ó ů`]c@s›dZddlmZddlZddlmZddlmZmZm Z m Z ddl m Z ddl mZddlmZejd ƒZejd ƒZejd ƒZejd ejƒZejd ejƒZejdejƒZejdƒZdefd„ƒYZe defd„ƒYƒZd„Zd„Zd„Zd„Z d„Z!dd„Z#e$d„Z%edƒd„ƒZ&e%dƒZ'dS(u CCG Lexicons i˙˙˙˙(tunicode_literalsN(t defaultdict(tPrimitiveCategoryt DirectiontCCGVartFunctionalCategory(tpython_2_unicode_compatible(t deprecated(t Expressionu([A-Za-z]+)(\[[A-Za-z,]+\])?u"([A-Za-z]+(?:\[[A-Za-z,]+\])?)(.*)u([\\/])([.,]?)([.,]?)(.*)u([\S_]+)\s*(::|[-=]+>)\s*(.+)u([^{}]*[^ {}])\s*(\{[^}]+\})?u \{([^}]+)\}u([^#]*)(?:#.*)?tTokencBs>eZdZdd„Zd„Zd„Zd„Zd„ZRS(uÄ 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) cCs||_||_||_dS(N(t_tokent_categt _semantics(tselfttokentcategt semantics((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pyt__init__=s  cCs|jS(N(R (R ((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pyRBscCs|jS(N(R (R ((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pyREscCsDd}|jdk r/dt|jƒd}ndt|jƒ|S(Nuu {u}(R tNonetstrR (R t semantics_str((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pyt__str__HscCs;t|tƒsdSt|j|jf|jƒ|jƒƒS(Ni˙˙˙˙(t isinstanceR tcmpR R RR(R tother((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pyt__cmp__NsN( t__name__t __module__t__doc__RRRRRR(((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pyR 1s      t CCGLexiconcBs2eZdZd„Zd„Zd„Zd„ZRS(uâ 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 cCs.t|ƒ|_||_||_||_dS(N(Rt_startt _primitivest _familiest_entries(R tstartt primitivestfamiliestentries((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pyR^s  cCs |j|S(u@ Returns all the possible categories for a word (R!(R tword((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pyt categoriesdscCs|jS(u; Return the target category for the parser (R(R ((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pyR"jscCsd}t}x}t|jƒD]l}|s5|d}n||d}t}x<|j|D]-}|sp|d}nt}|d|}qWWqW|S(uK String representation of the lexicon. Used for debugging. uu u => u | u%s(tTruetsortedR!tFalse(R tstringtfirsttidenttcat((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pyRps  (RRRRR'R"R(((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pyRTs    cCs˛|d}d}xf|dkrx|jdƒ rx|jdƒr]t|ƒ\}}||}q||d}|d}qW|jdƒrš|d|dfStd|dƒ‚dS( ub Separate the contents matching the first set of brackets from the rest of the input. iu(uu)iuUnmatched bracket in string 'u'N(t startswitht matchBracketstAssertionError(R+tresttinsidetpart((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pyR0Šs  cCs,|jdƒrt|ƒStj|ƒjƒS(ub Separate the string for the next portion of the category from the rest of the string u((R/R0t NEXTPRIM_REtmatchtgroups(R+((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pyt nextCategoryžs cCst|d|dƒS(u' Parse an application operator ii(R(tapp((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pytparseApplication¨scCs|r|dd!jdƒSgS(u7 Parse the subscripts for a primitive category ii˙˙˙˙u,(tsplit(tsubscr((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pytparseSubscriptsŻscCsé|ddkrE|ddkrE|dkr8tƒ}n||fSn|d}||kr˘||\}}|dkr€|}n|j||fgƒ}||fS||krŃt|dƒ}t||ƒ|fStd|dƒ‚dS(uƒ Parse a primitive category If the primitive is the special category 'var', replace it with the correct `CCGVar`. iuvariuString 'u-' is neither a family nor primitive category.N(RRt substituteR=RR1(tchunksR#R$tvartcatstrR.tcvartsubscrs((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pytparsePrimitiveCategory¸s"         c CsAt|ƒ\}}|jdƒrFt|dd!|||ƒ\}}n*ttj|ƒjƒ|||ƒ\}}xÄ|dkr6tj|ƒjƒ}t|dd!ƒ}|d}t|ƒ\}}|jdƒr÷t|dd!|||ƒ\} }n*ttj|ƒjƒ|||ƒ\} }t || |ƒ}qsW||fS(u{ Parse a string representing a category, and returns a tuple with (possibly) the CCG variable for the category u(ii˙˙˙˙uii( R8R/taugParseCategoryRDtPRIM_RER6R7tAPP_RER:R( tlineR#R$R@t cat_stringR2tresR9t directiontarg((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pyREÖs %' %'cCsŹtjƒg}i}ttƒ}xp|jƒD]b}tj|ƒjƒdjƒ}|dkrfq/n|j dƒrŽ|g|djƒj dƒD]}|jƒ^q’}q/t j|ƒjƒ\}}} t j| ƒjƒ\} } t | ||ƒ\} } |dkr| | f|| 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 ((Rt __future__Rtret collectionsRt nltk.ccg.apiRRRRt nltk.compatRtnltk.internalsRtnltk.sem.logicRtcompileRFR5RGtUNICODERRRSRURPtobjectR RR0R8R:R=RDRRER*RTR\topenccg_tinytiny(((s/lib/python2.7/site-packages/nltk/ccg/lexicon.pyt s6 "#5    ! 1*