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

   1 #!/usr/bin/env python
   2 # vim:ts=4:sts=4:sw=4:noet
   3
   4 from common import grab_json, grab_xml, download_rtmp, Node
   5
   6 import collections
   7
   8 BASE = "http://www.sbs.com.au"
   9 MENU_URL = "/api/video_feed/f/dYtmxB/%s?startIndex=%d"
  10 VIDEO_URL = BASE + "/api/video_feed/f/dYtmxB/CxeOeDELXKEv/%s?form=json"
  11
  12 NS = {
  13         "smil": "http://www.w3.org/2005/SMIL21/Language",
  14 }
  15
  16 SECTIONS = [
  17         "section-sbstv",
  18         "section-programs",
  19 ]
  20
  21 CATEGORY_MAP = {
  22         "Factual": "Documentary",
  23 }
  24
  25
  26 class SbsNode(Node):
  27         def __init__(self, title, parent, video_id):
  28                 Node.__init__(self, title, parent)
  29                 self.title = title
  30                 self.video_id = video_id.split("/")[-1]
  31                 self.can_download = True
  32
  33         def download(self):
  34                 doc = grab_json(VIDEO_URL % self.video_id, 0)
  35                 best_url = None
  36                 best_bitrate = 0
  37                 for d in doc["media$content"]:
  38                         bitrate = d["plfile$bitrate"]
  39                         if bitrate > best_bitrate or best_url is None:
  40                                 best_bitrate = bitrate
  41                                 best_url = d["plfile$url"]
  42
  43                 doc = grab_xml(best_url, 3600)
  44                 vbase = doc.xpath("//smil:meta/@base", namespaces=NS)[0]
  45                 vpath = doc.xpath("//smil:video/@src", namespaces=NS)[0]
  46                 ext = vpath.rsplit(".", 1)[1]
  47                 filename = self.title + "." + ext
  48
  49                 return download_rtmp(filename, vbase, vpath)
  50
  51 def fill_entry(get_catnode, entry):
  52         title = entry["title"]
  53         video_id = entry["id"]
  54         info = collections.defaultdict(list)
  55         for d in entry["media$categories"]:
  56                 if not d.has_key("media$scheme"):
  57                         continue
  58                 info[d["media$scheme"]].append(d["media$name"])
  59
  60         if "Section/Promos" in info.get("Section", []):
  61                 # ignore promos
  62                 return
  63
  64         for category in info.get("Genre", ["$UnknownCategory$"]):
  65                 category = CATEGORY_MAP.get(category, category)
  66                 parent_node = get_catnode(category)
  67                 SbsNode(title, parent_node, video_id)
  68
  69
  70 def fill_section(get_catnode, section):
  71         index = 1
  72         while True:
  73                 doc = grab_json(BASE + MENU_URL % (section, index), 3600)
  74                 if len(doc.get("entries", [])) == 0:
  75                         break
  76                 for entry in doc["entries"]:
  77                         fill_entry(get_catnode, entry)
  78                 index += doc["itemsPerPage"]
  79
  80 def fill_nodes(root_node):
  81         catnodes = {}
  82         def get_catnode(name):
  83                 try:
  84                         return catnodes[name]
  85                 except KeyError:
  86                         n = Node(name, root_node)
  87                         catnodes[name] = n
  88                         return n
  89
  90         for section in SECTIONS:
  91                 fill_section(get_catnode, section)
  92
  93