o rZh`@s@dZddlZddlZddlmZddlmZGdddeZdS)z An NLTK interface to the VerbNet verb lexicon For details about VerbNet see: https://verbs.colorado.edu/~mpalmer/projects/verbnet.html N) defaultdict)XMLCorpusReaderc@s0eZdZdZd@ddZedZ edZ edZ dAd d Z dAd d Z dBd dZ ddZ dAddZddZddZddZddZddZddZdd Zd!d"Zd#d$Zd%d&Zd'd(Zd)d*Zd+d,ZdCd.d/ZdCd0d1ZdCd2d3ZdCd4d5ZdCd6d7Z dCd8d9Z!dCd:d;Z"dCdd?Z$dS)DVerbnetCorpusReadera An NLTK interface to the VerbNet verb lexicon. From the VerbNet site: "VerbNet (VN) (Kipper-Schuler 2006) is the largest on-line verb lexicon currently available for English. It is a hierarchical domain-independent, broad-coverage verb lexicon with mappings to other lexical resources such as WordNet (Miller, 1990; Fellbaum, 1998), XTAG (XTAG Research Group, 2001), and FrameNet (Baker et al., 1998)." For details about VerbNet see: https://verbs.colorado.edu/~mpalmer/projects/verbnet.html FcCsBt||||tt|_ tt|_ i|_ i|_|dSN) r__init__rlist_lemma_to_class_wordnet_to_class_class_to_fileid_shortid_to_longid _quick_index)selfrootfileidsZ wrap_etreerI/var/www/auris/lib/python3.10/site-packages/nltk/corpus/reader/verbnet.pyr%s   zVerbnetCorpusReader.__init__z([^\-\.]*)-([\d+.\-]+)$z [\d+.\-]+$zH]+>|NcCs>|dur t|jSt|tr||}dd|dDS)zx Return a list of all verb lemmas that appear in any class, or in the ``classid`` if specified. NcSg|]}|dqS)nameget.0memberrrr Ssz.VerbnetCorpusReader.lemmas..MEMBERS/MEMBER)sortedrkeys isinstancestrvnclassfindallr rrrrlemmasHs   zVerbnetCorpusReader.lemmascCsD|dur t|jSt|tr||}tdd|dDgS)z| Return a list of all wordnet identifiers that appear in any class, or in ``classid`` if specified. Ncss |] }|ddVqdS)wnN)rsplitrrrr as  z1VerbnetCorpusReader.wordnetids..r)rr rrrrsumr r!rrr wordnetidsUs  zVerbnetCorpusReader.wordnetidscsxdurfdd|jDS|dur|j|S|dur"|j|S|dur5||}dd|dDSt|jS)a Return a list of the VerbNet class identifiers. If a file identifier is specified, then return only the VerbNet class identifiers for classes (and subclasses) defined by that file. If a lemma is specified, then return only VerbNet class identifiers for classes that contain that lemma as a member. If a wordnetid is specified, then return only identifiers for classes that contain that wordnetid as a member. If a classid is specified, then return only identifiers for subclasses of the specified VerbNet class. If nothing is specified, return all classids within VerbNet Ncsg|] \}}|kr|qSrr)rcffileidrrrvsz0VerbnetCorpusReader.classids..cSrIDrrsubclassrrrr}sSUBCLASSES/VNSUBCLASS)r itemsrr rr rr)r ZlemmaZ wordnetidr,classidxmltreerr+rclassidshs    zVerbnetCorpusReader.classidscCs||jvr ||S||}||jvr?|j||}||}||dkr*|S|dD] }||dkr<|Sq/Jtd|)aReturns VerbNet class ElementTree Return an ElementTree containing the xml for the specified VerbNet class. :param fileid_or_classid: An identifier specifying which class should be returned. Can be a file identifier (such as ``'put-9.1.xml'``), or a VerbNet class identifier (such as ``'put-9.1'``) or a short VerbNet class identifier (such as ``'9.1'``). r.z .