def setUp(self):
BaseTestCase.setUp(self)
for area in areas:
area.put()
for word in excludeWords:
ExcludeWordCharCache.put(word)
address.put()
def setUp(self):
BaseTestCase.setUp(self)
for area in areas:
area.put()
for word in excludeWords:
ExcludeWordCharCache.put(word)
address.put()
def _hasAreaName(cls, areaName, string):
""" 判断在string中是否包含区域名称
"""
position = string.find(areaName)
if position >= 0:
followString = string[position + len(areaName):]
if ExcludeWordCharCache.isStartWith(followString):
return cls._hasAreaName(areaName, followString)
else:
return True
def testLongest(self):
self.assertTrue(ExcludeWordCharCache.isStartWith(u"西街路口"))
def testOverlap(self):
self.assertTrue(ExcludeWordCharCache.isStartWith(u"中街"))
def testNoOverlap(self):
self.assertTrue(ExcludeWordCharCache.isStartWith(u"西路23号"))
def testMatchInMiddle(self):
self.assertFalse(ExcludeWordCharCache.isStartWith(u"上中路口"))
def testPartMatch(self):
self.assertFalse(ExcludeWordCharCache.isStartWith(u"中"))
def testNoOverlap(self):
self.assertTrue(ExcludeWordCharCache.isStartWith(u"西路23号"))
def testEmptyReader(self):
self.assertFalse(ExcludeWordCharCache.isStartWith(u""))
def clear(cls):
cls.deleteAll()
ExcludeWordCharCache.clear()
def put(self):
db.Model.put(self)
ExcludeWordCharCache.put(self.word)
def testEmptyReader(self):
self.assertFalse(ExcludeWordCharCache.isStartWith(u""))
def setUp(self):
BaseTestCase.setUp(self)
for word in excludeWords:
ExcludeWordCharCache.put(word)
def testMatch(self):
self.prepareService(u'{"word":"南路"}').post()
self.assertTrue(ExcludeWordCharCache.isStartWith(u"南路"))
def testLongest(self):
self.assertTrue(ExcludeWordCharCache.isStartWith(u"西街路口"))
def testOverlap(self):
self.assertTrue(ExcludeWordCharCache.isStartWith(u"中街"))
def testMatch(self):
self.prepareService(u'{"word":"南路"}').post()
self.assertTrue(ExcludeWordCharCache.isStartWith(u"南路"))
def setUp(self):
BaseTestCase.setUp(self)
for word in excludeWords:
ExcludeWordCharCache.put(word)
def testPartMatch(self):
self.assertFalse(ExcludeWordCharCache.isStartWith(u"中"))
def parse(cls, address, parent=None):
""" 从指定地址中分析出包含的区域
"""
for i in range(len(address)):
# 获取匹配的区域， 如果指定parent， 则匹配的区域必须是parent的下级区域
(areas, depth) = AreaCharCache.getMatchedAreas(address[i:])
areas = [Area.getByCode(code) for code in areas]
areas = [
area for area in areas
if (not parent) or cls._isChild(parent, area)
]
logging.debug("got areas[%s] for %s" %
(",".join([area.name for area in areas]), address))
if len(areas) > 0:
followStart = i + depth
#如果区域后面跟着特定的词语，如”路“， 则忽略这个区域
#如”湖南路“就不应该认为是”湖南省“
if ExcludeWordCharCache.isStartWith(address[followStart:]):
continue
#如果区域还有下级区域的话，则在后面字符串中继续查找下级区域，然后用找到的下级区域代替上级区域
childrenAreas = []
for area in areas:
if area.hasChild:
children = cls.parse(address[followStart:], area)
if len(children) > 0:
childrenAreas.extend(children)
if len(childrenAreas) > 0: areas = childrenAreas
logging.debug("got children areas[%s] for %s" %
(",".join([area.name
for area in areas]), address))
#如果在结果集中同时存在上级和下级区域，则去除上级区域
parentAreas = []
for area in areas:
parentAreas.extend(cls._getParents(area))
for area in parentAreas:
if area in areas:
areas.remove(area)
logging.debug("areas[%s] after removed parents for %s" %
(",".join([area.name
for area in areas]), address))
#如果存在多个结果的时候，在后续字符中查找上级区域，找到的必然是正确的结果
if len(areas) > 1:
matchedParents = [
area for area in areas
if cls._hasParent(area, address[followStart:])
]
if len(matchedParents) > 0: areas = matchedParents
logging.debug("areas[%s] after matched parents for %s" %
(",".join([area.name
for area in areas]), address))
return areas
'''record unparsed address --- BENSON'''
unparsedAddress = Address(name=address)
unparsedAddress.put()
UnparsedAddressSending.sendMail(unparsedAddress)
return []
def testMatchInMiddle(self):
self.assertFalse(ExcludeWordCharCache.isStartWith(u"上中路口"))