Check-in [2793a3e6f8]
Overview
Comment: | Consolidate bitrate filter in main update_streams() method. Fix conjoined category strings. |
---|---|
Downloads: | Tarball | ZIP archive | SQL archive |
Timelines: | family | ancestors | descendants | both | trunk |
Files: | files | file ages | folders |
SHA1: |
2793a3e6f8bce2b6984bbd828632d8e5 |
User & Date: | mario on 2015-05-03 09:25:02 |
Other Links: | manifest | tags |
Context
2015-05-03
| ||
09:35 | Implement "buffy" mode for just keeping Xiphs YP.XML in memory once traversed. check-in: 0afbbc5c26 user: mario tags: trunk | |
09:25 | Consolidate bitrate filter in main update_streams() method. Fix conjoined category strings. check-in: 2793a3e6f8 user: mario tags: trunk | |
09:16 | Fix server search var reference. check-in: b09ca3e24f user: mario tags: trunk | |
Changes
Modified channels/xiph.py from [e7ab9f6f02] to [73380fcbf4].
︙ | ︙ | |||
32 33 34 35 36 37 38 39 40 41 42 43 44 45 | # → "Clunky XML" fetches the olden YP.XML, which is really # slow, then slices out genres. No search. # # → "Forbidden Fruits" extracts from dir.xiph.org HTML pages, # with homepages and listener/max infos available. Search # is also possible. # from config import * from uikit import uikit import ahttp from channels import * import xml.dom.minidom | > > > | 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 | # → "Clunky XML" fetches the olden YP.XML, which is really # slow, then slices out genres. No search. # # → "Forbidden Fruits" extracts from dir.xiph.org HTML pages, # with homepages and listener/max infos available. Search # is also possible. # # The bitrate filter can strip any low-quality entries, but # retains `0` entries (which just lack meta information and # aren't necessarily low-bitrate.) from config import * from uikit import uikit import ahttp from channels import * import xml.dom.minidom |
︙ | ︙ | |||
70 71 72 73 74 75 76 77 78 | ] # entries contain no "|" search patterns anymore # Switch to JSON, XML or HTML extractor def update_streams(self, cat=None, search=None): if cat: cat = cat.lower() if conf.xiph_source in ("cache", "json"): log.PROC("Xiph mode: processing api.dir.xiph.org JSON (via api.include-once.org cache)") | > > | | | > > > > > > > | 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 | ] # entries contain no "|" search patterns anymore # Switch to JSON, XML or HTML extractor def update_streams(self, cat=None, search=None): if cat: cat = cat.lower() # retrieval module if conf.xiph_source in ("cache", "json"): log.PROC("Xiph mode: processing api.dir.xiph.org JSON (via api.include-once.org cache)") r = self.from_json_cache(cat, search) elif conf.xiph_source in ("xml", "buffy"): log.PROC("Xiph mode: xml.dom.minidom to traverse yp.xml") r = self.from_yp_xml(cat, search) else: log.PROC("Xiph mode: extract from dir.xiph.org HTML listings") r = self.from_raw_html(cat, search) # filter bitrate if conf.xiph_min_bitrate: r = [row for row in r if row.get("bitrate", 0) <= 10 or row.get("bitrate", 0) >= int(conf.xiph_min_bitrate)] return r # Retrieve partial stream list from api.include-once.org cache / JSON API wrapper # # The server interface is specifically designed for streamtuner2. It refurbishes # Xiphs JSOL dump (which is impossible to fix in Python, but easier per PHP). |
︙ | ︙ | |||
102 103 104 105 106 107 108 | #log.DATA(data) #-- extract l = [] data = json.loads(data) for e in data: #log.DATA(e) | < < | | | | | | | | | | | | | 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 | #log.DATA(data) #-- extract l = [] data = json.loads(data) for e in data: #log.DATA(e) if not len(l) or l[-1]["title"] != e["stream_name"]: l.append({ "title": e["stream_name"], "url": e["listen_url"], "format": e["type"], "bitrate": int(e["bitrate"]), "genre": e["genre"], "playing": e["current_song"], "listeners": 0, "max": 0, "homepage": (e["homepage"] if ("homepage" in e) else ""), }) # send back the list return l # Extract complete YP.XML, but just filter for genre/cat |
︙ | ︙ | |||
