RKc!@sldZddkZddklZddkTddklZddklZl Z l Z l Z l Z de fdYZd Zd e efd YZd e fd YZdefdYZdeefdYZde fdYZddklZdeefdYZdefdYZde fdYZde fdYZde fdYZdefd YZd!Zd"Zd#Zd$Zd%Z d&Z!dddd'Z#ed(d)Z$ei%d*ei&Z'ei%d+ei&Z(ei%d,ei&Z)ei%d-ei&Z*e+d.Z,e+d/Z-ei%d0ei&Z.d1Z/ei%d2ei&Z0ei%d3Z1d4Z2d5Z3d6Z4e d7Z5e d8Z6d9Z7d:Z8d;Z9d<Z:d=Z;e<d>jo e;ndd?d ddddddd@dAdBdCdDdEdFdGdHdIdJdKdLdMdNdOdPdQgZ=dS(Rs Basic data classes for representing context free grammars. A X{grammar} specifies which trees can represent the structure of a given text. Each of these trees is called a X{parse tree} for the text (or simply a X{parse}). In a X{context free} grammar, the set of parse trees for any piece of a text can depend only on that piece, and not on the rest of the text (i.e., the piece's context). Context free grammars are often used to find possible syntactic structures for sentences. In this context, the leaves of a parse tree are word tokens; and the node values are phrasal categories, such as C{NP} and C{VP}. The L{ContextFreeGrammar} class is used to encode context free grammars. Each C{ContextFreeGrammar} consists of a start symbol and a set of productions. The X{start symbol} specifies the root node value for parse trees. For example, the start symbol for syntactic parsing is usually C{S}. Start symbols are encoded using the C{Nonterminal} class, which is discussed below. A Grammar's X{productions} specify what parent-child relationships a parse tree can contain. Each production specifies that a particular node can be the parent of a particular set of children. For example, the production C{ -> } specifies that an C{S} node can be the parent of an C{NP} node and a C{VP} node. Grammar productions are implemented by the C{Production} class. Each C{Production} consists of a left hand side and a right hand side. The X{left hand side} is a C{Nonterminal} that specifies the node type for a potential parent; and the X{right hand side} is a list that specifies allowable children for that parent. This lists consists of C{Nonterminals} and text types: each C{Nonterminal} indicates that the corresponding child may be a C{TreeToken} with the specified node type; and each text type indicates that the corresponding child may be a C{Token} with the with that type. The C{Nonterminal} class is used to distinguish node values from leaf values. This prevents the grammar from accidentally using a leaf value (such as the English word "A") as the node of a subtree. Within a C{ContextFreeGrammar}, all node values are wrapped in the C{Nonterminal} class. Note, however, that the trees that are specified by the grammar do B{not} include these C{Nonterminal} wrappers. Grammars can also be given a more procedural interpretation. According to this interpretation, a Grammar specifies any tree structure M{tree} that can be produced by the following procedure: - Set M{tree} to the start symbol - Repeat until M{tree} contains no more nonterminal leaves: - Choose a production M{prod} with whose left hand side M{lhs} is a nonterminal leaf of M{tree}. - Replace the nonterminal leaf with a subtree, whose node value is the value wrapped by the nonterminal M{lhs}, and whose children are the right hand side of M{prod}. The operation of replacing the left hand side (M{lhs}) of a production with the right hand side (M{rhs}) in a tree (M{tree}) is known as X{expanding} M{lhs} to M{rhs} in M{tree}. iN(t deprecated(t*(tImmutableProbabilisticMixIn(t FeatStructtFeatDicttFeatStructParsertSLASHtTYPEt NonterminalcBsheZdZdZdZdZdZdZdZdZ dZ d Z d Z RS( s: A non-terminal symbol for a context free grammar. C{Nonterminal} is a wrapper class for node values; it is used by C{Production}s to distinguish node values from leaf values. The node value that is wrapped by a C{Nonterminal} is known as