Of5%SrSSKrSSKJr SSKJrJrJrJr SSK J r SSK J r \R"S5r\R"S5r\R"S 5r\R"S \R"5r\R"S \R"5r\R"S \R"5r\R"S 5r"SS5r"SS5rSrSrSrSrSrSSjrSSjr\ "S5S5r\"S5r g)z CCG Lexicons N) defaultdict)CCGVar DirectionFunctionalCategoryPrimitiveCategory) deprecated) Expressionz([A-Za-z]+)(\[[A-Za-z,]+\])?z"([A-Za-z]+(?:\[[A-Za-z,]+\])?)(.*)z([\\/])([.,]?)([.,]?)(.*)z([\S_]+)\s*(::|[-=]+>)\s*(.+)z([^{}]*[^ {}])\s*(\{[^}]+\})?z \{([^}]+)\}z([^#]*)(?:#.*)?c:\rSrSrSrS SjrSrSrSrSr S r g) Token-z Class representing a token. token => category {semantics} e.g. eat => S\var[pl]/var {\x y.eat(x,y)} * `token` (string) * `categ` (string) * `semantics` (Expression) Nc(XlX lX0lgN)_token_categ _semantics)selftokencateg semanticss 2/usr/lib/python3/dist-packages/nltk/ccg/lexicon.py__init__Token.__init__9s  #cUR$r)rrs rr Token.categ>s {{rcUR$r)rrs rrToken.semanticsAs rcSnURbS[UR5-S-nS[UR5-U-$)Nz {})rstrr)r semantics_strs r__str__ Token.__str__DsB ?? & 3t#77#=MC $$}44rc[U[5(dg[URUR4UR 5UR 55$)N) isinstancer cmprrrr)rothers r__cmp__ Token.__cmp__Js;%''DKK15;;=%//BSTTr)rrrr) __name__ __module__ __qualname____firstlineno____doc__rrrr$r+__static_attributes__rrr r -s! $ 5 Urr c0\rSrSrSrSrSrSrSrSr g) CCGLexiconPz Class representing a lexicon for CCG grammars. * `primitives`: The list of primitive categories for the lexicon * `families`: Families of categories * `entries`: A mapping of words to possible categories cH[U5UlX lX0lX@lgr)r_start _primitives _families_entries)rstart primitivesfamiliesentriess rrCCGLexicon.__init__Ys'. %! rc URU$)z0 Returns all the possible categories for a word )r;)rwords r categoriesCCGLexicon.categories_s}}T""rcUR$)z+ Return the target category for the parser )r8rs rr<CCGLexicon.startes{{rcSnSn[UR5HEnU(dUS-nX-S-nSnURUHnU(dUS-nOSnUSU--nM MG U$)z; String representation of the lexicon. Used for debugging. r T z => z | Fz%s)sortedr;)rstringfirstidentcats rr$CCGLexicon.__str__kswDMM*E$^f,FE}}U+#e^F!E$*, , + r)r;r:r9r8N) r-r.r/r0r1rrCr<r$r2r3rrr5r5Ps #  rr5cNUSSnSnUS:wahURS5(dRURS5(a[U5up1X#-nO X!S-nUSSnUS:waURS5(dMRURS5(a US-USS4$[SU-S-5e) zV Separate the contents matching the first set of brackets from the rest of the input. N(r )rzUnmatched bracket in string '') startswith matchBracketsAssertionError)rJrestinsideparts rrUrUs !":D F "*T__S11 ??3  (.LT]F1g%F8D "*T__S11 s d12h'' 86ACG HHrcURS5(a [U5$[RU5R 5$)zV Separate the string for the next portion of the category from the rest of the string rQ)rTrU NEXTPRIM_REmatchgroups)rJs r nextCategoryr^s: V$$   V $ + + --rc&[USUSS5$)z Parse an application operator rrPN)r)apps rparseApplicationras SVSW %%rc<U(aUSSRS5$/$)z/ Parse the subscripts for a primitive category rPr',)split)subscrs rparseSubscriptsrfs#a|!!#&& IrcUSS:XaUScUc [5nX34$USnXB;a$X$upVUcUnXS4$URXc4/5nXS4$XA;a[US5n[XG5U4$[ SU-S-5e)zs Parse a primitive category If the primitive is the special category 'var', replace it with the correct `CCGVar`. rvarrPzString 'z-' is neither a family nor primitive category.)r substituterfrrV)chunksr=r>rhcatstrrMcvarsubscrss rparsePrimitiveCategoryrnsayE !9 {h:  AYF &  ;Cz..4+/Cz !&),!&2C88 VMM rc@[U5upEURS5(a[USSXU5upcO0[[R U5R 5XU5upcUS:wa[R U5R 5n[USS5nUSn[U5upEURS5(a[USSXU5upO0[[R U5R 5XU5up[XiU5nUS:waMXc4$)zo Parse a string representing a category, and returns a tuple with (possibly) the CCG variable for the category rQrPr'r r) r^rTaugParseCategoryrnPRIM_REr\r]APP_RErar) liner=r>rh cat_stringrWresr` directionargs rrqrqs &d+ZS!!%j2&6 cR c, MM* % , , . c  "*ll4 '')$S1X. 1v)$/   % %)*Qr*:JRUVJS#/ j)002J#JS!95 "* :rc[R"5 /n0n[[5nUR 5GHqn[ R U5R5SR5nUS:XaM@URS5(aEUUSSR5RS5Vs/sHofR5PM sn-nM[R U5R5upxn [R U 5R5up[XU5upUS:XaX4X7'GMSnUSLaKU c[US -5e[R "["R U 5R5S5nXGR%['X|U55 GMt [)USX#U5$s snf) z8 Convert string representation into a lexicon for CCGs. rr z:-Nrcz::Tz@ must contain semantics because include_semantics is set to True)rreset_idrlist splitlines COMMENTS_REr\r]striprTrdLEX_RERHS_RErqrVr fromstring SEMANTICS_REappendr r5)lex_strinclude_semanticsr=r>r?rtprimrLseprhsrkr#rMrhrs rrrs OOJH$G""$  &--/288: 2:  ??4 $)-ab)9)?)?)D')D )D'J !' T 2 9 9 ; U&,ll3&7&>&>&@ #V)&hGJSd{$'* $,$,, `a %/$9$9(..}=DDFqI% %%eE &BCK%L jmZ7 CC9's4F?zUse fromstring() instead.c[U5$r)r)rs r parseLexiconr"s g ra # Rather minimal lexicon based on the openccg `tinytiny' grammar. # Only incorporates a subset of the morphological subcategories, however. :- S,NP,N # Primitive categories Det :: NP/N # Determiners Pro :: NP IntransVsg :: S\NP[sg] # Tensed intransitive verbs (singular) IntransVpl :: S\NP[pl] # Plural TransVsg :: S\NP[sg]/NP # Tensed transitive verbs (singular) TransVpl :: S\NP[pl]/NP # Plural the => NP[sg]/N[sg] the => NP[pl]/N[pl] I => Pro me => Pro we => Pro us => Pro book => N[sg] books => N[pl] peach => N[sg] peaches => N[pl] policeman => N[sg] policemen => N[pl] boy => N[sg] boys => N[pl] sleep => IntransVsg sleep => IntransVpl eat => IntransVpl eat => TransVpl eats => IntransVsg eats => TransVsg see => TransVpl sees => TransVsg r)F)!r1re collectionsr nltk.ccg.apirrrrnltk.internalsrnltk.sem.logicr compilerrr[rsUNICODErrrr~r r5rUr^rarfrnrqrropenccg_tinytinyr3rrrs #QQ%% **8 9jjBC  4 5 8"** E 8"** Ezz,bjj9 jj./  U UF--jI(.&<@.Db '())+r