1
0
Fork 0
mirror of https://gitlab.wikimedia.org/ladsgroup/Phabricator-maintenance-bot synced 2024-11-24 21:12:38 +01:00

Start new_wikis_handler.py

Starting the script to eliminate the toil in creating new wikis:
https://landing.google.com/sre/sre-book/chapters/eliminating-toil/

For now, it only takes a bug id and outputs the checklist of stuff that
are done and stuff that needs to be done.
This commit is contained in:
Amir Sarabadani 2020-05-19 11:16:25 +02:00
parent 4e6cac1ba5
commit 016fac156a

224
new_wikis_handler.py Normal file
View file

@ -0,0 +1,224 @@
import base64
import json
import re
import sys
import requests
from lib import Client
def get_file_from_gerrit(path):
gerrit_url = 'https://gerrit.wikimedia.org/g/'
url = gerrit_url + '{0}?format=TEXT'.format(path)
return base64.b64decode(requests.get(url).text).decode('utf-8')
def handle_non_special_wikis(parts, language_code):
if language_code != parts[0]:
return
dns_file = get_file_from_gerrit('operations/dns/+/master/templates/helpers/langlist.tmpl')
return f"'{language_code}'" in dns_file
def handle_special_wiki_dns(parts):
dns_file = get_file_from_gerrit('operations/dns/+/master/templates/wikimedia.org')
name = parts[0]
return f"\n{name}" in dns_file
def handle_special_wiki_apache(parts):
apache_file = get_file_from_gerrit('operations/puppet/+/master/modules/mediawiki/manifests/web/prod_sites.pp')
url = '.'.join(parts)
return url in apache_file
def post_a_comment(comment):
comment = 'Hello, I am helping on creating this wiki. ' + comment + \
' ^_^ Sincerely, your Fully Automated Resource Tackler'
pass
def create_subticket(text, projects, task_phid):
pass
def create_non_special_wikis_dns_subticket(parts, task_details):
pass
def create_special_wikis_dns_subticket(parts, task_details):
pass
def handle_subticket_for_cloud(ticket_phid, task_details, db_name):
pass
def create_apache_config_subticket(parts, task_details):
pass
def main():
print('\n------\n**Pre-install automatic checklist:**')
client = Client.newFromCreds()
gerrit_path = 'https://gerrit.wikimedia.org/g/'
task_details = client.taskDetails(client.lookupPhid(sys.argv[2]))
language_code = re.findall(r'\n- *?\*\*Language code:\*\* *?(\S+)', task_details['description'])
if not language_code:
return
language_code = language_code[0]
url = re.findall(r'\n- *?\*\*Site URL:\*\* *?(\S+)', task_details['description'])
if not url:
return
url = url[0]
parts = url.split('.')
if len(parts) != 3 or parts[2] != 'org':
return
if parts[1] == 'wikimedia':
dns = handle_special_wiki_dns(parts)
special = True
dns_url = gerrit_path + 'operations/dns/+/master/templates/wikimedia.org'
else:
dns = handle_non_special_wikis(parts, language_code)
dns_url = gerrit_path + 'operations/dns/+/master/templates/helpers/langlist.tmpl'
special = False
if not dns:
print(' [] [[{}|DNS]]'.format(dns_url))
if special:
create_special_wikis_dns_subticket(parts, task_details)
else:
create_non_special_wikis_dns_subticket(parts, task_details)
post_a_comment('It seems that there is not DNS entry for this wiki, '
'I am creaing a subticket, Please make a patch.')
else:
print(' [x] [[{}|DNS]]'.format(dns_url))
if parts[1] == 'wikipedia':
db_name = language_code + 'wiki'
else:
db_name = language_code + parts[1]
handle_subticket_for_cloud(client.lookupPhid('T251371'), task_details, db_name)
if special:
apache_url = gerrit_path + 'operations/puppet/+/master/modules/mediawiki/manifests/web/prod_sites.pp'
if not handle_special_wiki_apache(parts):
apache = False
print(' [] [[{}|Apache config]]'.format(apache_url))
create_apache_config_subticket(parts, task_details)
else:
apache = True
print(' [x] [[{}|Apache config]]'.format(apache_url))
else:
apache = True
print(' [x] Apache config (Not needed)')
langdb_url = 'https://raw.githubusercontent.com/wikimedia/language-data/master/data/langdb.yaml'
r = requests.get(langdb_url)
if re.search(r'\n *?' + language_code + ':', r.text):
langdb = True
print(' [x] [[{}|Language configuration in language data repo]]'.format(langdb_url))
else:
langdb = False
print(' [] [[{}|Language configuration in language data repo]]'.format(langdb_url))
