Compare commits
4 Commits
article_id
...
master
| Author | SHA1 | Date | |
|---|---|---|---|
|
befb87c8f5 | ||
|
|
1a27b68061 | ||
|
|
e32e826083 | ||
|
|
0c4cfcdfcf |
27
cron-wikipedia_revisions.sh
Normal file → Executable file
27
cron-wikipedia_revisions.sh
Normal file → Executable file
@ -1,30 +1,21 @@
|
||||
#!/bin/bash -x
|
||||
|
||||
WORKING_DIR="/home/SOC.NORTHWESTERN.EDU/bmh1867/covid19"
|
||||
cd $WORKING_DIR
|
||||
|
||||
TZ="UTC"
|
||||
date_string=$(date +%Y%m%d)
|
||||
|
||||
revs_log="enwp-revisions-${date_string}.log"
|
||||
./wikipedia/scripts/wikiproject_scraper.py 2> >(tee wikipedia/logs/${revs_log})
|
||||
|
||||
wd_log="wd-page-crawler-${date_string}.log"
|
||||
python3 ./real-time-wiki-covid-tracker/PageCrawler.py -a "./wikipedia/resources/enwp_wikiproject_covid19_articles.txt" 2> >(tee wikipedia/logs/${wd_log})
|
||||
|
||||
./wikipedia/scripts/fetch_revisions.py 2> >(tee -a wikipedia/logs/${revs_log})
|
||||
./wikipedia/scripts/fetch_enwiki_revisions.py 2> >(tee -a wikipedia/logs/${revs_log})
|
||||
mv wikipedia/logs/${revs_log} /var/www/covid19/wikipedia/logs/
|
||||
|
||||
python3 ./wikipedia/scripts/copy_revisions_data.py ${date_string}
|
||||
revs_tsv="digobs_covid19-wikipedia-enwiki_revisions-${date_string}.tsv"
|
||||
mv wikipedia/data/${revs_tsv} /var/www/covid19/wikipedia
|
||||
|
||||
cd wikipedia/data
|
||||
xz */${date_string}/*revisions*.json
|
||||
revs_json="digobs_covid19-wikipedia-enwiki_revisions-${date_string}.json"
|
||||
xz wikipedia/data/${revs_json}
|
||||
mv wikipedia/data/${revs_json}.xz /var/www/covid19/wikipedia
|
||||
|
||||
find */${date_string}/*revisions*.xz | while read line; do
|
||||
mkdir -p /var/www/covid9/wikipedia/$line
|
||||
mv $line /var/www/covid19/wikipedia/$line
|
||||
done
|
||||
|
||||
find */${date_string}/*revisions*.tsv | while read line; do
|
||||
mkdir -p /var/www/covid19/wikipedia/$line
|
||||
mv $line /var/www/covid19/wikipedia/$line
|
||||
done
|
||||
|
||||
cd ../..
|
||||
|
||||
24
cron-wikipedia_views.sh
Normal file → Executable file
24
cron-wikipedia_views.sh
Normal file → Executable file
@ -1,27 +1,19 @@
|
||||
#!/bin/bash -x
|
||||
|
||||
WORKING_DIR="/home/SOC.NORTHWESTERN.EDU/bmh1867/covid19"
|
||||
cd $WORKING_DIR
|
||||
|
||||
TZ="UTC"
|
||||
date_string=${OVERRIDE_DATE_STRING:-$(date +%Y%m%d)}
|
||||
|
||||
view_log="daily_views-${date_string}.log"
|
||||
view_log="enwp-daily_views-${date_string}.log"
|
||||
./wikipedia/scripts/wikiproject_scraper.py 2> >(tee wikipedia/logs/${view_log})
|
||||
|
||||
wd_log="wd-page-crawler-${date_string}.log"
|
||||
python3 ./real-time-wiki-covid-tracker/PageCrawler.py -a "./wikipedia/resources/enwp_wikiproject_covid19_articles.txt" 2> >(tee wikipedia/logs/${wd_log})
|
||||
|
||||
# get the list of files
|
||||
./wikipedia/scripts/fetch_daily_views.py -d "${date_string}" 2> >(tee -a wikipedia/logs/${view_log})
|
||||
./wikipedia/scripts/fetch_enwiki_daily_views.py -d "${date_string}" 2> >(tee -a wikipedia/logs/${view_log})
|
||||
mv wikipedia/logs/${view_log} /var/www/covid19/wikipedia/logs/${view_log}
|
||||
mv wikipedia/data/digobs_covid19-wikipedia-enwiki_dailyviews-${date_string}.tsv /var/www/covid19/wikipedia/
|
||||
|
||||
cd wikipedia/data
|
||||
find */${date_string}/*dailyviews*.tsv | while read line; do
|
||||
mkdir -p /var/www/covid19/wikipedia/$line
|
||||
mv $line /var/www/covid19/wikipedia/$line
|
||||
done
|
||||
# xz wikipedia/data/digobs_covid19-wikipedia-enwiki_dailyviews-${date_string}.json
|
||||
mv wikipedia/data/digobs_covid19-wikipedia-enwiki_dailyviews-${date_string}.json /var/www/covid19/wikipedia/
|
||||
|
||||
find */${date_string}/*dailyviews*.json | while read line; do
|
||||
mkdir -p /var/www/covid19/wikipedia/$line
|
||||
mv $line /var/www/covid19/wikipedia/$line
|
||||
done
|
||||
|
||||
cd ../..
|
||||
|
||||
3
wikipedia/scripts/.gitignore
vendored
3
wikipedia/scripts/.gitignore
vendored
@ -1,3 +0,0 @@
|
||||
\#*\#
|
||||
.\#*
|
||||
*~
|
||||
@ -1,17 +1,8 @@
|
||||
#!/usr/bin/env python3
|
||||
from requests import Request
|
||||
from datetime import datetime
|
||||
|
||||
import sys
|
||||
import subprocess
|
||||
import logging
|
||||
import itertools
|
||||
import requests
|
||||
from functools import partial
|
||||
from csv import DictWriter
|
||||
import json
|
||||
import mwapi as api
|
||||
|
||||
user_agent = "COVID-19 Digital Observatory, a Community Data Science Collective project. (https://github.com/CommunityDataScienceCollective/COVID-19_Digital_Observatory)"
|
||||
|
||||
def git_hash(short=False):
|
||||
if short:
|
||||
@ -19,67 +10,6 @@ def git_hash(short=False):
|
||||
else:
|
||||
subprocess.check_output(['git', 'rev-parse', 'HEAD']).decode().strip()
|
||||
|
||||
|
||||
def init_logging(args):
|
||||
#handle -W
|
||||
if args.logging_destination:
|
||||
logging.basicConfig(filename=args.logging_destination, filemode='a', level=args.logging_level)
|
||||
else:
|
||||
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.now())
|
||||
|
||||
logging.info(f"Starting at {export_time}.")
|
||||
logging.info(f"Last commit: {export_git_hash}")
|
||||
return(logging)
|
||||
|
||||
def call_view_api(page, project, query_date):
|
||||
url= f"https://wikimedia.org/api/rest_v1/metrics/pageviews/per-article/{project}/all-access/all-agents/{page}/daily/{query_date}00/{query_date}00"
|
||||
response = requests.get(url)
|
||||
if response.ok:
|
||||
return response.json().get('items',[None])[0]
|
||||
|
||||
else:
|
||||
logging.warning(f"Failure: {response.status_code} from {url}")
|
||||
print(response.json().get("detail",None))
|
||||
return None
|
||||
|
||||
# This function writes out the view data to a json file (j_outfile)
|
||||
# and a tsv file (t_outfile), keeps track of failures,
|
||||
# and returns the number of successes and failures
|
||||
def process_view_responses(responses, j_outfile, t_outfile, logging = None):
|
||||
|
||||
failures = len(list(itertools.takewhile(lambda r: r is None, responses)))
|
||||
successes = 1
|
||||
|
||||
try:
|
||||
first_response = next(responses)
|
||||
except StopIteration:
|
||||
logging.error("No valid responses")
|
||||
exit()
|
||||
|
||||
dw = DictWriter(t_outfile, sorted(first_response.keys()), delimiter='\t')
|
||||
dw.writeheader()
|
||||
json.dump(first_response, j_outfile)
|
||||
dw.writerow(first_response)
|
||||
|
||||
for response in responses:
|
||||
if response is None:
|
||||
failures = failures + 1
|
||||
continue
|
||||
else:
|
||||
successes = successes + 1
|
||||
|
||||
if logging is not None:
|
||||
logging.debug(f"printing data: {response}")
|
||||
|
||||
json.dump(response, j_outfile)
