import logging
import os.path
import json
-import subprocess
import datetime
from requests import Request
from csv import DictWriter
from mw import api
+import digobs
def parse_args():
parser = argparse.ArgumentParser(description='Call the views API to collect Wikipedia revision data.')
parser.add_argument('-o', '--output_folder', help='Where to save output', default="wikipedia/data", type=str)
parser.add_argument('-i', '--article_file', help='File listing article names', default="wikipedia/resources/enwp_wikiproject_covid19_articles.txt", type=str)
- parser.add_argument('-L', '--logging_level', help='Logging level. Options are debug, info, warning, error, critical. Default: info.', default='info', type=str),
+ parser.add_argument('-L', '--logging_level', help='Logging level. Options are debug, info, warning, error, critical. Default: info.', default='info', type=digobs.get_loglevel),
parser.add_argument('-W', '--logging_destination', help='Logging destination file. (default: standard error)', type=str),
args = parser.parse_args()
return(args)
output_path = args.output_folder
article_filename = args.article_file
- #handle -L
- loglevel_mapping = { 'debug' : logging.DEBUG,
- 'info' : logging.INFO,
- 'warning' : logging.WARNING,
- 'error' : logging.ERROR,
- 'critical' : logging.CRITICAL }
-
- if args.logging_level in loglevel_mapping:
- loglevel = loglevel_mapping[args.logging_level]
- else:
- print("Choose a valid log level: debug, info, warning, error, or critical")
- exit
-
#handle -W
if args.logging_destination:
- logging.basicConfig(filename=args.logging_destination, filemode='a', level=loglevel)
+ logging.basicConfig(filename=args.logging_destination, filemode='a', level=args.logging_level)
else:
- logging.basicConfig(level=loglevel)
+ logging.basicConfig(level=args.logging_level)
- export_git_hash = subprocess.check_output(['git', 'rev-parse', 'HEAD']).decode().strip()
- export_git_short_hash = subprocess.check_output(['git', 'rev-parse', '--short', 'HEAD']).decode().strip()
export_time = str(datetime.datetime.now())
export_date = datetime.datetime.today().strftime("%Y%m%d")
logging.info(f"Starting run at {export_time}")
- logging.info(f"Last commit: {export_git_hash}")
+ logging.info(f"Last commit: {digobs.git_hash()}")
json_output_filename = os.path.join(output_path, f"digobs_covid19-wikipedia-enwiki_revisions-{export_date}.json")
tsv_output_filename = os.path.join(output_path, f"digobs_covid19-wikipedia-enwiki_revisions-{export_date}.tsv")
'sha1' : 'sha1',
'contentmodel' : 'contentmodel',
'tags' : 'tags',
+ 'flags' : 'flags',
'comment' : 'comment',
'content' : 'content' }
- exclude_from_tsv = ['tags', 'comment', 'content']
+ exclude_from_tsv = ['tags', 'comment', 'content', 'flags']
# load the list of articles
with open(article_filename, 'r') as infile:
- article_list = [art.strip() for art in list(infile)]
+ article_list= list(map(str.strip, infile))
def get_revisions_for_page(title):
return api_session.revisions.query(properties=rv_props.values(),
tsv_fields = [e for e in tsv_fields if e not in exclude_from_tsv]
# add special export fields
- tsv_fields = tsv_fields + ['url', 'export_timestamp', 'export_commit']
+ tsv_fields = tsv_fields + ['anon', 'minor', 'url', 'export_timestamp', 'export_commit']
- export_info = { 'git_commit' : export_git_hash,
+ export_info = { 'git_commit' : digobs.git_hash(),
'timestamp' : export_time }
with open(json_output_filename, 'w') as json_output, \
if "sha1" not in rev:
rev["sha1"] = ""
+ if "userhidden" in rev:
+ rev["user"] = ""
+ rev["userid"] = ""
+
+ # recode anon so it's true or false instead of present/missing
+ if "anon" in rev:
+ rev["anon"] = True
+ else:
+ rev["anon"] = False
+
+ # let's recode "minor" in the same way
+ if "minor" in rev:
+ rev["minor"] = True
+ else:
+ rev["minor"] = False
+
# add page title information
rev['title'] = rev['page']['title']
rev['pageid'] = rev['page']['pageid']
'oldid' : rev['revid']}).prepare().url
rev['export_timestamp'] = export_time
- rev['export_commit'] = export_git_short_hash
+ rev['export_commit'] = digobs.git_hash(short=True)
tsv_writer.writerow({k: rev[k] for k in tsv_fields})
- break
if __name__ == "__main__":
-
main()