def apply_domain_limit(entries, domain_limit):
""" Takes a list of sanitised URLs and looks at their domains. If a
domain count exceeds the limit, the urls from that domain are
removed and replaced with a single 'domain: url' entry. The counts
of the new 'domain:' entries, and the remaining links are also
returned.
"""
stripped_urls = subdomains(entries)
new_url_set = set()
exceeded_domains = set()
domains_seen = set()
domains_count = dict()
for url in stripped_urls:
if url not in domains_seen:
domains_seen.add(url)
domains_count[url] = 1
else:
domains_count[url] += 1
for url in stripped_urls:
if domains_count[url] >= domain_limit:
exceeded_domains.add(url)
for entry in entries:
if (subdomain(entry) not in exceeded_domains and
rootdomain(entry) not in exceeded_domains):
new_url_set.add(entry)
applied_domain_limit = namedtuple('applied_domain_limit', 'remaining_urls exceeded_domains')
return applied_domain_limit(new_url_set, exceeded_domains)
def apply_domain_limit(entries, domain_limit):
""" Takes a list of sanitised URLs and looks at their domains. If a
domain count exceeds the limit, the urls from that domain are
removed and replaced with a single 'domain: url' entry. The counts
of the new 'domain:' entries, and the remaining links are also
returned.
"""
stripped_urls = subdomains(entries)
new_url_set = set()
exceeded_domains = set()
domains_seen = set()
domains_count = dict()
for url in stripped_urls:
if url not in domains_seen:
domains_seen.add(url)
domains_count[url] = 1
else:
domains_count[url] += 1
for url in stripped_urls:
if domains_count[url] >= domain_limit:
exceeded_domains.add(url)
for entry in entries:
if (subdomain(entry) not in exceeded_domains
and rootdomain(entry) not in exceeded_domains):
new_url_set.add(entry)
applied_domain_limit = namedtuple('applied_domain_limit',
'remaining_urls exceeded_domains')
return applied_domain_limit(new_url_set, exceeded_domains)
def import_from_file_contents(file_contents):
""" Takes a string, such as that from extract_file_contents(), splits
on new lines, and returns a dictionary of 'domain:' entries, and
standalone URL entries. Handles comments by ignoring them.
"""
urls = []
domains = []
file_contents = file_contents.splitlines()
for lineraw in file_contents:
if (not lineraw.isspace()) and (lineraw != ""):
if lineraw.startswith('"') and lineraw.endswith('"'):
lineraw = lineraw[1:-1]
if lineraw[0] == '#':
# comment entry
continue
# checking if url is valid
if not normalize(lineraw):
continue
else:
if lineraw[:7] == "domain:":
# domain entry
line = re.sub("\n", "", lineraw[7:])
# checking if domain url is valid
if not normalize(line):
continue
else:
if line.startswith('"') and line.endswith('"'):
line = line[1:-1]
# We run the domain extract here, as sometimes people accidentally
# put full URLs in domain entries. We assume they mean to exclude
# the domain (which is often now recommended anyway - "no good
# links from bad domains").
domain = subdomain(line)
domains.append(domain)
else:
# not a domain entry
line = re.sub("\n", "", lineraw)
urls.append(line)
return {'urls': urls, 'domains': domains}
def import_from_file_contents(file_contents):
""" Takes a string, such as that from extract_file_contents(), splits
on new lines, and returns a dictionary of 'domain:' entries, and
standalone URL entries. Handles comments by ignoring them.
"""
urls = []
domains = []
file_contents = file_contents.splitlines()
for lineraw in file_contents:
if (not lineraw.isspace()) and (lineraw != ""):
if lineraw.startswith('"') and lineraw.endswith('"'):
lineraw = lineraw[1:-1]
if lineraw[0] == '#':
# comment entry
continue
# checking if url is valid
if not normalize(lineraw):
continue
else:
if lineraw[:7] == "domain:":
# domain entry
line = re.sub("\n", "", lineraw[7:])
# checking if domain url is valid
if not normalize(line):
continue
else:
if line.startswith('"') and line.endswith('"'):
line = line[1:-1]
# We run the domain extract here, as sometimes people accidentally
# put full URLs in domain entries. We assume they mean to exclude
# the domain (which is often now recommended anyway - "no good
# links from bad domains").
domain = subdomain(line)
domains.append(domain)
else:
# not a domain entry
line = re.sub("\n", "", lineraw)
urls.append(line)
return {'urls': urls, 'domains': domains}
def apply_disavow(disavow_entries, urls_list):
""" Using a disavow file, tests which of a file of urls would be
disavowed and which wouldn't.
