def write_header_comment_file(self):
if self.header_written: return
comment_file = 'comment_' + self.outfile
wrtr_file = open(comment_file,'ab')
wrtr = UnicodeWriter(wrtr_file)
wrtr.writerow([u'Language', u'Url', u'Published', u'Country', u'ThreadId', u'Inserted', u'PostSize',
u'Subject', u'Text'])
wrtr_file.close()
self.header_written = True
def __init__(self, key, outfile, filter_thread):
self.url = 'http://api.boardreader.com/v1/Blogs/Thread'
self.key = key
self.outfile = outfile
self.filter_thread = filter_thread
self.response = requests.get(self.url, {'key': self.key, 'rt': 'json', 'filter_thread' :self.filter_thread})
wrtr_file = open(self.outfile,'ab')
wrtr = UnicodeWriter(wrtr_file)
for item in self.response.json()['response']['Matches']['Match']:
text = re.sub('[\,\n\t\b-]','',item['Text'])
text = text.replace(',','')
wrtr.writerow([item[u'Language'], item[u'Url'], item[u'Published'], item['Country'],
item[u'ThreadId'], item[u'Inserted'], str(item['PostSize']), item['Subject'].replace(',',''), text])
wrtr_file.close()
def save_response(self):
while self.query_limit > 0:
self.filter_inserted_to = int(time.mktime(time.strptime(self.response['Matches']['Match'][-1]['Published'],"%Y-%m-%d %H:%M:%S")))
self.response['Matches']['Match'].extend(self.get_response()['Matches']['Match'])
wrtr_file = open(self.outfile,'wb')
wrtr = UnicodeWriter(wrtr_file)
wrtr.writerow([u'Language', u'Url', u'Published', u'Country', u'ThreadId', u'Inserted', u'PostSize',
u'Subject', u'Text'])
for item in self.response['Matches']['Match']:
text = re.sub('[\,\n\t\b-]','',item['Text'])
text = text.replace(',','')
wrtr.writerow([item[u'Language'], item[u'Url'], item[u'Published'], item['Country'],
item[u'ThreadId'], item[u'Inserted'], str(item['PostSize']), item['Subject'].replace(',',''), text])
wrtr_file.close()
for item in self.response['Matches']['Match']:
if item[u'CommentsInThread'] > 0:
self.write_header_comment_file()
Comment(key=self.key, outfile='comment_' + self.outfile, filter_thread = item[u'ThreadId'])
def main():
"""docstring"""
LOG.debug('Started')
with open(HTML_DIR + CDC_FILENAME, 'rU') as csv_file:
LOG.info("Reading URLs from %s", HTML_DIR+CDC_FILENAME)
reader = UnicodeReader(csv_file)
# skip the header
reader.next()
for rec in reader:
state = rec[STATE_INDEX]
filename = state.replace(' ', '_')
url = rec[URL_INDEX]
LOG.info("Processing %s", state)
# retrieve the webpage data
webpage_data = get_info_on_webpage(url)
# get the addresses from the local csv files
csv_addresses = get_addresses_from_csv(CSV_DIR + filename + ".csv")
unique_in_webpage = [rec for rec in webpage_data
if normalize_address(rec[ADDRESS_INDEX])
not in csv_addresses]
unique_in_webpage = sorted(unique_in_webpage,
key=operator.itemgetter(CITY_INDEX))
try:
with open(OUTPUT_DIR + filename + ".csv", "wb") as csv_f:
writer = UnicodeWriter(csv_f)
writer.writerows(unique_in_webpage)
except UnicodeEncodeError as error:
print "UnicodeEncode Error"
LOG.debug('Finished')
if phone_info is None: continue
phone = phone_info.text
address = info.find('address').find('span').text
wrtr.writerow([city,subcategory,name,phone,address])
#print name, phone, address, city, subcategory
if data.find('ul','pager'):
if data.find('ul','pager').find('li','next'):
page = page + 1
writerecord(base_url, subcategory,city, page)
if __name__ == '__main__':
base_url = 'http://yellowpages.sulekha.com/clothing-accessories_delhi_clistings'
cities = ['chennai']
outfile = open('chennai_sulekha1.csv','wb')
wrtr = UnicodeWriter(outfile,delimiter=';')
wrtr.writerow(['City','Category','Name','Phone','Address'])
for city in cities:
url = base_url.replace('delhi',city)
next_iter = True
links = BeautifulSoup(urlopen(url),'html.parser')
for line in links.find('ol','business-clisting').findAll('div','blockTitle'):
new_url = line.find('a')['href']
if new_url == 'http://yellowpages.sulekha.com/tie-manufacturers_chennai':
next_iter = False
if next_iter: continue
incity = '_' + city
subcategory = new_url.split('/')[-1].replace('-',' ').replace(incity,'')
writerecord(new_url,subcategory,city)
outfile.close()
data = BeautifulSoup(urlopen(url),"html.parser")
#print url
for item in data.findAll('div','card'):
if (item.find('div','name') == None): break
name = item.find('div','name').text.strip()
place = item.find('div','place').text.strip()
phone = item.find('a','mob-link').text.strip()
if len(phone) == 0: continue
wrtr.writerow([category,name,place,phone])
def locations(city_file):
in_file = open(city_file)
reader = csv.reader(in_file)
for line in reader:
yield (line[0], line[1])
if __name__ == '__main__':
base_url = 'https://www.askme.com/search?q='
outfile = open('askme_mumbai.csv','w')
categories = map(lambda x: x.strip(),open('categories').readlines())
wrtr = UnicodeWriter(outfile)
wrtr.writerow(['category','name','address','phone'])
for locality,city in locations('mumbai_localities.csv'):
loc = locality.replace('.','').replace('-','').replace(' ','+')
for category in categories:
cat = category.strip().replace(' ','+') + '+'
url = base_url+cat+'in+'+loc+'&type=outlets&city='+city
writerecord(url,wrtr,category)
outfile.close()
def file_Write(self,outputList):
with open('translated.csv', 'wb') as files:
writer = UnicodeWriter(files,quoting=csv.QUOTE_ALL)
writer.writerows(outputList)
