many bug fixes

[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..7ce9c5a
--- a/bin/fetch_daily_views.py
+++ b/bin/fetch_daily_views.py
@@ -5,7 +5,8 @@
 # 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 
 # 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 a (3) JSON, TSV, and 
+# Feather file.

 #
 #
 ###############################################################################
 #
 #
 ###############################################################################


@@ -30,7 +31,7 @@ import datetime
 def parse_args():
 
     parser = argparse.ArgumentParser(description='Call the views API repeatedly.')
 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()
     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()


@@ -44,42 +45,62 @@ def main():
 
     outputPath = args.output_folder
     articleFile = args.article_file
 
     outputPath = args.output_folder
     articleFile = args.article_file

-    if (query_date):
+
+    if (args.query_date):

         queryDate = args.query_date
     else:
         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:
     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:
         for currentLine in theInfile:

-            articleList.append(currentLine["Article"])
+            articleList.append(currentLine)

 
 

-    with open(outputPath, 'w') as outfile:
+    j_Out = outputPath + "dailyviews" + queryDate + ".json"
+    with open(j_Out, 'w') as outfile:

         outfile.write("[")
 
     i = 0 #iterator to deal with end of file
 
     for a in articleList:
         outfile.write("[")
 
     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)
         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")
+        if response.ok:
+
+            #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
                 else: #at end of file

-                    outfile.write("\n")
+                    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)

 
 

-            time.sleep(1)
+            time.sleep(.1)

 
 

+    with open(j_Out, 'a') as j_outfile:
+        j_outfile.write("]")

 
 

-    with open(outputPath, 'a') as outfile:
-        outfile.write("]")
+    #read the json back in and make a feather file?