its X{symbol}. Symbols are typically strings representing phrasal categories (such as C{"NP"} or C{"VP"}). However, more complex symbol types are sometimes used (e.g., for lexicalized grammars). Since symbols are node values, they must be immutable and hashable. Two C{Nonterminal}s are considered equal if their symbols are equal. @see: L{ContextFreeGrammar} @see: L{Production} @type _symbol: (any) @ivar _symbol: The node value corresponding to this C{Nonterminal}. This value must be immutable and hashable. cCs||_t||_dS(s Construct a new non-terminal from the given symbol. @type symbol: (any) @param symbol: The node value corresponding to this C{Nonterminal}. This value must be immutable and hashable. N(t_symbolthasht_hash(tselftsymbol((s)/home/andreas/coglang/dop1/dop_grammar.pyt__init__gs cCs|iS(sf @return: The node value corresponding to this C{Nonterminal}. @rtype: (any) (R (R ((s)/home/andreas/coglang/dop1/dop_grammar.pyR sscCsBy'|i|ijot||iSWntj otSXdS(s @return: True if this non-terminal is equal to C{other}. In particular, return true iff C{other} is a C{Nonterminal} and this non-terminal's symbol is equal to C{other}'s symbol. @rtype: C{boolean} N(R t isinstancet __class__tAttributeErrortFalse(R tother((s)/home/andreas/coglang/dop1/dop_grammar.pyt__eq__zs cCs ||j S(s @return: True if this non-terminal is not equal to C{other}. In particular, return true iff C{other} is not a C{Nonterminal} or this non-terminal's symbol is not equal to C{other}'s symbol. @rtype: C{boolean} ((R R((s)/home/andreas/coglang/dop1/dop_grammar.pyt__ne__scCs&yt|i|iSWndSXdS(Ni(tcmpR (R R((s)/home/andreas/coglang/dop1/dop_grammar.pyt__cmp__scCs|iS(N(R (R ((s)/home/andreas/coglang/dop1/dop_grammar.pyt__hash__scCs4t|itod|ifSd|ifSdS(se @return: A string representation for this C{Nonterminal}. @rtype: C{string} s%ss%rN(RR t basestring(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyt__repr__scCs4t|itod|ifSd|ifSdS(se @return: A string representation for this C{Nonterminal}. @rtype: C{string} s%ss%rN(RR R(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyt__str__scCstd|i|ifS(sa @return: A new nonterminal whose symbol is C{M{A}/M{B}}, where C{M{A}} is the symbol for this nonterminal, and C{M{B}} is the symbol for rhs. @rtype: L{Nonterminal} @param rhs: The nonterminal used to form the right hand side of the new nonterminal. @type rhs: L{Nonterminal} s%s/%s(RR (R trhs((s)/home/andreas/coglang/dop1/dop_grammar.pyt__div__s cCs t|iS(N(tlenR (R ((s)/home/andreas/coglang/dop1/dop_grammar.pyt__len__s( t__name__t __module__t__doc__RR RRRRRRRR(((s)/home/andreas/coglang/dop1/dop_grammar.pyRTs     cCsWd|jo|id}n |i}g}|D]}|t|iq7~S(s Given a string containing a list of symbol names, return a list of C{Nonterminals} constructed from those symbols. @param symbols: The symbol name string. This string can be delimited by either spaces or commas. @type symbols: C{string} @return: A list of C{Nonterminals} constructed from the symbol names given in C{symbols}. The C{Nonterminals} are sorted in the same order as the symbols names. @rtype: C{list} of L{Nonterminal} t,(tsplitRtstrip(tsymbolst symbol_listt_[1]ts((s)/home/andreas/coglang/dop1/dop_grammar.pyt nonterminalss  tFeatStructNonterminalcBs eZdZdZdZRS(s|A feature structure that's also a nonterminal. It acts as its own symbol, and automatically freezes itself when hashed.cCs|iti|S(N(tfreezeRR(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyRs cCs|S(N((R ((s)/home/andreas/coglang/dop1/dop_grammar.pyR s(R R!R"RR (((s)/home/andreas/coglang/dop1/dop_grammar.pyR+s t ProductioncBszeZdZdZdZdZdZdZdZdZ dZ d Z d Z d Z d ZRS( s A grammar production. Each production maps a single symbol on the X{left-hand side} to a sequence of symbols on the X{right-hand side}. (In the case of context-free