def start(self):
print("######################################################")
print("############# PHASE 3 INITIALIZING QUERY #############")
print("######################################################" + '\n')
print("######################################################")
print("############# REVIEW LOOKUP SYSTEM #############")
#print("############# " + "Type 'q!' to quit" + " #############")
print("######################################################" + '\n')
self.reviewsDB = IndexDB('rw.idx')
self.ptermsDB = IndexDB('pt.idx')
self.rtermsDB = IndexDB('rt.idx')
self.scoresDB = IndexDB('sc.idx')
print("Type 'q!' to exit")
class Phase3:
reviewsDB = None
ptermsDB = None
rtermsDB = None
scoresDB = None
rgx = None
firstIntersectFlag = False
def __init__(self):
self.rgx = rgxHandler()
def start(self):
print("######################################################")
print("############# PHASE 3 INITIALIZING QUERY #############")
print("######################################################" + '\n')
print("######################################################")
print("############# REVIEW LOOKUP SYSTEM #############")
#print("############# " + "Type 'q!' to quit" + " #############")
print("######################################################" + '\n')
self.reviewsDB = IndexDB('rw.idx')
self.ptermsDB = IndexDB('pt.idx')
self.rtermsDB = IndexDB('rt.idx')
self.scoresDB = IndexDB('sc.idx')
print("Type 'q!' to exit")
def main(self):
while(1):
query = input("Please provide a Query: ")
print("")
if query == "q!":
self.reviewsDB.close()
self.ptermsDB.close()
self.rtermsDB.close()
self.scoresDB.close()
exit()
parsedQuery = self.queryParser(query)
# print(parsedQuery)
listOfReviews = self.getReviews(parsedQuery)
# print(listOfReviews)
self.displayReviews(listOfReviews)
def displayReviews(self, listOfReviews):
i = 0
for reviewKey in listOfReviews:
i += 1
reviewValue = self.reviewsDB.get(reviewKey)[0]
#print(reviewValue)
print("######################################################")
print("################# REVIEW " + str(i) + " #################")
print("######################################################" + '\n')
reviewValue = self.rgx.putLineTitlesBack(reviewValue)
for line in reviewValue:
if( "review/time" in line):
time = datetime.fromtimestamp(float(line.split(":")[1].strip("\n").strip()))
print("review/time: " + time.strftime("%b %d %Y")+ "\n")
else:
print(line, end='')
print('\n')
def getReviews(self, parsedQuery):
"""
Using the parsedQuery data, intersects the conditional filters amongs the reviews.
Until a filtered list of results is generated.
>>> p3 = Phase3()
>>> p3.start()
######################################################
############# PHASE 3 INITIALIZING QUERY #############
######################################################
<BLANKLINE>
######################################################
############# REVIEW LOOKUP SYSTEM #############
######################################################
<BLANKLINE>
Type 'q!' to exit
>>> parsedQuery = ([], [], [], [])
>>> p3.getReviews(parsedQuery)
[]
>>> parsedQuery = ([], [], [('r', 'ago')], [])
>>> p3.getReviews(parsedQuery)
['9']
>>> parsedQuery = (['ago'], [], [], [])
>>> p3.getReviews(parsedQuery)
['9']
>>> parsedQuery = (['again'], [], [], [])
>>> p3.getReviews(parsedQuery)
['8', '10']
>>> parsedQuery = (['again', 'used'], [], [], [])
>>> p3.getReviews(parsedQuery)
['10']
>>> parsedQuery = ([], ['ag'], [], [])
>>> p3.getReviews(parsedQuery)
