def yahoo_key_phrase(self, text): url = 'http://jlp.yahooapis.jp/KeyphraseService/V1/extract?appid=%s&sentence=%s' % ( self.yahoo_japan_app_id, text) result_page = WebPage(url) result_page.fetch_xml() key_phrases = result_page.pick_key_phrases() return key_phrases
def search_in_clueweb_with_expanded_query(): search_engine = SearchEngine() search_engine.action_word = request.form['action_word'] search_engine.hint_word = request.form['hint_word'] search_engine.find_related_action_words_with_google() search_engine.count_action_words() search_engine.sort_action_words_count() search_engine.pick_sorted_action_words_more_than_1_count() results = [] for elem in search_engine.sorted_action_words_more_than_1_count: elem['expanded_query'] = search_engine.action_word + ' ' + search_engine.hint_word + ' ' + elem['word'] url = 'http://karen.dl.local:8983/solr/ClueWeb09ja/select?q=' + elem['expanded_query'] + '&wt=xml' web_page = WebPage(url) web_page.fetch_xml() web_page.pick_texts_to_result_pages() # クエリ1つごとに結果xmlページがある # 結果xmlページの内容を1ページずつWebPageオブジェクトにしてresult_pagesとして1クエリに対応する結果ページに持たせる for result_page in web_page.result_pages: # result_page.text_body result_page.set_lines_from_texts() result_page.set_line_nums_with_word(search_engine.action_word) result_page.set_line_nums_around_action_word() result_page.set_line_clusters_around_action_word() # web_page.result_pages[0].line_clusters_around_action_word results.append({'pages': web_page.result_pages, 'expanded_query': elem['expanded_query']}) return render_template('search_in_clueweb_with_expanded_query.tmpl', results=results)
def yahoo_key_phrase(self, text): url = "http://jlp.yahooapis.jp/KeyphraseService/V1/extract?appid=%s&sentence=%s" % ( self.yahoo_japan_app_id, text, ) result_page = WebPage(url) result_page.fetch_xml() key_phrases = result_page.pick_key_phrases() return key_phrases
def clue_web_search(self, query): """ 検索するだけ。 rows=50の値を変えることで検索結果件数を変えられる。 返り値にlistでtextsを渡す """ options = '&rows=50' url = constants.CLUE_WEB_URL_HEAD + query + options + constants.CLUE_WEB_URL_TAIL clue_web_result_page = WebPage(url) clue_web_result_page.fetch_xml() clue_web_result_page.pick_texts() return clue_web_result_page.texts
def clue_web_search(self, query): """ 検索するだけ。 rows=50の値を変えることで検索結果件数を変えられる。 返り値にlistでtextsを渡す """ options = "&rows=50" url = constants.CLUE_WEB_URL_HEAD + query + options + constants.CLUE_WEB_URL_TAIL clue_web_result_page = WebPage(url) clue_web_result_page.fetch_xml() clue_web_result_page.pick_texts() return clue_web_result_page.texts