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

⌈⌋ ⎇ branch:  streamtuner2


Check-in [61a51c29f9]

Overview
Comment:Shoutcast: retry regex/dom really alternatively now (not just on exceptions, but also empty result sets)
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1: 61a51c29f9c7c26c536f9ec9834e7424503a405e
User & Date: mario on 2014-05-26 20:21:17
Other Links: manifest | tags
Context
2014-05-26
21:18
Reorganized Jamendo plugin to simplify API calls. check-in: ca359689bf user: mario tags: trunk
20:21
Shoutcast: retry regex/dom really alternatively now (not just on exceptions, but also empty result sets) check-in: 61a51c29f9 user: mario tags: trunk
19:59
Retry regex after PyQuery extraction mode (or other way round). check-in: 696a0ab060 user: mario tags: trunk
Changes
Hide Diffs Unified Diffs Ignore Whitespace Patch

Modified channels/jamendo.py from [75f35740e6] to [c602d9f7b7]. 

1
2
3
4
5

6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

# api: streamtuner2
# title: Jamendo
# description: A license-free music collection and artist hub.
# type: channel

# category: radio
# depends: json
# priority: default
#
# Now utilizes the Jamendo /v3.0/ API.
#
# Radio station lists are fixed for now. Querying the API twice per station
# doesn't seem overly sensible.
#
# Albums and Playlists are limited to 200 entries. Adding a cursor is
# feasible however.
#
# Tracks are queried by genre, where currently there's just a small built-in
# tag list in ST2.
#
#
# The v3.0 streaming URLs don't seem to work. Therefore some /get2 URLs will
# be used.
#





>









<
<
<







1
2
3
4
5
6
7
8
9
10
11
12
13
14
15



16
17
18
19
20
21
22

# api: streamtuner2
# title: Jamendo
# description: A license-free music collection and artist hub.
# type: channel
# version: 2.2
# category: radio
# depends: json
# priority: default
#
# Now utilizes the Jamendo /v3.0/ API.
#
# Radio station lists are fixed for now. Querying the API twice per station
# doesn't seem overly sensible.
#



# Tracks are queried by genre, where currently there's just a small built-in
# tag list in ST2.
#
#
# The v3.0 streaming URLs don't seem to work. Therefore some /get2 URLs will
# be used.
#

Modified channels/shoutcast.py from [ab1cca6f24] to [a7dc35c676]. 

124
125
126
127
128
129
130

131
132
133
134
135
136
137
138
139
140
141
142

143
144
145
146
147
148
149
150
            """

            # With the new shallow <td> lists it doesn't make much sense to use
            # the pyquery DOM traversal. There aren't any sensible selectors to
            # extract values; it's just counting the tags.
            # And there's a bug in PyQuery 1.2.4 and CssSelector. So make two
            # attempts, alternate between regex and DOM; user preference first.

            use_regex = not conf.get("pyquery") or not pq
            retry = 2
            while retry:
               retry -= 1
               try:
                  if use_regex:
                      return self.with_regex(html)
                  else:
                      return self.with_dom(html)
               except Exception as e:
                  use_regex ^= 1
                  __print__(dbg.ERR, e)

            return []


        # Extract using regex
        def with_regex(self, html):
            __print__(dbg.PROC, "channels.shoutcast.update_streams: regex scraping mode")
            rx_stream = re.compile(
                """







>


|



|

|

<

>
|







124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141

142
143
144
145
146
147
148
149
150
151
            """

            # With the new shallow <td> lists it doesn't make much sense to use
            # the pyquery DOM traversal. There aren't any sensible selectors to
            # extract values; it's just counting the tags.
            # And there's a bug in PyQuery 1.2.4 and CssSelector. So make two
            # attempts, alternate between regex and DOM; user preference first.
            entries = []
            use_regex = not conf.get("pyquery") or not pq
            retry = 2
            while retry and not entries:
               retry -= 1
               try:
                  if use_regex:
                      entries = self.with_regex(html)
                  else:
                      entries = self.with_dom(html)
               except Exception as e:

                  __print__(dbg.ERR, e)
               use_regex ^= 1
            return entries


        # Extract using regex
        def with_regex(self, html):
            __print__(dbg.PROC, "channels.shoutcast.update_streams: regex scraping mode")
            rx_stream = re.compile(
                """