Skip to content

Instantly share code, notes, and snippets.

@dyerw
Created November 4, 2015 23:01
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 dyerw/96e8a0a02ee70ed6b7a2 to your computer and use it in GitHub Desktop.
Save dyerw/96e8a0a02ee70ed6b7a2 to your computer and use it in GitHub Desktop.
Scrapes hearthpwn.com for decklist data
import urllib2
from bs4 import BeautifulSoup
PAGES_TO_SCRAPE = 10
# Get HTML for all deck listing pages
htmls = []
for x in range(PAGES_TO_SCRAPE):
print "SCRAPING PAGE " + str(x + 1)
response = urllib2.urlopen('http://hearthpwn.com/decks?filter-deck-tag=3&page=' + str(x + 1))
html = response.read()
htmls += [html]
# Parse all pages for links to each deck
links = []
for html in htmls:
soup = BeautifulSoup(html, 'html.parser')
deck_table = soup.find(attrs={'class' :'listing-decks'})
rows = deck_table.tbody.find_all('tr')
for row in rows:
data = row.find_all('td')
for link in data[0].find_all('a'):
if '/decks/' in link['href']:
links += [link['href']]
# Grab the decklists from each deck link
decklists = []
for link in links:
endpoint = link.split('/')[2]
deck_id = endpoint.split('-')[0]
print deck_id
response = urllib2.urlopen('http://hearthpwn.com/decks/' + deck_id + '/export/1')
deck_list = response.read()
decklists += [deck_list]
# Parse each decklist into python datastructures
def parse_cockatrice_export(decklist):
lines = decklist.split('\n')
deck = {}
for line in lines:
if line and line[0].isdigit():
deck[line[2:-1]] = line[0]
return deck
parsed_decks = map(parse_cockatrice_export, decklists)
with open('deckdata.csv', 'w') as f:
for i, deck in enumerate(parsed_decks):
for card, number in deck.iteritems():
f.write(','.join([str(i), card, number]) + '\n')
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment