Index: contrib/radiolist.py ================================================================== --- contrib/radiolist.py +++ contrib/radiolist.py @@ -1,24 +1,25 @@ # encoding: UTF-8 # api: streamtuner2 # title: radiolist.net # description: Station list by continent+country # url: http://radiolist.net/ -# version: 0.4 +# version: 0.5 # type: channel # category: radio -# priority: extra +# priority: obsolete # png: # iVBORw0KGgoAAAANSUhEUgAAABgAAAAYBAMAAAASWSDLAAAAFVBMVEVKb61qibyDnMegs9S6yeDV4O37/vyx66abAAAAAWJLR0QAiAUdSAAAAAlwSFlzAAALEwAACxMB # AJqcGAAAAAd0SU1FB+ECDBAgLJqgZW4AAADoSURBVBjTNdBNj4MgEAbgqdLeZdo9C5NwFmo5Y7Wedauc1y/+/09YdLskkDwJmZl3IOxnON4A8frQhdc/7mG2cv3gx29X # rdUfZuVHQ3JHEzZ7GSuNXxFV/FYYwryO6MOiZqEdnQPUC/fsXZaMuxa6MFfOVYN7kIWpHZClyJGLFjbbC617KaRUEJ4r4fU7IqNYrW5f2kgU5gZInG6MZ086eejcyIvO # 1KwoLayoJjqnuWO5giW8msxVmBQXD5PttSlRm8TG2fDNZS3rRO/opeSCMnPa82xSmNgkfRxJ5yZxlPrPDmLu+7GqX4lERq4G0UEyAAAAAElFTkSuQmCC # extraction-method: regex # # Radio station list grouped by continents and countries. -# Some categories return no results, because web players are -# filtered out. +# +# NO LONGER LISTS STREAMING URLS (~ 2019), thus isn't interesting +# enough anymore for extended support. import re import action import ahttp @@ -26,86 +27,72 @@ from channels import * # radiolist.net # -# · Groups stations by continents and countries. Where Europe seems to be the -# main category (empty "" path), while U.S. is labeled "/world", and Canada -# and Asia etc. again a subpath "/world/canada" even. The .catmap{} assigns -# paths to titles. -# -# · Playlist formats vary wildly. Therefore this module comes with a guessing -# method (super crude) of its own. -# -# · The audio-format-from-URL guessing should be generalized out of here perhaps. -# -# · Each station is in a … block. Invidual regexps are used for field -# extraction afterwards (instead of a block match). -# -# · Entries may contain more than one streaming url. Each accompanied by a -# bitrate. → Therefore the .best_url() sorting method. -# -# · Later versions might of course use multi-urls again… -# -class radiolist (ChannelPlugin, action.heuristic_funcs): +# · Groups stations by continents and countries. +# +# · Only fetches titles/homepages henceforth. +# +class radiolist (ChannelPlugin): # module attributes - listformat = "pls" + listformat = "href" has_search = False - categories = ["Europe", "America", "Canada", "Oceania", "Asia"] - catmap = {"Europe":"", "America":"world", "Canada":"world/canada", "Oceania":"world/oceania", "Asia":"world/asia"} + categories = ["Europe", "America", "Canada", "Australia"] + catmap = {"Albania": "albania", "America": "us", "Andorra": "andorra", "Australia": "au", "Austria": "austria", "Belarus": "belarus", "Belgium": "belgium", "Bulgaria": "bulgaria", "Canada": "can", "Croatia": "croatia", "Denmark": "denmark", "Estonia": "estonia", "Europe": "", "Finland": "finland", "France": "france", "Germany": "germany", "Greece": "greece", "Hungary": "hungary", "Iceland": "iceland", "Ireland": "ireland", "Italy": "italy", "Latvia": "latvia", "Liechtenstein": "liechtenstein", "Lithuania": "lithuania", "Luxembourg": "luxembourg", "Macedonia": "macedonia", "Malta": "malta", "Moldova": "moldova", "Monaco": "monaco", "Montenegro": "montenegro", "Netherlands": "netherlands", "New Zealand": "nz", "Norway": "norway", "Poland": "poland", "Portugal": "portugal", "Romania": "romania", "Russia": "russia", "Serbia": "serbia", "Slovakia": "slovakia", "Slovenia": "slovenia", "South America": "sa", "Spain": "spain", "Sweden": "sweden", "Switzerland": "switzerland", "Ukraine": "ukraine"} titles = dict( genre="Genre", title="Station", playing="Location", bitrate="Bitrate", listeners=False ) # just a static list for now def update_categories(self): - self.catmap = {"Europe":"", "America":"world", "Canada":"world/canada", "Oceania":"world/oceania", "Asia":"world/asia"} + self.catmap = {"Europe":"", "America":"us", "Canada":"ca", "Australia":"au", "New Zealand":"nz", "South America":"sa"} c = []# rx_links = re.compile(r""" - ([\w\s-]+)[^<]* - """, re.X) - for title in ["Europe", "America", "Canada", "Oceania", "Asia"]: +
  • \s+ + \s+ ]*>\s*([\w\s-]+?)\s*< + """, re.X|re.S) + for title in self.catmap.keys(): c.append(title) html = ahttp.get("http://www.radiolist.net/" + self.catmap[title]) sub = [] for p,t in re.findall(rx_links, html): + log.I(p,t) if t in ["Terms", "About Us", "Donation", "United States"]: continue sub.append(t) self.catmap[t] = p c.append(sorted(sub)) self.categories = c + + # extraction rules + recipe = { + "block": """(.+?)
    • """, + "split": None, + "fields": { + "title": 'data-item-title="(.+?)"', + "url": 'data-item-link="(http.+?)"', + "homepage": 'data-item-link="(.+?)"', + "favicon": ']+src="(.+?)"', + "description": '(.+?)

    ' + } + } # extract stream urls def update_streams(self, cat): - rx_title = re.compile(']+)"[^>]+target="_blank"[^>]*>(.+?)', re.I) - rx_urls = re.compile(']+)">(\d+)(?: Kbps)*', re.I) - rx_genre = re.compile(']+>(\w*[^<>]*)\s*]+>(\w+[^<>]+)\s*$', re.I) entries = [] html = ahttp.get("http://radiolist.net/" + self.catmap[cat]) - for block in re.findall("(.+?)", html, re.S): - ut = re.findall(rx_title, block) # homepage+title - uu = re.findall(rx_urls, block) # urls+bitrates - lg = re.findall(rx_genre, block) # location+genre - #print ut, uu, lg - if ut and uu and lg: - url, br = self.best_url(uu) - entries.append(dict( - homepage = ut[0][0], - title = unhtml(ut[0][1]), - url = url, - bitrate = br, - format = self.mime_guess(url, "audio/mpeg"), - listformat = self.list_guess(url), - playing = lg[0][0], - genre = lg[0][1] - )) + for block in re.findall(self.recipe["block"], html, re.S): + log.HTML(block) + e = {"genre":"-", "playing":cat, "format":"text/html"} + for id,rx in self.recipe["fields"].iteritems(): + uu = re.findall(rx, block) + log.RX(id,rx,uu) + if uu: + e[id] = unhtml(uu[0]) + if "url" in e and "title" in e: + entries.append(e) # done [log.DATA(e) for e in entries] return entries - # pick highest rated URL from [(url,bitrate),…] tuples - def best_url(self, urls): - r = dict([(u, to_int(b)) for u,b in urls]) # {url: bitrate, …} - best = sorted(r, key=r.get, reverse=True) - return best[0], r[best[0]] -