productions, the left-hand side must be a C{Nonterminal}, and the right-hand side is a sequence of terminals and C{Nonterminals}.) X{terminals} can be any immutable hashable object that is not a C{Nonterminal}. Typically, terminals are strings representing words, such as C{"dog"} or C{"under"}. @see: L{ContextFreeGrammar} @see: L{DependencyGrammar} @see: L{Nonterminal} @type _lhs: L{Nonterminal} @ivar _lhs: The left-hand side of the production. @type _rhs: C{tuple} of (C{Nonterminal} and (terminal)) @ivar _rhs: The right-hand side of the production. cCs]t|ttfotdn||_t||_t|i|if|_dS(s Construct a new C{Production}. @param lhs: The left-hand side of the new C{Production}. @type lhs: L{Nonterminal} @param rhs: The right-hand side of the new C{Production}. @type rhs: sequence of (C{Nonterminal} and (terminal)) s9production right hand side should be a list, not a stringN( Rtstrtunicodet TypeErrort_lhsttuplet_rhsR R (R tlhsR((s)/home/andreas/coglang/dop1/dop_grammar.pyRs  cCs|iS(sc @return: the left-hand side of this C{Production}. @rtype: L{Nonterminal} (R1(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyR4scCs|iS(s @return: the right-hand side of this C{Production}. @rtype: sequence of (C{Nonterminal} and (terminal)) (R3(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyRscCs t|iS(sX @return: the length of the right-hand side. @rtype: C{integer} (RR3(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyRscCs1tg}|iD]}|t|tq~S(sl @return: True if the right-hand side only contains C{Nonterminal}s @rtype: C{bool} (tallR3RR(R R(tn((s)/home/andreas/coglang/dop1/dop_grammar.pyt is_nonlexicalscCs |i S(sm @return: True if the right-hand contain at least one terminal token @rtype: C{bool} (R7(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyt is_lexicalscCs9d|if}x"|iD]}|d|f7}qW|S(sv @return: A verbose string representation of the C{Production}. @rtype: C{string} s%r ->s %r(R1R3(R R.telt((s)/home/andreas/coglang/dop1/dop_grammar.pyR#s  cCsd|S(sw @return: A concise string representation of the C{Production}. @rtype: C{string} s%s((R ((s)/home/andreas/coglang/dop1/dop_grammar.pyR.scCs6t||io#|i|ijo|i|ijS(sf @return: true if this C{Production} is equal to C{other}. @rtype: C{boolean} (RRR1R3(R R((s)/home/andreas/coglang/dop1/dop_grammar.pyR6scCs ||j S(N((R R((s)/home/andreas/coglang/dop1/dop_grammar.pyR?scCs=t||ipdSt|i|if|i|ifS(Ni(RRRR1R3(R R((s)/home/andreas/coglang/dop1/dop_grammar.pyRBscCs|iS(sU @return: A hash value for the C{Production}. @rtype: C{int} (R (R ((s)/home/andreas/coglang/dop1/dop_grammar.pyRFs(R R!R"RR4RRR7R8RRRRRR(((s)/home/andreas/coglang/dop1/dop_grammar.pyR-s         tDependencyProductioncBseZdZdZRS(s A dependency grammar production. Each production maps a single head word to an unordered list of one or more modifier words. cCs9d|if}x"|iD]}|d|f7}qW|S(s @return: A verbose string representation of the C{DependencyProduction}. @rtype: C{string} s'%s' ->s '%s'(R1R3(R R.R9((s)/home/andreas/coglang/dop1/dop_grammar.pyRSs  (R R!R"R(((s)/home/andreas/coglang/dop1/dop_grammar.pyR:NstWeightedProductioncBs;eZdZdZdZdZdZdZRS(s A probabilistic context free grammar production. PCFG C{WeightedProduction}s are essentially just C{Production}s that have probabilities associated with them. These probabilities are used to record how likely it is that a given production will be used. In particular, the probability of a C{WeightedProduction} records the likelihood that its right-hand side is the correct instantiation for any given occurance of its left-hand side. @see: L{Production} cKs'ti||ti|||dS(s{ Construct a new C{WeightedProduction}. @param lhs: The left-hand side of the new C{WeightedProduction}. @type lhs: L{Nonterminal} @param rhs: The right-hand side of the new C{WeightedProduction}. @type