def get_json(url):
""" Find and load "standardized" json document containing recipe """
return_value = None
user_agent = {'User-agent': 'Mozilla/5.0'}
page = requests.get(url, headers = user_agent)
match = re.search(r'<script[^>]*type=.?application/ld\+json.?[^>]*>', page.text)
if match:
print_debug("Found an occurance of 'application/ld+json'")
soup = BeautifulSoup(page.text, 'html5lib')
scripts = soup.findAll('script', attrs = {'type':'application/ld+json'})
for script in scripts:
json_stripped=re.sub('^[^\{\[]*', '', script.text)
raw_json = json.loads(json_stripped)
if type(raw_json) == list:
return_value = json_find_array_element(raw_json, '@type', 'Recipe')
try:
return_value['publisher'] = json_clean_value(json_clean_value(source_json, 'publisher', json.loads('{}'), 'name', ''))
if return_value['publisher'] == '':
return_value['publisher'] = json_clean_value(json_find_array_element(raw_json, '@type', 'Organization'), 'name', url2publisher(url))
except:
if not return_value is None:
return_value['publisher'] = url2publisher(url)
elif '@graph' in raw_json and type(raw_json['@graph']) == list:
return_value = json_find_array_element(raw_json['@graph'], '@type', 'Recipe')
try:
return_value['publisher'] = json_clean_value(json_clean_value(source_json, 'publisher', json.loads('{}'), 'name', ''))
if return_value['publisher'] == '':
return_value['publisher'] = json_clean_value(json_find_array_element(raw_json['@graph'], '@type', 'Organization'), 'name', url2publisher(url))
except:
if not return_value is None:
return_value['publisher']=url2publisher(url)
else:
if return_value is None:
try:
if raw_json['@type'] == 'Recipe' and 'recipeIngredient' in raw_json:
return_value = raw_json
else:
return_value = None
except:
return_value = None
try:
return_value['publisher'] = json_clean_value(json_clean_value(source_json, 'publisher', json.loads('{}')), 'name', url2publisher(url))
except:
if not return_value is None:
return_value['publisher']=url2publisher(url)
if (not return_value is None) and ('recipeIngredient' in return_value):
pass
else:
return_value = None
return return_value
def output_group(json_obj,
group_key,
item_key,
item_prefix,
item_wrap=False,
format='rst',
base_level=2):
""" returns string containg formated groups/lists """
out_string = ''
group_count = len(json_clean_value(recipe_json, group_key))
for group_index, group in enumerate(
json_clean_value(recipe_json, group_key)):
group_title = json_clean_value(group, 'title')
if group_title != '':
if group_index > 0:
out_string += '\n'
out_string += output_header(group_title,
format=format,
level=(base_level + 1))
for item_count, item in enumerate(
json_clean_value(group, item_key), 1):
if item_prefix == '#':
prefix = str(item_count).strip() + '. '
else:
prefix = item_prefix.strip() + ' '
if item_wrap:
item_lines = textwrap.wrap(
item,
width=75,
initial_indent=prefix,
subsequent_indent=re.sub('.', ' ', prefix))
for line in item_lines:
out_string += line + '\n'
else:
out_string += prefix.strip() + ' ' + str(
item) + '\n'
return out_string
def get_json(args, url):
""" Find and load "standardized" json document containing recipe """
return_value = None
page = requests.get(url)
page = BeautifulSoup(requests.get(url).text, 'html5lib')
scripts = page.findAll('script')
for script in scripts:
match = re.search(r'root\.__INITIAL_STATE__\.store', script.text)
if match:
for line in iter(script.text.splitlines()):
match = re.search(r'root\.__INITIAL_STATE__\.store', line)
