cleaning out commented code

[covid19.git] / bin / fetch_daily_views.py

diff --git a/bin/fetch_daily_views.py b/bin/fetch_daily_views.py
index 7ce9c5ac9597bca4dad0abc8cc28cd34d5a8b5d2..5ce989f4f0e2e9e7cb96439d6d7708ff89d1556c 100755 (executable)
--- a/bin/fetch_daily_views.py
+++ b/bin/fetch_daily_views.py
@@ -5,19 +5,11 @@
 # 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) JSON, TSV, and 
-# Feather file.
-#
+# fetch view information for each article. Output is to (3) JSON and TSV.
 #
 ###############################################################################
 
 
-#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
@@ -25,6 +17,7 @@ import csv
 import time
 import os.path
 import datetime
+#import feather
 
 
 
@@ -56,52 +49,48 @@ def main():
 
 
     articleList = []
+#1 Load up the list of article names
+
     with open(articleFile, 'r') as infileHandle:
-        #theInfile = csv.reader(infileHandle, quotechar='"')
         theInfile = csv.reader(infileHandle)
         next(theInfile) #skip header
         for currentLine in theInfile:
             articleList.append(currentLine)
 
     j_Out = outputPath + "dailyviews" + queryDate + ".json"
-    with open(j_Out, 'w') as outfile:
-        outfile.write("[")
+    t_Out = outputPath + "dailyviews" + queryDate + ".tsv"
+
+    j = []
 
     i = 0 #iterator to deal with end of file
 
+#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
-
-
         response = requests.get(url)
         if response.ok:
+            jd = json.loads(response.content)
+            j.append(jd["items"][0])
+            time.sleep(.1)
 
-            #do json entry
-            j=json.loads(response.content)
-            with open(j_Out, 'a') as j_outfile: 
-                json.dump(j, j_outfile)
-                if i < len(articleList):
-                    j_outfile.write(",\n")
-                else: #at end of file
-                    j_outfile.write("\n")
-
-            #do tsv entry
-            #with open(outputPath + "dailyviews" + queryDate + ".tsv", 'a') as t_outfile: 
-            #    dw = csv.DictWriter(t_outfile, sorted(j[0].keys()), delimiter='\t')
-            #    if i==1:
-            #        dw.writeheader()
-            #    dw.writerows(j)
+#3 Save results as a JSON and TSV
 
-            time.sleep(.1)
+    #all data in j now, make json file
+    with open(j_Out, 'w') as j_outfile: 
+        json.dump(j, j_outfile, indent=2)
 
-    with open(j_Out, 'a') as j_outfile:
-        j_outfile.write("]")
+    with open(t_Out, 'w') as t_outfile:
+        dw = csv.DictWriter(t_outfile, sorted(j[0].keys()), delimiter='\t')
+        dw.writeheader()
+        dw.writerows(j)
 
-    #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__":