['8', '9', '10']
>>> parsedQuery = (['again'], ['ag'], [], [])
>>> p3.getReviews(parsedQuery)
['8', '10']
>>> parsedQuery = ([], [], [], [('rdate', '<', '2000/01/01')])
>>> p3.getReviews(parsedQuery)
['4', '5']
>>> parsedQuery = ([], [], [], [('rdate', '<', '2000/01/01'), ('pprice', '<', '17')])
>>> p3.getReviews(parsedQuery)
['5']
>>> parsedQuery = (['cross'], [], [], [])
>>> p3.getReviews(parsedQuery)
['5', '7', '8', '9', '10']
>>> parsedQuery = ([], [], [('r', 'cross')], [])
>>> p3.getReviews(parsedQuery)
['5', '7', '8', '10']
>>> parsedQuery = ([], [], [('p', 'cross')], [])
>>> p3.getReviews(parsedQuery)
['7', '8', '9', '10']
>>> parsedQuery = ([], ['not'], [], [])
>>> p3.getReviews(parsedQuery)
['1', '2', '8', '9']
>>> parsedQuery = ([], ['not'], [('r', 'cross')], [])
>>> p3.getReviews(parsedQuery)
['8']
>>> parsedQuery = ([], [], [], [('rscore', '<', '5')])
>>> p3.getReviews(parsedQuery)
['1', '3', '4']
>>> parsedQuery = ([], [], [], [('rscore', '>', '4')])
>>> p3.getReviews(parsedQuery)
['2', '5', '6', '7', '8', '9', '10']
>>> parsedQuery = (['find'], [], [], [('rscore', '<', '5')])
>>> p3.getReviews(parsedQuery)
['1', '4']
>>> parsedQuery = ([], [], [], [('pprice', '<', '16')])
>>> p3.getReviews(parsedQuery)
['5', '6']
>>> parsedQuery = (['old'], [], [], [('pprice', '<', '16')])
>>> p3.getReviews(parsedQuery)
['6']
>>> parsedQuery = ([], [], [], [('rdate', '<', '2000/01/01')])
>>> p3.getReviews(parsedQuery)
['4', '5']
>>> parsedQuery = (['find'], [], [], [('rdate', '<', '2000/01/01')])
>>> p3.getReviews(parsedQuery)
['4']
>>> parsedQuery = ([], [], [], [('rdate', '>', '2000/01/01')])
>>> p3.getReviews(parsedQuery)
['1', '2', '3', '6', '7', '8', '9', '10']
>>> parsedQuery = ([], [], [], [('rdate', '>', '2009/01/01'), ('pprice', '>', '16'), ('pprice', '<', '18')])
>>> p3.getReviews(parsedQuery)
['2']
>>> parsedQuery = (['shazam'], [], [], [('rdate', '>', '2009/01/01'), ('pprice', '>', '16'), ('pprice', '<', '18')])
>>> p3.getReviews(parsedQuery)
[]
"""
self.firstIntersectFlag = False
reviewList = []
tmpList = []
#Select by selections, selector = (selector, searchTerm)
for entry in parsedQuery[2]:
selector = entry[0]
term = entry[1]
if(selector == "r"):
subList = self.rtermsDB.get(term)
for i in subList:
tmpList.append(i)
elif(selector == "p"):
subList = self.ptermsDB.get(term)
for i in subList:
tmpList.append(i)
reviewList = self.ourIntersect(reviewList, tmpList)
tmpList = []
#Select by words, word = (searchTerm)
for entry in parsedQuery[0]:
subList = self.rtermsDB.get(entry)
for i in subList:
tmpList.append(i)
subList = self.ptermsDB.get(entry)
for i in subList:
tmpList.append(i)
reviewList = self.ourIntersect(reviewList, tmpList)
tmpList = []
#Select by wilds, wild = (searchTerm)
for entry in parsedQuery[1]:
subList = self.rtermsDB.getWild(entry)
for i in subList:
tmpList.append(i)
subList = self.ptermsDB.getWild(entry)
for i in subList:
tmpList.append(i)
reviewList = self.ourIntersect(reviewList, tmpList)
tmpList = []
#Select by comparator, comparator = (comparator, operator, value)
#pprice < 20
#rdate > 2007/06/20
#rscore < 3
#product/price: unknown
#review/score: 5.0
#review/time: 1075939200
for entry in parsedQuery[3]:
comparator = entry[0]
oper = entry[1]
value = entry[2]
ops = {"<": operator.lt, ">": operator.gt}
if(comparator == "rdate"):
comparator = "rtime"
year,month,day = value.split("/")
try:
value = datetime(int(year), int(month), int(day))
except:
print("Invalid Date Provided. No Results Found.")
return []
else:
value = value + ".0"
keys = self.reviewsDB.getAllReviewKeys()
for key in keys:
item = self.rgx.putLineTitlesBack( self.reviewsDB.get(key)[0] )
itemPrice = item[2].split(":")[1].strip("\n").strip()
itemScore = item[6].split(":")[1].strip("\n").strip()
itemDate = datetime.fromtimestamp( float(item[7].split(":")[1].strip("\n").strip() ))
# print(itemPrice)
# print(itemScore)
# print(itemDate)
# print("")
comp_to_val = {"pprice": itemPrice, "rscore": itemScore, "rtime": itemDate }
if ops[oper](comp_to_val[comparator], value) :
tmpList.append(key)
reviewList = self.ourIntersect(reviewList, tmpList)
tmpList = []
# print(reviewList)
return sorted(reviewList, key=float)
def ourIntersect(self, b1, b2):
if(not self.firstIntersectFlag):
self.firstIntersectFlag = True
return list(set(b2))
else:
return list(set(b1).intersection(b2))
def queryParser(self, query):
"""
Parser returns tuples containing 4 lists containng tuples.
([words], [wilds], [selectors], [comparators])
word = (searchTerm)
wild = (searchTerm)
selector = (selector, searchTerm)
comparator = (comparator, operator, value)
>>> p3 = Phase3()
>>> p3.queryParser("")
([], [], [], [])
>>> result = p3.queryParser("P:caMeRa")
>>> result[2]
[('p', 'camera')]
>>> result = p3.queryParser("r:grEaT")
>>> result[2]
[('r', 'great')]
>>> result =p3.queryParser("cAmeRa")
>>> result[0]
['camera']
>>> result = p3.queryParser("cam%")
>>> result[1]
['cam']
>>> result = p3.queryParser("r:great cam%")
>>> result[1]
['cam']
>>> result[2]
[('r', 'great')]
>>> result = p3.queryParser("rscore > 4")
>>> result[3] == [('rscore', '>', '4')]
True
>>> result = p3.queryParser("camera rscore < 3")
>>> result[0]
['camera']
>>> result[3] == [('rscore', '<', '3')]
True
>>> result = p3.queryParser("pprice < 60 camera")
>>> result[0]
['camera']
>>> result[3] == [('pprice', '<', '60')]
True
>>> result = p3.queryParser("camera rdate > 2007/06/20")
>>> result[0]
['camera']
>>> result[3] == [('rdate', '>', '2007/06/20')]
True
>>> result = p3.queryParser("camera rdate > 2007/06/20 pprice > 20 pprice < 60")
>>> result[0]
['camera']
>>> result[3] == [('rdate', '>', '2007/06/20'), ('pprice', '>', '20'), ('pprice', '<', '60')]
True
"""
query = query.strip().lower().rstrip('\r\n')
searchTerms = [] #(searchTerm)
wildCardTerms = [] #(searchTerm)
selectors = [] #(selector, searchTerm)
comparators = [] #(comparator, operator, value)
selector = re.compile(r"(r:|p:)[a-z]*")
wild = re.compile(r"[a-z]*%")
comparator = re.compile(r"\w*\s(<|>)\s[\w/]*")
word = re.compile(r"[a-z]+")
while(query != ""):
# time.sleep(3)
# print("looping query: " + query)
query.strip().rstrip('\r\n')
if(comparator.search(query)):
found = comparator.search(query).group(0)
query = query.replace(found, "")
if(">" in found):
comparators.append( (found.split(">")[0].strip(),">",found.split(">")[1].strip()) )
else:
comparators.append( (found.split("<")[0].strip(),"<",found.split("<")[1].strip()) )
continue
elif (selector.match(query)):
# print("Selector found")
found = selector.search(query).group(0)
query = query.replace(found, "")
selectors.append((found.split(":")[0],found.split(":")[1]))
continue
elif (wild.search(query)):
# print("wild found")
found = wild.search(query).group(0)
query = query.replace(found, "")
wildCardTerms.append(found.strip("%"))
continue
elif (word.search(query)):
# print("Word found")
found = word.search(query).group(0)
query = query.replace(found, "")
searchTerms.append(found)
continue
else:
break
return (searchTerms,wildCardTerms,selectors,comparators)