core_messages_url = 'https://raw.githubusercontent.com/wikimedia/mediawiki/master/languages/messages/Messages{}.php'.format(
language_code[0].upper() + language_code[1:]
)
r = requests.get(core_messages_url)
if r.status_code == 200:
core_lang = True
print(' [x] [[{}|Language configuration in mediawiki core]]'.format(core_messages_url))
else:
core_lang = False
print(' [] [[{}|Language configuration in mediawiki core]]'.format(core_messages_url))
path = 'mediawiki/extensions/WikimediaMessages/+/master/i18n/wikimediaprojectnames/en.json'
wikimedia_messages_data = get_file_from_gerrit(path)
wikimedia_messages_data = json.loads(wikimedia_messages_data)
if 'project-localized-name-' + db_name in wikimedia_messages_data:
wikimedia_messages_one = True
print(' [x] [[{}|Wikimedia messages configuration]]'.format(gerrit_path + path))
else:
wikimedia_messages_one = False
print(' [] [[{}|Wikimedia messages configuration]]'.format(gerrit_path + path))
url = 'https://en.wikipedia.org/wiki/MediaWiki:Project-localized-name-' + db_name
r = requests.get(url)
if 'Wikipedia does not have a' not in r.text:
wikimedia_messages_one_deployed = True
print(' [x] [[{}|deployed]]'.format(url))
else:
wikimedia_messages_one_deployed = False
print(' [] [[{}|deployed]]'.format(url))
if parts[1] == 'wikipedia':
path = 'mediawiki/extensions/WikimediaMessages/+/master/i18n/wikimediainterwikisearchresults/en.json'
search_messages_data = get_file_from_gerrit(path)
search_messages_data = json.loads(search_messages_data)
if 'search-interwiki-results-' + db_name in search_messages_data:
wikimedia_messages_two = True
print(' [x] [[{}|Wikimedia messages (interwiki search result) configuration]]'.format(gerrit_path + path))
else:
wikimedia_messages_two = False
print(' [] [[{}|Wikimedia messages (interwiki search result) configuration]]'.format(gerrit_path + path))
url = 'https://en.wikipedia.org/wiki/MediaWiki:Search-interwiki-results-' + db_name
r = requests.get(url)
if 'Wikipedia does not have a' not in r.text:
wikimedia_messages_two_deployed = True
print(' [x] [[{}|deployed]]'.format(url))
else:
wikimedia_messages_two_deployed = False
print(' [] [[{}|deployed]]'.format(url))
else:
wikimedia_messages_two = True
wikimedia_messages_two_deployed = True
print(' [x] Wikimedia messages (interwiki search result) configuration (not needed)')
if dns and apache and langdb and core_lang and wikimedia_messages_one and wikimedia_messages_one_deployed and wikimedia_messages_two and wikimedia_messages_two_deployed:
print('**The Wiki is ready to be created.**')
else:
print('**The creation is blocked until these part are all done.**')
print('\n-------\n**Post install automatic checklist:**')
path = 'mediawiki/services/restbase/deploy/+/master/scap/vars.yaml'
restbase = get_file_from_gerrit(path)
if '.'.join(parts) in restbase:
print(' [x] [[{}|RESTbase]]'.format(gerrit_path + path))
else:
print(' [] [[{}|RESTbase]]'.format(gerrit_path + path))
path = 'mediawiki/services/cxserver/+/master/config/languages.yaml'
cxconfig = get_file_from_gerrit(path)
if '\n- ' + language_code in cxconfig:
print(' [x] [[{}|CX Config]]'.format(gerrit_path + path))
else:
print(' [] [[{}|CX Config]]'.format(gerrit_path + path))
path = 'analytics/refinery/+/master/static_data/pageview/whitelist/whitelist.tsv'
refinery_whitelist = get_file_from_gerrit(path)
if '.'.join(parts[:2]) in refinery_whitelist:
print(' [x] [[{}|Analytics refinery]]'.format(gerrit_path + path))
else:
print(' [] [[{}|Analytics refinery]]'.format(gerrit_path + path))
url = 'pywikibot/core/+/master/pywikibot/families/{}_family.py'.format(parts[1])
pywikibot = get_file_from_gerrit(url)
if f"'{language_code}'" in pywikibot:
print(' [x] [[{}|Pywikibot]]'.format(gerrit_path + url))
else:
print(' [] [[{}|Pywikibot]]'.format(gerrit_path + url))
url = 'https://www.wikidata.org/w/api.php?action=help&modules=wbgetentities'
wikiata_help_page = requests.get(url).text
if db_name in wikiata_help_page:
print(' [x] [[{}|Wikidata]]'.format(url))
else:
print(' [] [[{}|Wikidata]]'.format(url))
print(' [] Import from Incubator')
print(' [] Clean up old interwiki links')
main()