³ò B_Kc @sêddkZddkZddklZddklZddkTddkTdefd„ƒYZ de fd„ƒYZ d e fd „ƒYZ d e fd „ƒYZ d e fd„ƒYZde fd„ƒYZde fd„ƒYZdS(iÿÿÿÿN(tTree(t ElementTree(t*tPropbankCorpusReadercBs_eZdZdeeeed„Zed„Zd„Zd„Zd„Zd„Z d„Z RS( s— Corpus reader for the propbank corpus, which augments the Penn Treebank with information about the predicate argument structure of every verb instance. The corpus consists of two parts: the predicate-argument annotations themselves, and a set of X{frameset files} which define the argument labels used by the annotations, on a per-verb basis. Each X{frameset file} contains one or more predicates, such as C{'turn'} or C{'turn_on'}, each of which is divided into coarse-grained word senses called X{rolesets}. For each X{roleset}, the frameset file provides descriptions of the argument roles, along with examples. tcCs€t|tƒot||ƒ}nt|ƒ}ti||||g||ƒ||_||_||_||_ ||_ dS(sÛ @param root: The root directory for this corpus. @param propfile: The name of the file containing the predicate- argument annotations (relative to C{root}). @param framefiles: A list or regexp specifying the frameset fileids for this corpus. @param parse_fileid_xform: A transform that should be applied to the fileids in this corpus. This should be a function of one argument (a fileid) that returns a string (the new fileid). @param parse_corpus: The corpus containing the parse trees corresponding to this corpus. These parse trees are necessary to resolve the tree pointers used by propbank. N( t isinstancet basestringtfind_corpus_fileidstlistt CorpusReadert__init__t _propfilet _framefilest _verbsfilet_parse_fileid_xformt _parse_corpus(tselftroottpropfilet framefilest verbsfiletparse_fileid_xformt parse_corpustencoding((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyR s     cCsk|tjo |i}nt|tƒo |g}ntg}|D]}||i|ƒiƒqE~ƒS(sV @return: the text contents of the given fileids, as a single string. (tNonet_fileidsRRtconcattopentread(Rtfileidst_[1]tf((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pytraw>scCs.t|i|iƒ|id|i|iƒƒS(s… @return: a corpus view that acts as a list of L{PropbankInstance} objects, one for each verb in the corpus. R(tStreamBackedCorpusViewtabspathR t_read_instance_blockR(R((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyt instancesFs cCs+t|i|iƒtd|i|iƒƒS(s @return: a corpus view that acts as a list of strings, one for each line in the predicate-argument annotation file. R(R!R"R tread_line_blockR(R((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pytlinesOscCsµ|idƒd}d|}||ijotd|ƒ‚nti|i|ƒiƒƒiƒ}xI|idƒD]"}|i d|jo|SququWtd||fƒ‚dS( sE @return: the xml description for the given roleset. t.is frames/%s.xmlsFrameset file for %s not foundspredicate/rolesettidsRoleset %s not found in %sN( tsplitR t ValueErrorRtparseR"Rtgetroottfindalltattrib(Rt roleset_idtlemmat framefiletetreetroleset((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyR3Xs $ cCs+t|i|iƒtd|i|iƒƒS(s @return: a corpus view that acts as a list of all verb lemmas in this corpus (from the verbs.txt file). R(R!R"R R%R(R((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pytverbslscCs`g}xStdƒD]E}|iƒiƒ}|o&|iti||i|iƒƒqqW|S(Nid(trangetreadlinetstriptappendtPropbankInstanceR+RR(Rtstreamtblocktitline((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyR#us   ( t__name__t __module__t__doc__RR R R$R&R3R4R#(((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRs    R9cBsVeZed„Zd„Zd„Zd„ZeeddƒZe eed„ƒZ RS(c Cs[||_||_||_||_||_||_||_t|ƒ|_| |_ dS(N( tfileidtsentnumtwordnumttaggerR3t inflectiont predicatettuplet argumentsR( RRARBRCRDR3RERFRHR((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyR ˆs        cCsd|i|i|ifS(Ns((RARBRC(R((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyt__repr__´scCs|d|i|i|i|i|i|if}|i|idff}x.t|ƒD] \}}|d||f7}qTW|S(Ns%s %s %s %s %s %strels %s-%s( RARBRCRDR3RERHRFtsorted(Rtstitemstargloctargid((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyt__str__¸s cCsS|idjodSn|i|iiƒjodSn|ii|iƒ|iS(N(RRRARt parsed_sentsRB(R((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyt _get_treeÀs!tdocsv The parse tree corresponding to this instance, or C{None} if the corresponding tree is not available.cCs°|iƒ}t|ƒdjotd|ƒ‚n|d \}}}}}} g} |dD]!} | idƒo | | q^q^~ } g} |dD]!} | idƒp | | q—q—~ }t| ƒdjotd|ƒ‚n|dj o||ƒ}nt|ƒ}t|ƒ}ti| ƒ} ti| dd ƒ}g}xB|D]:}|iddƒ\}}|i ti|ƒ|fƒqPWt |||||| |||ƒ S( Nis!Badly formatted propbank line: %ris-reliiiüÿÿÿt-( R)tlenR*tendswithRtinttPropbankInflectionR+tPropbankTreePointerR8R9(RLRRtpiecesRARBRCRDR3RERtpRJt_[2]targsRFRHtargRNRO((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyR+Ès, 99     ( R>R?RR RIRPRRtpropertyttreet staticmethodR+(((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyR9†s +     tPropbankPointercBseZdZd„ZRS(sn A pointer used by propbank to identify one or more constituents in a parse tree. C{PropbankPointer} is an abstract base class with three concrete subclasses: - L{PropbankTreePointer} is used to point to single constituents. - L{PropbankSplitTreePointer} is used to point to 'split' constituents, which consist of a sequence of two or more C{PropbankTreePointer}s. - L{PropbankChainTreePointer} is used to point to entire trace chains in a tree. It consists of a sequence of pieces, which can be C{PropbankTreePointer}s or C{PropbankSplitTreePointer}s. cCs$|itjotdƒ‚ndS(Ns)PropbankPointer is an abstract base class(t __class__tPropbankPoitnertAssertionError(R((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyR ýs(R>R?R@R (((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRbïs tPropbankChainTreePointercBs,eZd„Zd„Zd„Zd„ZRS(cCs ||_dS(N(RZ(RRZ((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyR s cCsdid„|iDƒƒS(NRcssx|]}d|VqWdS(s%sN((t.0R[((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pys s(tjoinRZ(R((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRPscCsd|S(Ns((R((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRI scCsQ|djotdƒ‚ntdg}|iD]}||i|ƒq1~ƒS(NsParse tree not avaialables*CHAIN*(RR*RRZtselect(RR`RR[((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRi s(R>R?R RPRIRi(((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRfs   tPropbankSplitTreePointercBs,eZd„Zd„Zd„Zd„ZRS(cCs ||_dS(N(RZ(RRZ((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyR s cCsdid„|iDƒƒS(Nt,cssx|]}d|VqWdS(s%sN((RgR[((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pys s(RhRZ(R((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRPscCsd|S(Ns((R((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRIscCsQ|djotdƒ‚ntdg}|iD]}||i|ƒq1~ƒS(NsParse tree not avaialables*SPLIT*(RR*RRZRi(RR`RR[((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRis(R>R?R RPRIRi(((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRjs   RYcBsSeZdZd„Zed„ƒZd„Zd„Zd„Zd„Z d„Z RS(sD wordnum:height*wordnum:height*... wordnum:height, cCs||_||_dS(N(RCtheight(RRCRl((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyR $s c Csÿ|idƒ}t|ƒdjo2tg}|D]}|ti|ƒq0~ƒSn|idƒ}t|ƒdjo2tg}|D]}|ti|ƒq„~ƒSn|idƒ}t|ƒdjotd|ƒ‚ntt|dƒt|dƒƒS(NRiRkt:isbad propbank pointer %ri(R)RURfRYR+RjR*RW(RLRZRteltR\((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyR+(s ( ('cCsd|i|ifS(Ns%s:%s(RCRl(R((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRP;scCsd|i|ifS(NsPropbankTreePointer(%d, %d)(RCRl(R((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRI>scCsx(t|ttfƒo|id}qWt|tƒptt|ƒt|ƒƒSnt|i|i f|i|i fƒS(Ni( RRfRjRZRYtcmpR(RCRl(Rtother((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyt__cmp__As  cCs.|djotdƒ‚n||i|ƒS(NsParse tree not avaialable(RR*ttreepos(RR`((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRiLscCs*|djotdƒ‚n|g}g}d}xñtoét|dtƒoŠt|ƒt|ƒjo|idƒn|dcd7<|dt|dƒjo|i|d|dƒq!|iƒ|iƒq5||ijo#t |t|ƒ|i d ƒSq5|d7}|iƒq5WdS(s} Convert this pointer to a standard 'tree position' pointer, given that it points to the given tree. sParse tree not avaialableiiÿÿÿÿiN( RR*tTrueRRRUR8tpopRCRGRl(RR`tstackRrRC((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRrPs"   # ( R>R?R@R RaR+RPRIRqRiRr(((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRYs    RXcBs¤eZdZdZdZdZdZdZdZdZ dZ dZ dZ d Z dZd Zd d d d d d „Zd „Zd „ZeidƒZed„ƒZRS(R<tgR[tvRtntotbt3taRTcCs1||_||_||_||_||_dS(N(tformttensetaspecttpersontvoice(RR}R~RR€R((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyR Šs     cCs#|i|i|i|i|iS(N(R}R~RR€R(R((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRP‘scCsd|S(Ns((R((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRI”ss"[igpv\-][fpn\-][pob\-][3\-][ap\-]$cCset|tƒptdƒ‚nt|ƒdjptii|ƒ otd|ƒ‚nt|ŒS(Nsexpected a stringis!Bad propbank inflection string %r(RRt TypeErrorRURXt _VALIDATEtmatchR*(RL((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyR+™s (R>R?t INFINITIVEtGERUNDt PARTICIPLEtFINITEtFUTUREtPASTtPRESENTtPERFECTt PROGRESSIVEtPERFECT_AND_PROGRESSIVEt THIRD_PERSONtACTIVEtPASSIVEtNONER RPRItretcompileRƒRaR+(((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyRXss&  (R“tcodecst nltk.treeRt nltk.etreeRtutiltapiR RtobjectR9RbRfRjRYRX(((s1/p/zhu/06/nlp/nltk/nltk/corpus/reader/propbank.pyss    uiU