def getMovieCodeByAPI(urlOpener, mTitle, mYear):
findList = []
sUrlAdd = urllib.urlencode({"q": mTitle})
urlAdr = IMDBbyTitleAPI + sUrlAdd
try:
url = urlOpener.open(urlAdr)
except:
print("UrlOpener error: Unable to open: " + urlAdr)
return "ce0000000", "Connection error", 0
IMDBfoundAPI = minidom.parseString(url.read())
movies = IMDBfoundAPI.getElementsByTagName("ImdbEntity")
for movie in movies:
description = movie.getElementsByTagName("Description")
for a in description:
if (a.childNodes[0].nodeValue[:4]).isnumeric:
# Some times isnumeric is not working ¿unicode type problem?
sYear = a.childNodes[0].nodeValue[:4]
sYear = filter(type(sYear).isdigit, sYear)
sYear = "0" + sYear
year = int(sYear)
movieFoundTitle = movie.childNodes[0].nodeValue
movieFoundYear = year
movieFoundCode = movie.getAttribute("id")
titleRatio = sDiff(None, mTitle, movieFoundTitle).ratio()
# print("API: " + str(movieFoundTitle) + " | " + str(movieFoundYear) + " | " + str(movieFoundCode) + " | Ratio: " + str(titleRatio))
findList.append({"code": movieFoundCode, "title": movieFoundTitle, "year": movieFoundYear, "ratio": titleRatio})
# Get the best match
bestResult = {"code": "nm0000000", "title": "No match", "year": "0", "ratio": 0}
for movie in findList:
if abs(int(movie["year"]) - int(mYear)) <= 1:
if bestResult["ratio"] < movie["ratio"]:
bestResult["code"] = movie["code"]
bestResult["title"] = movie["title"]
bestResult["year"] = movie["year"]
bestResult["ratio"] = movie["ratio"]
if bestResult["ratio"] > 0.5:
return bestResult["code"], bestResult["title"], bestResult["year"]
else:
# This is a very optimistic way of think, but for my movie list it works nice.
if str(findList[0]["year"]) == mYear:
optimistic_code = findList[0]["code"]
optimistic_title = findList[0]["title"]
optimistic_year = findList[0]["year"]
return optimistic_code, optimistic_title, optimistic_year
else:
# No result.
return None, None, None
def checkTitle(htmlCode, title, year):
print("FIRE!")
start = htmlCode.find("(")
htmlCode = htmlCode[start + 1 : -1]
try:
htmlCode = ast.literal_eval(htmlCode)
except:
print("EVAL Error: ", htmlCode)
return 0, "tt0000000", "Unknow Title", "0000"
# "d" list contain an dictionary with all suggestion. Nice! this is what we want!
for a in htmlCode["d"]:
print(a["id"], a["l"], a["y"])
if "y" in a:
ratio = sDiff(None, title, a["l"]).ratio()
if a["y"] == int(year) and ratio > 0.8:
print("Found: ", a["id"], a["l"], a["y"])
return 1, a["id"], a["l"], a["y"]
return 0, "tt0000000", "Unknow Title", "0000"
def checkTitle(htmlCode, title, year):
print("FIRE!")
start = htmlCode.find("(")
htmlCode = htmlCode[start + 1:-1]
try:
htmlCode = ast.literal_eval(htmlCode)
except:
print("EVAL Error: ", htmlCode)
return 0, "tt0000000", "Unknow Title", "0000"