"""
disavow_links = []
disavow_domains = []
output_dict = {}
if 'urls' in disavow_entries:
disavow_links_details = normalize_and_dedupe_with_counts(
disavow_entries['urls'])
disavow_links = disavow_links_details.clean_urls
output_dict[
'disavow_links_entered'] = disavow_links_details.urls_entered
output_dict[
'unique_disavow_links_entered'] = disavow_links_details.unique_urls_entered
if 'domains' in disavow_entries:
disavow_domains_details = normalize_and_dedupe_with_counts(
disavow_entries['domains'])
disavow_domains = subdomains(disavow_domains_details.clean_urls)
urls_to_test_details = normalize_and_dedupe_with_counts(urls_list)
urls = urls_to_test_details.clean_urls
disavowed_urls = []
non_disavowed_urls = []
for url in urls:
if (url in disavow_links) or (subdomain(url) in disavow_domains):
disavowed_urls.append(url)
else:
non_disavowed_urls.append(url)
total_disavowed_links = len(disavowed_urls)
total_remaining_links = len(non_disavowed_urls)
output_dict.update({
'disavowed': disavowed_urls,
'non_disavowed': non_disavowed_urls,
'domains_entered': disavow_domains_details.urls_entered,
'unique_domains_entered': disavow_domains_details.unique_urls_entered,
'urls_entered_to_test': urls_to_test_details.urls_entered,
'unique_urls_entered_to_test':
urls_to_test_details.unique_urls_entered,
'total_disavowed_links': total_disavowed_links,
'total_remaining_links': total_remaining_links
})
return output_dict
def apply_disavow(disavow_entries, urls_list):
""" Using a disavow file, tests which of a file of urls would be
disavowed and which wouldn't.
"""
disavow_links = []
disavow_domains = []
output_dict = {}
if 'urls' in disavow_entries:
disavow_links_details = normalize_and_dedupe_with_counts(disavow_entries['urls'])
disavow_links = disavow_links_details.clean_urls
output_dict['disavow_links_entered'] = disavow_links_details.urls_entered
output_dict['unique_disavow_links_entered'] = disavow_links_details.unique_urls_entered
if 'domains' in disavow_entries:
disavow_domains_details = normalize_and_dedupe_with_counts(disavow_entries['domains'])
disavow_domains = subdomains(disavow_domains_details.clean_urls)
urls_to_test_details = normalize_and_dedupe_with_counts(urls_list)
urls = urls_to_test_details.clean_urls
disavowed_urls = []
non_disavowed_urls = []
for url in urls:
if (url in disavow_links) or (subdomain(url) in disavow_domains):
disavowed_urls.append(url)
else:
non_disavowed_urls.append(url)
total_disavowed_links = len(disavowed_urls)
total_remaining_links = len(non_disavowed_urls)
output_dict.update({
'disavowed': disavowed_urls,
'non_disavowed': non_disavowed_urls,
'domains_entered': disavow_domains_details.urls_entered,
'unique_domains_entered': disavow_domains_details.unique_urls_entered,
'urls_entered_to_test': urls_to_test_details.urls_entered,
'unique_urls_entered_to_test': urls_to_test_details.unique_urls_entered,
'total_disavowed_links': total_disavowed_links,
'total_remaining_links': total_remaining_links
})
return output_dict
def combine_with_original_disavow(file_contents, disavow_entries):
""" Takes the disavow file passed to disavow_file_to_dict() and it's
resulting output and combines them to create a .txt file with the
relevant 'domain:' entries and individual links to be disavowed,
while maintaining the order and the comments from the original
document.