rhs: sequence of (C{Nonterminal} and (terminal)) @param prob: Probability parameters of the new C{WeightedProduction}. N(RRR-(R R4Rtprob((s)/home/andreas/coglang/dop1/dop_grammar.pyRks cCsti|d|iS(Ns [%s](R-RR<(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyRxscCsOt||io<|i|ijo)|i|ijo|i|ijS(N(RRR1R3R<(R R((s)/home/andreas/coglang/dop1/dop_grammar.pyR{scCs ||j S(N((R R((s)/home/andreas/coglang/dop1/dop_grammar.pyRscCst|i|i|ifS(N(R R1R3R<(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyRs(R R!R"RRRRR(((s)/home/andreas/coglang/dop1/dop_grammar.pyR;_s     tContextFreeGrammarcBseZdZdZdZdZddedZdZ dZ dZ e dd Z d Zd Zd Zd ZdZdZdZdZdZdZdZRS(s$ A context-free grammar. A grammar consists of a start state and a set of productions. The set of terminals and nonterminals is implicitly specified by the productions. If you need efficient key-based access to productions, you can use a subclass to implement it. cCsC||_||_td|D|_|i|idS(sG Create a new context-free grammar, from the given start state and set of C{Production}s. @param start: The start symbol @type start: L{Nonterminal} @param productions: The list of productions that defines the grammar @type productions: C{list} of L{Production} cssx|]}|iVqWdS(N(R4(t.0tprod((s)/home/andreas/coglang/dop1/dop_grammar.pys s N(t_startt _productionstsett _categoriest_calculate_indexest_calculate_grammar_forms(R tstartt productions((s)/home/andreas/coglang/dop1/dop_grammar.pyRs   c Csh|_h|_h|_h|_x|iD]}|i}||ijog|i|tcat((s)/home/andreas/coglang/dop1/dop_grammar.pys s css(x!|]}|t|gfVqWdS(N(RB(R>RH((s)/home/andreas/coglang/dop1/dop_grammar.pys s (t _lhs_indext _rhs_indext _empty_indext_lexical_indexRAR1tappendR3R4RR.t setdefaultRBtaddRRtdictRCt _leftcornerst_leftcorner_parentstTrueRRtissubsettupdate( R R?R4trhs0ttokentimmediate_leftcornerstchildt leftwordst leftparentst once_againtparent((s)/home/andreas/coglang/dop1/dop_grammar.pyRDs^         + *     cCs|iS(sY @return: The start symbol of the grammar @rtype: L{Nonterminal} (R@(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyRFscCs|o|otdn| o(| o |p|iS|iiSn|oJ| oB|p|ii|gS||ijo|i|gSgSnr|o| o|ii|gSg}|ii|gD]-}||ii|gjo ||qq~SdS(s Return the grammar productions, filtered by the left-hand side or the first item in the right-hand side. @param lhs: Only return productions with the given left-hand side. @param rhs: Only return productions with the given first item in the right-hand side. @param empty: Only return productions with an empty right-hand side. @return: A list of productions matching the given constraints. @rtype: C{list} of C{Production} sCYou cannot select empty and non-empty productions at the same time.N(t ValueErrorRARKtvaluesRItgetRJ(R R4RtemptyR(R?((s)/home/andreas/coglang/dop1/dop_grammar.pyRGs  cCs|ii|tS(s Return the set of all words that the given category can start with. Also called the I{first set} in compiler construction. (RQR`RB(R RH((s)/home/andreas/coglang/dop1/dop_grammar.pyt leftcornersscCs|ii|tS(si Return the set of all categories for which the given category is a left corner. (RRR`RB(R RH((s)/home/andreas/coglang/dop1/dop_grammar.pytleftcorner_parentsscCspg}|D]$}|ii|p ||q q ~}|o-did|D}td|ndS(sz Check whether the grammar rules cover the given list of tokens. If not, then raise an exception. s, css x|]}d|fVqWdS(s%rN((R>tw((s)/home/andreas/coglang/dop1/dop_grammar.pys s s3Grammar does not cover some of the input words: %r.N(RLR`tjoinR^(R ttokensR(ttoktmissing((s)/home/andreas/coglang/dop1/dop_grammar.pytcheck_coverages 's.Use ContextFreeGrammar.check_coverage instead.cCs-x&|D]}|ii|otSqWtS(s Check whether the grammar rules cover the given list of tokens. @param tokens: the given list of tokens. @type tokens: a C{list} of C{string} objects. @return: True/False (RLR`RRS(R RfRW((s)/home/andreas/coglang/dop1/dop_grammar.pytcovers$s  c Cs|i}tg}|D]}||iq~|_tg}|D]*}t|djo||iqGqG~|_td|D|_t d|D|_ tg}|D]*}t|djo||iqq~|_ dS(s@ Pre-calculate of which form(s) the grammar is. icssx|]}t|VqWdS(N(R(R>tp((s)/home/andreas/coglang/dop1/dop_grammar.pys :s cssx|]}t|VqWdS(N(R(R>Rk((s)/home/andreas/coglang/dop1/dop_grammar.pys ;s N( RAR5R8t _is_lexicalRR7t_is_nonlexicaltmint_min_lentmaxt_max_lent_all_unary_are_lexical(R tprodsR(Rkt_[2]t_[3]((s)/home/andreas/coglang/dop1/dop_grammar.pyRE2s 03cCs|iS(s: True if all productions are lexicalised. (Rl(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyR8?scCs|iS(s True if all lexical rules are "preterminals", that is, unary rules which can be separated in a preprocessing step. This means that all productions are of the forms A -> B1 ... Bn (n>=0), or A -> "s". Note: is_lexical() and is_nonlexical() are not opposites. There are grammars which are neither, and grammars which are both. (Rm(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyR7Es cCs|iS(sP The right-hand side length of the shortest grammar production. (Ro(R ((s)/home/andreas/coglang/dop1/dop_grammar.pytmin_lenRscCs|iS(sO The right-hand side length of the longest grammar production. (Rq(R ((s)/home/andreas/coglang/dop1/dop_grammar.pytmax_lenXscCs |idjS(s9 True if there are no empty productions. i(Ro(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyt is_nonempty^scCs |idjS(s True if all productions are at most binary. Note that there can still be empty and unary productions. i(Rq(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyt is_binariseddscCs$|io|io |iS(sa True if all productions are of the forms A -> B C, A -> B, or A -> "s". (RxR7Ry(R ((s)/home/andreas/coglang/dop1/dop_grammar.pytis_flexible_chomsky_normal_formkscCs|io|iS(sx A grammar is of Chomsky normal form if all productions are of the forms A -> B C, or A -> "s". (RzRr(R ((s)/home/andreas/coglang/dop1/dop_grammar.pytis_chomsky_normal_formrs cCsdt|iS(Ns(RRA(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyRzscCsJdt|i}|d|i7}x|iD]}|d|7}q.W|S(NsGrammar with %d productionss (start state = %r)s %s(RRAR@(R R.t production((s)/home/andreas/coglang/dop1/dop_grammar.pyR}s  N(R R!R"RRDRFtNoneRRGRbRcRiRRjRER8R7RvRwRxRyRzR{RR(((s)/home/andreas/coglang/dop1/dop_grammar.pyR=s(  2 )           (t DeprecatedtGrammarcBseZdZRS(s$Use nltk.ContextFreeGrammar instead.(R R!R"(((s)/home/andreas/coglang/dop1/dop_grammar.pyRstFeatureGrammarcBsMeZdZdZdZddedZdZdZ dZ RS(s, A feature-based grammar. This is equivalent to a L{ContextFreeGrammar} whose nonterminals are L{FeatStructNonterminal}s. A grammar consists of a start state and a set of productions. The set of terminals and nonterminals is implicitly specified by the productions. cCsti|||dS(sS Create a new feature-based grammar, from the given start state and set of C{Production}s. @param start: The start symbol @type start: L{FeatStructNonterminal} @param productions: The list of productions that defines the grammar @type productions: C{list} of L{Production} N(R=R(R RFRG((s)/home/andreas/coglang/dop1/dop_grammar.pyRs cCs<h|_h|_h|_h|_x|iD]}|i|i}||ijog|i|(R(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyRscCs)ttt|pt|i|iS(N(RRttypeR(R R((s)/home/andreas/coglang/dop1/dop_grammar.pyRscCs|iS(N(R (R ((s)/home/andreas/coglang/dop1/dop_grammar.pyR s(R R!R"RRRR(((s)/home/andreas/coglang/dop1/dop_grammar.pyRs    tDependencyGrammarcBs;eZdZdZdZdZdZdZRS(s A dependency grammar. A DependencyGrammar consists of a set of productions. Each production specifies a head/modifier relationship between a pair of words. cCs ||_dS(s Create a new dependency grammar, from the set of C{Production}s. @param productions: The list of productions that defines the grammar @type productions: C{list} of L{Production} N(RA(R RG((s)/home/andreas/coglang/dop1/dop_grammar.pyRscCsNxG|iD]<}x3|iD](}|i|jo||jotSqWq WtS(sB @param head: A head word. @type head: C{string}. @param mod: A mod word, to test as a modifier of 'head'. @type mod: C{string}. @return: true if this C{DependencyGrammar} contains a C{DependencyProduction} mapping 'head' to 'mod'. @rtype: C{boolean}. (RAR3R1RSR(R theadtmodR|t possibleMod((s)/home/andreas/coglang/dop1/dop_grammar.pytcontainss   cCsNxG|iD]<}x3|iD](}|i|jo||jotSqWq WtS(sB @param head: A head word. @type head: C{string}. @param mod: A mod word, to test as a modifier of 'head'. @type mod: C{string}. @return: true if this C{DependencyGrammar} contains a C{DependencyProduction} mapping 'head' to 'mod'. @rtype: C{boolean}. (RAR3R1RSR(R RRR|R((s)/home/andreas/coglang/dop1/dop_grammar.pyt __contains__-s   cCs9dt|i}x|iD]}|d|7}qW|S(s| @return: A verbose string representation of the C{DependencyGrammar} @rtype: C{string} s&Dependency grammar with %d productionss %s(RRA(R R.R|((s)/home/andreas/coglang/dop1/dop_grammar.pyRJs  cCsdt|iS(sb @return: A concise string representation of the C{DependencyGrammar} s&Dependency grammar with %d productions(RRA(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyRUs(R R!R"RRRRR(((s)/home/andreas/coglang/dop1/dop_grammar.pyR s    tStatisticalDependencyGrammarcBs2eZdZdZdZdZdZRS(s cCs||_||_||_dS(N(RAt_eventst_tags(R RGteventsttags((s)/home/andreas/coglang/dop1/dop_grammar.pyRbs  cCsNxG|iD]<}x3|iD](}|i|jo||jotSqWq WtS(sB @param head: A head word. @type head: C{string}. @param mod: A mod word, to test as a modifier of 'head'. @type mod: C{string}. @return: true if this C{DependencyGrammar} contains a C{DependencyProduction} mapping 'head' to 'mod'. @rtype: C{boolean}. (RAR3R1RSR(R RRR|R((s)/home/andreas/coglang/dop1/dop_grammar.pyRgs   c Csdt|i}x|iD]}|d|7}qW|d7}x,|iD]!}|d|i||f7}qIW|d7}x,|iD]!}|d||i|f7}qW|S(s @return: A verbose string representation of the C{StatisticalDependencyGrammar} @rtype: C{string} s2Statistical dependency grammar with %d productionss %ss Events:s %d:%ss Tags:s %s: (%s)(RRARR(R R.R|teventttag_word((s)/home/andreas/coglang/dop1/dop_grammar.pyRxs     cCsdt|iS(sm @return: A concise string representation of the C{StatisticalDependencyGrammar} s2Statistical Dependency grammar with %d productions(RRA(R ((s)/home/andreas/coglang/dop1/dop_grammar.pyRs(R R!R"RRRR(((s)/home/andreas/coglang/dop1/dop_grammar.pyR]s    tWeightedGrammarcBseZdZdZdZRS(s A probabilistic context-free grammar. A Weighted Grammar consists of a start state and a set of weighted productions. The set of terminals and nonterminals is implicitly specified by the productions. PCFG productions should be C{WeightedProduction}s. C{WeightedGrammar}s impose the constraint that the set of productions with any given left-hand-side must have probabilities that sum to 1. If you need efficient key-based access to productions, you can use a subclass to implement it. @type EPSILON: C{float} @cvar EPSILON: The acceptable margin of error for checking that productions with a given left-hand side have probabilities that sum to 1. g{Gz?cCsti|||h}x:|D]2}|i|id|i||i B C in a PCFG is: | count(A -> B C) | P(B, C | A) = --------------- where * is any right hand side | count(A -> *) @param start: The start symbol @type start: L{Nonterminal} @param productions: The list of productions that defines the grammar @type productions: C{list} of L{Production} iiR<(R`R4R;RtfloatR(RFRGtpcounttlcountR?R(RkRs((s)/home/andreas/coglang/dop1/dop_grammar.pyt induce_pcfgs&McCs t|tS(s< @return: a C{list} of context-free L{Production}s. (tparse_productiontstandard_nonterm_parser(tinput((s)/home/andreas/coglang/dop1/dop_grammar.pytparse_cfg_productionscCs"t|t\}}t||S(s @return: a L{ContextFreeGrammar}. @param input: a grammar, either in the form of a string or else as a list of strings. (t parse_grammarRR=(RRFRG((s)/home/andreas/coglang/dop1/dop_grammar.pyt parse_cfgscCst|tdtS(s< @return: a C{list} of PCFG L{WeightedProduction}s. t probabilistic(RRRS(R((s)/home/andreas/coglang/dop1/dop_grammar.pytparse_pcfg_productionscCs(t|tdt\}}t||S(s @return: a probabilistic L{WeightedGrammar}. @param input: a grammar, either in the form of a string or else as a list of strings. R(RRRSR(RRFRG((s)/home/andreas/coglang/dop1/dop_grammar.pyt parse_pcfgs cCs t||S(s= @return: a C{list} of feature-based L{Production}s. (R(Rtfstruct_parser((s)/home/andreas/coglang/dop1/dop_grammar.pytparse_fcfg_productionscCs|djottf}n|djot|td|}n|dj otdnt||i\}}t||S(s @return: a feature structure based L{FeatureGrammar}. @param input: a grammar, either in the form of a string or else as a list of strings. @param features: a tuple of features (default: SLASH, TYPE) @param logic_parser: a parser for lambda-expressions (default: LogicParser()) @param fstruct_parser: a feature structure parser (only if features and logic_parser is None) t logic_parsers8'logic_parser' and 'fstruct_parser' must not both be setN( R}RRRR+t ExceptionRt partial_parseR(RtfeaturesRRRFRG((s)/home/andreas/coglang/dop1/dop_grammar.pyt parse_fcfgs     sUse nltk.parse_fcfg() instead.cCs t|S(N(R(R((s)/home/andreas/coglang/dop1/dop_grammar.pyt parse_featcfg1ss \s* -> \s*s( \[ [\d\.]+ \] ) \s*s( "[^"]+" | \'[^\']+\' ) \s*s\| \s*c CsHd}|||\}}ti||}|ptdn|i}dg}gg}xr|t|jo^ti||}|oc|o\|i}t|iddd!|d<|ddjotd|dfqqe||djoZti||}|ptd n|di |iddd!|i}qe||d jo<t i||}|i d|i g|i}qe|||\}}|di |qeW|o>g} t ||D]"\} } | t || d | q~ Sg} |D]} | t || q'~ Sd S( sX Parse a grammar rule, given as a string, and return a list of productions. isExpected an arrowgiig?s9Production probability %f, should not be greater than 1.0s'"sUnterminated stringt|R<N(t _ARROW_REtmatchR^tendRt_PROBABILITY_RERtgroupt _TERMINAL_RERMt_DISJUNCTION_REtzipR;R-( tlinetnonterm_parserRtposR4tmt probabilitiestrhsidestnontermR(Rt probabilityRt((s)/home/andreas/coglang/dop1/dop_grammar.pyR<sD     !  7c Cst|to|id}n|}d}g}d}xTt|D]F\}}||i}|idp |djoqHn|ido|d id}qHnd}y|ddjou|d idd \} } | d jo<|| d\}} | t | jot d q>qXt d n|t |||7}WqHt j o&} t d |d || fqHXqHW|p t dn|p|di }n||fS(s @return: a pair of - a starting category - a list of C{Production}s @param input: a grammar, either in the form of a string or else as a list of strings. @param nonterm_parser: a function for parsing nonterminals. It should take a C{(string,position)} as argument and return a C{(nonterminal,position)} as result. @param probabilistic: are the grammar rules probabilistic? s tt#s\it it%iRFsBad argument to start directives Bad directivesUnable to parse line %s: %s %ssNo productions found!N( RR.R$R}t enumerateR%t startswithtendswithtrstripRR^RR4( RRRtlinesRFRGt continue_linetlinenumRt directivetargsRte((s)/home/andreas/coglang/dop1/dop_grammar.pyRss@     s( [\w/]+ ) \s*cCsPti||}|ptd||nt|id|ifS(NsExpected a nonterminal, found: i(t_STANDARD_NONTERM_RERR^RRR(tstringRR((s)/home/andreas/coglang/dop1/dop_grammar.pyRs sH^\s* # leading whitespace ('[^']+')\s* # single-quoted lhs (?:[-=]+>)\s* # arrow (?:( # rhs: "[^"]+" # doubled-quoted terminal | '[^']+' # single-quoted terminal | \| # disjunction ) \s*) # trailing space *$s"('[^']'|[-=]+>|"[^"]+"|'[^']+'|\|)cCsg}xt|idD]y\}}|i}|idp |djoqny|t|7}Wqtj otd||fqXqWt|djo tdnt|S(Ns RRsUnable to parse line %s: %sisNo productions found!