# "d" list contain an dictionary with all suggestion. Nice! this is what we want!
for a in htmlCode["d"]:
print(a["id"], a["l"], a["y"])
if ("y" in a):
ratio = sDiff(None, title, a["l"]).ratio()
if (a["y"] == int(year) and ratio > 0.8):
print("Found: ", a["id"], a["l"], a["y"])
return 1, a["id"], a["l"], a["y"]
return 0, "tt0000000", "Unknow Title", "0000"
def getMovieCode(urlOpener, mTitle, mYear):
findList = []
sUrlAdd = urllib.urlencode({"q": mTitle, "s": "all"})
urlAdr = IMDBakas + "find?" + sUrlAdd
try:
url = urlOpener.open(urlAdr)
except:
return "ee0000000", "Connection error", 0
urlAdrRed = url.geturl()
if urlAdrRed.find("/title/tt") != -1:
mCode = urlAdrRed[
urlAdrRed.find("/title/tt") + len("/title/tt") - 2 : urlAdrRed.find("/title/tt") + len("/title/tt") + 7
]
pattern = re.compile(
'itemprop="name">([\W\w]*?)<span class="nobr">[\n\w\W]*?\((<a href="\/year\/\d+\/">)?(\d+)(<\/a>)?\)<\/span>'
)
match = pattern.search(url.read())
try:
movieTitle = match.group(1)
movieTitle = movieTitle.replace("\n", "")
movieTitle = htmlFilter(movieTitle)
except:
movieTitle = "Not Found"
try:
movieYear = match.group(2)
movieYear = movieYear.replace("\n", "")
except:
movieYear = "0"
return mCode, movieTitle, movieYear
else:
pattern = re.compile("link=\/title\/tt(\d+)\/" + "'" + ';">([\w\W]+?)<\/a>\s\((\d+)[\w\W]*?\)([\w\W]+?)<\/td>')
iterator = pattern.finditer(url.read())
for result in iterator:
movieCode = "tt" + str(result.group(1))
movieYear = result.group(3)
movieTitle = result.group(2)
movieTitle = htmlFilter(movieTitle)
badCapture = movieTitle.find(';">')
while badCapture != -1:
# While used to fix problems with results with images (Remember KISS).
movieTitle = movieTitle[badCapture + 3 :]
badCapture = movieTitle.find(';">')
titleRatio = sDiff(None, mTitle, movieTitle).ratio()
findList.append({"code": movieCode, "title": movieTitle, "year": movieYear, "ratio": titleRatio})
# Find AKAS tittles.
pattern = re.compile('aka "([\w\W]+?)"')
akasIterator = pattern.finditer(result.group(4))
for akaResult in akasIterator:
akaTitle = akaResult.group(1)
akaTitle = htmlFilter(akaTitle)
titleRatio = sDiff(None, mTitle, akaTitle).ratio()
findList.append({"code": movieCode, "title": akaTitle, "year": movieYear, "ratio": titleRatio})
bestResult = {"code": "nm0000000", "title": "No match", "year": "0", "ratio": 0}
for movie in findList:
if abs(int(movie["year"]) - int(mYear)) <= 1:
if bestResult["ratio"] < movie["ratio"]:
bestResult["code"] = movie["code"]
bestResult["title"] = movie["title"]
bestResult["year"] = movie["year"]
bestResult["ratio"] = movie["ratio"]
if bestResult["ratio"] > 0.5:
return bestResult["code"], bestResult["title"], bestResult["year"]
else:
return "bm0000000", "Bad matches", 0
def getMovieCodeByAPI(urlOpener, mTitle, mYear):
findList = []
sUrlAdd = urllib.urlencode({'q': mTitle})
urlAdr = IMDBbyTitleAPI + sUrlAdd
try:
url = urlOpener.open(urlAdr)
except:
print("UrlOpener error: Unable to open: " + urlAdr)
return "ce0000000", "Connection error", 0
IMDBfoundAPI = minidom.parseString(url.read())
movies = IMDBfoundAPI.getElementsByTagName("ImdbEntity")
for movie in movies:
description = movie.getElementsByTagName("Description")
for a in description:
if (a.childNodes[0].nodeValue[:4]).isnumeric:
# Some times isnumeric is not working ¿unicode type problem?
sYear = a.childNodes[0].nodeValue[:4]
sYear = filter(type(sYear).isdigit, sYear)
sYear = "0" + sYear
year = int(sYear)
movieFoundTitle = movie.childNodes[0].nodeValue
movieFoundYear = year
movieFoundCode = movie.getAttribute("id")
titleRatio = sDiff(None, mTitle, movieFoundTitle).ratio()
# print("API: " + str(movieFoundTitle) + " | " + str(movieFoundYear) + " | " + str(movieFoundCode) + " | Ratio: " + str(titleRatio))
findList.append({
'code': movieFoundCode,
'title': movieFoundTitle,
'year': movieFoundYear,
'ratio': titleRatio
})
# Get the best match
bestResult = {
'code': 'nm0000000',
'title': 'No match',
'year': '0',
'ratio': 0
}
for movie in findList:
if abs(int(movie['year']) - int(mYear)) <= 1:
if bestResult['ratio'] < movie['ratio']:
bestResult['code'] = movie['code']
bestResult['title'] = movie['title']
bestResult['year'] = movie['year']
bestResult['ratio'] = movie['ratio']
if bestResult['ratio'] > 0.5:
return bestResult['code'], bestResult['title'], bestResult['year']
else:
# This is a very optimistic way of think, but for my movie list it works nice.
if str(findList[0]['year']) == mYear:
optimistic_code = findList[0]['code']
optimistic_title = findList[0]['title']
optimistic_year = findList[0]['year']
return optimistic_code, optimistic_title, optimistic_year
else:
# No result.
return None, None, None
def getMovieCode(urlOpener, mTitle, mYear):
findList = []
sUrlAdd = urllib.urlencode({'q': mTitle, 's': 'all'})
urlAdr = IMDBakas + "find?" + sUrlAdd
try:
url = urlOpener.open(urlAdr)
except:
return "ce0000000", "Connection error", 0
urlAdrRed = url.geturl()
urlHTML = url.read()
if urlAdrRed.find("/title/tt") != -1:
mCode = urlAdrRed[urlAdrRed.find("/title/tt") + len("/title/tt") -
2:urlAdrRed.find("/title/tt") + len("/title/tt") + 7]
pattern = re.compile(
'itemprop="name">([\W\w]*?)<span class="nobr">[\n\w\W]*?\((<a href="\/year\/\d+\/">)?(\d+)(<\/a>)?\)<\/span>'
)
match = pattern.search(urlHTML)
try:
movieTitle = match.group(1)
movieTitle = movieTitle.replace("\n", "")
movieTitle = htmlFilter(movieTitle)
except:
movieTitle = "Not Found"
try:
movieYear = match.group(2)
movieYear = movieYear.replace("\n", "")
except:
movieYear = "0"
return mCode, movieTitle, movieYear
else:
pattern = re.compile(
'"result_text"> <a href="\/title\/(tt\d+).*?>(.+?)<\/a>.*?\((\d+)\)',
re.MULTILINE | re.DOTALL)
iterator = pattern.finditer(urlHTML)
for result in iterator:
movieCode = str(result.group(1))
movieYear = result.group(3)
movieTitle = result.group(2)
movieTitle = htmlFilter(movieTitle)
badCapture = movieTitle.find(';">')
while badCapture != -1:
# While used to fix problems with results with images (Remember KISS).
movieTitle = movieTitle[badCapture + 3:]
badCapture = movieTitle.find(';">')
titleRatio = sDiff(None, mTitle, movieTitle).ratio()
findList.append({
'code': movieCode,
'title': movieTitle,
'year': movieYear,
'ratio': titleRatio
})
# Find AKAS tittles.
pattern = re.compile(
'"result_text"> <a href="\/title\/(tt\d+).*?\((\d+)\).*?aka.*?"(.*?)"',
re.MULTILINE | re.DOTALL)
akasIterator = pattern.finditer(urlHTML)
for akaResult in akasIterator:
akaTitle = akaResult.group(3)
akaTitle = htmlFilter(akaTitle)
akaYear = akaResult.group(2)
akaMovieCode = akaResult.group(1)
titleRatio = sDiff(None, mTitle, akaTitle).ratio()
findList.append({
'code': akaMovieCode,
'title': akaTitle,
'year': akaYear,
'ratio': titleRatio
})
# Get the best match
bestResult = {
'code': 'nm0000000',
'title': 'No match',
'year': '0',
'ratio': 0
}
for movie in findList:
if abs(int(movie['year']) - int(mYear)) <= 1:
if bestResult['ratio'] < movie['ratio']:
bestResult['code'] = movie['code']
bestResult['title'] = movie['title']
bestResult['year'] = movie['year']
bestResult['ratio'] = movie['ratio']
if bestResult['ratio'] > 0.5:
return bestResult['code'], bestResult['title'], bestResult['year']
else:
return "bm0000000", "Bad matches", 0
def getMovieCode(self, mTitle, mYear):
findList = []
intento = 0
while intento < MAX_RETRY:
try:
webResponse = self.webSession.get(
self.IMDBakas + "find", params={"ref_": "nv_sr_fn", "q": mTitle.encode("utf-8"), "s": "all"}
)
intento = 99
except:
intento = intento + 1
if intento == MAX_RETRY:
print ("ERROR FOUND: Connection failed at imdb.getMovieCode() - " + mTitle + "(" + mYear + ")")
return self.ImdbFoundMovie(result=self.ImdbFoundMovie.RESULT_BAD_MATCH)
urlAdrRed = webResponse.url
urlHTML = webResponse.text
if urlAdrRed.find("/title/tt") != -1:
mCode = urlAdrRed[
urlAdrRed.find("/title/tt") + len("/title/tt") - 2 : urlAdrRed.find("/title/tt") + len("/title/tt") + 7
]
pattern = re.compile(
'itemprop="name">([\W\w]*?)<\/span>[\n\w\W]*?<span class="nobr">[\n\w\W]*?\(<a href="[\w\W]+?>(\d+)<\/a>'
)
match = pattern.search(urlHTML)
try:
movieTitle = match.group(1)
movieTitle = movieTitle.replace("\n", "")
except:
movieTitle = "Not Found"
try:
movieYear = match.group(2)
movieYear = movieYear.replace("\n", "")
except:
movieYear = "0"
return self.ImdbFoundMovie(mCode, movieTitle, movieYear)
else:
pattern = re.compile(
'"result_text"> <a href="\/title\/(tt\d+).*?>(.+?)<\/a>.*?\((\d+)\)', re.MULTILINE | re.DOTALL
)
iterator = pattern.finditer(urlHTML)
for result in iterator:
movieCode = str(result.group(1))
movieYear = result.group(3)
movieTitle = result.group(2)
badCapture = movieTitle.find(';">')
while badCapture != -1:
movieTitle = movieTitle[badCapture + 3 :]
badCapture = movieTitle.find(';">')
titleRatio = sDiff(None, mTitle, movieTitle).ratio()
findList.append(self.ImdbFoundMovie(code=movieCode, title=movieTitle, year=movieYear, ratio=titleRatio))
# Find AKAS tittles.
pattern = re.compile(
'"result_text"> <a href="\/title\/(tt\d+).*?\((\d+)\).*?aka.*?"(.*?)"', re.MULTILINE | re.DOTALL
)
akasIterator = pattern.finditer(urlHTML)
for akaResult in akasIterator:
akaTitle = akaResult.group(3)
akaYear = akaResult.group(2)
akaMovieCode = akaResult.group(1)
titleRatio = sDiff(None, mTitle, akaTitle).ratio()
if akaTitle.find(mTitle) != -1:
titleRatio2 = 0.6
else:
titleRatio2 = 0
titleRatio = max(titleRatio, titleRatio2)
findList.append(self.ImdbFoundMovie(code=akaMovieCode, title=akaTitle, year=akaYear, ratio=titleRatio))
# Get the best match
bestResult = self.ImdbFoundMovie(ratio=0, result=self.ImdbFoundMovie.RESULT_NO_MATCH)
for movie in findList:
if abs(int(movie.get_year()) - int(mYear)) <= 1:
if bestResult.get_ratio() < movie.get_ratio():
bestResult = movie
if bestResult.get_ratio() > 0.5:
return bestResult
else:
return self.ImdbFoundMovie(result=self.ImdbFoundMovie.RESULT_BAD_MATCH)
def getMovieCodeByAPI(self, mTitle, mYear):
findList = []
intento = 0
while intento < MAX_RETRY:
try:
webResponse = self.webSession.get(self.IMDBbyTitleAPI, params={"q": mTitle.encode("utf-8")})
intento = 99
except:
intento = intento + 1
if intento == MAX_RETRY:
print ("ERROR FOUND: Connection failed at imdb.getMovieCodeByAPI() - " + mTitle + " (" + mYear + ")")
return self.ImdbFoundMovie(result=self.ImdbFoundMovie.RESULT_BAD_MATCH)
# urlHTML = webResponse.text
# Let's some magic to happen
soup = BeautifulSoup(webResponse.text)
urlHTML = soup.get_text()
# Magic is done
# I will be very glad if someone explain me why does it work
IMDBfoundAPI = minidom.parseString(urlHTML)
movies = IMDBfoundAPI.getElementsByTagName("ImdbEntity")
for movie in movies:
description = movie.getElementsByTagName("Description")
for a in description:
rc = []
for node in a:
if node.nodeType == node.TEXT_NODE:
rc.append(node.data)
print "".join(rc)
if (a.childNodes[0].nodeValue[:4]).isnumeric:
# Some times isnumeric is not working ¿unicode type problem?
sYear = a.childNodes[0].nodeValue[:4]
sYear = filter(type(sYear).isdigit, sYear)
sYear = "0" + sYear
year = int(sYear)
movieFoundTitle = movie.childNodes[0].nodeValue
movieFoundYear = year
movieFoundCode = movie.getAttribute("id")
titleRatio = sDiff(None, mTitle, movieFoundTitle).ratio()
# print("API: " + str(movieFoundTitle) + " | " + str(movieFoundYear) + " | " + str(movieFoundCode) + " | Ratio: " + str(titleRatio))
findList.append(
self.ImdbFoundMovie(code=movieFoundCode, title=movieFoundTitle, year=movieFoundYear, ratio=titleRatio)
)
# Get the best match
bestResult = self.ImdbFoundMovie(ratio=0, result=self.ImdbFoundMovie.RESULT_NO_MATCH)
for movie in findList:
if abs(int(movie.get_year()) - int(mYear)) <= 1:
if bestResult.get_ratio() < movie.get_ratio():
bestResult = movie
if bestResult.get_ratio() > 0.5:
return bestResult
else:
# This is a very optimistic way of think, but for my movie list it works nice.
if len(findList) > 0:
if str(findList[0].get_year()) == mYear:
return findList[0]
else:
# No result.
return self.ImdbFoundMovie(result=self.ImdbFoundMovie.RESULT_BAD_MATCH)
else:
# No result.
return self.ImdbFoundMovie(result=self.ImdbFoundMovie.RESULT_BAD_MATCH)
def getMovieCodeByAPI(self, mTitle, mYear):
findList = []
sUrlAdd = urllib.urlencode({'q': mTitle.encode('utf-8')})
urlAdr = self.IMDBbyTitleAPI + sUrlAdd
intento = 0
while intento < MAX_RETRY:
try:
webResponse = self.webSession.open(urlAdr)
intento = 99
except:
intento = intento + 1
if intento == MAX_RETRY:
print("ERROR FOUND: Connection failed at imdb.getMovieCodeByAPI() - " + mTitle + " (" + mYear + ")")
return self.ImdbFoundMovie(result=self.ImdbFoundMovie.RESULT_BAD_MATCH)
urlHTML = webResponse.read()
# urlHTML = unicode(urlHTML, 'utf-8')
IMDBfoundAPI = minidom.parseString(urlHTML)
movies = IMDBfoundAPI.getElementsByTagName("ImdbEntity")
for movie in movies:
description = movie.getElementsByTagName("Description")
for a in description:
if (a.childNodes[0].nodeValue[:4]).isnumeric:
# Some times isnumeric is not working ¿unicode type problem?
sYear = a.childNodes[0].nodeValue[:4]
sYear = filter(type(sYear).isdigit, sYear)
sYear = "0" + sYear
year = int(sYear)
movieFoundTitle = movie.childNodes[0].nodeValue
movieFoundYear = year
movieFoundCode = movie.getAttribute("id")
titleRatio = sDiff(None, mTitle, movieFoundTitle).ratio()
# print("API: " + str(movieFoundTitle) + " | " + str(movieFoundYear) + " | " + str(movieFoundCode) + " | Ratio: " + str(titleRatio))
findList.append(self.ImdbFoundMovie(code=movieFoundCode, title=movieFoundTitle, year=movieFoundYear,
ratio=titleRatio))
# Get the best match
bestResult = self.ImdbFoundMovie(ratio=0, result=self.ImdbFoundMovie.RESULT_NO_MATCH)
for movie in findList:
if abs(int(movie.get_year()) - int(mYear)) <= 1:
if bestResult.get_ratio() < movie.get_ratio():
bestResult = movie
if bestResult.get_ratio() > 0.5:
return bestResult
else:
# This is a very optimistic way of think, but for my movie list it works nice.
if len(findList) > 0:
if str(findList[0].get_year()) == mYear:
return findList[0]
else:
# No result.
return self.ImdbFoundMovie(result=self.ImdbFoundMovie.RESULT_BAD_MATCH)
else:
# No result.
return self.ImdbFoundMovie(result=self.ImdbFoundMovie.RESULT_BAD_MATCH)
def get_title_ratio(self):
try:
titleRatio = sDiff(None, self.__search_title, self.__title).ratio()
except:
raise
return titleRatio
