svtplay: they changed the website.

fixes: #1134

lib/svtplay_dl/service/svtplay.py

@@ -5,19 +5,19 @@ import re
 import copy
 import json
 import hashlib
+import datetime
 import logging
 from urllib.parse import urljoin, urlparse, parse_qs
 from operator import itemgetter
 
 from svtplay_dl.service import Service, MetadataThumbMixin
 from svtplay_dl.utils.text import filenamify
-from svtplay_dl.fetcher.hds import hdsparse
 from svtplay_dl.fetcher.hls import hlsparse
 from svtplay_dl.fetcher.dash import dashparse
 from svtplay_dl.subtitle import subtitle
 from svtplay_dl.error import ServiceError
 
-URL_VIDEO_API = "http://api.svt.se/videoplayer-api/video/"
+URL_VIDEO_API = "https://api.svt.se/video/"
 
 
 class Svtplay(Service, MetadataThumbMixin):
@@ -63,28 +63,24 @@ class Svtplay(Service, MetadataThumbMixin):
             return
         janson = json.loads(match.group(1))["videoPage"]
 
-        if "programTitle" not in janson["video"]:
+        self.visibleid = list(janson['visible'].keys())[0]
-            yield ServiceError("Can't find any video on that page.")
+        match = re.search(r"__svtplay_apollo'] = ({.*});", urldata)
-            return
-
-        if self.access:
-            for i in janson["video"]["versions"]:
-                if i["accessService"] == self.access:
-                    url = urljoin("http://www.svtplay.se", i["contentUrl"])
-                    res = self.http.get(url)
-                    match = re.search("__svtplay'] = ({.*});", res.text)
         if not match:
             yield ServiceError("Can't find video info.")
             return
-                    janson = json.loads(match.group(1))["videoPage"]
 
-        self.outputfilename(janson["video"])
+        janson = json.loads(match.group(1))
-        self.extrametadata(janson)
+        for key in janson["ROOT_QUERY"].keys():
+            if "listablesByEsceni" in key:
+                esceni = key
+                break
+
+        self.type_name = janson["ROOT_QUERY"][esceni][0]["typename"]
+        vid = janson["{}:{}".format(self.type_name, self.visibleid)]["videoSvtId"]
+
+        self.outputfilename(janson)
+        self.extrametadata(janson, self.type_name, self.visibleid)
 
-        if "programVersionId" in janson["video"]:
-            vid = janson["video"]["programVersionId"]
-        else:
-            vid = janson["video"]["id"]
         res = self.http.get(URL_VIDEO_API + vid)
         try:
             janson = res.json()
@@ -119,14 +115,6 @@ class Svtplay(Service, MetadataThumbMixin):
                     if alt:
                         alt_streams = hlsparse(self.config, self.http.request("get", alt.request.url), alt.request.url, output=self.output)
 
-                elif i["format"] == "hds":
-                    match = re.search(r"\/se\/secure\/", i["url"])
-                    if not match:
-                        streams = hdsparse(self.config, self.http.request("get", i["url"], params={"hdcore": "3.7.0"}),
-                                           i["url"], output=self.output)
-                        if alt:
-                            alt_streams = hdsparse(self.config, self.http.request("get", alt.request.url, params={"hdcore": "3.7.0"}),
-                                                   alt.request.url, output=self.output)
                 elif i["format"] == "dash264" or i["format"] == "dashhbbtv":
                     streams = dashparse(self.config, self.http.request("get", i["url"]), i["url"], output=self.output)
                     if alt:
@@ -155,7 +143,7 @@ class Svtplay(Service, MetadataThumbMixin):
         for i in dataj["gridPage"]["content"]:
             videos.append(i["contentUrl"])
         page += 1
-        self._last_chance(videos, page, pages)
+        videos.extend(self._last_chance(videos, page, pages))
         return videos
 
     def _genre(self, jansson):
@@ -178,34 +166,53 @@ class Svtplay(Service, MetadataThumbMixin):
 
         videos = []
         tab = None
-        match = re.search("__svtplay'] = ({.*});", self.get_urldata())
-        if re.search("sista-chansen", parse.path):
-            videos = self._last_chance(videos, 1)
-        elif not match:
-            logging.error("Couldn't retrieve episode list.")
-            return
-        else:
-            dataj = json.loads(match.group(1))
-            if re.search("/genre", parse.path):
-                videos = self._genre(dataj)
-            else:
         if parse.query:
             query = parse_qs(parse.query)
             if "tab" in query:
                 tab = query["tab"][0]
 