|
||||
dw.writerow(response)
|
||||
|
||||
return (successes,failures)
|
||||
|
||||
def get_loglevel(arg_loglevel):
|
||||
loglevel_mapping = { 'debug' : logging.DEBUG,
|
||||
'info' : logging.INFO,
|
||||
@ -95,74 +25,3 @@ def get_loglevel(arg_loglevel):
|
||||
return logging.INFO
|
||||
|
||||
|
||||
def get_revisions_for_page(title, api_session, logging, rv_props):
|
||||
|
||||
result = api_session.get(action='query',
|
||||
prop='revisions',
|
||||
rvprop=rv_props.values(),
|
||||
titles = {title},
|
||||
rvdir='newer',
|
||||
rvslots='*',
|
||||
continuation=True)
|
||||
return result
|
||||
|
||||
def get_pages_revisions(titles, project, logging, rv_props):
|
||||
logging.info(f"pulling revisions for: {project}")
|
||||
|
||||
api_session = api.Session(f"https://{project}.org/w/api.php",
|
||||
user_agent=user_agent
|
||||
)
|
||||
|
||||
return itertools.chain(* map(partial(get_revisions_for_page, api_session = api_session, logging = logging, rv_props = rv_props), titles))
|
||||
|
||||
def rev_batch_to_json(rev, export_info, json_output = None):
|
||||
rev['exported'] = export_info
|
||||
if json_output is None:
|
||||
return json.dumps(rev)
|
||||
else:
|
||||
json.dump(rev, json_output)
|
||||
|
||||
def rev_batch_to_tsv(batch, tsv_fields, export_info, project, tsv_writer=None):
|
||||
batch = batch.get('query',dict())
|
||||
pages = batch.get('pages',dict())
|
||||
for pageid, page in pages.items():
|
||||
pageid = page.get('pageid',None)
|
||||
ns = page.get('ns',None)
|
||||
title = page.get('title','')
|
||||
logging.info(f"pulling revisions for: {title}")
|
||||
revs = page.get('revisions',[])
|
||||
for rev in revs:
|
||||
|
||||
# handle missing data
|
||||
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'] = title
|
||||
rev['pageid'] = pageid
|
||||
rev['namespace'] = ns
|
||||
rev['contentmodel'] = rev['slots']['main']['contentmodel']
|
||||
# construct a URL
|
||||
rev['url'] = Request('GET', f'https://{project}.org/w/index.php',
|
||||
params={'title' : rev['title'].replace(" ", "_"),
|
||||
'oldid' : rev['revid']}).prepare().url
|
||||
|
||||
rev['export_timestamp'] = export_info['export_timestamp']
|
||||
rev['export_commit'] = export_info['export_commit']
|
||||
tsv_writer.writerow({k: rev[k] for k in tsv_fields})
|
||||
|
||||
@ -1,85 +0,0 @@
|
||||
#!/usr/bin/env python3
|
||||
import argparse
|
||||
import sqlite3
|
||||
import requests
|
||||
from datetime import datetime, timedelta
|
||||
import logging
|
||||
import digobs
|
||||
from os import path, mkdir
|
||||
from functools import partial
|
||||
from itertools import chain
|
||||
|
||||
if __name__ == "__main__":
|
||||
|
||||
parser = argparse.ArgumentParser(description="Get a list of pages related to COVID19, pandemic, and SARS-COV2 virus related entities.")
|
||||
parser.add_argument('-o', '--output_folder', help='Where to save output', default="wikipedia/data", 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=argparse.FileType('a'))
|
||||
parser.add_argument('-d', '--query_date', help='Date if not yesterday, in YYYYMMDD format.', type=lambda s: datetime.strptime(s, "%Y%m%d"))
|
||||
|
||||
parser.add_argument('-i', '--input_file', help="Input a file of page names from the English Wikiproject.", type=argparse.FileType('r'), default='./wikipedia/resources/enwp_wikiproject_covid19_articles.txt')
|
||||
|
||||
parser.add_argument('-b', '--input_db', help="Input a path to a sqlite3 database from the real-time-covid-tracker project", type = sqlite3.connect, default='real-time-wiki-covid-tracker/AllWikidataItems.sqlite')
|
||||
|
||||
args = parser.parse_args()
|
||||
conn = args.input_db
|
||||
conn.row_factory = sqlite3.Row
|
||||
|
||||
#handle -d
|
||||
if args.query_date:
|
||||
query_date = args.query_date.strftime("%Y%m%d")
|
||||
else:
|
||||
yesterday = datetime.today() - timedelta(days=1)
|
||||
query_date = yesterday.strftime("%Y%m%d")
|
||||
|
||||
digobs.init_logging(args)
|
||||
|
||||
logging.info(f"Destructively outputting results to {args.output_folder}")
|
||||
|
||||
#1 Load up the list of article names
|
||||
|
||||
logging.info("loading info from database")
|
||||
projects = [row['project'] for row in conn.execute("SELECT DISTINCT project from pagesPerProjectTable;").fetchall()]
|
||||
|
||||
successes = 0
|
||||
failures = 0
|
||||
|
||||
for project in projects:
|
||||
project_folder = path.join(args.output_folder, project)
|
||||
if not path.exists(project.folder):
|
||||
mkdir(project_folder)
|
||||
|
||||
dump_folder = path.join(projct.folder, export_date)
|
||||
if not path.exists(dump_folder):
|
||||
mkdir(dump_folder)
|
||||
|
||||
logging.info(f"Getting page views for {project}")
|
||||
rows = conn.execute(f"SELECT DISTINCT page from pagesPerProjectTable WHERE project='{project}';").fetchall()
|
||||
pages = (row['page'] for row in rows)
|
||||
|
||||
# special case for english, we have a wikiproject input file
|
||||
if project == "en.wikipedia":
|
||||
pages = chain(pages, map(str.strip,args.input_file))
|
||||
|
||||
call_view_api = partial(digobs.call_view_api, project=project, query_date = query_date)
|
||||
|
||||
responses = map(call_view_api, pages)
|
||||
|
||||
j_outfilename = path.join(j_output_folder, f"digobs_covid19_{project}_dailyviews-{query_date}.json")
|
||||
t_outfilename = path.join(t_output_folder, f"digobs_covid19_{project}_dailyviews-{query_date}.tsv")
|
||||
|
||||
with open(j_outfilename, 'w') as j_outfile, \
|
||||
open(t_outfilename, 'w') as t_outfile:
|
||||
|
||||
proj_successes, proj_failures = digobs.process_view_responses(responses, j_outfile, t_outfile, logging)
|
||||
logging.info(f"(Processed {proj_successes} successes and {proj_failures} for {project}")
|
||||
successes = proj_successes + successes
|
||||
failures = proj_failures + failures
|
||||
|
||||
conn.close()
|
||||
# f_Out = outputPath + "dailyviews" + query_date + ".feather"
|
||||
# read the json back in and make a feather file?
|
||||
logging.debug(f"Run complete at {datetime.now()}")
|
||||
logging.info(f"Processed {successes} successful URLs and {failures} failures.")
|
||||
|
||||
|
||||
110
wikipedia/scripts/fetch_enwiki_daily_views.py
Executable file
110
wikipedia/scripts/fetch_enwiki_daily_views.py
Executable file
@ -0,0 +1,110 @@
|
||||
#!/usr/bin/env python3
|
||||
|
||||
###############################################################################
|
||||
#
|
||||
# This script assumes the presence of the COVID-19 repo.
|
||||
#
|
||||
# It (1) reads in the article list and then (2) calls the Wikimedia API to
|
||||
# fetch view information for each article. Output is to (3) JSON and TSV.
|
||||
#
|
||||
###############################################################################
|
||||
|
||||
import sys
|
||||
import requests
|
||||
import argparse
|
||||
import json
|
||||
import time
|
||||
import os.path
|
||||
import datetime
|
||||
import logging
|
||||
from csv import DictWriter
|
||||
import digobs
|
||||
#import feather #TBD
|
||||
|
||||
def parse_args():
|
||||
parser = argparse.ArgumentParser(description='Call the views API to collect Wikipedia view 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('-d', '--query_date', help='Date if not yesterday, in YYYYMMDD format.', 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)
|
||||
|
||||
def main():
|
||||
|
||||
args = parse_args()
|
||||
|
||||
outputPath = args.output_folder
|
||||
articleFile = args.article_file
|
||||
|
||||
#handle -d
|
||||
if args.query_date:
|
||||
query_date = args.query_date
|
||||
else:
|
||||
yesterday = datetime.datetime.today() - datetime.timedelta(days=1)
|
||||
query_date = yesterday.strftime("%Y%m%d")
|
||||
|
||||
#handle -W
|
||||
if args.logging_destination:
|
||||
logging.basicConfig(filename=args.logging_destination, filemode='a', level=args.logging_level)
|
||||
else:
|
||||
logging.basicConfig(level=args.logging_level)
|
||||
|
||||
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: {digobs.git_hash()}")
|
||||
|
||||
#1 Load up the list of article names
|
||||
j_outfilename = os.path.join(outputPath, f"digobs_covid19-wikipedia-enwiki_dailyviews-{query_date}.json")
|
||||
t_outfilename = os.path.join(outputPath, f"digobs_covid19-wikipedia-enwiki_dailyviews-{query_date}.tsv")
|
||||
|
||||
with open(articleFile, 'r') as infile:
|
||||
articleList = list(map(str.strip, infile))
|
||||
|
||||
success = 0 #for logging how many work/fail
|
||||
failure = 0
|
||||
|
||||
#3 Save results as a JSON and TSV
|
||||
with open(j_outfilename, 'w') as j_outfile, \
|
||||
open(t_outfilename, 'w') as t_outfile:
|
||||
|
||||
#2 Repeatedly call the API with that list of names
|
||||
for a in articleList:
|
||||
url= f"https://wikimedia.org/api/rest_v1/metrics/pageviews/per-article/en.wikipedia/all-access/all-agents/{a}/daily/{query_date}00/{query_date}00"
|
||||
|
||||
response = requests.get(url)
|
||||
if response.ok:
|
||||
jd = response.json()["items"][0]
|
||||
success = success + 1
|
||||
else:
|
||||
failure = failure + 1
|
||||
logging.warning(f"Failure: {response.status_code} from {url}")
|
||||
continue
|
||||
|
||||
# start writing the CSV File if it doesn't exist yet
|
||||
try:
|
||||
dw
|
||||
except NameError:
|
||||
dw = DictWriter(t_outfile, sorted(jd.keys()), delimiter='\t')
|
||||
dw.writeheader()
|
||||
|
||||
logging.debug(f"printing data: {jd}")
|
||||
|
||||
# write out the line of the json file
|
||||
print(json.dumps(jd), file=j_outfile)
|
||||
|
||||
# write out of the csv file
|
||||
dw.writerow(jd)
|
||||
|
||||
# f_Out = outputPath + "dailyviews" + query_date + ".feather"
|
||||
# read the json back in and make a feather file?
|
||||
logging.debug(f"Run complete at {datetime.datetime.now()}")
|
||||
logging.info(f"Processed {success} successful URLs and {failure} failures.")
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
|
||||
main()
|
||||
148
wikipedia/scripts/fetch_enwiki_revisions.py
Executable file
148
wikipedia/scripts/fetch_enwiki_revisions.py
Executable file
@ -0,0 +1,148 @@
|
||||
#!/usr/bin/env python3
|
||||
|
||||
###############################################################################
|
||||
#
|
||||
# This script assumes the presence of the COVID-19 repo.
|
||||
#
|
||||
# It (1) reads in the article list and then (2) calls the Wikimedia API to
|
||||
# fetch view information for each article. Output is to (3) JSON and TSV.
|
||||
#
|
||||
###############################################################################
|
||||
|
||||
import argparse
|
||||
import logging
|
||||
import os.path
|
||||
import json
|
||||
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=digobs.get_loglevel),
|
||||
parser.add_argument('-W', '--logging_destination', help='Logging destination file. (default: standard error)', type=str),
|
||||
args = parser.parse_args()
|
||||
return(args)
|
||||
|
||||
def main():
|
||||
args = parse_args()
|
||||
|
||||
output_path = args.output_folder
|
||||
article_filename = args.article_file
|
||||
|
||||
#handle -W
|
||||
if args.logging_destination:
|
||||
logging.basicConfig(filename=args.logging_destination, filemode='a', level=args.logging_level)
|
||||
else:
|
||||
logging.basicConfig(level=args.logging_level)
|
||||
|
||||
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: {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")
|
||||
|
||||
api_session = api.Session("https://en.wikipedia.org/w/api.php")
|
||||
|
||||
# list of properties from the API we want to gather (basically all of
|
||||
# them supported by mediawik-utilities)
|
||||
|
||||
rv_props = {'revid' : 'ids',
|
||||
'timestamp' : 'timestamp',
|
||||
'user' : 'user',
|
||||
'userid' : 'userid',
|
||||
'size' : 'size',
|
||||
'sha1' : 'sha1',
|
||||
'contentmodel' : 'contentmodel',
|
||||
'tags' : 'tags',
|
||||
'flags' : 'flags',
|
||||
'comment' : 'comment',
|
||||
'content' : 'content' }
|
||||
|
||||
exclude_from_tsv = ['tags', 'comment', 'content', 'flags']
|
||||
|
||||
# load the list of articles
|
||||
with open(article_filename, 'r') as infile:
|
||||
article_list= list(map(str.strip, infile))
|
||||
|
||||
def get_revisions_for_page(title):
|
||||
return api_session.revisions.query(properties=rv_props.values(),
|
||||
titles={title},
|
||||
direction="newer")
|
||||
|
||||
tsv_fields = ['title', 'pageid', 'namespace']
|
||||
tsv_fields = tsv_fields + list(rv_props.keys())
|
||||
|
||||
# drop fields that we identified for exclusion
|
||||
tsv_fields = [e for e in tsv_fields if e not in exclude_from_tsv]
|
||||
|
||||
# add special export fields
|
||||
tsv_fields = tsv_fields + ['anon', 'minor', 'url', 'export_timestamp', 'export_commit']
|
||||
|
||||
export_info = { 'git_commit' : digobs.git_hash(),
|
||||
'timestamp' : export_time }
|
||||
|
||||
with open(json_output_filename, 'w') as json_output, \
|
||||
open(tsv_output_filename, 'w') as tsv_output:
|
||||
|
||||
tsv_writer = DictWriter(tsv_output, fieldnames=tsv_fields, delimiter="\t")
|
||||
tsv_writer.writeheader()
|
||||
|
||||
for article in article_list:
|
||||
logging.info(f"pulling revisions for: {article}")
|
||||
for rev in get_revisions_for_page(article):
|
||||
logging.debug(f"processing raw revision: {rev}")
|
||||
|
||||
# add export metadata
|
||||
rev['exported'] = export_info
|
||||
|
||||
# save the json version of the code
|
||||
print(json.dumps(rev), file=json_output)
|
||||
|
||||
# handle missing data
|
||||
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']
|
||||
rev['namespace'] = rev['page']['ns']
|
||||
|
||||
# construct a URL
|
||||
rev['url'] = Request('GET', 'https://en.wikipedia.org/w/index.php',
|
||||
params={'title' : rev['title'].replace(" ", "_"),
|
||||
'oldid' : rev['revid']}).prepare().url
|
||||
|
||||
rev['export_timestamp'] = export_time
|
||||
rev['export_commit'] = digobs.git_hash(short=True)
|
||||
|
||||
tsv_writer.writerow({k: rev[k] for k in tsv_fields})
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
@ -1,120 +0,0 @@
|
||||
#!/usr/bin/env python3
|
||||
|
||||
###############################################################################
|
||||
#
|
||||
# This script assumes the presence of the COVID-19 repo.
|
||||
#
|
||||
# It (1) reads in the article list and then (2) calls the Wikimedia API to
|
||||
# fetch view information for each article. Output is to (3) JSON and TSV.
|
||||
#
|
||||
###############################################################################
|
||||
|
||||
import argparse
|
||||
import logging
|
||||
from os import path, mkdir
|
||||
import json
|
||||
import datetime
|
||||
import sqlite3
|
||||
from functools import partial
|
||||
from itertools import chain
|
||||
from csv import DictWriter
|
||||
import digobs
|
||||
|
||||
def main():
|
||||
|
||||
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', '--input_file', help="Input a file of page names from the English Wikiproject.", type=argparse.FileType('r'), default='./wikipedia/resources/enwp_wikiproject_covid19_articles.txt')
|
||||
parser.add_argument('-d', '--input_db', help="Input a path to a sqlite3 database from the real-time-covid-tracker project", type = sqlite3.connect, default='real-time-wiki-covid-tracker/AllWikidataItems.sqlite')
|
||||
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=argparse.FileType('a'))
|
||||
|
||||
args = parser.parse_args()
|
||||
|
||||
logging = digobs.init_logging(args)
|
||||
|
||||
conn = args.input_db
|
||||
conn.row_factory = sqlite3.Row
|
||||
|
||||
projects = (row['project'] for row in conn.execute("SELECT DISTINCT project from pagesPerProjectTable;").fetchall())
|
||||
|
||||
tsv_fields = ['title', 'pageid', 'namespace']
|
||||
|
||||
# list of properties from the API we want to gather (basically all of
|
||||
# them supported by mediawik-utilities)
|
||||
|
||||
rv_props = {'revid' : 'ids',
|
||||
'timestamp' : 'timestamp',
|
||||
'user' : 'user',
|
||||
'userid' : 'userid',
|
||||
'size' : 'size',
|
||||
'sha1' : 'sha1',
|
||||
'contentmodel' : 'contentmodel',
|
||||
'tags' : 'tags',
|
||||
'flags' : 'flags',
|
||||
'comment' : 'comment',
|
||||
'content' : 'content' }
|
||||
|
||||
def get_project_pages(project):
|
||||
return (row['page'] for row in conn.execute(f"SELECT DISTINCT page FROM pagesPerProjectTable WHERE project == '{project}';").fetchall())
|
||||
|
||||
def get_project_revisions(project):
|
||||
pages = get_project_pages(project)
|
||||
if project=="en.wikipedia":
|
||||
pages = chain(pages, map(str.strip,args.input_file))
|
||||
return digobs.get_pages_revisions(pages, project=project, logging=logging, rv_props=rv_props)
|
||||
|
||||
tsv_fields = tsv_fields + list(rv_props.keys())
|
||||
|
||||
exclude_from_tsv = ['tags', 'comment', 'content', 'flags']
|
||||
|
||||
# drop fields that we identified for exclusion
|
||||
tsv_fields = [e for e in tsv_fields if e not in exclude_from_tsv]
|
||||
|
||||
# add special export fields
|
||||
tsv_fields = tsv_fields + ['anon', 'minor', 'url', 'export_timestamp', 'export_commit']
|
||||
|
||||
export_time = str(datetime.datetime.now())
|
||||
|
||||
rev_batch_to_tsv = partial(digobs.rev_batch_to_tsv,
|
||||
tsv_fields = tsv_fields,
|
||||
export_info={'export_timestamp':export_time,
|
||||
'export_commit':digobs.git_hash(short=True)})
|
||||
|
||||
export_info = { 'git_commit' : digobs.git_hash(),
|
||||
'timestamp' : export_time }
|
||||
|
||||
export_date = datetime.datetime.today().strftime("%Y%m%d")
|
||||
|
||||
rev_batch_to_json = partial(digobs.rev_batch_to_json,
|
||||
export_info = export_info)
|
||||
|
||||
def write_project_pages(project):
|
||||
project_folder = path.join(args.output_folder, project)
|
||||
if not path.exists(project_folder):
|
||||
mkdir(project_folder)
|
||||
|
||||
dump_folder = path.join(project_folder, export_date)
|
||||
if not path.exists(dump_folder):
|
||||
mkdir(dump_folder)
|
||||
|
||||
project_revs = get_project_revisions(project)
|
||||
|
||||
json_output_filename = path.join(dump_folder, f"digobs_covid19_{project}_revisions-{export_date}.json")
|
||||
tsv_output_filename = path.join(dump_folder, f"digobs_covid19_{project}_revisions-{export_date}.tsv")
|
||||
|
||||
with open(json_output_filename, 'w') as json_output, \
|
||||
open(tsv_output_filename, 'w') as tsv_output:
|
||||
tsv_writer = DictWriter(tsv_output, fieldnames=tsv_fields, delimiter="\t")
|
||||
tsv_writer.writeheader()
|
||||
|
||||
for rev_batch in project_revs:
|
||||
logging.debug(f"processing raw revision: {rev_batch}")
|
||||
rev_batch_to_json(rev_batch, json_output=json_output)
|
||||
rev_batch_to_tsv(rev_batch, project=project, tsv_writer=tsv_writer)
|
||||
|
||||
for project in projects:
|
||||
write_project_pages(project)
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
Loading…
Reference in New Issue
Block a user