Internet radio browser GUI for music/video streams from various directory services.

โŒˆโŒ‹ โŽ‡ branch:  streamtuner2


radionet.py at [059815af7b]

File contrib/radionet.py artifact bb2c46f76a part of check-in 059815af7b


# encoding: UTF-8
# api: streamtuner2
# title: radio.net
# description: Europe's biggest radio platform
# url: http://radio.net/
# version: 0.8
# type: channel
# category: radio
# png:
#   iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAAt0lEQVR42mNgYGD4r+Ar/F/BDwkD+SBxojBMs1mLPBArgGlFqEEENYMNQNLsukIDYkirAvGu
#   ABsA1OC6XOP/5f8nwIaYAg0k2gBFsAsgTgcZkvnfDugFEeK9AFKsCPMG0CU6eZJgQ4R1eP8H7LLEivWyFJANQcQCLPBAmkGG4MJohmA6C6QA5gI5OxEUDNII
#   MwSvASBFIA3ociCxkWQAKMDICkSQIpgh2LDnSmP80YhsCFEJiRIMADpmeUOpqgjRAAAAAElFTkSuQmCC
# priority: optional
# extraction-method: regex
#
# Radio.net lists around 20.000 worldwide radio stations.
# A maximum of three pages from each genre are fetched here,
# some of the empty categories already omitted.
#
# The website heavily depends on JavaScript, a Flash player,
# some social tracking cookies. Now requires additional page
# request to get real streaming url, but at least no more
# expiring access key.


import time
import json
import re
from config import *
from channels import *
import ahttp
import action


# hook special JSON format in to avoid grepping images by generic handler
action.playlist_fmt_prio.insert(5, "rnjs")
action.playlist_content_map.insert(7, ("rnjs", r'"logo175x175rounded"'))
action.extract_playlist.extr_urls["rnjs"] = dict(
    url   = r" (?x) \"streamUrl\" \s*:\s* \"(\w+:\\?/\\?/[^\"]+)\" ",
    title = r" (?x) \"(?:description|seoTitle)\" \s*:\s* \"([^\"]+)\" ",
    unesc = "json",
)


# Radio.net
#
# ยท Uses HTML block-wise regex extraction.
#   โ†’ <a href="stationname.radio.net"> <imgโ€ฆ> <strong>โ€ฆ</strong> <small>โ€ฆ</small>
#
# ยท There's an API key in each page listing, contained in a script block
#   as `apiKey: 'โ€ฆ'?`
#
# ยท Which is needed for generating the station info JSON urls:
#   โ†’ https://api.radio.net/info/v2/search/station?apikey=โ€ฆ&pageindex=1&station=STNAME
#
# ยท To extract these JSON info targets, a custom extraction recipie is injected
#   into the action module.
#   โ†’ "streamUrl": and "description": are scanned for.
#
class radionet (ChannelPlugin):

    # control flags
    has_search = False
    audioformat = "audio/mpeg"
    listformat = "href"
    titles = dict(listeners=False, playing="Description")

    # sources
    apiPrefix = "https://api.radio.net/info/v2"
    genre_url = "http://www.radio.net/genre/{}"
    apiKey = None
    
    
    # Retrieve cat list and map
    def update_categories(self):
        html = ahttp.get("http://www.radio.net/")
        self.set_key(html)
        ls = re.findall("""<li><a class="language-info".*?>([\w\s']+)</a>""", html)
        self.categories = [i for i in ls][0:-18]


    # Fetch entries
    def update_streams(self, cat, search=None):

        # category page, get key
        html = ahttp.get(self.genre_url.format(cat))
        for p in range(2, 4):
            if html.find('"?p={}">'.format(p)) >= 0:
                html += ahttp.get(self.genre_url.format(cat) + "?p={}".format(p))
        self.set_key(html)
        r = []


        # split station blocks
        for row in re.findall("""<div class="stationinfo(.+?)</div>""", html, re.S):
            log.FOR_rOW(1)
        
            # extract text fields
            d = re.findall("""
              <a\s+href="(?:https?:)?(//(?:[\w-]+)\.radio\.net/s/([^"]+)/?)" .*?
              <img[^>]+ src="([^<">]+)" .*?
              <strong[^>]*>(.*?)</strong> .*?
              <small[^>]*>\s*(.*?)\s*</small>
            """, row, re.X|re.S)
#            log.DATA_ROW(d)
            
            # refurbish extracted strings
            if d and len(d) and len(d[0]) == 5:
                href, name, img, title, desc = d[0]
                r.append(dict(
                    name = name,
                    genre = cat,
                    title = unhtml(title),
                    playing = unhtml(desc),
                    url = "urn:radionet:"+name, #self._url(name),
                    homepage = "http:{}".format(href),
                    img = img,
                ));
        return r

    # api search is gone, now requires to fetch streamUrl from per-radio homepage
    def resolve_urn(self, row):
        if row.get("url", "-").find("urn:radionet:") != 0:
            return
        html = ahttp.get(row["homepage"])
        stream = re.findall('"stream[s:[{"\s]+url"[\s:]+"([^"]+)"', html, re.S|re.I)
        if stream:
            row["url"] = stream[0]
            return row
    

    # extract JavaScript key from any HTML blob (needed for station query)
    def set_key(self, html):
        ls = re.findall("""apiKey: '(\w+)'""", html)
        if ls:
            self.apiKey = ls[0]