python 全文检索引擎是什么

fromwhoosh.indeximportcreate_in

fromwhoosh.fieldsimport*

fromwhoosh.analysisimportRegexAnalyzer

fromwhoosh.analysisimportTokenizer,Token

中文分词解析器

classChineseTokenizer(Tokenizer):

  """

  中文分词解析器

  """

  def__call__(self, value, positions=False, chars=False,

         keeporiginal=True, removestops=True, start_pos=0, start_char=0,

         mode='',**kwargs):

    assertisinstance(value, text_type),"%r is not unicode "%value

    t=Token(positions, chars, removestops=removestops, mode=mode,**kwargs)

    list_seg=jieba.cut_for_search(value)

    forwinlist_seg:

      t.original=t.text=w

      t.boost=0.5

      ifpositions:

        t.pos=start_pos+value.find(w)

      ifchars:

        t.startchar=start_char+value.find(w)

        t.endchar=start_char+value.find(w)+len(w)

      yieldt

defchinese_analyzer():

  returnChineseTokenizer()

@staticmethod

  defcreate_index(document_dir):

    analyzer=chinese_analyzer()

    schema=Schema(titel=TEXT(stored=True, analyzer=analyzer), path=ID(stored=True),

            content=TEXT(stored=True, analyzer=analyzer))

    ix=create_in("./", schema)

    writer=ix.writer()

    forparents, dirnames, filenamesinos.walk(document_dir):

      forfilenameinfilenames:

        title=filename.replace(".txt", "").decode('utf8')

        printtitle

        content=open(document_dir+'/'+filename,'r').read().decode('utf-8')

        path=u"/b"

        writer.add_document(titel=title, path=path, content=content)

    writer.commit()

@staticmethod

 defsearch(search_str):

   title_list=[]

   print'here'

   ix=open_dir("./")

   searcher=ix.searcher()

   printsearch_str,type(search_str)

   results=searcher.find("content", search_str)

   forhitinresults:

     printhit['titel']

     printhit.score

     printhit.highlights("content", top=10)

     title_list.append(hit['titel'])

   print'tt',title_list

   returntitle_list