³ò dhKc @s·ddkZddkZddkZddkZddklZddkZddklZddk l Z ddk l Z defd„ƒYZ d„Zed jo eƒndS( iÿÿÿÿN(tadd(tParserI(tDependencyGraph(t find_binaryt MaltParsercBsSeZed„Zeed„Zed„Zed„Zed„Zed„Z RS(c Csj|iƒd|_t|_|dj o ||_n1tiidddddddddg ƒ|_dS(Nt malt_temps^-?[0-9]+(.[0-9]+)?$tCDs(The|the|A|a|An|an)$tATs.*able$tJJs.*ness$tNNs.*ly$tRBs.*s$tNNSs.*ing$tVBGs.*ed$tVBDs.*(s^-?[0-9]+(.[0-9]+)?$R(s(The|the|A|a|An|an)$sAT(s.*able$R(s.*ness$R (s.*ly$R (s.*s$R (s.*ing$R (s.*ed$R (s.*R ( t config_malttmcotFalset_trainedtNonettaggertnltkttagt RegexpTagger(tselfR((s%/p/zhu/06/nlp/nltk/nltk/parse/malt.pyt__init__s      c Csjddddddddg}tttti|ƒƒ}td|d |d d gd d d|ƒ|_dS(s“ Configure NLTK's interface to the C{malt} package. This searches for a directory containing the malt jar @param bin: The full path to the C{malt} binary. If not specified, then nltk will search the system for a C{malt} binary; and if one is not found, it will raise a C{LookupError} exception. @type bin: C{string} t.s/usr/lib/malt-1*s/usr/share/malt-1*s/usr/local/bins/usr/local/malt-1*s/usr/local/bin/malt-1*s/usr/local/share/malt-1*smalt.jart searchpathtenv_varstMALTPARSERHOMEturls/http://w3.msi.vxu.se/~jha/maltparser/index.htmltverboseN(treduceRtmaptglobRt _malt_bin(RtbinRt _malt_patht malt_path((s%/p/zhu/06/nlp/nltk/nltk/parse/malt.pyR*s  c Cs®|iptdƒ‚n|iptdƒ‚ntiitiƒdƒ}tiitiƒdƒ}d}|p'|dtiitiƒdƒ7}nd}zít |dƒ}xet |i i |i ƒƒƒD]E\}\}} |id |d |d | | d d d d d f ƒqÕW|idƒ|iƒdd|idtiƒd|id|d|dg} |i| d|ƒti|ƒSWd|o|iƒnXdS(så Use MaltParser to parse a sentence @param sentence: Input sentence to parse @type sentence: L{str} @return: C{DependencyGraph} the dependency graph representation of the sentence sAMaltParser location is not configured. Call config_malt() first.s1Parser has not been trained. Call train() first.smalt_input.conllsmalt_output.conlls-java -jar %s -w %s -c %s -i %s -o %s -m parses > smalt.outtws%s %s %s %s %s %s %s %s %s %s it_t0tas tjavas-jar %ss-w %ss-c %ss-i %ss-o %ss-m parsetparseN(R"t ExceptionRtostpathtjointtempfilet gettempdirRtopent enumerateRRtsplittwritetcloseRt_executeRtload( RtsentenceRt input_filet output_filetexecute_stringtftitwordRtcmd((s%/p/zhu/06/nlp/nltk/nltk/parse/malt.pyR+Js.  ' .  !c Csžtiitiƒdƒ}d}zMt|dƒ}|idig}|D]}||idƒqJ~ƒƒWd|o|i ƒnX|i |d|ƒdS(sŸ Train MaltParser from a list of C{DependencyGraph}s @param depgraphs: C{list} of C{DependencyGraph}s for training input data smalt_train.conllR&s i NR( R-R.R/R0R1RR2R5tto_conllR6ttrain_from_file(Rt depgraphsRR:R=t_[1]tdg((s%/p/zhu/06/nlp/nltk/nltk/parse/malt.pyttrainps>cCsá|iptdƒ‚nd }t|dƒo]|}tiitiƒdƒ}|i ƒi ƒ}t |dƒ}|i |ƒ|i ƒndd|idtiƒd|i d |d g}|i|d |ƒt|_d S( s… Train MaltParser from a file @param conll_file: C{str} for the filename of the training input data sAMaltParser location is not configured. Call config_malt() first.tzipfilesmalt_train.conllR&R*s-jar %ss-w %ss-c %ss-i %ss-m learnRFN(R"R,RthasattrR-R.R/R0R1R2treadR5R6RR7tTrueR(Rt conll_fileRR=tzip_conll_filet conll_strR@((s%/p/zhu/06/nlp/nltk/nltk/parse/malt.pyRBs  cCs]|p:tiitiƒdƒ}|id||fdƒntidi|ƒƒ}dS(Nts! > %smalt_%s.out 2> %smalt_%s.errit (R-R.R/R0R1tappendtsystem(RR@ttypeRttemp_dirt malt_exit((s%/p/zhu/06/nlp/nltk/nltk/parse/malt.pyR7 s( t__name__t __module__RRRRR+RFRBR7(((s%/p/zhu/06/nlp/nltk/nltk/parse/malt.pyRs   &  cCs„tdƒ}tdƒ}t}tƒ}|i||gd|ƒ|idd|ƒiƒiƒGH|idd|ƒiƒiƒGHdS(NsI1 John _ NNP _ _ 2 SUBJ _ _ 2 sees _ VB _ _ 0 ROOT _ _ 3 a _ DT _ _ 4 SPEC _ _ 4 dog _ NN _ _ 2 OBJ _ _ s£1 John _ NNP _ _ 2 SUBJ _ _ 2 walks _ VB _ _ 0 ROOT _ _ RsJohn sees Marys a man runs(RRRRFR+ttreetpprint(tdg1tdg2Rt maltParser((s%/p/zhu/06/nlp/nltk/nltk/parse/malt.pytdemo§s    t__main__(R-R0t subprocessR!toperatorRRtapiRtdependencygraphRtnltk.internalsRRR\RU(((s%/p/zhu/06/nlp/nltk/nltk/parse/malt.pys s     “