//VNSUBCLASSFzUnknown identifier )_fileidsxmllongidr rr ValueError)r Zfileid_or_classidr3r,treer0rrrrs    zVerbnetCorpusReader.vnclasscs<|durjSt|trj|gSfdd|DS)z Return a list of fileids that make up this corpus. If ``vnclass_ids`` is specified, then return the fileids that make up the specified VerbNet class(es). Ncsg|] }j|qSr)r r8)rZ vnclass_idr rrrz/VerbnetCorpusReader.fileids..)r6rrr r8)r Z vnclass_idsrr;rrs  zVerbnetCorpusReader.fileidsc Cs\t|tr ||}g}|d}|D]}|||||||||dq|S)apGiven a VerbNet class, this method returns VerbNet frames The members returned are: 1) Example 2) Description 3) Syntax 4) Semantics :param vnclass: A VerbNet class identifier; or an ElementTree containing the xml contents of a VerbNet class. :return: frames - a list of frame dictionaries z FRAMES/FRAME)example descriptionsyntax semantics) rrrr append_get_example_within_frame_get_description_within_frame _get_syntactic_list_within_frame_get_semantics_within_frame)r rframesZvnframesvnframerrrrFs  zVerbnetCorpusReader.framescCs,t|tr ||}dd|dD}|S)aAReturns subclass ids, if any exist Given a VerbNet class, this method returns subclass ids (if they exist) in a list of strings. :param vnclass: A VerbNet class identifier; or an ElementTree containing the xml contents of a VerbNet class. :return: list of subclasses cSrr-rr/rrrrs z2VerbnetCorpusReader.subclasses..r1)rrrr )r r subclassesrrrrHs zVerbnetCorpusReader.subclassescCsRt|tr ||}g}|dD]}||ddd|dDdq|S)abReturns thematic roles participating in a VerbNet class Members returned as part of roles are- 1) Type 2) Modifiers :param vnclass: A VerbNet class identifier; or an ElementTree containing the xml contents of a VerbNet class. :return: themroles: A list of thematic roles in the VerbNet class zTHEMROLES/THEMROLEtypecS"g|] }|d|ddqSValuerIvaluerIrrrestrrrrrz1VerbnetCorpusReader.themroles..SELRESTRS/SELRESTR)rI modifiers)rrrr rAr)r r themrolesZtrolerrrrTs  zVerbnetCorpusReader.themrolescCs"|jD] }||||qdS)aC Initialize the indexes ``_lemma_to_class``, ``_wordnet_to_class``, and ``_class_to_fileid`` by scanning through the corpus fileids. This is fast if ElementTree uses the C implementation (<0.1 secs), but quite slow (>10 secs) if only the python implementation is available. N)r6 _index_helperr7)r r,rrr_indexs zVerbnetCorpusReader._indexcCs|d}||j|<||j||<|dD] }|j|d||ddD] }|j||q,q|dD]}| ||q=dS)zHelper for ``_index()``r.rrr#r$r1N) rr r shortidr rrAr%r rU)r r4r,rrr#r0rrrrU s  z!VerbnetCorpusReader._index_helperc Cs|jD]x}|dd}||j|<||j||<||V}|j|D]E}|}|ddurN|j |d ||d D] }|j | |qBq&|ddurh||j|d<|d}||j||<q&JdWdn1svwYqdS)a Initialize the indexes ``_lemma_to_class``, ``_wordnet_to_class``, and ``_class_to_fileid`` by scanning through the corpus fileids. This doesn't do proper xml parsing, but is good enough to find everything in the standard VerbNet corpus -- and it runs about 30 times faster than xml parsing (with the python ElementTree; only 2-3 times faster if ElementTree uses the C implementation). NrFzunexpected match condition) r6r r rWopen _INDEX_REfinditerreadgroupsrrAr%r )r r,rfpmr_r#rrrr s*     z VerbnetCorpusReader._quick_indexc Cs\|j|r|S|j|std|z|j|WSty-}ztd||d}~ww)zReturns longid of a VerbNet class Given a short VerbNet class identifier (eg '37.10'), map it to a long id (eg 'confess-37.10'). If ``shortid`` is already a long id, then return it as-isvnclass identifier %r not foundN) _LONGID_REmatch _SHORTID_REr9r KeyError)r rWerrrr8;s    zVerbnetCorpusReader.longidcCs6|j|r|S|j|}|r|dStd|)zReturns shortid of a VerbNet class Given a long VerbNet class identifier (eg 'confess-37.10'), map it to a short id (eg '37.10'). If ``longid`` is already a short id, then return it as-is.rZrb)rerdrcgroupr9)r