m3nu / upodder Goto Github PK

Hello,

somtimes a podcast disappears. But the url still works and upodder downloads it, tryies to parse it and fails:

Traceback (most recent call last): File "/home/user/.pyenv/versions/2.7.10/bin/upodder", line 11, in <module> sys.exit(main()) File "/home/user/.pyenv/versions/2.7.10/lib/python2.7/site-packages/upodder/upodder.py", line 193, in main process_feed(url) File "/home/user/.pyenv/versions/2.7.10/lib/python2.7/site-packages/upodder/upodder.py", line 168, in process_feed l.info("Checking feed: %s"%feed.feed.title) File "/home/user/.pyenv/versions/2.7.10/lib/python2.7/site-packages/feedparser.py", line 400, in __getattr__ raise AttributeError, "object has no attribute '%s'" % key AttributeError: object has no attribute 'title'
That is, it looks for some attribute and it does not exit.
As a suggestion, at least upodder should show which feed is parsing in order to make easier to delete it. Something like this:

if url and url[0] not in CONFIGCOMMENT: l.debug("Url: %s" % url) process_feed(url)
It would be nice to be able to capture the exception and to allow upodder to continue downloading podcasts...

Thank you and keep up the good work!

Could you add a command line argument (e.g. -s) to create one subdirectory per feed inside the -p path? That would be ideal for organization when downloading/archiving huge amounts of podcast. If that option already exists I could not find it in the documentation.

Currently we only show the URL. Title is more useful.

Too old for us: Celebrating (or not) the Centenary of the Russian Revolution [Audio]
Already seen: Celebrity [Audio]
Downloading The French Election and the Left [Audio] from http://media.rawvoice.com/lse_publiclecturesandevents/richmedia.lse.ac.uk/publiclecturesandevents/20170509_1830_theFrenchElectionAndTheLeft.mp3
100%|████████████████████████████████████████████████| 42.5M/42.5M [00:25<00:00, 1.66MB/s]
Moving {/tmp/upodder/cadce213ff5c92391888d1e784a4180c93cad59d} to {/home/Jadamso/Desktop/MiscPersonal/Audio/Podx/The_French_Election_and_the_Left_Audio_.mp3}
Traceback (most recent call last):
File "/usr/bin/upodder", line 9, in
load_entry_point('upodder==0.6.9', 'console_scripts', 'upodder')()
File "/usr/lib/python2.7/site-packages/upodder/upodder.py", line 212, in main
process_feed(url)
File "/usr/lib/python2.7/site-packages/upodder/upodder.py", line 192, in process_feed
EntryProcessor(entry, feed)
File "/usr/lib/python2.7/site-packages/upodder/upodder.py", line 98, in init
SeenEntry( pub_date=self.pub_date, hashed=self.hashed)
File "/usr/lib64/python2.7/site-packages/sqlobject/main.py", line 1301, in init
self._create(id, **kw)
File "/usr/lib64/python2.7/site-packages/sqlobject/main.py", line 1352, in _create
self._SO_finishCreate(id)
File "/usr/lib64/python2.7/site-packages/sqlobject/main.py", line 1381, in _SO_finishCreate
self._init(id)
File "/usr/lib64/python2.7/site-packages/sqlobject/main.py", line 986, in _init
self._SO_selectInit(selectResults)
File "/usr/lib64/python2.7/site-packages/sqlobject/main.py", line 1227, in _SO_selectInit
colValue = _col.to_python(colValue, self._SO_validatorState)
File "/usr/lib64/python2.7/site-packages/sqlobject/col.py", line 1251, in to_python
return datetime.timedelta(seconds=value.seconds)
AttributeError: 'str' object has no attribute 'seconds'

podx.txt

Please consider adding support for 'audio/x-m4a' filetype in order to improve downloading of some m4a podcast files on feeds like these:

http://feeds.soundcloud.com/users/soundcloud:users:205124646/sounds.rss
http://allearsenglish.libsyn.com/rss

