makes TSV

[covid19.git] / bin / fetch_daily_views.py

diff --git a/bin/fetch_daily_views.py b/bin/fetch_daily_views.py
old mode 100644 (file)
new mode 100755 (executable)
index 3e8cb5a..ac925ed
--- a/bin/fetch_daily_views.py
+++ b/bin/fetch_daily_views.py
@@ -5,8 +5,7 @@
 # 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.
-#
+# fetch view information for each article. Output is to (3) JSON and TSV.
 #
 ###############################################################################
 
@@ -24,13 +23,14 @@ import csv
 import time
 import os.path
 import datetime
+#import feather
 
 
 
 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('-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)
     args = parser.parse_args()
@@ -44,43 +44,55 @@ def main():
 
     outputPath = args.output_folder
     articleFile = args.article_file
-    if (query_date):
+
+    if (args.query_date):
         queryDate = args.query_date
     else:
-        queryDate = datetime.datetime.today().strftime("%Y%m%d")
+        yesterday = datetime.datetime.today() - datetime.timedelta(days=1)
+        queryDate = yesterday.strftime("%Y%m%d")
 
+    queryDate = queryDate + "00" #requires specifying hours
 
+
+    articleList = []
     with open(articleFile, 'r') as infileHandle:
-        theInfile = csv.reader(infileHandle, quotechar='"')
+        #theInfile = csv.reader(infileHandle, quotechar='"')
+        theInfile = csv.reader(infileHandle)
+        next(theInfile) #skip header
         for currentLine in theInfile:
-            articleList.append(currentLine["Article"])
+            articleList.append(currentLine)
+
+    j_Out = outputPath + "dailyviews" + queryDate + ".json"
+    t_Out = outputPath + "dailyviews" + queryDate + ".tsv"
 
-    with open(outputPath, 'w') as outfile:
-        outfile.write("[")
+    j = []
 
     i = 0 #iterator to deal with end of file
 
     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 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")
+        if response.ok:
+            jd = json.loads(response.content)
+            j.append(jd["items"][0])
+            time.sleep(.1)
+
 
-            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(t_Out, 'w') as t_outfile:
+        dw = csv.DictWriter(t_outfile, sorted(j[0].keys()), delimiter='\t')
+        dw.writeheader()
+        dw.writerows(j)
 
-    with open(outputPath, 'a') as outfile:
-        outfile.write("]")
 
+    f_Out = outputPath + "dailyviews" + queryDate + ".feather"
+    #read the json back in and make a feather file? 
 
 
 if __name__ == "__main__":