-                if dataj["relatedVideoContent"]:
+        if re.search("sista-chansen", parse.path):
-                    items = dataj["relatedVideoContent"]["relatedVideosAccordion"]
+            videos = self._last_chance(videos, 1)
-                    for i in items:
-                        if tab:
-                            if i["slug"] == tab:
-                                videos = self.videos_to_list(i["videos"], videos)
         else:
-                            if "klipp" not in i["slug"] and "kommande" not in i["slug"]:
+            match = re.search(r"__svtplay'] = ({.*});", self.get_urldata())
-                                videos = self.videos_to_list(i["videos"], videos)
+            if not match:
-                        if self.config.get("include_clips"):
+                logging.error("Can't find video info.")
-                            if i["slug"] == "klipp":
+                return videos
-                                videos = self.videos_to_list(i["videos"], videos)
+            janson = json.loads(match.group(1))["videoPage"]
+            self.visibleid = list(janson['visible'].keys())[0]
+            match = re.search(r"__svtplay_apollo'] = ({.*});", self.get_urldata())
+            if not match:
+                logging.error("Can't find video info.")
+                return videos
+            janson = json.loads(match.group(1))
+            episode = janson["Variant:{}".format(self.visibleid)]
+            associatedContent = episode["associatedContent({\"include\":[\"season\",\"productionPeriod\",\"clips\",\"upcoming\"]})"]
+
+            keys = []
+            videos = []
+            videos.append(janson[episode["urls"]["id"]]["svtplay"])
+            for i in associatedContent:
+                if tab:
+                    section = "Selection:{}".format(tab)
+                    if section == i["id"]:
+                        keys.append(section)
+                else:
+                    if i["id"] == "Selection:upcoming":
+                        continue
+                    elif self.config.get("include_clips") and "Selection:clips" in i["id"]:
+                        keys.append(i["id"])
+                    elif "Selection:clips" not in i["id"]:
+                        keys.append(i["id"])
+
+            for i in keys:
+                for n in janson[i]["items"]:
+                    epi = janson[janson[n["id"]]["item"]["id"]]
+                    if "variants" in epi:
+                        for z in epi["variants"]:
+                            if janson[janson[z["id"]]["urls"]["id"]]["svtplay"] not in videos:
+                                videos.append(janson[janson[z["id"]]["urls"]["id"]]["svtplay"])
+                    if janson[epi["urls"]["id"]]["svtplay"] not in videos:
+                        videos.append(janson[epi["urls"]["id"]]["svtplay"])
 
         episodes = [urljoin("http://www.svtplay.se", x) for x in videos]
 
@@ -231,16 +238,11 @@ class Svtplay(Service, MetadataThumbMixin):
     def outputfilename(self, data):
         name = None
         desc = None
-        if "programTitle" in data and data["programTitle"]:
+        pid = data["{}:{}".format(self.type_name, self.visibleid)]["parent"]["id"]
-            name = filenamify(data["programTitle"])
-        elif "titleSlug" in data and data["titleSlug"]:
-            name = filenamify(data["titleSlug"])
-        other = data["title"]
 
-        if "programVersionId" in data:
+        name = data[pid]["slug"]
-            vid = str(data["programVersionId"])
+        other = data["{}:{}".format(self.type_name, self.visibleid)]["slug"]
-        else:
+        vid = data["{}:{}".format(self.type_name, self.visibleid)]["id"]
-            vid = str(data["id"])
         id = hashlib.sha256(vid.encode("utf-8")).hexdigest()[:7]
 
         if name == other:
@@ -250,10 +252,10 @@ class Svtplay(Service, MetadataThumbMixin):
             other = None
 
         season, episode = self.seasoninfo(data)
-        if "accessService" in data:
+        if "accessibility" in data["{}:{}".format(self.type_name, self.visibleid)]:
-            if data["accessService"] == "audioDescription":
+            if data["{}:{}".format(self.type_name, self.visibleid)]["accessibility"] == "AudioDescribed":
                 desc = "syntolkat"
-            if data["accessService"] == "signInterpretation":
+            if data["{}:{}".format(self.type_name, self.visibleid)]["accessibility"] == "SignInterpreted":
                 desc = "teckentolkat"
 
         if not other:
@@ -261,7 +263,7 @@ class Svtplay(Service, MetadataThumbMixin):
         elif desc:
             other += "-{}".format(desc)
 