if match:
raw_json_text = re.sub('[^}]*$','', line)
raw_json_text = re.sub('^[^{]*', '', raw_json_text)
raw_json_text = re.sub('"email":{"regExp":.*,"password"', '"email":{"regExp":"","password"', raw_json_text)
raw_json_text = re.sub('"password":{"regExp":.*,"messages"', '"password":{"regExp":""},"messages"', raw_json_text)
raw_json = json.loads(raw_json_text)
return_value = json_clean_value(raw_json, 'content', json.loads('{}'))
#print_debug(json.dumps(return_value, indent=4))
return return_value
def recipe_json2doc(args, recipe_json, format='rst', base_level=1):
""" Build reStructuredText from recipe JSON """
def format2text(format):
""" Formats output ext to human readable format name """
format_text = ''
if format == 'json':
format_text = 'JSON'
elif format == 'md':
format_text = 'Markdown'
elif format == 'rst':
format_text = 'reStructuredText'
else:
format_text = "Unknown format [%s]" % (format)
print_warning("Unknown format [%s]" % (format))
#raise ("ERROR: Unknown format [%s]" % (format))
return format_text
def output_header(header_text, format='rst', level=1):
""" returns string containg formated header """
out_string = ''
if format == 'md':
out_string += '#' * (level + 1)
out_string += ' '
out_string += header_text + '\n'
if format == 'rst':
level_chars = ['=', '-', '^']
level_char = level_chars[level - 1]
out_string += re.sub('.', level_char, header_text) + '\n'
out_string += '\n'
return out_string
def output_group(json_obj,
group_key,
item_key,
item_prefix,
item_wrap=False,
format='rst',
base_level=2):
""" returns string containg formated groups/lists """
out_string = ''
group_count = len(json_clean_value(recipe_json, group_key))
for group_index, group in enumerate(
json_clean_value(recipe_json, group_key)):
group_title = json_clean_value(group, 'title')
if group_title != '':
if group_index > 0:
out_string += '\n'
out_string += output_header(group_title,
format=format,
level=(base_level + 1))
for item_count, item in enumerate(
json_clean_value(group, item_key), 1):
if item_prefix == '#':
prefix = str(item_count).strip() + '. '
else:
prefix = item_prefix.strip() + ' '
if item_wrap:
item_lines = textwrap.wrap(
item,
width=75,
initial_indent=prefix,
subsequent_indent=re.sub('.', ' ', prefix))
for line in item_lines:
out_string += line + '\n'
else:
out_string += prefix.strip() + ' ' + str(
item) + '\n'
return out_string
format_prefix = '-'
if format == 'md':
format_prefix = '*'
print_debug("Building " + format2text(format) +
" from recipe JSON...")
print_debug(recipe_json)
output = output_header(json_clean_value(recipe_json, 'title'),
format)
recipe_yield = json_clean_value(recipe_json, 'yield')
preptime = json_clean_value(recipe_json, 'preptime')
cooktime = json_clean_value(recipe_json, 'cooktime')
totaltime = json_clean_value(recipe_json, 'totaltime')
info = "| "
if preptime != '':
info += 'Prep: ' + preptime + ' | '
if totaltime != '':
info += 'Total: ' + totaltime + ' | '
if recipe_yield != '':
info += 'Yield: ' + str(recipe_yield) + ' | '
info = info.strip()
if info != '|':
divider_line = re.sub('[^|]', '-', info)
if format == 'rst':
divider_line = re.sub('[|]', '+', divider_line)
output += divider_line + '\n' + info + '\n' + divider_line + '\n\n'
# TODO: make this work with markdown and missing URL
url = json_clean_value(recipe_json, 'url')
author = json_clean_value(recipe_json, 'author')
if url is None or url == '':
if not author is None and author != '':