(RR$R%Rtparse_dependency_productionR^RR(R)RGRR((s)/home/andreas/coglang/dop1/dop_grammar.pytparse_dependency_grammars   c Csti|p tdnti|}g}t|D](\}}|ddjo ||q=q=~}|did}gg}xJ|dD]>}|djo|igq|di|idqWg}|D]} |t|| q~S(NsBad production stringiiis'"Ri( t _PARSE_DG_RERR^t _SPLIT_DG_RER$RR%RMR:( R)tpiecesR(tiRktlhsideRtpieceRttrhside((s)/home/andreas/coglang/dop1/dop_grammar.pyRs B   c Csddkl}l}l}|d\}}}}|d\}}} } ||} dG||||||| | ||g GHdG|i GHH|||gGH|d} dG| GHd G| i GHd G| i id d d dGHHdGH| iddgGH| iddgGHdS(sU A demonstration showing how C{ContextFreeGrammar}s can be created and used. i(R*R-Rs S, NP, VP, PPs N, V, P, DetsSome nonterminals:s S.symbol() =>s S -> NP VP PP -> P NP NP -> Det N | NP PP VP -> V NP | VP PP Det -> 'a' | 'the' N -> 'dog' | 'cat' V -> 'chased' | 'sat' P -> 'on' | 'in' s A Grammar:s grammar.start() =>s grammar.productions() =>R#s, Ris%Coverage of input words by a grammar:tatdogttoyN( tnltkR*R-RR RFRGtreplaceRj( R*R-RtStNPtVPtPPtNtVtPtDett VP_slash_NPtgrammar((s)/home/andreas/coglang/dop1/dop_grammar.pytcfg_demos$ (   s7 S -> NP VP [1.0] NP -> Det N [0.5] | NP PP [0.25] | 'John' [0.1] | 'I' [0.15] Det -> 'the' [0.8] | 'my' [0.2] N -> 'man' [0.5] | 'telescope' [0.5] VP -> VP PP [0.1] | V NP [0.7] | V [0.2] V -> 'ate' [0.35] | 'saw' [0.65] PP -> P NP [1.0] P -> 'with' [0.61] | 'under' [0.39] s S -> NP VP [1.0] VP -> V NP [.59] VP -> V [.40] VP -> VP PP [.01] NP -> Det N [.41] NP -> Name [.28] NP -> NP PP [.31] PP -> P NP [1.0] V -> 'saw' [.21] V -> 'ate' [.51] V -> 'ran' [.28] N -> 'boy' [.11] N -> 'cookie' [.12] N -> 'table' [.13] N -> 'telescope' [.14] N -> 'hill' [.5] Name -> 'Jack' [.52] Name -> 'Bob' [.48] P -> 'with' [.61] P -> 'under' [.39] Det -> 'the' [.41] Det -> 'a' [.31] Det -> 'my' [.28] c Csddkl}ddkl}ddkl}ddkl}ti}|d}dG| GHdG|i GHd G|i GHd G|i GHHt }d G| GHd G|i GHd G|i iddddGHHdGH|iddgGH|iddgGHdGHg}x_|id D]P}xG|i|D]6} | idt| idd|| i7}q2WqWtd} || |}|GHHdGH|i|} | id|iddi} | GHx| i| D] } | GHqWdS(sR A demonstration showing how C{WeightedGrammar}s can be created and used. i(ttreebank(ttreetransforms(R(tpchartisA PCFG production:s pcfg_prod.lhs() =>s pcfg_prod.rhs() =>s pcfg_prod.prob() =>sA PCFG grammar:s grammar.start() =>s grammar.productions() =>R#s, Ris%Coverage of input words by a grammar:Rtboytgirls'Induce PCFG grammar from treebank data:t collapsePOSt horzMarkovRs%Parse sentence using induced grammar:is wsj_0001.mrgiN(t nltk.corpusRRRRt nltk.parseRt toy_pcfg1RGR4RR<t toy_pcfg2RFRRjRt parsed_sentstcollapse_unaryRtchomsky_normal_formRtInsideChartParserttracetleavest nbest_parse(RRRRt pcfg_prodst pcfg_prodRRGRttreeRtparsertsenttparse((s)/home/andreas/coglang/dop1/dop_grammar.pyt pcfg_demo,sR       cCs(ddk}|iid}|GHHdS(Nis!grammars/book_grammars/feat0.fcfg(t nltk.datatdatatload(Rtg((s)/home/andreas/coglang/dop1/dop_grammar.pyt fcfg_demois cCstd}|GHdS(s] A demonstration showing the creation and inspection of a C{DependencyGrammar}. sP 'scratch' -> 'cats' | 'walls' 'walls' -> 'the' 'cats' -> 'the' N(R(R((s)/home/andreas/coglang/dop1/dop_grammar.pytdg_demoos cCs'td}|i}|iGHdS(sh A demonstration of how to read a string representation of a CoNLL format dependency tree. sg 1 Ze ze Pron Pron per|3|evofmv|nom 2 su _ _ 2 had heb V V trans|ovt|1of2of3|ev 0 ROOT _ _ 3 met met Prep Prep voor 8 mod _ _ 4 haar haar Pron Pron bez|3|ev|neut|attr 5 det _ _ 5 moeder moeder N N soort|ev|neut 3 obj1 _ _ 6 kunnen kan V V hulp|ott|1of2of3|mv 2 vc _ _ 7 gaan ga V V hulp|inf 6 vc _ _ 8 winkelen winkel V V intrans|inf 11 cnj _ _ 9 , , Punc Punc komma 8 punct _ _ 10 zwemmen zwem V V intrans|inf 11 cnj _ _ 11 of of Conj Conj neven 7 vc _ _ 12 terrassen terras N N soort|mv|neut 11 cnj _ _ 13 . . Punc Punc punt 12 punct _ _ N(tDependencyGraphRtpprint(tdgR((s)/home/andreas/coglang/dop1/dop_grammar.pytsdg_demo{s  cCs'tttttdS(N(RRRRR (((s)/home/andreas/coglang/dop1/dop_grammar.pytdemos t__main__R*RRRRRRRRRRRR RRRR RR(>R"tretnltk.internalsRt nltk.compattnltk.probabilityRtnltk.featstructRRRRRtobjectRR*R+R-R:R;R=R~RRRRRRRRRRRRR}RRtcompiletVERBOSERRRRRRRRRRRRRRRRRRRR R R t__all__(((s)/home/andreas/coglang/dop1/dop_grammar.pytFs  (j  r,tP46 &    7 2     )   =