o 1)hZt @sLUdZddlZddlZddlmZddlmZddlm Z ddl m Z ddl mZdd lmZejre d/Z?e d0Z@e d1ZAe d2ZBe d3ZCe d4ZDe d5ZEe d6ZFe d7ZGe d8ZHe d9ZIe d:ZJe d;ZKe d<ZLe d=ZMe d>ZNe d?ZOe d@ZPe dAZQe dBZRe dCZSidDe#dEe;dFe'dGe*dHe3dIe2dJe6dKe<dLe.dMe8dNe/dOe9dPe-dQe7dRe)dSe4dTe+e,e0e1e$e(e%e5e&e:dU ZTdVdWeTUDZVeWeTeWeVksJdXedNdYXdZd[eYeTd\d]d^DdOZZe[eIeKeJe=eNeOePgZ\e[e=eQeKePgZ]d_e^d`e^fdadbZ_dcddd`e^fdedfZ`dge^d`e^fdhdiZadje^d`ebfdkdlZcdmdnd`ejdeje^e^ffdodpZeGdqdrdrZfGdsddddejgZhGdtduduZiGdvdwdwZjddxdyZkGdzd{d{elZmGd|d}d}ejgZnGd~d d ZodS)zImplements a Jinja / Python combination lexer. The ``Lexer`` class is used to do some preprocessing. It filters out invalid operators like the bitshift operators we don't allow in templates. It separates template code and python code in expressions. N) literal_eval)deque)intern)pattern)TemplateSyntaxError)LRUCache) Environment2Lexer _lexer_cachez\s+z (\r\n|\r|\n)z7('([^'\\]*(?:\\.[^'\\]*)*)'|"([^"\\]*(?:\\.[^"\\]*)*)")z ( 0b(_?[0-1])+ # binary | 0o(_?[0-7])+ # octal | 0x(_?[\da-f])+ # hex | [1-9](_?\d)* # decimal | 0(_?0)* # decimal zero ) z (?) z>=srBzoperators droppedr:ccs|]}t|VqdSN)reescaper>xr=r=rA rHcCs t| SrC)lenrGr=r=rAs rL)key token_typereturncCsL|tvrt|Stdtdtdtdtdtdtdtdt dt d t d t d i ||S) Nzbegin of commentzend of commentr%zbegin of statement blockzend of statement blockzbegin of print statementzend of print statementzbegin of line statementzend of line statementztemplate data / textzend of template)reverse_operatorsTOKEN_COMMENT_BEGINTOKEN_COMMENT_END TOKEN_COMMENTTOKEN_LINECOMMENTTOKEN_BLOCK_BEGINTOKEN_BLOCK_ENDTOKEN_VARIABLE_BEGINTOKEN_VARIABLE_ENDTOKEN_LINESTATEMENT_BEGINTOKEN_LINESTATEMENT_END TOKEN_DATA TOKEN_EOFget)rNr=r=rA_describe_token_types" r^tokenTokencCs|jtkr|jSt|jS)z#Returns a description of the token.)type TOKEN_NAMEvaluer^)r_r=r=rAdescribe_tokens  rdexprcCs8d|vr|dd\}}|tkr|St|S|}t|S)z0Like `describe_token` but for token expressions.r9r)splitrbr^)rerarcr=r=rAdescribe_token_exprsrgrccCstt|S)zsCount the number of newline characters in the string. This is useful for extensions that filter a stream. )rJ newline_refindall)rcr=r=rAcount_newlinesrj environmentr cCstj}t|jt||jft|jt||jft|jt||jfg}|j dur8| t|j t d||j f|j durM| t|j t d||j fddt|ddDS)zACompiles all the rules from the environment into a list of rules.Nz ^[ \t\v]*z(?:^|(?<=\S))[^\S\r\n]*cSsg|]}|ddqS)rNr=rFr=r=rA sz!compile_rules..T)reverse)rDrErJcomment_start_stringrQblock_start_stringrUvariable_start_stringrWline_statement_prefixappendrYline_comment_prefixTOKEN_LINECOMMENT_BEGINsorted)rlerulesr=r=rA compile_ruless:    ryc@sLeZdZdZefdedejeddfddZde d ej edd fd d Z dS) FailurezjClass that raises a `TemplateSyntaxError` if called. Used by the `Lexer` to specify known errors. messageclsrONcCs||_||_dSrC)r{ error_class)selfr{r|r=r=rA__init__s zFailure.__init__linenofilenamez te.NoReturncCs||j||rC)r}r{)r~rrr=r=rA__call__ szFailure.__call__) __name__ __module__ __qualname____doc__rstrtTyperintOptionalrr=r=r=rArzs rzc@sXeZdZUeed<eed<eed<defddZdedefdd Zd edefd d Z d S)r`rrarcrOcCt|SrC)rdr~r=r=rA__str__z Token.__str__recCs2|j|krdSd|vr|dd|j|jgkSdS)zTest a token against a token expression. This can either be a token type or ``'token_type:token_value'``. This can only test against string values and types. Tr9rF)rarfrcr~rer=r=rAtests z Token.testiterablecstfdd|DS)z(Test against multiple token expressions.c3s|]}|VqdSrC)r)r>rerr=rArH&rIz!Token.test_any..)any)r~rr=rrAtest_any$szToken.test_anyN) rrrr__annotations__rrboolrrr=r=r=rAr` s c@s2eZdZdZd ddZd dd Zdefd d ZdS)TokenStreamIteratorz`The iterator for tokenstreams. Iterate over the stream until the eof token is reached. stream TokenStreamrONcCs ||_dSrC)r)r~rr=r=rAr.s zTokenStreamIterator.__init__cCs|SrCr=rr=r=rA__iter__1szTokenStreamIterator.__iter__cCs.|jj}|jtur|jtt|j|SrC)rcurrentrar\close StopIterationnextr~r_r=r=rA__next__4s    zTokenStreamIterator.__next__)rrrON)rOr)rrrrrrr`rr=r=r=rAr)s   rc@seZdZdZdejedejedejefddZ de fdd Z de fd d Z ede fd d ZdeddfddZdefddZd#deddfddZdedejefddZdede fddZdefddZd$dd Zdedefd!d"ZdS)%rzA token stream is an iterable that yields :class:`Token`\s. The parser however does not iterate over it but calls :meth:`next` to go one token ahead. The current active token is stored as :attr:`current`. generatorr"rcCs>t||_t|_||_||_d|_tdtd|_ t |dS)NFr) iter_iterr_pushedr"rclosedr` TOKEN_INITIALrr)r~rr"rr=r=rArEs  zTokenStream.__init__rOcCrrC)rrr=r=rArSrzTokenStream.__iter__cCst|jp |jjtuSrC)rrrrar\rr=r=rA__bool__VszTokenStream.__bool__cCs| S)z Are we at the end of the stream?r=rr=r=rAeosYszTokenStream.eosr_NcCs|j|dS)z Push a token back to the stream.N)rrsrr=r=rApush^szTokenStream.pushcCs"t|}|j}||||_|S)zLook at the next token.)rrr)r~Z old_tokenresultr=r=rAlookbs  zTokenStream.lookrncCst|D]}t|qdS)zGot n tokens ahead.N)ranger)r~r_r=r=rAskipjs  zTokenStream.skiprecCs|j|r t|SdS)zqPerform the token test and return the token if it matched. Otherwise the return value is `None`. N)rrrrr=r=rAnext_ifos zTokenStream.next_ifcCs||duS)z8Like :meth:`next_if` but only returns `True` or `False`.N)rrr=r=rAskip_ifxszTokenStream.skip_ifcCs\|j}|jr|j|_|S|jjtur,z t|j|_W|Sty+|Y|Sw|S)z|Go one token ahead and return the old one. Use the built-in :func:`next` instead of calling this directly. ) rrpopleftrar\rrrr)r~rvr=r=rAr|s   zTokenStream.