-        self.output["title"] = name
+        self.output["title"] = filenamify(name)
         self.output["id"] = id
         self.output["season"] = season
         self.output["episode"] = episode
@@ -269,34 +271,34 @@ class Svtplay(Service, MetadataThumbMixin):
 
     def seasoninfo(self, data):
         season, episode = None, None
-        if "season" in data and data["season"]:
+
-            season = "{:02d}".format(data["season"])
+        if "episode" not in data["{}:{}".format(self.type_name, self.visibleid)]:
-            if int(season) == 0:
-                season = None
-        if "episodeNumber" in data and data["episodeNumber"]:
-            episode = "{:02d}".format(data["episodeNumber"])
-            if int(episode) == 0:
-                episode = None
-        if episode is not None and season is None:
-            # Missing season, happens for some barnkanalen shows assume first and only
-            season = "01"
             return season, episode
 
-    def extrametadata(self, data):
+        episodeid = data["{}:{}".format(self.type_name, self.visibleid)]["episode"]["id"]
+        if "positionInSeason" not in data[episodeid]:
+            return season, episode
+
+        match = re.search(r"Säsong (\d+) — Avsnitt (\d+)", data[episodeid]["positionInSeason"])
+        if not match:
+            return season, episode
+
+        season = "{:02d}".format(match.group(1))
+        episode = "{:02d}".format(match.group(2))
+
+        return season, episode
+
+    def extrametadata(self, data, type_name, visibleid):
+        episode = data["{}:{}".format(type_name, visibleid)]
+
         self.output["tvshow"] = (self.output["season"] is not None and self.output["episode"] is not None)
-        try:
+        if "validFrom" in episode:
-            self.output["publishing_datetime"] = data["video"]["broadcastDate"] / 1000
+            self.output["publishing_datetime"] = int(datetime.datetime.strptime(episode["validFrom"], "%Y-%m-%dT%H:%M:%S%z").strftime('%s'))
-        except KeyError:
+
-            pass
+        self.output["title_nice"] = data[data["{}:{}".format(type_name, visibleid)]["parent"]["id"]]["name"]
-        try:
-            title = data["video"]["programTitle"]
-            self.output["title_nice"] = title
-        except KeyError:
-            title = data["video"]["titleSlug"]
-            self.output["title_nice"] = title
 
         try:
-            t = data['state']["titleModel"]["thumbnail"]
+            t = data[data[episode["parent"]["id"]]["image"]["id"]]
         except KeyError:
             t = ""
         if isinstance(t, dict):
@@ -307,10 +309,7 @@ class Svtplay(Service, MetadataThumbMixin):
             url = t.format(format="large")
             self.output["showthumbnailurl"] = url
         try:
-            t = data["video"]["thumbnailXL"]
+            t = data[episode["image"]["id"]]
-        except KeyError:
-            try:
-                t = data["video"]["thumbnail"]
         except KeyError:
             t = ""
         if isinstance(t, dict):
@@ -320,11 +319,9 @@ class Svtplay(Service, MetadataThumbMixin):
             # Get the image if size/format is not specified in the URL set it to large
             url = t.format(format="large")
             self.output["episodethumbnailurl"] = url
-        try:
+
-            self.output["showdescription"] = data['state']["titleModel"]["description"]
+        if "longDescription" in data[episode["parent"]["id"]]:
-        except KeyError:
+            self.output["showdescription"] = data[episode["parent"]["id"]]["longDescription"]
-            pass
+
-        try:
+        if "longDescription" in episode:
-            self.output["episodedescription"] = data["video"]["description"]
+            self.output["episodedescription"] = episode["longDescription"]
-        except KeyError:
-            pass