³ò B_Kc@s“ddkZddklZlZddkTddkTeidƒZeidƒZ eidƒZ de fd„ƒYZ d e fd „ƒYZ dS( iÿÿÿÿN(t bracket_parsetTree(t*s\(([^\s()]+) ([^\s()]+)\)s\([^\s()]+ ([^\s()]+)\)s \s*\(\s*\(tBracketParseCorpusReadercBsSeZdZedeed„Zd„Zd„Zd„Zed„Z d„Z RS(sT Reader for corpora that consist of parenthesis-delineated parse trees. tunindented_parencCs5ti||||ƒ||_||_||_dS(s @param root: The root directory for this corpus. @param fileids: A list or regexp specifying the fileids in this corpus. @param comment_char: The character which can appear at the start of a line to indicate that the rest of the line is a comment. @param detect_blocks: The method that is used to find blocks in the corpus; can be 'unindented_paren' (every unindented parenthesis starts a new parse) or 'sexpr' (brackets are matched). N(t CorpusReadert__init__t _comment_chart_detect_blockst_tag_mapping_function(tselftroottfileidst comment_chart detect_blockstencodingttag_mapping_function((s6/p/zhu/06/nlp/nltk/nltk/corpus/reader/bracket_parse.pyRs   cCsÒ|idjot|d|iƒSn¨|idjot|ƒSnŠ|idjoht|ddƒ}|ioDg}|D],}|tidti|iƒd|ƒq|~}n|Snd p td ‚dS( NtsexprR t blanklineRtstart_res^\(s (?m)^%s.*tisbad block type( Rtread_sexpr_blockRtread_blankline_blocktread_regexp_blocktretsubtescapetAssertionError(R tstreamttokst_[1]ttok((s6/p/zhu/06/nlp/nltk/nltk/corpus/reader/bracket_parse.pyt _read_block0s =cCsUti|ƒo|iƒdd!}ntidd|ƒ}tidd|ƒ}|S(Niiÿÿÿÿs\((.)\)s(\1 \1)s"\(([^\s()]+) ([^\s()]+) [^\s()]+\)s(\1 \2)(tEMPTY_BRACKETStmatchtstripRR(R tt((s6/p/zhu/06/nlp/nltk/nltk/corpus/reader/bracket_parse.pyt _normalizeAs c Csæyt|i|ƒƒSWnÈtj o¼}tiidƒ|id jomxjtddƒD]U}y9t|i|d|ƒƒ}tiid|ƒ|SWqZtj oqZXqZWntiidƒtd|i |ƒƒSnXdS( Ns(Bad tree detected; trying to recover... smismatched parensiit)s( Recovered by adding %d close paren(s) s' Recovered by returning a flat parse. tS(smismatched parens( RR%t ValueErrortsyststderrtwritetargstrangeRt_tag(R R$tetntv((s6/p/zhu/06/nlp/nltk/nltk/corpus/reader/bracket_parse.pyt_parseLs cCs„g}ti|i|ƒƒD]\}}|||fq~}|o:g}|D]"\}}|||i|ƒfqQ~}n|S(N(tTAGWORDtfindallR%R (R R$t simplify_tagsRtwt tagged_sentt_[2]((s6/p/zhu/06/nlp/nltk/nltk/corpus/reader/bracket_parse.pyR._s ?3cCsti|i|ƒƒS(N(tWORDR4R%(R R$((s6/p/zhu/06/nlp/nltk/nltk/corpus/reader/bracket_parse.pyt_wordfs( t__name__t __module__t__doc__tNoneRR R%R2tFalseR.R:(((s6/p/zhu/06/nlp/nltk/nltk/corpus/reader/bracket_parse.pyRs    tAlpinoCorpusReadercBs&eZdZeed„Zd„ZRS(s/ Reader for the Alpino Dutch Treebank. c Cs)ti||dddd|d|ƒdS(Ns alpino\.xmlRRRR(RR(R R RR((s6/p/zhu/06/nlp/nltk/nltk/corpus/reader/bracket_parse.pyRmscCs†|d djodSntidd|ƒ}tidd|ƒ}tidd |ƒ}tid d|ƒ}tid d|ƒ}|S( Ni s s(\1s- s(\1 \2)s R&s.*s(RR(R R$((s6/p/zhu/06/nlp/nltk/nltk/corpus/reader/bracket_parse.pyR%ss(R;R<R=R>RR%(((s6/p/zhu/06/nlp/nltk/nltk/corpus/reader/bracket_parse.pyR@is(R)t nltk.treeRRtutiltapiRtcompileR3R9R!tSyntaxCorpusReaderRR@(((s6/p/zhu/06/nlp/nltk/nltk/corpus/reader/bracket_parse.pys s   P