3 ­d‡&ã@sdZddlZddlmZddlmZmZmZmZddl m Z ddl m Z ej dƒZej dƒZej d ƒZej d ejƒZej d ejƒZej d ejƒZej d ƒZGdd„dƒZGdd„dƒZdd„Zdd„Zdd„Zdd„Zdd„Zd%dd„Zd&dd „Ze d!ƒd"d#„ƒZed$ƒZ dS)'z CCG Lexicons éN)Ú defaultdict)ÚCCGVarÚ DirectionÚFunctionalCategoryÚPrimitiveCategory)Ú deprecated)Ú Expressionz([A-Za-z]+)(\[[A-Za-z,]+\])?z"([A-Za-z]+(?:\[[A-Za-z,]+\])?)(.*)z([\\/])([.,]?)([.,]?)(.*)z([\S_]+)\s*(::|[-=]+>)\s*(.+)z([^{}]*[^ {}])\s*(\{[^}]+\})?z \{([^}]+)\}z([^#]*)(?:#.*)?c@s:eZdZdZd dd„Zdd„Zdd„Zd d „Zd d „ZdS)ÚTokenzÄ 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) NcCs||_||_||_dS)N)Z_tokenÚ_categÚ _semantics)ÚselfÚtokenÚcategÚ semantics©rú0/tmp/pip-build-v9q4h5k9/nltk/nltk/ccg/lexicon.pyÚ__init__9szToken.__init__cCs|jS)N)r )r rrrr>sz Token.categcCs|jS)N)r )r rrrrAszToken.semanticscCs2d}|jdk r dt|jƒd}dt|jƒ|S)NÚz {Ú})r Ústrr )r Ú semantics_strrrrÚ__str__Ds z Token.__str__cCs*t|tƒsdSt|j|jf|jƒ|jƒƒS)Nééÿÿÿÿ)Ú isinstancer Úcmpr r rr)r ÚotherrrrÚ__cmp__Js z Token.__cmp__)N) Ú__name__Ú __module__Ú __qualname__Ú__doc__rrrrrrrrrr -s   r c@s0eZdZdZdd„Zdd„Zdd„Zdd „Zd S) Ú CCGLexiconzâ 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)rÚ_startZ _primitivesZ _familiesÚ_entries)r ÚstartÚ primitivesÚfamiliesÚentriesrrrrYs zCCGLexicon.__init__cCs |j|S)z@ Returns all the possible categories for a word )r$)r ÚwordrrrÚ categories_szCCGLexicon.categoriescCs|jS)z; Return the target category for the parser )r#)r rrrr%eszCCGLexicon.startcCsnd}d}x`t|jƒD]R}|s$|d}||d}d}x0|j|D]"}|sR|d}nd}|d|}q@WqW|S)zK String representation of the lexicon. Used for debugging. rTÚ z => z | Fz%s)Úsortedr$)r ÚstringÚfirstÚidentÚcatrrrrks  zCCGLexicon.__str__N)rrr r!rr*r%rrrrrr"Ps r"cCs”|dd…}d}xP|dkr`|jdƒ r`|jdƒrFt|ƒ\}}||}q||d}|dd…}qW|jdƒr€|d|dd…fStd|dƒ‚dS) zb Separate the contents matching the first set of brackets from the rest of the input. rNú(rú)rzUnmatched bracket in string 'ú')Ú startswithÚ matchBracketsÚAssertionError)r-ÚrestZinsideÚpartrrrr5…s      r5cCs |jdƒrt|ƒStj|ƒjƒS)zb Separate the string for the next portion of the category from the rest of the string r1)r4r5Ú NEXTPRIM_REÚmatchÚgroups)r-rrrÚ nextCategory™s r<cCst|d|dd…ƒS)z' Parse an application operator rrN)r)ÚapprrrÚparseApplication£sr>cCs|r|dd…jdƒSgS)z7 Parse the subscripts for a primitive category rú,r)Úsplit)ZsubscrrrrÚparseSubscriptsªsrAcCs¦|ddkr.|ddkr.|dkr&tƒ}||fS|d}||krp||\}}|dkrX|}n|j||fgƒ}||fS||kr’t|dƒ}t||ƒ|fStd|dƒ‚dS)zƒ Parse a primitive category If the primitive is the special category 'var', replace it with the correct `CCGVar`. rÚvarrNzString 'z-' is neither a family nor primitive category.)rÚ substituterArr6)Úchunksr&r'rBÚcatstrr0ZcvarZsubscrsrrrÚparsePrimitiveCategory³s"    rFc Csät|ƒ\}}|jdƒr2t|dd…|||ƒ\}}nttj|ƒjƒ|||ƒ\}}xŒ|dkrÚtj|ƒjƒ}t|dd…ƒ}|d}t|ƒ\}}|jdƒr°t|dd…|||ƒ\} }nttj|ƒjƒ|||ƒ\} }t || |ƒ}qPW||fS)z{ Parse a string representing a category, and returns a tuple with (possibly) the CCG variable for the category r1rrrérr) r<r4ÚaugParseCategoryrFÚPRIM_REr:r;ÚAPP_REr>r) Úliner&r'rBZ cat_stringr7Úresr=Ú directionÚargrrrrHÑs      rHFcCs(tjƒg}i}ttƒ}xü|jƒD]ð}tj|ƒjƒdjƒ}|dkrFq"|j dƒrv|dd„|dd…jƒj dƒDƒ}q"t j|ƒjƒ\}}}t j|ƒjƒ\} } t | ||ƒ\} } |d krÂ| | f||<q"d} |d krü| dkrät|d ƒ‚ntjtj| ƒjƒdƒ} ||jt|| | ƒƒq"Wt|d|||ƒS) z@ Convert string representation into a lexicon for CCGs. rrz:-cSsg|] }|jƒ‘qSr)Ústrip)Ú.0Zprimrrrú szfromstring..éNr?z::Tz@ must contain semantics because include_semantics is set to True)rZreset_idrÚlistÚ splitlinesÚ COMMENTS_REr:r;rOr4r@ÚLEX_REÚRHS_RErHr6rÚ fromstringÚ SEMANTICS_REÚappendr r")Úlex_strZinclude_semanticsr&r'r(rKr/ÚsepÚrhsrErr0rBrrrrrXñs4 $ rXzUse fromstring() instead.cCst|ƒS)N)rX)r[rrrÚ parseLexicon"sr^a¡ # 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 )N)F)!r!ÚreÚ collectionsrZ nltk.ccg.apirrrrZnltk.internalsrZnltk.sem.logicrÚcompilerIr9rJÚUNICODErVrWrYrUr r"r5r<r>rArFrHrXr^Zopenccg_tinytinyrrrrÚ s0       #5   1*