This is my current FILE_TYPES dictionary:

# Dict of possible file types
FILE_TYPES = {
    'audio/mpeg': 'mp3',
    'audio/x-m4a': 'm4a',
    'video/x-m4v': 'm4v',
    'audio/x-opus': 'opus',
    'audio/x-ogg': 'ogg',
    'audio/aac': 'aac',
    'audio/mp4': 'm4a',
    'audio/mp3': 'mp3'
    }

For the following feed, it fails to show or download the latest 3 episodes:
https://www.relay.fm/clockwise/feed

There is no error message shown. I have not had a chance to debug this feed. Seems like a valid feed on feedvalidator.com

What steps will reproduce the problem?
1. Using this rss feed: 
http://downloads.bbc.co.uk/podcasts/radio4/fricomedy/rss.xml
2. Run upodder

What is the expected output? What do you see instead?

Upodder reports a problem with the feed, but there is nothing wrong with the 
feed but feedparser reports that it has overridden the HTTP encoding.  
Feedparser sets bozo and sets bozo_exception to CharacterEncodingOverride in 
this case.  Below is a small patch to address this case.


What version of the product are you using? On what operating system?
r7 version.

Patch to address this issue.

Index: upodder.py
===================================================================
--- upodder.py  (revision 7)
+++ upodder.py  (working copy)
@@ -164,7 +164,7 @@
 def manageFeed(url):
        """Let's deal with this podcast feed"""
        feed = feedparser.parse(url)
-       if feed.bozo:
+       if feed.bozo and type(feed.bozo_exception) != 
type(feedparser.CharacterEncodingOverride())::
                l.error("Erroneous feed URL: %s (%s)"%(url,feed.bozo_exception))
                return
        l.info("Checking feed: {%s}"%feed.feed.title)



Please provide any additional information below.

Original issue reported on code.google.com by [email protected] on 24 Sep 2011 at 9:44

https://news.ycombinator.com/item?id=19063727

Sometimes hangs on unreliable connections. Resume support?

Some podcast hosting services (blubrry.com for example) do not allow default 
urllib user agent. It would be nice to allow some customization here.

Original issue reported on code.google.com by [email protected] on 2 Jun 2014 at 3:34

Hey there! Thanks for picking up maintenance of upodder!

I ran into a problem when using upodder from PyPI. If your /tmp/ directory is on a different partition than your target directory (as is very common with *nix machines), you'll see an error like this:

(snip)
python2.7/site-packages/upodder/upodder.py", line 127, in _download_enclosure
    os.rename(downloadto, moveto)
OSError: [Errno 18] Cross-device link

Moves/renames can't be executed across partitions/devices like that. The usual response is to either:

1. Copy the target to destination, then delete the original copy of the target.
2. Use a location on the same partition as the temporary file.

The latter is preferable in most of my experience, so I created the AtomicSaver (docs, code). Give it a glance and let me know if you'd like any help fixing this issue.

Add support for downloading latest only

At the moment you start upodder, it stays silent for a bit (downloading the feed no doubt), then shows the urls "too old" and "already seen".

Would be nice if could show the name of the feed it is processing before it shows the urls...

I would like to not get debug logging but --log-level doesn't appear to do anything...

the oldness argument is supposed to allow the script to ignore feed items older than X number of days (default=30). If I try to use this option, it always fails because the number I enter on the commandline is a string instead of an int.

The fix is easy: please add type=int to the add_argument() function

The default=30 always works, because 30 is an int. But commandline arguments are interpreted as strings by default.

Near as I can tell, the only way to modify the output file name is to edit upodder.py and change FILENAME.

A nice enhancement would be to add a command line switch
--filename="{entry_date} {entry_title}.{filename_extension}"

With a -d / --delete flag, any episode that exists locally and is marked as read should be deleted.