__next__cCs&t|jjtd|_td|_d|_dS)zClose the stream.rr=TN)r`rrr\rrrrr=r=rArs  zTokenStream.closecCsn|j|s3t|}|jjturtd|d|jj|j|jtd|dt |j|jj|j|jt |S)z}Expect a given token type and return it. This accepts the same argument as :meth:`jinja2.lexer.Token.test`. z%unexpected end of template, expected r8zexpected token z, got ) rrrgrar\rrr"rrdrrr=r=rAexpects   zTokenStream.expect)r)rON)rrrrrIterabler`rrrrrrrpropertyrrrrrrrrrrr=r=r=rAr?s*   rc CsZ|j|j|j|j|j|j|j|j|j|j |j |j f }t |}|dur+t|t |<}|S)z(Return a lexer which is probably cached.N)rpblock_end_stringrqvariable_end_stringrocomment_end_stringrrrt trim_blocks lstrip_blocksnewline_sequencekeep_trailing_newliner r]r )rlrMlexerr=r=rA get_lexers" rcs$eZdZdZdZfddZZS)OptionalLStripzWA special tuple for marking a point in the state that can have lstrip applied. r=cst||SrC)super__new__)r|memberskwargs __class__r=rArszOptionalLStrip.__new__)rrrr __slots__r __classcell__r=r=rrArsrc@sNeZdZUejeed<ejeejedfeje fed<ej eed<dS)_Ruler.tokenscommandN) rrrrPatternrrUnionTuplerzrr=r=r=rArs $rc@seZdZdZdddZdedefd d Z   dd ed ejed ejedejede f ddZ  ddej ej e eefd ejed ejedejefddZ  dd ed ejed ejedejedejej e eeff ddZdS)r a Class that implements a lexer for a given environment. Automatically created by the environment class, usually you don't have to do that. Note that the lexer is not automatically bound to an environment. Multiple environments can share the same lexer. rlr rONc Cs&tj}dtdtjtfdd}tttdttt dtt t dtt t dtttdtttdg}t|}||j}||j}||j}||j} |jrLdnd} |j|_|j|_|j|_d|d|d |d } d | gd d |D} dt|d| dttddt|dtdgtt|d|d|d || d tt fdt|dt!dfdgt"t|d|d|d || dt#dg|t$t|d| d | t%dg|t&t|d|d|d|d || d ttt'dt|dt!dfdgt(t|dt)dg|t*t|dt+t,fdgi|_-dS)NrGrOcSst|tjtjBSrC)rDcompileMSrKr=r=rAcszLexer.__init__..cz\n?rz(?Pz(\-|\+|)\s*raw\s*(?:\-z\s*|z))r:cSs"g|] \}}d|d|dqS)z(?Prrr=r=rArm s"z"Lexer.__init__..rootz(.*?)(?:r2#bygroupz.+z (.*?)((?:\+z|\-#popz(.)zMissing end of comment tagz(?:\+z\-z (.*?)((?:z(\-|\+|))\s*endraw\s*(?:\+zMissing end of raw directivez \s*(\n|$)z(.*?)()(?=\n|$)).rDrErrrr whitespace_reTOKEN_WHITESPACEfloat_re TOKEN_FLOAT integer_re TOKEN_INTEGERname_rerb string_re TOKEN_STRING operator_reTOKEN_OPERATORryrprrrrrrrjoinrr[rQrSrRrzrUrVrWrXTOKEN_RAW_BEGIN TOKEN_RAW_ENDrYrZrurTTOKEN_LINECOMMENT_ENDrx) r~rlrwrZ tag_rulesZroot_tag_rulesZblock_start_reZ block_end_reZcomment_end_reZvariable_end_reZblock_suffix_reZ root_raw_reZ root_parts_rer=r=rArs                  zLexer.__init__rccCst|j|S)z`Replace all newlines with the configured sequence in strings and template data. )rhrr)r~rcr=r=rA_normalize_newlinesVrkzLexer._normalize_newlinessourcer"rstatecCs&|||||}t||||||S)z:Calls tokeniter + tokenize and wraps it in a token stream.) tokeniterrwrap)r~rr"rrrr=r=rAtokenize\szLexer.tokenizerc csT|D]\}}}|tvr q|}|tkrt}n|tkrt}n|ttfvr$q|tkr.||}nr|dkr5|}nk|t krG|}| sFt d|||nY|t kr}z||dd ddd}WnBty|}zt|dd} t | ||||d }~ww|tkrt|d d d }n|tkrt|d d }n|tkrt|}t|||Vqd S) zThis is called with the stream as returned by `tokenize` and wraps every token in a :class:`Token` and converts the value. keywordzInvalid character in identifierrasciibackslashreplacezunicode-escaper9Nrrr)ignored_tokensrYrUrZrVrrr[rrb isidentifierrrencodedecode Exceptionrrfstriprrreplacerrr operatorsr`) r~rr"rrr_Z value_strrcrwmsgr=r=rArgsR   z Lexer.wrapccs"t|ddd}|js|ddkr|d=d|}d}d}dg}|dur:|dkr:|d vs3Jd ||d |j|d} t|} g} d} d } | D]\}}}|||}|dur_qN| ri|tt t fvriqNt |t r2| }t |tr|d}td d|dddD}|dkr|}|t|dd} |g|dd}n.|dkr|jr|ts|dd}|dks| rt||r|d|g|dd}t|D]W\}}t |tr||||dkr|D]\}}|dur |||fV||d7}nqt|dq||}|s|tvr%|||fV||d| 7}d} qni|}|tkr|dkrF| dn@|dkrQ| dn5|dkr\| dn*|dvr| sot d|d|||| !}||krt d|d|d||||s|tvr|||fV||d7}|dddk} |"}|dur|dkr|!n*|dkr|D]\}}|dur||n qt|dn|||j|d} n ||krt|d|}n|| krdSt d ||d!||||qL)"aThis method tokenizes the text and returns the tokens in a generator. Use this method if you just want to tokenize a template. .. versionchanged:: 3.0 Only ``\n``, ``\r\n`` and ``\r`` are treated as line breaks. Nrr rrr)variableblockz invalid stateZ_beginTcss|] }|dur|VqdSrCr=)r>gr=r=rArHsz"Lexer.tokeniter..r*r)rz= wanted to resolve the token dynamically but no group matchedr3r4r1r2r/r0)r4r2r0z unexpected ''z ', expected 'rzA wanted to resolve the new state dynamically but no group matchedz* yielded empty string without stack changezunexpected char z at )#rhrfrrrsrxrJmatchrXrVrZ isinstancetuplegroupsrrrstripcountr groupdictr]rWrfindr fullmatch enumeraterzitems RuntimeErrorignore_if_emptygrouprrpopend)r~rr"rrlinesposrstack statetokensZ source_lengthZbalancing_stackZnewlines_strippedZ line_startingregexr new_statemrtextZ strip_signstrippedZl_posidxr_rMrcr&Z expected_oppos2r=r=rArs                                zLexer.tokeniter)rlr rON)NNN)NN)rrrrrrrrrrrrrrIteratorr`rrr=r=r=rAr sT w   :)rlr rOr )prrDtypingrastr collectionsrsysrZ _identifierrr exceptionsrutilsr TYPE_CHECKINGZtyping_extensionsterlr r MutableMappingrrrrrhrr IGNORECASEVERBOSErrZ TOKEN_ADDZ TOKEN_ASSIGNZ TOKEN_COLONZ TOKEN_COMMAZ TOKEN_DIVZ TOKEN_DOTZTOKEN_EQZTOKEN_FLOORDIVZTOKEN_GTZ TOKEN_GTEQZ TOKEN_LBRACEZTOKEN_LBRACKETZ TOKEN_LPARENZTOKEN_LTZ TOKEN_LTEQZ TOKEN_MODZ TOKEN_MULZTOKEN_NEZ TOKEN_PIPEZ TOKEN_POWZ TOKEN_RBRACEZTOKEN_RBRACKETZ TOKEN_RPARENZTOKEN_SEMICOLONZ TOKEN_SUBZ TOKEN_TILDErrrrbrrrUrVrWrXrrrQrRrSrYrZrurrTr[rr\rr rPrJrrvr frozensetrr rr^rdrgrrjListryrz NamedTupler`rrrrrrr r=r=r=rAs.                &  ") m