137 138 139 140 141 142 143 | buffy = [] # Get XML blob yp = ahttp.get(self.xml_url, statusmsg="Brace yourselves, still downloading the yp.xml blob.") log.DATA("returned") self.status("Yes, XML parsing isn't much faster either.", timeout=20) for entry in xml.dom.minidom.parseString(yp).getElementsByTagName("entry"): | < < | | | | | | | | | | | | | | 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 | buffy = [] # Get XML blob yp = ahttp.get(self.xml_url, statusmsg="Brace yourselves, still downloading the yp.xml blob.") log.DATA("returned") self.status("Yes, XML parsing isn't much faster either.", timeout=20) for entry in xml.dom.minidom.parseString(yp).getElementsByTagName("entry"): buffy.append({ "title": x(entry, "server_name"), "url": x(entry, "listen_url"), "format": self.mime_fmt(x(entry, "server_type")[6:]), "bitrate": bitrate(x(entry, "bitrate")), "channels": x(entry, "channels"), "samplerate": x(entry, "samplerate"), "genre": x(entry, "genre"), "playing": x(entry, "current_song"), "listeners": 0, "max": 0, "homepage": "", }) self.status("This. Is. Happening. Now.") # Filter out a single subtree l = [] if cat: rx = re.compile(cat.lower()) l = [] |
︙ | ︙ | |||
213 214 215 216 217 218 219 | .*? "listeners">\[(\d+) .*? "stream-description">(.*?)< .*? Tags: (.*?) </div> .*? href="(/listen/\d+/listen.xspf)" .*? class="format"\s+title="([^"]+)" .*? /by_format/([^"]+) """, html, re.X|re.S) | < | 221 222 223 224 225 226 227 228 229 230 231 232 233 234 | .*? "listeners">\[(\d+) .*? "stream-description">(.*?)< .*? Tags: (.*?) </div> .*? href="(/listen/\d+/listen.xspf)" .*? class="format"\s+title="([^"]+)" .*? /by_format/([^"]+) """, html, re.X|re.S) # Assemble for homepage, title, listeners, playing, tags, url, bits, fmt in ls: r.append(dict( genre = unhtml(tags), title = unhtml(title), homepage = ahttp.fix_url(homepage), |
︙ | ︙ | |||
243 244 245 246 247 248 249 | "top40", "90s", "80s", "britpop", "disco", "urban", "party", "mashup", "kpop", "jpop", "lounge", "softpop", "top", "popular", "schlager", ], "rock", [ "alternative", "electro", "country", "mixed", "metal", | | | | | | 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 | "top40", "90s", "80s", "britpop", "disco", "urban", "party", "mashup", "kpop", "jpop", "lounge", "softpop", "top", "popular", "schlager", ], "rock", [ "alternative", "electro", "country", "mixed", "metal", "eclectic", "folk", "anime", "hardcore", "pure", "jrock" ], "dance", [ "electronic", "deephouse", "dancefloor", "elektro", "eurodance", "rnb", ], "hits", [ "russian", "hit", "star" ], "radio", [ "live", "community", "student", "internet", "webradio", ], "classic", [ "classical", "ebu", "vivaldi", "piano", "opera", "classix", "chopin", "renaissance", "classique", ], "talk", [ "news", "politics", "medicine", "health", "sport", "education", "entertainment", "podcast", ], "various", [ "hits", "ruhit", "mega" ], "house", |
︙ | ︙ | |||
328 329 330 331 332 333 334 | "world", # REGIONAL [ "france", "greek", "german", "westcoast", "bollywood", "indian", "nederlands", "europa", "italia", "brazilian", "tropical", "korea", "seychelles", "black", "japanese", "ethnic", "country", "americana", "western", "cuba", "afrique", "paris", "celtic", "ambiance", "francais", "liberte", "anglais", "arabic", | | | 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 | "world", # REGIONAL [ "france", "greek", "german", "westcoast", "bollywood", "indian", "nederlands", "europa", "italia", "brazilian", "tropical", "korea", "seychelles", "black", "japanese", "ethnic", "country", "americana", "western", "cuba", "afrique", "paris", "celtic", "ambiance", "francais", "liberte", "anglais", "arabic", "hungary", "folklore", "latin", "dutch", "italy" ], "artist", # ARTIST NAMES [ "mozart", "beatles", "michael", "nirvana", "elvis", "britney", "abba", "madonna", "depeche", ], "salsa", |
︙ | ︙ | |||
418 419 420 421 422 423 424 | "musica", "old", "emisora", "img", "rockabilly", "charts", [ | | | 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 | "musica", "old", "emisora", "img", "rockabilly", "charts", [ "best80", "70er", "80er", "60er", "chart", ], "other", [ "varios" ], "soulful", "listening", |
︙ | ︙ |