for testing
authorKaylea Champion <kaylea@uw.edu>
Fri, 27 Mar 2020 23:00:36 +0000 (16:00 -0700)
committerKaylea Champion <kaylea@uw.edu>
Fri, 27 Mar 2020 23:00:36 +0000 (16:00 -0700)
bin/fetch_daily_views.py [new file with mode: 0644]

diff --git a/bin/fetch_daily_views.py b/bin/fetch_daily_views.py
new file mode 100644 (file)
index 0000000..3e8cb5a
--- /dev/null
@@ -0,0 +1,88 @@
+#!/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 a (3) TSV file.
+#
+#
+###############################################################################
+
+
+#1 Load up the list of article names
+
+#2 Repeatedly call the API with that list of names
+
+#3 Save results as a TSV
+
+import requests
+import argparse
+import json
+import csv
+import time
+import os.path
+import datetime
+
+
+
+def parse_args():
+
+    parser = argparse.ArgumentParser(description='Call the views API repeatedly.')
+    parser.add_argument('-o', '--output_folder', help='Where to save output', default="", type=str)
+    parser.add_argument('-i', '--article_file', help='File listing article names', default="../resources/articles.txt", type=str)
+    parser.add_argument('-d', '--query_date', help='Date if not yesterday, in YYYYMMDD format please.', type=str)
+    args = parser.parse_args()
+
+    return(args)
+
+
+def main():
+
+    args = parse_args()
+
+    outputPath = args.output_folder
+    articleFile = args.article_file
+    if (query_date):
+        queryDate = args.query_date
+    else:
+        queryDate = datetime.datetime.today().strftime("%Y%m%d")
+
+
+    with open(articleFile, 'r') as infileHandle:
+        theInfile = csv.reader(infileHandle, quotechar='"')
+        for currentLine in theInfile:
+            articleList.append(currentLine["Article"])
+
+    with open(outputPath, 'w') as outfile:
+        outfile.write("[")
+
+    i = 0 #iterator to deal with end of file
+
+    for a in articleList:
+        i = i+1
+        url= "https://wikimedia.org/api/rest_v1/metrics/pageviews/per-article/en.wikipedia/all-access/all-agents/"
+        url= url + a + "/daily/" + queryDate + "/" + queryDate #for now, single date at a time
+
+
+        response = requests.get(url)
+        if raw_response.ok:
+            with open(outputPath, 'a') as outfile:
+                json.dump(json.loads(ident_response.content), outfile)
+                if i < len(revList):
+                    outfile.write(",\n")
+                else: #at end of file
+                    outfile.write("\n")
+
+            time.sleep(1)
+
+
+    with open(outputPath, 'a') as outfile:
+        outfile.write("]")
+
+
+
+if __name__ == "__main__":
+
+    main()

Community Data Science Collective || Want to submit a patch?