code.delx.au - webdl/blob - sbs.py

   1 import requests_cache
   2 from common import grab_html, grab_json, grab_xml, download_hls, Node, append_to_qs
   3
   4 import json
   5 import logging
   6 import os
   7 import sys
   8
   9 BASE = "https://www.sbs.com.au"
  10 FULL_VIDEO_LIST = BASE + "/api/video_feed/f/Bgtm9B/sbs-section-programs/"
  11 VIDEO_SMIL_URL = BASE + "/api/v3/video_smil?id="
  12
  13 NS = {
  14     "smil": "http://www.w3.org/2005/SMIL21/Language",
  15 }
  16
  17
  18 class SbsVideoNode(Node):
  19     def __init__(self, title, parent, url):
  20         Node.__init__(self, title, parent)
  21         self.video_id = url.split("/")[-1]
  22         self.can_download = True
  23
  24     def download(self):
  25         filename = self.title + ".ts"
  26
  27         with requests_cache.disabled():
  28             doc = grab_xml(VIDEO_SMIL_URL + self.video_id)
  29             video_el = doc.xpath("//smil:video", namespaces=NS)
  30             if not video_el:
  31                 print("Cannot find video:", error)
  32                 return False
  33             video_url = video_el[0].attrib["src"]
  34
  35         return download_hls(filename, video_url)
  36
  37 class SbsNavNode(Node):
  38     def create_video_node(self, entry_data):
  39         SbsVideoNode(entry_data["title"], self, entry_data["id"])
  40
  41     def find_existing_child(self, path):
  42         for child in self.children:
  43             if child.title == path:
  44                 return child
  45
  46 class SbsRootNode(SbsNavNode):
  47     def __init__(self, parent):
  48         Node.__init__(self, "SBS", parent)
  49
  50     def fill_children(self):
  51         all_video_entries = self.load_all_video_entries()
  52         category_and_entry_data = self.explode_videos_to_unique_categories(all_video_entries)
  53         for category_path, entry_data in category_and_entry_data:
  54             nav_node = self.create_nav_node(self, category_path)
  55             nav_node.create_video_node(entry_data)
  56
  57     def load_all_video_entries(self):
  58         channels = [
  59             "Channel/NITV",
  60             "Channel/SBS1",
  61             "Channel/SBS Food",
  62             "Channel/SBS VICELAND",
  63             "Channel/SBS World Movies",
  64             "Channel/Web Exclusive",
  65         ]
  66
  67         all_entries = {}
  68         for channel in channels:
  69             self.load_all_video_entries_for_channel(all_entries, channel)
  70
  71         all_entries = list(all_entries.values())
  72         print(" SBS fetched", len(all_entries))
  73         return all_entries
  74
  75     def load_all_video_entries_for_channel(self, all_entries, channel):
  76         offset = 1
  77         page_size = 500
  78         duplicate_warning = False
  79
  80         while True:
  81             entries = self.fetch_entries_page(channel, offset, page_size)
  82             if len(entries) == 0:
  83                 break
  84
  85             for entry in entries:
  86                 guid = entry["guid"]
  87                 if guid in entries and not duplicate_warning:
  88                     # https://bitbucket.org/delx/webdl/issues/102/recent-sbs-series-missing
  89                     logging.warn("SBS returned a duplicate response, data is probably missing. Try decreasing page_size.")
  90                     duplicate_warning = True
  91
  92                 all_entries[guid] = entry
  93
  94             offset += page_size
  95             if os.isatty(sys.stdout.fileno()):
  96                 sys.stdout.write(".")
  97                 sys.stdout.flush()
  98
  99     def fetch_entries_page(self, channel, offset, page_size):
 100         url = append_to_qs(FULL_VIDEO_LIST, {
 101             "range": "%s-%s" % (offset, offset+page_size-1),
 102             "byCategories": channel,
 103         })
 104         data = grab_json(url)
 105         if "entries" not in data:
 106             raise Exception("Missing data in SBS response", data)
 107         return data["entries"]
 108
 109     def explode_videos_to_unique_categories(self, all_video_entries):
 110         for entry_data in all_video_entries:
 111             for category_data in entry_data["media$categories"]:
 112                 category_path = self.calculate_category_path(
 113                     category_data["media$scheme"],
 114                     category_data["media$name"],
 115                 )
 116                 if category_path:
 117                     yield category_path, entry_data
 118
 119     def calculate_category_path(self, scheme, name):
 120         if not scheme:
 121             return
 122         if scheme == name:
 123             return
 124         name = name.split("/")
 125         if name[0] != scheme:
 126             name.insert(0, scheme)
 127         return name
 128
 129     def create_nav_node(self, parent, category_path):
 130         if not category_path:
 131             return parent
 132
 133         current_path = category_path[0]
 134         current_node = parent.find_existing_child(current_path)
 135         if not current_node:
 136             current_node = SbsNavNode(current_path, parent)
 137         return self.create_nav_node(current_node, category_path[1:])
 138
 139 def fill_nodes(root_node):
 140     SbsRootNode(root_node)