use csv file instead of reading from csv page

update_films_by_year_page.py

@@ -6,9 +6,6 @@ from vcinema_utils import VCinemaUtils
 from wiki_utils import WikiUtils
 
 
-JACKNET_WIKI_URL = "https://wiki.jacknet.io"
-
-
 def build_table(films_by_year):
     films_by_year_descending = OrderedDict(sorted(films_by_year.items(), key=lambda t: t[0], reverse=True))
 

vcinema_utils/VCinemaUtils.py

@@ -1,22 +1,31 @@
-from bs4 import BeautifulSoup
 from progress.bar import Bar
 
 from imdb_utils import IMDbUtils
 from wiki_utils import WikiUtils
 
 
-def get_vcinema_viewings(token_id, token_secret):
-    # Page ID of /Vcinema/CSV
+JACKNET_WIKI_URL = "https://wiki.jacknet.io"
+
+
+def get_viewings_csv_attachment_id(token_id, token_secret):
+    attachments = WikiUtils.get_attachments(JACKNET_WIKI_URL, token_id, token_secret)
+
+    # Page ID of "https://wiki.jacknet.io/books/vcinema/page/csv")
     page_id = 11
+    viewings_csv_file_name = "vcinema.csv"
 
-    wiki_base_url = "https://wiki.jacknet.io"
+    return next((x['id'] for x in attachments if x['uploaded_to'] == page_id and x['name'] == viewings_csv_file_name), None)
 
-    html_page = WikiUtils.get_page_export_html(page_id, wiki_base_url, token_id, token_secret)
 
-    soup = BeautifulSoup(html_page, 'html.parser')
-    elements = soup.find("code").text.strip().split("\n")
-    headers = elements.pop(0).split(",")
-    viewings = [dict(zip(headers, row.split(","))) for row in elements]
+def get_vcinema_viewings(token_id, token_secret):
+    attachment_id = get_viewings_csv_attachment_id(token_id, token_secret)
+
+    viewings_csv = WikiUtils.get_attachment_contents(attachment_id, JACKNET_WIKI_URL, token_id, token_secret)
+    viewings_csv = viewings_csv.decode("utf-8")
+    viewings_csv_rows = viewings_csv.strip().split("\n")
+
+    headers = viewings_csv_rows.pop(0).split(",")
+    viewings = [dict(zip(headers, row.split(","))) for row in viewings_csv_rows]
 
     return viewings