r8rarrrrWJs    zVerbnetCorpusReader.shortidcCsNg}|dD]}dd|dD}||d||ddkdq|S) anReturns semantics within a single frame A utility function to retrieve semantics within a frame in VerbNet Members of the semantics dictionary: 1) Predicate value 2) Arguments :param vnframe: An ElementTree containing the xml contents of a VerbNet frame. :return: semantics: semantics dictionary zSEMANTICS/PREDcSrJ)rIrN)rIrNr)rargrrrrjrQzCVerbnetCorpusReader._get_semantics_within_frame..zARGS/ARGrNbool!)predicate_value argumentsnegated)r rAr)r rGZsemantics_within_single_framepredrmrrrrE\s  z/VerbnetCorpusReader._get_semantics_within_framecCs$|d}|dur|j}|Sd}|S)a'Returns example within a frame A utility function to retrieve an example within a frame in VerbNet. :param vnframe: An ElementTree containing the xml contents of a VerbNet frame. :return: example_text: The example sentence for this particular frame zEXAMPLES/EXAMPLENr$)findtext)r rGZexample_elementZ example_textrrrrBws z-VerbnetCorpusReader._get_example_within_framecCs"|d}|jd|dddS)abReturns member description within frame A utility function to retrieve a description of participating members within a frame in VerbNet. :param vnframe: An ElementTree containing the xml contents of a VerbNet frame. :return: description: a description dictionary with members - primary and secondary Z DESCRIPTIONprimary secondaryr$)rrrs)rpattribr)r rGZdescription_elementrrrrCs  z1VerbnetCorpusReader._get_description_within_framecCsg}|dD]6}|j}t}d|jvr|dnd|d<dd|dD|d<dd|d D|d <|||d q|S) a[Returns semantics within a frame A utility function to retrieve semantics within a frame in VerbNet. Members of the syntactic dictionary: 1) POS Tag 2) Modifiers :param vnframe: An ElementTree containing the xml contents of a VerbNet frame. :return: syntax_within_single_frame ZSYNTAXrNr$cSrJrKrrOrrrrrQzHVerbnetCorpusReader._get_syntactic_list_within_frame..rR selrestrscSrJrKrrOrrrrrQzSYNRESTRS/SYNRESTR synrestrs)pos_tagrS)rptagdictrtrr rA)r rGZsyntax_within_single_frameeltrwrSrrrrDs   z4VerbnetCorpusReader._get_syntactic_list_within_framecCst|tr ||}|dd}||j|ddd7}||j|ddd7}|d7}||j|ddd7}|d7}||j|dd7}|S)a%Returns pretty printed version of a VerbNet class Return a string containing a pretty-printed representation of the given VerbNet class. :param vnclass: A VerbNet class identifier; or an ElementTree containing the xml contents of a VerbNet class. r.  indentz Thematic roles:  z Frames: )rrrrpprint_subclassespprint_memberspprint_themroles pprint_frames)r rsrrrpprints zVerbnetCorpusReader.pprintr$cCLt|tr ||}||}|sdg}dd|}tj|d||ddS)a>Returns pretty printed version of subclasses of VerbNet class Return a string containing a pretty-printed representation of the given VerbNet class's subclasses. :param vnclass: A VerbNet class identifier; or an ElementTree containing the xml contents of a VerbNet class. (none)z Subclasses:  Fr|initial_indentsubsequent_indent)rrrrHjointextwrapfill)r rr~rHrrrrr   z%VerbnetCorpusReader.pprint_subclassescCr)a?Returns pretty printed version of members in a VerbNet class Return a string containing a pretty-printed representation of the given VerbNet class's member verbs. :param vnclass: A VerbNet class identifier; or an ElementTree containing the xml contents of a VerbNet class. rz Members: rrr|r)rrrr"rrr)r rr~membersrrrrrrz"VerbnetCorpusReader.pprint_memberscCsxt|tr ||}g}||D]%}|d|d}dd|dD}|r1|dd|7}||qd|S) aHReturns pretty printed version of thematic roles in a VerbNet class Return a string containing a pretty-printed representation of the given VerbNet class's thematic roles. :param vnclass: A VerbNet class identifier; or an ElementTree containing the xml contents of a VerbNet class. * rIcSsg|] }|d|dqSrMr)rmodifierrrrrr<z8VerbnetCorpusReader.pprint_themroles..rS[{}]rr{)rrrrTrformatrrA)r rr~piecesZthemrolepiecerSrrrrs   z$VerbnetCorpusReader.pprint_themrolescCsDt|tr ||}g}||D] }||||qd|S)a?Returns pretty version of all frames in a VerbNet class Return a string containing a pretty-printed representation of the list of frames within the VerbNet class. :param vnclass: A VerbNet class identifier; or an ElementTree containing the xml contents of a VerbNet class. r{)rrrrFrA_pprint_single_framer)r rr~rrGrrrrs  z!VerbnetCorpusReader.pprint_framescCsd|||d}||||dd7}||||dd7}||d7}||||d7}|S)aReturns pretty printed version of a single frame in a VerbNet class Returns a string containing a pretty-printed representation of the given frame. :param vnframe: An ElementTree containing the xml contents of a VerbNet frame. r{rz Syntax: z Semantics: r) _pprint_description_within_frame_pprint_example_within_frame_pprint_syntax_within_frame_pprint_semantics_within_frame)r rGr~Z frame_stringrrrrs  z(VerbnetCorpusReader._pprint_single_framecCs|dr |d|dSdS)a&Returns pretty printed version of example within frame in a VerbNet class Return a string containing a pretty-printed representation of the given VerbNet frame example. :param vnframe: An ElementTree containing the xml contents of a Verbnet frame. r=z Example: Nr)r rGr~rrrr0s z0VerbnetCorpusReader._pprint_example_within_framecCs6||dd}|ddr|d|dd7}|S)aReturns pretty printed version of a VerbNet frame description Return a string containing a pretty-printed representation of the given VerbNet frame description. :param vnframe: An ElementTree containing the xml contents of a VerbNet frame. r>rrrsz ({})r)r rGr~r>rrrr<s z4VerbnetCorpusReader._pprint_description_within_framecCsg}|dD]A}|d}g}d|dvr#|ddr#||dd|dd|dd|ddD7}|rB|d d |7}||q|d |S) a&Returns pretty printed version of syntax within a frame in a VerbNet class Return a string containing a pretty-printed representation of the given VerbNet frame syntax. :param vnframe: An ElementTree containing the xml contents of a VerbNet frame. r?rwrNrScSs g|] }d|d|dqS)z{}{}rNrIrrOrrrrYszCVerbnetCorpusReader._pprint_syntax_within_frame..rurvrr)rArr)r rGr~relementrZ modifier_listrrrrJs    z/VerbnetCorpusReader._pprint_syntax_within_framec slg}|dD]#}dd|dD}||drdnd|dd d |d qd fd d|DS)a,Returns a pretty printed version of semantics within frame in a VerbNet class Return a string containing a pretty-printed representation of the given VerbNet frame semantics. :param vnframe: An ElementTree containing the xml contents of a VerbNet frame. r@cSsg|]}|dqS)rNr)rargumentrrrrqszFVerbnetCorpusReader._pprint_semantics_within_frame..rmrn¬r$rl(z, )r{c3s|] }d|VqdS)rNr)rrr}rrr&uszEVerbnetCorpusReader._pprint_semantics_within_frame..)rAr)r rGr~r predicatermrr}rrfs (z2VerbnetCorpusReader._pprint_semantics_within_frame)Fr)NNNN)r$)%__name__ __module__ __qualname____doc__rrecompilercrer\r"r(r5rrrFrHrTrVrUr r8rWrErBrCrDrrrrrrrrrrrrrrrsL      !  ""       r)rrr collectionsrZnltk.corpus.reader.xmldocsrrrrrrs