output += 'Source: ' + author + '\n\n'
else:
if author is None or author == '':
author = url2domain(url)
if format == 'md':
output += 'Source: [' + author + '](' + url + ')\n\n'
elif format == 'rst':
output += 'Source: `' + author + ' <' + url + '>`__\n\n'
else:
output += 'Source: ' + author + '\n\n'
description = textwrap.wrap(json_clean_value(
recipe_json, 'description'),
width=75)
for line in description:
output += line + '\n'
output += '\n'
output += output_header('Ingredients', format=format, level=2)
output += output_group(recipe_json,
'ingredient_groups',
'ingredients',
format_prefix,
format=format,
base_level=2)
output += '\n'
output += output_header('Directions', format=format, level=2)
output += output_group(recipe_json,
'direction_groups',
'directions',
'#',
item_wrap=True,
format=format,
base_level=2)
notes = json_clean_value(recipe_json, 'notes')
if not notes is None and notes != '':
output += '\n'
output += output_header('Notes', format=format, level=2)
for note in notes:
note = re.sub('\*\*\*', '', note)
if len(notes) > 1:
not_prefic = format_prefix.strip() + ' '
for line in textwrap.wrap(note,
width=75,
initial_indent=note_prefix,
subsequent_indent=re.sub(
'.', ' ', note_prefix)):
output += line + '\n'
else:
for line in textwrap.wrap(note, width=75):
output += line + '\n'
output += '\n'
return output
def recipe_output_file(args, recipe_json, format=""):
""" Output recipe_json document in the desired format """
def recipe_json2doc(args, recipe_json, format='rst', base_level=1):
""" Build reStructuredText from recipe JSON """
def format2text(format):
""" Formats output ext to human readable format name """
format_text = ''
if format == 'json':
format_text = 'JSON'
elif format == 'md':
format_text = 'Markdown'
elif format == 'rst':
format_text = 'reStructuredText'
else:
format_text = "Unknown format [%s]" % (format)
print_warning("Unknown format [%s]" % (format))
#raise ("ERROR: Unknown format [%s]" % (format))
return format_text
def output_header(header_text, format='rst', level=1):
""" returns string containg formated header """
out_string = ''
if format == 'md':
out_string += '#' * (level + 1)
out_string += ' '
out_string += header_text + '\n'
if format == 'rst':
level_chars = ['=', '-', '^']
level_char = level_chars[level - 1]
out_string += re.sub('.', level_char, header_text) + '\n'
out_string += '\n'
return out_string
def output_group(json_obj,
group_key,
item_key,
item_prefix,
item_wrap=False,
format='rst',
base_level=2):
""" returns string containg formated groups/lists """
out_string = ''
group_count = len(json_clean_value(recipe_json, group_key))
for group_index, group in enumerate(
json_clean_value(recipe_json, group_key)):
group_title = json_clean_value(group, 'title')
if group_title != '':
if group_index > 0:
out_string += '\n'
out_string += output_header(group_title,
format=format,
level=(base_level + 1))
for item_count, item in enumerate(
json_clean_value(group, item_key), 1):
if item_prefix == '#':
prefix = str(item_count).strip() + '. '
else:
prefix = item_prefix.strip() + ' '
if item_wrap:
item_lines = textwrap.wrap(
item,
width=75,
initial_indent=prefix,
subsequent_indent=re.sub('.', ' ', prefix))
for line in item_lines:
out_string += line + '\n'
else:
out_string += prefix.strip() + ' ' + str(
item) + '\n'
return out_string
format_prefix = '-'
if format == 'md':
format_prefix = '*'
print_debug("Building " + format2text(format) +
" from recipe JSON...")
print_debug(recipe_json)
output = output_header(json_clean_value(recipe_json, 'title'),
format)
recipe_yield = json_clean_value(recipe_json, 'yield')
preptime = json_clean_value(recipe_json, 'preptime')
cooktime = json_clean_value(recipe_json, 'cooktime')
totaltime = json_clean_value(recipe_json, 'totaltime')
info = "| "
if preptime != '':
info += 'Prep: ' + preptime + ' | '
if totaltime != '':
info += 'Total: ' + totaltime + ' | '
if recipe_yield != '':
info += 'Yield: ' + str(recipe_yield) + ' | '
info = info.strip()
if info != '|':
divider_line = re.sub('[^|]', '-', info)
if format == 'rst':
divider_line = re.sub('[|]', '+', divider_line)
output += divider_line + '\n' + info + '\n' + divider_line + '\n\n'
# TODO: make this work with markdown and missing URL
url = json_clean_value(recipe_json, 'url')
author = json_clean_value(recipe_json, 'author')
if url is None or url == '':
if not author is None and author != '':
output += 'Source: ' + author + '\n\n'
else:
if author is None or author == '':
author = url2domain(url)
if format == 'md':
output += 'Source: [' + author + '](' + url + ')\n\n'
elif format == 'rst':
output += 'Source: `' + author + ' <' + url + '>`__\n\n'
else:
output += 'Source: ' + author + '\n\n'
description = textwrap.wrap(json_clean_value(
recipe_json, 'description'),
width=75)
for line in description:
output += line + '\n'
output += '\n'
output += output_header('Ingredients', format=format, level=2)
output += output_group(recipe_json,
'ingredient_groups',
'ingredients',
format_prefix,
format=format,
base_level=2)
output += '\n'
output += output_header('Directions', format=format, level=2)
output += output_group(recipe_json,
'direction_groups',
'directions',
'#',
item_wrap=True,
format=format,
base_level=2)
notes = json_clean_value(recipe_json, 'notes')
if not notes is None and notes != '':
output += '\n'
output += output_header('Notes', format=format, level=2)
for note in notes:
note = re.sub('\*\*\*', '', note)
if len(notes) > 1:
not_prefic = format_prefix.strip() + ' '
for line in textwrap.wrap(note,
width=75,
initial_indent=note_prefix,
subsequent_indent=re.sub(
'.', ' ', note_prefix)):
output += line + '\n'
else:
for line in textwrap.wrap(note, width=75):
output += line + '\n'
output += '\n'
return output
def output_filename(filename, ext=""):
""" Ensures filename has proper extension. """
ret_value = filename
if ext != "":
ret_value = (filename + "." + ext).replace(
"." + ext + "." + ext, "." + ext)
return ret_value
ret_value = 0
title = json_clean_value(recipe_json, "title")
if format == '':
if not args.outfile is None and args.outfile != '':
try:
format = (os.path.splitext(args.outfile)[1]).split(".")[-1]
except:
pass
if format == 'json':
output = json.dumps(recipe_json, indent=4)
elif format == 'md':
output = recipe_json2doc(args, recipe_json, format='md')
elif format == 'rst':
output = recipe_json2doc(args, recipe_json, format='rst')
else:
print_error("Unknown format [%s]" % (format))
raise ("ERROR: Unknown format [%s]" % (format))
if output is None or output.strip() == "":
print_error("Problem output is empty")
else:
if args.save_to_file:
if args.outfile is None or args.outfile == "":
savefile = output_filename(re.sub(r'\W+', '', title),
format)
else:
savefile = output_filename(args.outfile, format)
print_info("Writing output to %s..." % savefile)
text_file = open(savefile, "w")
ret_value = text_file.write(output)
text_file.close()
else:
print(output)
return ret_value
def found_paywall(source_json):
ret_value = False
paywall_json = list(json_find_key(source_json, 'paywall'))
if paywall_json or paywall_json[0] == 'TRUE' or json_clean_value(paywall_json[1], 'status') == "READY":
ret_value = True
def ci2json(args, url):
""" Loads Cook's Illustrated (and affiliated) URL and checks for
authentication and then builds Recipe JSON
"""
def get_json(args, url):
""" Get JSON from page """
import pickle
def find_script(source_html):
if source_html is None:
return None
else:
tree = html.fromstring(source_html)
script_element = tree.xpath('//script[@id="__NEXT_DATA__"]')[0]
return json.loads(script_element.text)
def found_paywall(source_json):
ret_value = False
paywall_json = list(json_find_key(source_json, 'paywall'))
if paywall_json or paywall_json[0] == 'TRUE' or json_clean_value(paywall_json[1], 'status') == "READY":
ret_value = True
def cookie_filename(url):
return '.' + url2domain(url) + '.cookies'
def save_cookies(requests_cookiejar, url):
""" save cookie jar """
filename = cookie_filename(url)
# Check if ~/.config/recipe-dl exists
path = os.path.expanduser('~') + "/.config/recipe-dl"
filename = path + '/' + filename
if not os.path.isdir(path):
# If not check for ~/.config and create recipe-dl
if os.path.isdir(os.path.expanduser('~') + "/.config"):
try:
os.makedirs(path)
except OSError:
if not os.path.isdir(path):
raise
else:
path = os.path.expanduser('.')
