Skip to content

Instantly share code, notes, and snippets.

@aleszu
Created April 26, 2018 18:21
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save aleszu/c78d64125ce2aa697f291ba3130bb095 to your computer and use it in GitHub Desktop.
Save aleszu/c78d64125ce2aa697f291ba3130bb095 to your computer and use it in GitHub Desktop.
import csv
import requests
import threading
import os
import sys
from Queue import Queue
from bs4 import BeautifulSoup as bs
# Globals
q = Queue()
lock = threading.Lock()
total_articles = 0
cur_article = 0
target_path = ''
# Represents Article
class Article():
def __init__(self, name, link):
self.headline = name
self.link = link
self.text = None
self.done = False
self.trys = 0
# Returns article text
def parse_text(self, html):
soup = bs(html, 'html.parser')
for remove in soup(["script", "stylings", "style"]):
remove.decompose()
text = soup.get_text()
text = ''.join([c for c in text if ord(c) < 128])
text = text.translate({ord(c):None for c in u'\r\n\t'})
if len(text):
self.done = True
return text
# Gets the text for a given article
def getText(self):
try:
r = requests.get(self.link)
self.text = self.parse_text(r.content)
except:
pass # ignore
self.trys += 1
if self.trys >= 2:
self.done = True
return self.done
# saves the article to s3 bucket
# add this instead to JSON
def save(self):
fname = (''.join([c for c in self.headline.split(' ')])).replace('/', '').replace('.', '').replace("'", "") + '.txt'
# write text to file
try:
f = open(target_path + fname, 'w')
f.write(self.text)
f.close()
except:
pass
# Attempt to open file
def openFile(path):
try:
f = open(path, 'r')
return f
except:
print 'file open failed, please check path.'
# skip headers if csv file has column headers
def skipHeaders(reader):
x = raw_input('does file have column headers(y/n): ')
if x == 'y' or x == '':
next(reader) # skip headers
# build queue
def init():
global total_articles
f = openFile(raw_input('relative path to the file: '))
reader = csv.reader(f)
skipHeaders(reader) # skip headers
url_col = int(raw_input('which column has the url [0-n]: '))
name_col = raw_input('which column has article name (press enter if none): ')
name_col = int(name_col) if name_col != '' else url_col
for row in reader:
if row[url_col]:
article = Article(row[name_col], row[url_col])
q.put(article)
total_articles += 1
# Set data destination
def setDestination():
global target_path
dst = raw_input('path of where to store text files: ')
if not os.path.exists(dst):
os.makedirs(dst)
target_path = dst
# updates task done count and prints update on progress
def updateQueue():
global cur_article
global total_articles
with lock:
q.task_done()
cur_article += 1
s = '%d/%d articles complete' % (cur_article, total_articles)
sys.stdout.write('\r' + s)
if cur_article == total_articles:
sys.stdout.write('\n')
sys.stdout.flush()
# thread function
def worker():
while not q.empty():
article = q.get()
resp = article.getText()
if resp:
updateQueue()
article.save()
else:
q.task_done()
q.put(article)
# build the threads
def initThreads():
for i in range(30):
t = threading.Thread(target=worker)
t.daemon = True
t.start()
if __name__ == '__main__':
init()
total_articles = q.qsize()
setDestination()
initThreads()
q.join()
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment