update compile script

[covid19.git] / wikipedia_views / scripts / fetch_daily_views.py

diff --git a/wikipedia_views/scripts/fetch_daily_views.py b/wikipedia_views/scripts/fetch_daily_views.py
index 5ce989f4f0e2e9e7cb96439d6d7708ff89d1556c..4f496fee7351963bc3d1ba5caad45da9de670c17 100755 (executable)
--- a/wikipedia_views/scripts/fetch_daily_views.py
+++ b/wikipedia_views/scripts/fetch_daily_views.py
@@ -17,8 +17,8 @@ import csv
 import time
 import os.path
 import datetime
-#import feather
-
+import logging
+#import feather #TBD
 
 
 def parse_args():
@@ -27,6 +27,8 @@ def parse_args():
     parser.add_argument('-o', '--output_folder', help='Where to save output', default="../data/", 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)
+    parser.add_argument('-L', '--logging_level', help='Logging level. Options are debug, info, warning, error, critical. Default: info.', default='info'), 
+    parser.add_argument('-W', '--logging_destination', help='Logging destination.', default='../logs/'), 
     args = parser.parse_args()
 
     return(args)
@@ -39,6 +41,7 @@ def main():
     outputPath = args.output_folder
     articleFile = args.article_file
 
+    #handle -d
     if (args.query_date):
         queryDate = args.query_date
     else:
@@ -47,39 +50,63 @@ def main():
 
     queryDate = queryDate + "00" #requires specifying hours
 
+    #handle -W
+    logHome = f"{args.logging_destination}dailylogrun{datetime.datetime.today().strftime('%Y%m%d')}"
+
+    #handle -L
+    loglevel = args.logging_level
+    if loglevel == 'debug':
+        logging.basicConfig(filename=logHome, filemode='a', level=logging.DEBUG)
+    elif loglevel == 'info':
+        logging.basicConfig(filename=logHome, filemode='a', level=logging.INFO)
+    elif loglevel == 'warning':
+        logging.basicConfig(filename=logHome, filemode='a', level=logging.WARNING)
+    elif loglevel == 'error':
+        logging.basicConfig(filename=logHome, filemode='a', level=logging.ERROR)
+    elif loglevel == 'critical':
+        logging.basicConfig(filename=logHome, filemode='a', level=logging.CRITICAL)
+    else: 
+        print("Choose a valid log level: debug, info, warning, error, or critical") 
+        exit
+
 
     articleList = []
-#1 Load up the list of article names
+    logging.debug(f"Starting run at {datetime.datetime.now()}")
 
-    with open(articleFile, 'r') as infileHandle:
-        theInfile = csv.reader(infileHandle)
-        next(theInfile) #skip header
-        for currentLine in theInfile:
-            articleList.append(currentLine)
+    #1 Load up the list of article names
 
-    j_Out = outputPath + "dailyviews" + queryDate + ".json"
-    t_Out = outputPath + "dailyviews" + queryDate + ".tsv"
+    j_Out = f"{outputPath}dailyviews{queryDate}.json"
+    t_Out = f"{outputPath}dailyviews{queryDate}.tsv"
 
-    j = []
+    with open(articleFile, 'r') as infile:
+        next(infile) #skip header
+        articleList = list(infile)
 
-    i = 0 #iterator to deal with end of file
+    j = []
+    success = 0 #for logging how many work/fail
+    failure = 0 
 
-#2 Repeatedly call the API with that list of names
+    #2 Repeatedly call the API with that list of names
 
     for a in articleList:
-        a = a[0] #destringify
-        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
+        a = a.strip("\"\n") #destringify
+        url= f"https://wikimedia.org/api/rest_v1/metrics/pageviews/per-article/en.wikipedia/all-access/all-agents/{a}/daily/{queryDate}/{queryDate}"
+
         response = requests.get(url)
         if response.ok:
             jd = json.loads(response.content)
             j.append(jd["items"][0])
             time.sleep(.1)
+            success = success + 1
+        else:
+            failure = failure + 1
+            logging.warning(f"Failure: {response.status_code} from {url}")
 
-#3 Save results as a JSON and TSV
+    #3 Save results as a JSON and TSV
 
     #all data in j now, make json file
+    logging.info(f"Processed {success} successful URLs and {failure} failures.")
+
     with open(j_Out, 'w') as j_outfile: 
         json.dump(j, j_outfile, indent=2)
 
@@ -88,9 +115,10 @@ def main():
         dw.writeheader()
         dw.writerows(j)
 
+    logging.debug(f"Run complete at {datetime.datetime.now()}")
 
-    f_Out = outputPath + "dailyviews" + queryDate + ".feather"
-    #read the json back in and make a feather file? 
+    # f_Out = outputPath + "dailyviews" + queryDate + ".feather"
+    # read the json back in and make a feather file? 
 
 
 if __name__ == "__main__":