"""
output = []
# extract = extract_file_contents(disavow_file)
file_contents = file_contents.splitlines()
urls_encountered = set()
domains_encountered = set()
for raw_entry in file_contents:
if (not raw_entry.isspace()) and (raw_entry != ""):
# Strip quotes for lines wrapped in quotes
if raw_entry.startswith('"') and raw_entry.endswith('"'):
raw_entry = raw_entry[1:-1]
if raw_entry[0] == '#':
# line is a comment, so we just keep it
output.append(raw_entry)
continue
if raw_entry[:7] == 'domain:':
# line is an domain entry
# clean the domain entry
domain_normalized = normalize(raw_entry[7:])
# check if it is valid, if not then include it is a comment
if not domain_normalized:
output.append('# invalid entry - ' + raw_entry)
else:
clean_domain = subdomain(domain_normalized)
if clean_domain in disavow_entries['domain_entries']:
if clean_domain not in domains_encountered:
output.append('domain:' + clean_domain)
domains_encountered.add(clean_domain)
else:
output.append('# domain entry already present - ' + clean_domain)
else:
# line is a url entry
# clean the url entry
url_normalized = normalize(raw_entry)
# check if link entry is valid
if not url_normalized:
output.append('# invalid entry - ' + raw_entry)
else:
url_subdomain = subdomain(url_normalized)
url_rootdomain = rootdomain(url_normalized)
if url_subdomain in disavow_entries['domain_entries']:
if url_subdomain not in domains_encountered:
domains_encountered.add(url_subdomain)
output.append('domain:' + url_subdomain)
else:
output.append('# link now disavowed via new domain entry - ' + raw_entry)
elif url_rootdomain in disavow_entries['domain_entries']:
if url_rootdomain not in domains_encountered:
domains_encountered.add(url_rootdomain)
output.append('domain:' + url_rootdomain)
else:
output.append('# link now disavowed via new domain entry - ' + raw_entry)
elif url_normalized in disavow_entries['url_entries']:
if url_normalized not in urls_encountered:
output.append(url_normalized)
urls_encountered.add(url_normalized)
else:
output.append('# link entry already present')
else:
output.append('# error occurred, not sure what to do with this - ' + raw_entry)
return output
def combine_with_original_disavow(file_contents, disavow_entries):
""" Takes the disavow file passed to disavow_file_to_dict() and it's
resulting output and combines them to create a .txt file with the
relevant 'domain:' entries and individual links to be disavowed,
while maintaining the order and the comments from the original
document.
"""
output = []
# extract = extract_file_contents(disavow_file)
file_contents = file_contents.splitlines()
urls_encountered = set()
domains_encountered = set()
for raw_entry in file_contents:
if (not raw_entry.isspace()) and (raw_entry != ""):
# Strip quotes for lines wrapped in quotes
if raw_entry.startswith('"') and raw_entry.endswith('"'):
raw_entry = raw_entry[1:-1]
if raw_entry[0] == '#':
# line is a comment, so we just keep it
output.append(raw_entry)
continue
if raw_entry[:7] == 'domain:':
# line is an domain entry
# clean the domain entry
domain_normalized = normalize(raw_entry[7:])
# check if it is valid, if not then include it is a comment
if not domain_normalized:
output.append('# invalid entry - ' + raw_entry)
else:
clean_domain = subdomain(domain_normalized)
if clean_domain in disavow_entries['domain_entries']:
if clean_domain not in domains_encountered:
output.append('domain:' + clean_domain)
domains_encountered.add(clean_domain)
else:
output.append('# domain entry already present - ' +
clean_domain)
else:
# line is a url entry
# clean the url entry
url_normalized = normalize(raw_entry)
# check if link entry is valid
if not url_normalized:
output.append('# invalid entry - ' + raw_entry)
else:
url_subdomain = subdomain(url_normalized)
url_rootdomain = rootdomain(url_normalized)
if url_subdomain in disavow_entries['domain_entries']:
if url_subdomain not in domains_encountered:
domains_encountered.add(url_subdomain)
output.append('domain:' + url_subdomain)
else:
output.append(
'# link now disavowed via new domain entry - '
+ raw_entry)
elif url_rootdomain in disavow_entries['domain_entries']:
if url_rootdomain not in domains_encountered:
domains_encountered.add(url_rootdomain)
output.append('domain:' + url_rootdomain)
else:
output.append(
'# link now disavowed via new domain entry - '
+ raw_entry)
elif url_normalized in disavow_entries['url_entries']:
if url_normalized not in urls_encountered:
output.append(url_normalized)
urls_encountered.add(url_normalized)
else:
output.append('# link entry already present')
else:
output.append(
'# error occurred, not sure what to do with this - '
+ raw_entry)
return output