Add parase query

ml/Cut.py

@@ -29,7 +29,7 @@ class Cut:
 					flag = 1
 					break
 				data = json.loads(line)
-				seg_list = jieba.cut(data['content'],cut_all=True)
+				seg_list = jieba.cut(data['content'],cut_all=False)
 				result = ' '.join(seg_list)
 				data['content'] = result
 				cut_file.write(json.dumps(data)+'\n')

ml/InverseIndex.py

@@ -42,7 +42,7 @@ class InverseIndex:
 			if not line:
 				break
 			data = json.loads(line)
-			seg_list = list(jieba.cut(data['title'], cut_all=True))
+			seg_list = list(jieba.cut(data['title'], cut_all=False))
 			count+=1
 			for w in seg_list:
 				if w not in self.worddict:
@@ -59,7 +59,7 @@ class InverseIndex:
 			if not line:
 				break
 			data = json.loads(line)
-			seg_list = list(jieba.cut(data['title'],cut_all=True))
+			seg_list = list(jieba.cut(data['title'],cut_all=False))
 			doc.append(seg_list)
 		return doc
 
@@ -72,8 +72,6 @@ class InverseIndex:
 			if not line:
 				break
 			data = json.loads(line)
-#	seg_list = jieba.cut(data['content'],cut_all=True)
-
 			keyword = analyse.extract_tags(data['content'],topK=20)
 			seg = " ".join(keyword)
 			print seg
@@ -132,6 +130,6 @@ class InverseIndex:
 		
 		
 #test
-ii = InverseIndex()
-ii.CalcTFIDF()
+#ii = InverseIndex()
+#ii.CalcTFIDF()
 #ii.loadDataFromCutFile(20)

ml/Search.py

@@ -19,32 +19,62 @@ class Search:
 		return kw_id
 
 
-	def QuerySingle(self,searchWord):
-		idx = self.kw_id[searchWord.decode('utf-8')]
+	#返回文档号
+	def QuerySingle(self,searchWord,ishow):
+		if self.kw_id.has_key(searchWord.decode('utf-8')):
+			idx = self.kw_id[searchWord.decode('utf-8')]
+			cut = Cut()
+			ii_line = cut.getInverseIndexRow(idx,Global.inverse_dir,Global.filesize)
+			record =json.loads(ii_line)
+			if ishow:
+				for rec in record:
+					line = cut.getRow(int(rec),Global.cutnews_origin_dir,Global.filesize)
+					data = json.loads(line)
+					print data['title'],'\n',data['time'],'\n',data['content'],'\n'
+		#返回单个词项对应的倒排记录表
+			return record
+		else:
+			if isshow:
+				print 'Not Exists Record!'
+			#调用该函数后需要对结果进行判断
+			return dict()
+		
+	
+	#'与'查询：先分词，再合并倒排记录,不考虑权重,返回文档号
+	def QueryPhrase(self,searchPhrase,ishow = True):
+		words = jieba.cut(searchPhrase.decode('utf-8'),cut_all=False)
 		cut = Cut()
-		ii_line = cut.getInverseIndexRow(idx,Global.inverse_dir,Global.filesize)
-		record =json.loads(ii_line)
-		for rec in record:
-			line = cut.getRow(int(rec),Global.cutnews_origin_dir,Global.filesize)
-			data = json.loads(line)
-			print data['title'],'\n',data['time'],'\n',data['content'],'\n'
+		result = set(range(1,100000))
+		for word in words:
+			if not self.kw_id.has_key(word):
+				print 'Not Exist Record'
+				return set()
+			idx = self.kw_id[word]
+			ii_line = cut.getInverseIndexRow(idx,Global.inverse_dir,Global.filesize)
+			record =json.loads(ii_line)
+			re = set()
+			for rec in record:
+				re.add(int(rec))
+			result = result & re
+		print result
+		if ishow:
+			if len(result) == 0:
+				print 'Not Exists Record!'
+			else:
+				for rst in result:
+					line = cut.getRow(int(rst),Global.cutnews_origin_dir,Global.filesize)
+					data = json.loads(line)
+					print data['title'],'\n',data['time'],'\n',data['content'],'\n'
+		return result
 
-	def getQueryItem(self,searchWord):
-		idx = self.kw_id[searchWord.decode('utf-8')]
-		cut = Cut()
-		ii_line = cut.getInverseIndexRow(idx,Global.inverse_dir,Global.filesize)
-		record =json.loads(ii_line)
-		for rec in record:
-			line = cut.getRow(int(rec),Global.cutnews_origin_dir,Global.filesize)
-			data = json.loads(line)
-			print data['title'],'\n',data['time'],'\n',data['content'],'\n'
-
-
-	def getInverseRecord(self,item):
+	#返回热点新闻
+	def QueryHotNews(self):
 		pass
 
-	def mergeInverseRecord(self,RecordList):
+	#返回最新新闻
+	def QueryByTime(self):
 		pass
 	
 search = Search()
-search.getQueryItem(sys.argv[1])
+#search.QueryPhrase(sys.argv[1])
+search.QueryPhrase(sys.argv[1])