Skip to content

Instantly share code, notes, and snippets.

@fredley
Created May 17, 2016 09:13
Show Gist options
  • Star 6 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save fredley/9308b931479a4fac7e46c22face4657f to your computer and use it in GitHub Desktop.
Save fredley/9308b931479a4fac7e46c22face4657f to your computer and use it in GitHub Desktop.
BBC recipe archive tool
"""
Save a copy of all BBC recipes as JSON files.
Instructions:
Run this file with Python 3 in an environment with BeautifulSoup4 and requests:
virtualenv -p python3 env
source env/bin/activate
pip install requests beautifulsoup4
python recipes.py
"""
import codecs
import os
import json
import requests
import string
from bs4 import BeautifulSoup
save_directory = 'recipes'
parser = 'html.parser'
root_url = 'http://www.bbc.co.uk'
if not os.path.exists(save_directory):
os.makedirs(save_directory)
def tryget(el):
try:
return el.text.strip()
except:
return "None"
for letter in list(string.ascii_lowercase):
letter_page = requests.get(root_url + '/food/dishes/by/letter/' + letter)
letter_soup = BeautifulSoup(letter_page.text, parser)
dishes = letter_soup.find_all('li', class_='resource food')
for dish in dishes:
dish_link = dish.find('a').get('href')
dish_page = requests.get(root_url + dish_link)
dish_soup = BeautifulSoup(dish_page.text, parser)
recipes = dish_soup.find_all('li', class_='with-image')
for recipe in recipes:
recipe_link = recipe.find('a').get('href')
recipe_page = requests.get(root_url + recipe_link)
recipe_soup = BeautifulSoup(recipe_page.text, parser)
name = recipe_soup.find('h1', class_='content-title__text').text.strip()
recipe_data = {
'name': name,
'prep_time': tryget(recipe_soup.find('p', class_='recipe-metadata__prep-time')),
'cook_time': tryget(recipe_soup.find('p', class_='recipe-metadata__cook-time')),
'serves': tryget(recipe_soup.find('p', class_='recipe-metadata__serving')),
'author': {
'name': tryget(recipe_soup.find('div', class_='chef__name')),
'programme': tryget(recipe_soup.find('div', class_='chef__programme-name')),
},
'ingredients': [],
'method': []
}
ingredients = recipe_soup.find_all('li', class_='recipe-ingredients__list-item')
for ingredient in ingredients:
recipe_data['ingredients'].append(ingredient.text.strip())
steps = recipe_soup.find_all('li', class_='recipe-method__list-item')
for step in steps:
recipe_data['method'].append(step.text.strip())
recipe_file = codecs.open(os.path.join(save_directory, name + ".json"), 'w', 'utf-8')
recipe_file.write(json.dumps(recipe_data, ensure_ascii=False, indent=2, separators=(',', ': ')))
recipe_file.close()
print('Wrote recipe file for {}'.format(name))
@gut5
Copy link

gut5 commented May 17, 2016

Only scraping 1235 files. Thought there were 11000?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment