Returns the source representation string.rrz\\\r5postfixrr;c34K|]}|jywr=)sourcers r)r:zToken.source..s2P444;;4r) r*rr2rFrAreprrDrrrrC)rLr*lengths r)rz Token.sources X TZZ( ( { "tzz? "  & #++E48 8YF 1 *&*1gnnf=="($q'..991%)!W^^VT!W^^LL"(#((2P42P*PQQr(c|jd}|jjd|jddz}|dk(rd|dzfS|||jjd|j dz fS)zCA tuple with the position of the token in terms of line and column.rN r5)rrrcountrindex)rL token_indexlines r)positionzToken.positionsyiil {{!!,;/55d;a? 19kAo% %[4;;#5#5l{#C#J#J4#PPPPr( Token[TK]c|jjj|jJd|jjd|j|j}|j |_|S)z$ !Fyy"bii7&: )&&--7*B"(L!% vx01#%tBx!$ &'-zz|$FH)&&--7*B"(L!%=*")&&--7*B"(L s*B"C7%C8C7C41C73C44C7messagercH|r |j|vr|j|yyr=r*rrLrrs r)expectedzToken.expectedhs( t{{'1##G, ,27r(cF|r|j|vr|j|yr=rrs r) unexpectedzToken.unexpectedls&$++0##G, ,1r(c|r t|S|jtvrtd|zS|jdk(rtd|jzS|jdk(rtd|jzS|jdk(rtd|jzS|jdk7rtd |jzS|jj jd k(r td Std S) Nz unexpected %srzinvalid literal %rr zunknown symbol %rrzunexpected name %rrzunexpected literal %rrzsource is emptyzunexpected end of source)r"r*rArFrrrLrs r)rzToken.wrong_syntaxps g& & [[ /o45 5 [[K '2TZZ?@ @ [[K '1DJJ>? ? [[H $2TZZ?@ @ [[G #5 BC C [[   % % 2/0 089 9r(ct|Sr=) TypeErrorrs r) wrong_typezToken.wrong_types !!r(ct|Sr=)rrs r) wrong_valuezToken.wrong_values '""r(r=)r+rTT)z invalid type)z invalid value)8r#r$r%r&r|rw__annotations__r}r*r2r~rrr __slots__r rxrrrMr rslicerrrrrrrcr`rurvrXpropertyrrrrrrrrrrrrrrrr"rrrrrr'r(r)r{r{ys BCLCLFCKE3!GXc]!3I H  C= S/)-V2V  V15V,S,R,, ?U?r':??U3:.R,,-U3:.34U3:.4  $$2$$$<< J#JFt s Mc M MRRR,Q%S/QQDD Pt PDD "R"""r"#%&S%&Xk%:%&N@D--x}--BF-3-#-$-:HSM:Z:$"#"9"#3#Z#r(r{ceZdZUeeed<eeed<eeed<eeeed<e eeefed<dede eedfd e eefd dffd Z xZ S) ParserMetatoken_base_classliterals_patternr tokenizerrrbases. namespacer+ctt| ||||}tj|j j jD]:\}}t|ts|j |j k(s1tdt|ds t|_ t|dstjd|_t|dstjd|_d|vrd|_d|vrBi|_|D]6}t|ds|j$j'|j$|S|S) Nz>@DAq!Z(Q\\S^^-K"#abbA s./#(C s./#%::O$C sN+!zz*:;C  i ' CM  *!C # :~6$$++J,C,CD $ r()r#r$r%rrrrr2rrrr r __classcell__)r^s@r)rrs3icl"#, %% d3i003uT#Y^'<cSVhr(rTK_coT)ry covariantc eZdZUdZeZdZeee e d<iZ e e e efe d<ee d<e e d<eee e d<ee d<ee d <eee e d <ee e d <ee e d <d Zd6dZde fdZde fdZdedefdZde defdZddde dee defdZde de fdZd7dedefdZedeeeffdZdefdZ defdZ!d8d ed!edefd"Z"e#d#e de fd$Z$e%d%e&e e efd&e'de efd'Z(e%d%e ddfd(Z)e%d%e d)e d&e'de efd*Z*e%d7d%e d+ede efd,Z+e%d7d%e d+ede efd-Z,e%d7d%e d+ede efd.Z-e%d7d%e d+ede efd/Z.e%d7d%e d+ede efd0Z/e%d7d%e d+ede efd1Z0e%d7d%e&e e efd+ede1e1d2e'fge1d2e'fffd3Z2e%d6d4Z3e%de4e e efdee fd5Z5y)9Parsera; Parser class for implementing a Top-Down Operator Precedence parser. :cvar symbol_table: a dictionary that stores the token classes defined for the language. :cvar token_base_class: the base class for creating language's token classes. :cvar tokenizer: the language tokenizer compiled regexp. Nrr _start_tokenrtokensr next_tokenrrr)rrrrrrr+c|j|jd|_td|_d|_|j d||_|jx|_|_ y)Nr>r'r) rbuildrrrrrrrrr_s r)rMzParser.__init__s\ >> ! JJL 2h 8D--i8>'+'8'88 T_r(cJd|jjt|fzSrrr_s r)r`zParser.__repr__rr(c4|jjdS)Nz())r^r#r_s r)rczParser.__str__s..))*"--r(rNct|txr3|j|juxr|j|jk(Sr=)rrrrrWs r)rXz Parser.__eq__sC%(4  ! !U%;%; ;4   !3!3 3 4r(cZ|jsJd t|jj||_||_|j|j}|jjd|td|_d|_ |jx|_|_ S#t$r;}|j d||}|j dj|d}~wwxYw#td|_d|_ |jx|_|_ wxYw)z Parses a source code of the formal language. This is the main method that has to be called for a parser's instance. :param source: The source string. :return: The root of the token's tree that parse the source. zParser tokenizer is not built!rzinvalid source type, {}Nrr')rrfinditerrrrrformatradvance expressionrrrrr)rLrerrr root_tokens r)parsez Parser.parses~~???~ = P"4>>#:#:6#BC !DK LLN*J OO $ $W -r(DK"DO+/+<+< //..w7 7__ #{{DO??. ..??((*224 + 9d//8>DO:: )-)?)?)A&w   L";$"3"3F";D"AHzz= qzU" g."?$"3"3J"?e"L6zz53'>P!'NE 'Cd&7&7 &B4&ODO&zz%R#G,E 'Ed&7&7 &DT5&QDOzz#A$"3"3K"@s7|"Tzz 9d//9$EDOzz  zzG L"I&Dd&7&7 &DT7&SDO//66s3x6HHI(I&Dd&7&7 &DT7&SDO//66s3x6HHIsEJ L7 M)A4LL M&!AM!!M&) N72AN22N7 stop_symbolsc |s|jjd|jjdk(r|jj|j|_g} t |j |_|jjd}|B|j}||vr|j|nF |j|||_n+|j|jj dj|S#t$r5|jd||_|jjwxYw#t$r|jd||_YwwxYw)a Advances until one of the symbols is found or the end of source is reached, returning the raw source string placed before. Useful for raw parsing of comments and references enclosed between specific symbols. :param stop_symbols: The symbols that have to be found for stopping advance. :return: The source string chunk enclosed between the initial position and the first stop symbol. z at least a stop symbol required!rrr r>)rrr*rrnextrrr(rrrr+ StopIterationrC)rLr1 source_chunkr*s r) advance_untilzParser.advance_until9sh//,,-OP P __ # #w .//..0 0__ "$  A"&t{{"3 ..q1%#\\^F\1$++F3A.Gd.?.?.G.MDO! !''(=(=(?@)*ww|$$ (A.Ld.?.? .LT.RDO"&//">">"@@A! "<$"3"3G"E#FFr}c2|j|jj}||jjkrE|j|jj |}||jjkrEt t|S)aP Pratt's function for parsing an expression. It calls token.nud() and then advances until the right binding power is less the left binding power of the next token, invoking the led() method on the following token. :param rbp: right binding power for the expression. :return: left token. )rrrrr|rrr)rLr}rs r)r zParser.expressionasm zz~~DOO''' LLN::>>$'DDOO'''E4  r(c.|jjS)z:Property that returns the current line and column indexes.)rrr_s r)rzParser.positionqszz"""r(c6|jjS)zu Returns `True` if the parser is positioned at the start of the source, ignoring the spaces. )rrr_s r)rzParser.is_source_startvs zz))++r(c6|jjS)zx Returns `True` if the parser is positioned at the start of a source line, ignoring the spaces. )rrr_s r)rzParser.is_line_start}s zz''))r(rrc:|jj||S)ay Returns `True` if the source has an extra space (whitespace, tab or newline) immediately before or after the current position of the parser. :param before: if `True` considers also the extra spaces before the current token symbol. :param after: if `True` considers also the extra spaces after the current token symbol. )rr)rLrrs r)rzParser.is_spacedszz##FE22r(string_literalcL|ddjddjddS)Nr5r6z\''z\"")rD)r<s r)r,zParser.unescapes)a#++E37??sKKr(r*kwargsc t|tr5d|vrtd|z|jd|} |j|}nmt|t2rt5|t6st9d t6d |d|}|jj|j:|urtd|z|j=D]V\}} |dk(r| |j>kDr| |_"|dk(r| |j@kDr| |_ >tC| sJt-||| X|S#t $r?||d<||d<|jdd}t|t r t|x}|d<d|vr|jd}nBdjt|t|jjdd}|jd |jf}|j|j|d d t!t"t$t'|||}||j|<t)j*|t-t.j0|j||YwxYw)a Register/update a token class in the symbol table. :param symbol: The identifier symbol for a new class or an existent token class. :param kwargs: Optional attributes/methods for the token class. :return: A token class. r;z&%r: a symbol can't contain whitespacesr~r*r class_namez_{}{}r>rN)r$r% __return__zA string or a z subclass requested, not r'z!Token class %r is not registered.r|r})"rr2rgetrr+tuplerIrrrGr3rDrr r$rrrrrregistersetattrrrtype issubclassr{rr~rr|r}callable) r r*r@r~ token_classrtoken_class_nametoken_class_baseskeyrFs r)rFzParser.registers? fc "f} !IF!RSS **]F;K T!..{; @FD)FE1JeU[\] ] K##F$6$67{J !D{!RSS ,,.JCe|  7"' %+//"9"' % S%0 )] T$*x (3}% 7H5eU+.8%.@@EF7O6)'-zz,'?$'.~~,V4E ((*223;($ %+JJw9M9M8O$P! "%..$4"& #K)9;Lf!U 1<  -((5 CNN35E{S; TsDEI'&I'c:|j|j=y)z/Unregister a token class from the symbol table.N)rr)r r*s r) unregisterzParser.unregisters   V\\^ ,r( new_symbolc |j|}|j|fi|}|jjD],\}}||vs|dvs|j dr t |||.|S)z*Duplicate a token class with a new symbol.)r*rr/)rrFrrrnrG)r r*rQr@rKnew_token_classrNrFs r) duplicatezParser.duplicatesw&&v. &#,,z.nudKr(_args_kwargsc|jSr=r)rLrYrZs r)rz Parser.literal..evaluates :: r(r.)rr|rr)r{rrrF)r r*rUrrs r)r.zParser.literals^ eEl uU|  5<    ||F)hTW|XXr(chdttdttfd}|j|d||S)zCRegister a token for a symbol that represents a *nullary* operator.rLr+c|Sr=r'r_s r)rzParser.nullary..nudrXr(operator)rr|rr{rrFr r*rUrs r)nullaryzParser.nullarys4 eEl uU| ||F*"#|FFr(cpdttdttffd }|j|d|S)zHRegister a token for a symbol that represents a *prefix* unary operator.rLr+cH|jjf|dd|SNr}rr )rLrUs r)rzParser.prefix..nuds&kk,,,45DGKr(zprefix operator)rr|r}rr_r`s ` r)prefixz Parser.prefixs; eEl uU| ||F*;QT|UUr(cdttdttdttfd}|j|d|||S)zIRegister a token for a symbol that represents a *postfix* unary operator.rLrr+c|f|dd|Sr=r'rs r)rzParser.postfix..ledseDGKr(zpostfix operatorrr|r}rr_r r*rUrs r)rzParser.postfixsE eEl %, 5< ||F*<""RU|VVr(cdttdttdttffd }|j|d|S)zIRegister a token for a symbol that represents an *infix* binary operator.rLrr+cJ||jjf|dd|SrdrfrLrrUs r)rzParser.infix..led s(DKK22r2::DGKr(r^rjr_rks ` r)infixz Parser.infixsD eEl %, 5< ||F*""#|NNr(cdttdttdttffd }|j|ddz |S)zJRegister a token for a symbol that represents an *infixr* binary operator.rLrr+cP||jjdz f|dd|S)Nr5rerfrns r)rzParser.infixr..leds,DKK22rAv2>>DGKr(r^r5rjr_rks ` r)infixrz Parser.infixrsH eEl %, 5< ||F*""q&c|RRr(.cz|j|d||dtdtfdtdtfffd }|S)z Register a token for a symbol that represents a custom operator or redefine a method for an existing token. r^)rr|r}func.r+c|jjdd}tt|st |dt |||S)Nr/rz is not a method of )r# partitionrJgetattrrrG)rt method_namerKs r)bindzParser.method..bind!sS--11#6q9KGK=>;/1Ek] STT Kd 3Kr()rFr r)r r*rUryrKs @r)methodz Parser.methodsGll6lL  xS) hsCx.@   r(c<d|jvr|jdd|jvr|jdd|jvr|jdd|jvr|jd|j|j|_y)z Builds the parser class. Checks if all declared symbols are defined and builds the regex tokenizer using the symbol related patterns. rrrr N)rrFcreate_tokenizerr)r s r)rz Parser.build)s C,, , LL # #** * LL ! c.. . LL % c.. . LL %,,S-=-=> r(c  g}g}g}t}|jD]\}}|tvr|j|j |j7|j j |%|jtj|wt|dk(r%|jtj||jtj|g}|r,|jdjt|d|r/|jdjdj||r;|jdjdjt|d|r |jdj|d j|jjdj||j j} tj| S) a Returns a regex based tokenizer built from a symbol table of token classes. The returned tokenizer skips extra spaces between symbols. A regular expression is created from the symbol table of the parser using a template. The symbols are inserted in the template putting the longer symbols first. Symbols and their patterns can't contain spaces. :param symbol_table: a dictionary containing the token classes of the formal language. r5|ct| Sr=rxs r)z)Parser.create_tokenizer..ZsTWXYTZSZr()rNz[{}]r>z\b(?:{})\b(?![\-\.])ct| Sr=rrs r)rz)Parser.create_tokenizer.._s c!fWr(z({})|({})|({})|(\S)|\s+)setrrAraddrrrrescaperrCsortedrrr ) r rcharacter_patternsstring_patterns name_patternscustom_patternsr*rKsymbols_patternstokenizer_patterns r)r|zParser.create_tokenizer;s  %#/#5#5#7 FK($$0##K$7$78!!''/;$$RYYv%67V!"))"))F*;<&&ryy'89$8')   # #CHHVOIZ-[$\ ]   # #FMM"'':L2M$N O   # #$;$B$B 3DEF%    # #CHH_$= >6==  ( ( HH% &    $ $  zz+,,r()r+N)rr)6r#r$r%r&r{rrrrr2rrr rrrrrrMr`rcrurvrXr#rr6rwr rrrrrr staticmethodr, classmethodrrrFrPrTr.rargrrorrr rzrrr|r'r(r)rrs(,Ix %,+-L$sDK'(- K U3Z  Ls$$cl"#,WI9J#J..4F4t4 =C=E=4?CAAhsmAuAF&%3&%3&%P!c!%! #%S/##,,*t* 3 3D 3D 3LLLL?eCe$45??e??B----sstE{YSYcY$u+YYGSGcG$u+GG VCVSVeVVWSWcW$u+WWO3OCOU OOSCSSSeSSE#tE{"23#s(+,hsCx.@@A ??"--N3U ;K,L--QXY\Q]----r(r) metaclass))r&rrabcr unicodedatarr1decimalrrtypingrrr r r r r rrrrrrrrrr frozensetrA SyntaxErrorr"r2rGrIrxr{rrrr'r(r)rs ,-'''''2 99DDD4$7$7N T&M#OB M#`""J uSzT:x-WU^zx-r(