X-Git-Url: https://code.delx.au/webdl/blobdiff_plain/8c49f0ff2a004225bbb02dd9f8772440dd3d7db0..992e749d40b94d33d1703f90afb28d8c8b548dd5:/sbs.py

diff --git a/sbs.py b/sbs.py
index dc353ca..c3cbe52 100644
--- a/sbs.py
+++ b/sbs.py
@@ -2,10 +2,12 @@ import requests_cache
 from common import grab_html, grab_json, grab_xml, download_hls, download_mpd, Node, append_to_qs
 
 import json
+import logging
+import os
 import sys
 
 BASE = "https://www.sbs.com.au"
-FULL_VIDEO_LIST = BASE + "/api/video_search/v2/?m=1&filters={section}{Programs}"
+FULL_VIDEO_LIST = BASE + "/api/video_feed/f/Bgtm9B/sbs-section-programs/"
 VIDEO_URL = BASE + "/ondemand/video/single/%s"
 
 NS = {
@@ -23,8 +25,13 @@ class SbsVideoNode(Node):
         with requests_cache.disabled():
             doc = grab_html(VIDEO_URL % self.video_id)
         player_params = self.get_player_params(doc)
-        release_url = player_params["releaseUrls"]["html"]
 
+        error = player_params.get("error", None)
+        if error:
+            print("Cannot download:", error)
+            return False
+
+        release_url = player_params["releaseUrls"]["html"]
         filename = self.title + ".ts"
 
         hls_url = self.get_hls_url(release_url)
@@ -76,22 +83,55 @@ class SbsRootNode(SbsNavNode):
             nav_node.create_video_node(entry_data)
 
     def load_all_video_entries(self):
+        channels = [
+            "Channel/SBS1",
+            "Channel/SBS Food",
+            "Channel/SBS VICELAND",
+            "Channel/SBS World Movies",
+            "Channel/Web Exclusive",
+        ]
+
+        all_entries = {}
+        for channel in channels:
+            self.load_all_video_entries_for_channel(all_entries, channel)
+
+        all_entries = list(all_entries.values())
+        print(" SBS fetched", len(all_entries))
+        return all_entries
+
+    def load_all_video_entries_for_channel(self, all_entries, channel):
         offset = 1
-        amount = 49
+        page_size = 500
+        duplicate_warning = False
+
         while True:
-            url = append_to_qs(FULL_VIDEO_LIST, {"range": "%s-%s" % (offset, offset+amount)})
-            data = grab_json(url)
-            if "entries" not in data:
-                raise Exception("Missing data in SBS response", data)
-            entries = data["entries"]
+            entries = self.fetch_entries_page(channel, offset, page_size)
             if len(entries) == 0:
                 break
+
             for entry in entries:
-                yield entry
-            offset += amount
-            sys.stdout.write(".")
-            sys.stdout.flush()
-        print()
+                guid = entry["guid"]
+                if guid in entries and not duplicate_warning:
+                    # https://bitbucket.org/delx/webdl/issues/102/recent-sbs-series-missing
+                    logging.warn("SBS returned a duplicate response, data is probably missing. Try decreasing page_size.")
+                    duplicate_warning = True
+
+                all_entries[guid] = entry
+
+            offset += page_size
+            if os.isatty(sys.stdout.fileno()):
+                sys.stdout.write(".")
+                sys.stdout.flush()
+
+    def fetch_entries_page(self, channel, offset, page_size):
+        url = append_to_qs(FULL_VIDEO_LIST, {
+            "range": "%s-%s" % (offset, offset+page_size-1),
+            "byCategories": channel,
+        })
+        data = grab_json(url)
+        if "entries" not in data:
+            raise Exception("Missing data in SBS response", data)
+        return data["entries"]
 
     def explode_videos_to_unique_categories(self, all_video_entries):
         for entry_data in all_video_entries: