dhKc @sdZddklZlZlZddklZlZddkl Z ddk l Z l Z ddk ZdddddZd efd YZd d ZdS( s Utility functions for parsers. i(tContextFreeGrammartFeatureGrammartWeightedGrammar(tChartt ChartParser(tInsideChartParser(t FeatureCharttFeatureChartParserNicKstii||}t|tptdnt|to1|djo t}n||d|d|Snt|t oH|djo t }n|djo t }n||d|d|SnE|djo t }n|djo t }n||d|d|SdS(s Load a grammar from a file, and build a parser based on that grammar. The parser depends on the grammar format, and might also depend on properties of the grammar itself. The following grammar formats are currently supported: - C{'cfg'} (CFGs: L{ContextFreeGrammar}) - C{'pcfg'} (probabilistic CFGs: L{WeightedGrammar}) - C{'fcfg'} (feature-based CFGs: L{ContextFreeGrammar}) @type grammar_url: C{str} @param grammar_url: A URL specifying where the grammar is located. The default protocol is C{"nltk:"}, which searches for the file in the the NLTK data package. @type trace: C{int} @param trace: The level of tracing that should be used when parsing a text. C{0} will generate no tracing output; and higher numbers will produce more verbose tracing output. @param parser: The class used for parsing; should be L{ChartParser} or a subclass. If C{None}, the class depends on the grammar format. @param chart_class: The class used for storing the chart; should be L{Chart} or a subclass. Only used for CFGs and feature CFGs. If C{None}, the chart class depends on the grammar format. @type beam_size: C{int} @param beam_size: The maximum length for the parser's edge queue. Only used for probabilistic CFGs. @param load_args: Keyword parameters used when loading the grammar. See L{data.load} for more information. s@The grammar must be a ContextFreeGrammar, or a subclass thereof.ttracet beam_sizet chart_classN(tnltktdatatloadt isinstanceRt ValueErrorRtNoneRRRRRR(t grammar_urlRtparserR R t load_argstgrammar((s%/p/zhu/06/nlp/nltk/nltk/parse/util.pyt load_parsers$"          t TestGrammarcBs)eZdZeedZedZRS(s Unit tests for CFG. cCs=||_t|dd|_||_||_||_dS(NRi(t test_grammarRtcptsuitet_acceptt_reject(tselfRRtaccepttreject((s%/p/zhu/06/nlp/nltk/nltk/parse/util.pyt__init__Vs    c Csx|iD]}|ddGxddgD]}x||D]}|i}|ii|}|o'|o H|GHx|D] }|GHqyWn|djo(|gjotd|qt}q:|otd|q:t} q:Wq)W|o| o dGHq q WdS( s| Sentences in the test suite are divided into two classes: - grammatical (C{accept}) and - ungrammatical (C{reject}). If a sentence should parse accordng to the grammar, the value of C{trees} will be a non-empty list. If a sentence should be rejected according to the grammar, then the value of C{trees} will be C{None}. tdoct:RRsSentence '%s' failed to parse'sSentence '%s' received a parse'sAll tests passed!N(RtsplitRtparseRtTrue( Rt show_treesttesttkeytsentttokensttreesttreetacceptedtrejected((s%/p/zhu/06/nlp/nltk/nltk/parse/util.pytrun_s0         (t__name__t __module__t__doc__RRtFalseR.(((s%/p/zhu/06/nlp/nltk/nltk/parse/util.pyRRs s#%;cCsg}x|idD]}|djp|d|joqn|idd}t}t|djoM|dd jo|dd j}|d}qt|d}|d}n|i}|gjoqn|||fg7}qW|S( sg Parses a string with one test sentence per line. Lines can optionally begin with: - a C{bool}, saying if the sentence is grammatical or not, or - an C{int}, giving the number of parse trees is should have, The result information is followed by a colon, and then the sentence. Empty lines and lines beginning with a comment char are ignored. @return: a C{list} of C{tuple} of sentences and expected results, where a sentence is a C{list} of C{str}, and a result is C{None}, or C{bool}, or C{int} @param comment_chars: L{str} of possible comment characters. s tiR!iiR$ttrueR2tfalse(sTruestruesFalsesfalse(sTruestrue(R"Rtlentint(tstringt comment_charst sentencestsentencet split_infotresultR)((s%/p/zhu/06/nlp/nltk/nltk/parse/util.pytextract_test_sentencess % (R1t nltk.grammarRRRtchartRRtpchartRt featurechartRRt nltk.dataR RRtobjectRR>(((s%/p/zhu/06/nlp/nltk/nltk/parse/util.pys s  </