o iG@sZddlZddlmZddlmZddlmZGdddeZddZe d kr+edSdS) N) defaultdict)reduce) CorpusReadercs^eZdZdZedZeddZdfdd Z dd d Z dd d Z dd dZ ddZ ZS)LinThesaurusCorpusReaderzEWrapper for the LISP-formatted thesauruses distributed by Dekang Lin.z \("?([^"]+)"? \(desc [0-9.]+\).+cCsttS)z6Factory for creating defaultdict of defaultdict(dict)s)rdictrrQ/var/www/edux/Edux_v2/venv/lib/python3.10/site-packages/nltk/corpus/reader/lin.py__defaultdict_factorysz.LinThesaurusCorpusReader.__defaultdict_factoryc st|dttj|_||_|jdddD]T\}}}t|C}d}|D]6}| }|r7tj d|} d}q%|dkr>d}q%| d} t | dkr[| \} } t| |j|| | d <q%Wd n1sfwYqd S) a Initialize the thesaurus. :param root: root directory containing thesaurus LISP files :type root: C{string} :param badscore: the score to give to words which do not appear in each other's sets of synonyms :type badscore: C{float} z sim[A-Z]\.lspT)include_encodinginclude_fileidz\1Fz)) "N)super__init__rr._LinThesaurusCorpusReader__defaultdict_factory _thesaurus _badscoreabspathsopenstrip_key_resubsplitlenfloat) selfrootbadscorepathencodingfileidlin_filefirstlinekey split_linengramscore __class__rrrs4    z!LinThesaurusCorpusReader.__init__Ncsfkr|rdSddjDS|r'j|vr$j|SjSfddjDS)a Returns the similarity score for two ngrams. :param ngram1: first ngram to compare :type ngram1: C{string} :param ngram2: second ngram to compare :type ngram2: C{string} :param fileid: thesaurus fileid to search in. If None, search all fileids. :type fileid: C{string} :return: If fileid is specified, just the score for the two ngrams; otherwise, list of tuples of fileids and scores. ?cSsg|]}|dfqS)r,r.0fidrrr Qsz7LinThesaurusCorpusReader.similarity..cs:g|]}|j|vrj|njfqSr)rrr-ngram1ngram2rrrr0Zs )_fileidsrr)rr2r3r"rr1r similarity?s z#LinThesaurusCorpusReader.similarityc,|r j|SfddjDS)a Returns a list of scored synonyms (tuples of synonyms and scores) for the current ngram :param ngram: ngram to lookup :type ngram: C{string} :param fileid: thesaurus fileid to search in. If None, search all fileids. :type fileid: C{string} :return: If fileid is specified, list of tuples of scores and synonyms; otherwise, list of tuples of fileids and lists, where inner lists consist of tuples of scores and synonyms. c"g|] }|j|fqSr)ritemsr.r"r(rrrr0uz.)rr8r4rr(r"rr:rscored_synonymsfs   z(LinThesaurusCorpusReader.scored_synonymscr6)a Returns a list of synonyms for the current ngram. :param ngram: ngram to lookup :type ngram: C{string} :param fileid: thesaurus fileid to search in. If None, search all fileids. :type fileid: C{string} :return: If fileid is specified, list of synonyms; otherwise, list of tuples of fileids and lists, where inner lists contain synonyms. cr7r)rkeysr9r:rrr0r;z5LinThesaurusCorpusReader.synonyms..)rr>r4r<rr:rsynonymszs   z!LinThesaurusCorpusReader.synonymscstfddjdS)z Determines whether or not the given ngram is in the thesaurus. :param ngram: ngram to lookup :type ngram: C{string} :return: whether the given ngram is in the thesaurus. cs|pj|vSN)r)accumr"r:rrsz7LinThesaurusCorpusReader.__contains__..F)rr4)rr(rr:r __contains__s  z%LinThesaurusCorpusReader.__contains__)r r@)__name__ __module__ __qualname____doc__recompiler staticmethodrrr5r=r?rC __classcell__rrr*rrs   $ ' rcCsddlm}d}d}td|t||td|t||td|t|j|dd td|t|j|dd td |d |d t|||dS) Nr) lin_thesaurusbusiness enterprisezGetting synonyms for zGetting scored synonyms for z5Getting synonyms from simN.lsp (noun subsection) for zsimN.lsp)r"zSimilarity score for z and :) nltk.corpusrLprintr?r=r5)thesword1word2rrrdemos     rU__main__) rH collectionsr functoolsrnltk.corpus.readerrrrUrDrrrrs