if os.path.isdir(path):
filename = path + '/' + filename
print_debug ('Saving cookies to ' + path )
with open(filename, 'wb') as f:
pickle.dump(requests_cookiejar, f)
def load_cookies(url):
""" Loads Cookie jar """
print_debug ('Loading cookies...')
filename = cookie_filename(url)
# First look in current directory
if not os.path.isfile(filename):
print_debug ("Unable to find " + filename + ' locally.')
# Next look in ~/.config/recipe-dl
path = os.path.expanduser('~') + "/.config/recipe-dl"
print_debug ("Searching " + path)
if os.path.isfile(path + '/' + filename):
filename = path + '/' + filename
else:
# Lastly look where the script is located.
print_debug ("Not found. Using script location.")
filename = os.path.dirname(os.path.abspath(__file__)) + "/" + filename
if os.path.isfile(filename):
print_debug("found.")
with open(filename, 'rb') as f:
return pickle.load(f)
else:
print_debug ("Unable to find " + filename)
return None
def get_credentials():
""" Retrieve Credentals """
def input_credential(prompt):
""" Prompt and input credentals """
credential = ''
while credential == '':
print_to_console(prompt)
credential = input()
return credential
credential_json = {}
credential_json['user'] = input_credential("Enter email address:")
credential_json['pass'] = input_credential("Enter password:"******""" Load page using existing cookies """
print_debug ("Getting page using cookies...")
recipe_page = None
if cookies is None:
#load cookies and do a request
cookies = load_cookies(url)
if not cookies is None:
print_debug ('cookies = ' + str(requests.utils.dict_from_cookiejar(cookies)))
recipe_page = requests.get(url, cookies=cookies).text
return recipe_page
def get_page_using_session(args, url):
print_debug ("Getting page using sessions...")
auth_json = get_credentials()
session_requests = requests.session()
domain = url2domain(url)
signin_url = "https://" + domain +"/sign_in?next=%2F"
signin_page = session_requests.get(signin_url)
tree = html.fromstring(signin_page.text)
action = tree.xpath('//form[@class="appForm"]/@action')[0]
payload={}
input_elements = tree.xpath('//form[@class="appForm"]//input')
for input_element in input_elements:
payload[input_element.name] = input_element.value
payload['utf8'] = '✓'
payload['user[email]'] = auth_json['user']
payload['user[password]'] = auth_json['pass']
# Perform login
authorize_url = "https://" + domain + action + "?next=%2F"
result = session_requests.post(authorize_url, data = payload, headers = dict(referer = signin_url))
save_cookies(session_requests.cookies, url)
# Grab page
recipe_page = session_requests.get(url, headers = dict(referer = url))
return recipe_page.text
raw_json = json.loads('{ "paywall": true }')
if not args.authorize_ci:
# Getting file using cookies
raw_html = get_page_using_cookie(args, url)
if not raw_html is None:
raw_json = find_script(raw_html)
if args.authorize_ci or raw_html is None or found_paywall(raw_json):
'Getting page using full authentication'
raw_html = get_page_using_session(args, url)
raw_json = find_script(raw_html)
raw_json = raw_json['props']['initialState']['content']['documents']
raw_json = raw_json[list(json.loads(json.dumps(raw_json)))[0]]
return raw_json
print_debug("Using Cook's Illustrated scraper...")
recipe_json={}
recipe_json['url'] = url
source_json = get_json(args, url)
if not source_json is None:
print_debug(str(source_json))
recipe_json['title'] = json_clean_value(source_json, 'title')
recipe_json['description'] = strip_tags(json_clean_value(source_json['metaData']['fields'], 'description'), strip_newline = True)
recipe_json['yield'] = json_clean_value(source_json, 'yields')
# Parse Times
time_note = json_clean_value(source_json, 'recipeTimeNote')
if time_note == '':
time_note = 'TBD'
recipe_json['preptime'] = ''
recipe_json['cooktime'] = ''
recipe_json['totaltime'] = time_note
author = json_clean_value(source_json['metaData']['fields'], 'source')
if author == '':
author = url2publisher(url)
recipe_json['author'] = author
# Ingredients
recipe_json['ingredient_groups'] = []
ingredient_groups = json_clean_value(source_json, "ingredientGroups")
for group in ingredient_groups:
group_json = json.loads('{"title":"","ingredients":[]}')
if len(ingredient_groups) > 1:
group_json['title'] = json_clean_value(group['fields'], 'title')
ingredients = json_clean_value(group['fields'], "recipeIngredientItems")
for ingredient in ingredients:
qty = json_clean_value(ingredient['fields'], "qty")
unit = json_clean_value(ingredient['fields'], "preText")
item = json_clean_value(json_clean_value(ingredient['fields'], "ingredient", json.loads('{"fields": ""}'))['fields'], 'title')
modifier = json_clean_value(ingredient['fields'], "postText")
group_json['ingredients'].append(strip_tags("%s %s %s%s" % (qty, unit, item, modifier), strip_newline = True))
recipe_json['ingredient_groups'].append(group_json)
# Directions
recipe_json['direction_groups'] = []
group_json = json.loads('{"group":"","directions":[]}')
steps = json_clean_value(source_json, "instructions")
for step in steps:
group_json['directions'].append(strip_tags(json_clean_value(step['fields'], "content"), strip_newline = True))
recipe_json['direction_groups'].append(group_json)
recipe_json['notes'] = []
recipe_json['notes'].append(strip_tags(json_clean_value(source_json, 'headnote'), strip_newline = True))
else:
raise UrlError(url, 'URL not supported.')
return recipe_json
def generic2json(args, url):
""" Loads generic URL and builds recipe JSON """
def get_json(url):
""" Find and load "standardized" json document containing recipe """
return_value = None
user_agent = {'User-agent': 'Mozilla/5.0'}
page = requests.get(url, headers = user_agent)
match = re.search(r'<script[^>]*type=.?application/ld\+json.?[^>]*>', page.text)
if match:
print_debug("Found an occurance of 'application/ld+json'")
soup = BeautifulSoup(page.text, 'html5lib')
scripts = soup.findAll('script', attrs = {'type':'application/ld+json'})
for script in scripts:
json_stripped=re.sub('^[^\{\[]*', '', script.text)
raw_json = json.loads(json_stripped)
if type(raw_json) == list:
return_value = json_find_array_element(raw_json, '@type', 'Recipe')
try:
return_value['publisher'] = json_clean_value(json_clean_value(source_json, 'publisher', json.loads('{}'), 'name', ''))
if return_value['publisher'] == '':
return_value['publisher'] = json_clean_value(json_find_array_element(raw_json, '@type', 'Organization'), 'name', url2publisher(url))
except:
if not return_value is None:
return_value['publisher'] = url2publisher(url)
elif '@graph' in raw_json and type(raw_json['@graph']) == list:
return_value = json_find_array_element(raw_json['@graph'], '@type', 'Recipe')
try:
return_value['publisher'] = json_clean_value(json_clean_value(source_json, 'publisher', json.loads('{}'), 'name', ''))
if return_value['publisher'] == '':
return_value['publisher'] = json_clean_value(json_find_array_element(raw_json['@graph'], '@type', 'Organization'), 'name', url2publisher(url))
except:
if not return_value is None:
return_value['publisher']=url2publisher(url)
else:
if return_value is None:
try:
if raw_json['@type'] == 'Recipe' and 'recipeIngredient' in raw_json:
return_value = raw_json
else:
return_value = None
except:
return_value = None
try:
return_value['publisher'] = json_clean_value(json_clean_value(source_json, 'publisher', json.loads('{}')), 'name', url2publisher(url))
except:
if not return_value is None:
return_value['publisher']=url2publisher(url)
if (not return_value is None) and ('recipeIngredient' in return_value):
pass
else:
return_value = None
return return_value
print_debug("Using generic scraper...")
recipe_json={}
recipe_json['url'] = url
source_json = get_json(url)
if source_json is None:
print_info("No application+ld json attempting to use recipe-scrapers...")
recipe_json = recipe_scraper2json(args, url)
else:
print_debug(json.dumps(source_json))
recipe_json['title'] = json_clean_value(source_json, 'headline', json_clean_value(source_json, 'name'))
recipe_json['description'] = json_clean_value(source_json, 'description')
if 'recipeYield' in source_json and type(source_json['recipeYield']) == list:
recipe_json['yield'] = max(source_json['recipeYield'])
else:
recipe_json['yield'] = json_clean_value(source_json, 'recipeYield')
# Parse Times
minutes_total = iso8601.to_minutes(json_clean_value(source_json, 'totalTime'))
minutes_cook = iso8601.to_minutes(json_clean_value(source_json, 'cookTime'))
minutes_prep = iso8601.to_minutes(json_clean_value(source_json, 'prepTime'))
if minutes_prep == 0 and minutes_total > 0 and minutes_cook > 0:
minutes_prep = minutes_total - minutes_cook
if minutes_total == 0 and (minutes_prep > 0 or minutes_cook > 0):
minutes_total = minutes_prep + minutes_cook
recipe_json['preptime'] = minutes2time(minutes_prep, '')
recipe_json['cooktime'] = minutes2time(minutes_cook, '')
recipe_json['totaltime'] = minutes2time(minutes_total)
# Parse Author
publisher = json_clean_value(source_json, 'publisher')
author = json_clean_value(source_json, 'author')
if type(author) == list:
if 'name' in author[0]:
author = author[0]['name']
elif 'name' in author:
author = author['name']
if publisher != "":
if author == "" or publisher == author:
author == publisher
else:
if not (publisher in author):
author = publisher + ' (' + author + ')'
recipe_json['author'] = author
# Ingredients
ingredients = list(json_find_key(source_json, "recipeIngredient"))[0]
recipe_json['ingredient_groups'] = []
recipe_json['ingredient_groups'].append(json.loads('{"title":"","ingredients":[]}'))
out_ingredients = []
for ingredient in ingredients:
out_ingredients.append(strip_tags(ingredient))
recipe_json['ingredient_groups'][0]['ingredients'] = out_ingredients
# Directions
out_instruction=[]
instructionsSection=list(json_find_key(source_json, 'recipeInstructions'))[0]
try:
instructions=list(json_find_key(source_json, 'itemListElement'))[0]
except IndexError:
instructions=instructionsSection
print_debug(str(instructions))
if str(instructions)[0] == '[':
for instruction in instructions:
try:
instruction_json = instruction
out_instruction.append(strip_tags(instruction_json['text']))
except:
out_instruction.append(strip_tags(str(instruction)))
else:
out_instruction.append(strip_tags(str(instructions)))
recipe_json['direction_groups'] = []
recipe_json['direction_groups'].append(json.loads('{"group":"","directions":[]}'))
recipe_json['direction_groups'][0]['directions'] = out_instruction
return recipe_json
def epicurious2json(args, url):
""" Loads Epicurious URL and builds recipe JSON """
def get_json(args, url):
""" Find and load "standardized" json document containing recipe """
return_value = None
page = requests.get(url)
page = BeautifulSoup(requests.get(url).text, 'html5lib')
scripts = page.findAll('script')
for script in scripts:
match = re.search(r'root\.__INITIAL_STATE__\.store', script.text)
if match:
for line in iter(script.text.splitlines()):
match = re.search(r'root\.__INITIAL_STATE__\.store', line)
if match:
raw_json_text = re.sub('[^}]*$','', line)
raw_json_text = re.sub('^[^{]*', '', raw_json_text)
raw_json_text = re.sub('"email":{"regExp":.*,"password"', '"email":{"regExp":"","password"', raw_json_text)
raw_json_text = re.sub('"password":{"regExp":.*,"messages"', '"password":{"regExp":""},"messages"', raw_json_text)
raw_json = json.loads(raw_json_text)
return_value = json_clean_value(raw_json, 'content', json.loads('{}'))
#print_debug(json.dumps(return_value, indent=4))
return return_value
print_debug("Using Epicurious scraper...")
recipe_json={}
recipe_json['url'] = url
source_json = get_json(args, url)
if not source_json is None:
recipe_json['title'] = json_clean_value(source_json, 'hed')
recipe_json['description'] = strip_tags(json_clean_value(source_json, 'dek'))
recipe_json['yield'] = json_clean_value(json_clean_value(source_json, 'servingSizeInfo',json.loads('{}')), 'servingSizeDescription')
# Parse Times
minutes_prep = iso8601.to_minutes(json_clean_value(source_json, 'formattedPrepTime'))
minutes_cook = iso8601.to_minutes(json_clean_value(source_json, 'formattedCookTime'))
minutes_total = minutes_prep + minutes_cook
if minutes_prep == 0 and minutes_total > 0 and minutes_cook > 0:
minutes_prep = minutes_total - minutes_cook
recipe_json['preptime'] = minutes2time(minutes_prep, '')
recipe_json['cooktime'] = minutes2time(minutes_cook, '')
recipe_json['totaltime'] = minutes2time(minutes_total)
# Parse Author
publisher = "Epicurious"
author = json_clean_value(source_json, 'author', '')
if type(author) == list:
if 'name' in author[0]:
author = author[0]['name']
elif 'name' in author:
author = author['name']
if publisher != "":
if author == "" or publisher == author:
author == publisher
else:
if not (publisher in author):
author = publisher + ' (' + author + ')'
recipe_json['author'] = author
# Ingredients
recipe_json['ingredient_groups'] = []
ingredient_groups = json_clean_value(source_json, "ingredientGroups")
for group in ingredient_groups:
group_json = json.loads('{"title":"","ingredients":[]}')
if len(ingredient_groups) > 1:
group_json['title'] = json_clean_value(group_json, "hed")
ingredients = json_clean_value(group, "ingredients")
for ingredient in ingredients:
group_json['ingredients'].append(strip_tags(json_clean_value(ingredient, "description")))
recipe_json['ingredient_groups'].append(group_json)
# Directions
recipe_json['direction_groups'] = []
direction_groups = json_clean_value(source_json, "preparationGroups")
for group in direction_groups:
group_json = json.loads('{"group":"","directions":[]}')
if len(direction_groups) > 1:
group_json['group'] = strip_tags(json_clean_value(group_json, "hed"))
steps = json_clean_value(group, "steps")
for step in steps:
group_json['directions'].append(strip_tags(json_clean_value(step, "description")))
recipe_json['direction_groups'].append(group_json)
else:
raise UrlError(url, 'URL not supported.')
return recipe_json
