Why won't podcast notice already downloaded podcasts?

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Why won't podcast notice already downloaded podcasts?

SnerkRabbledauber
On a new installation of Linux Mint 17.2 I configured Banshee to use a
podcast directory in which I had already downloaded many podcasts while
running Banshee on my old computer. Now when I subscribe to a podcast it
shows no archived shows, even though they are there in the podcast
directory.

Is there any way to make Banshee see those existing podcasts instead of
downloading copies of all of them?

Thanks.
_______________________________________________
banshee-list mailing list
[hidden email]
https://mail.gnome.org/mailman/listinfo/banshee-list  (unsubscribe here)
Reply | Threaded
Open this post in threaded view
|

Re: Why won't podcast notice already downloaded podcasts?

David Nielsen
Getting Banshee to do that is harder than it sounds. Banshee doesn’t even attempt to do this, no media player with podcast support I know of does this, it is largely down to how Podcasts work today.

tl;dr:
There is no reliable way to do this. Podcasts rely on extremely frail technology. Importing an existing Banshee database is your best bet.

Longer, more depressing:

Each Show/Network publishes a feed (typically RSS) which lists the episodes they currently have available. The norm is to have a history that goes back to the very first episode, but in some cases e.g. the TWiT network (hosting popular shows like Security Now!, This Week in Tech, and friends) limits their feed to the most recent X episodes. This is also the case for feeds which bundle up previously aired seasons to sell, in which case these will be removed from the published feed.

Hence if the feed no longer lists the episodes, they cannot be matched to an existing feed and you are out of luck.

We could attach helpful data to the downloaded files, but this would cause problems for episodes that change e.g. to correct an audio problem, publishers will occasionally replace the file and clients then redownload it. Alternatively we could keep such records separately, but that would be no different than you importing an existing Banshee database which should retain your subscriptions.

The podcast feed does have useful information that could be applied towards this task, e.g. if there was a published checksum we could calculate that for files found in the Podcast folder or specified by the user and match them that way. After all checksum must be how we determine if the file was downloaded correctly, right?

Bingo you are thinking, Banshee sucks for not doing this and you’d be right.. was it not for the depressing state of Podcast feeds. Sadly no podcast publisher actually fills out the fields in their feeds correctly and the data gotten from a feed is reliable outside of providing a valid download URL, only in a few cases (forget things like checksums, unique entry IDs, contact emails.. you know, useful stuff). Feel free to confirm this by running your favorite feeds through Feed Validator (http://feedvalidator.org) and watch the horror.

My expressed opinion for quite a while has been that Podcasters need better tools for publishing their work and manage their feeds. Podcasts are simply, sadly, broken at this point in time, and parsing currently published feeds have been known to summon the Great Old Ones with greater reliability than provide the information Podcast clients need. 

- David


Sent from Mailbox


On Mon, Jul 27, 2015 at 6:48 PM, Kevin Bowersox <[hidden email]> wrote:

On a new installation of Linux Mint 17.2 I configured Banshee to use a
podcast directory in which I had already downloaded many podcasts while
running Banshee on my old computer. Now when I subscribe to a podcast it
shows no archived shows, even though they are there in the podcast
directory.

Is there any way to make Banshee see those existing podcasts instead of
downloading copies of all of them?

Thanks.
_______________________________________________
banshee-list mailing list
[hidden email]
https://mail.gnome.org/mailman/listinfo/banshee-list (unsubscribe here)



_______________________________________________
banshee-list mailing list
[hidden email]
https://mail.gnome.org/mailman/listinfo/banshee-list  (unsubscribe here)
Reply | Threaded
Open this post in threaded view
|

Re: Why won't podcast notice already downloaded podcasts?

Adam  Tauno Williams
On Tue, 2015-07-28 at 07:15 -0700, [hidden email] wrote:
> tl;dr:
> There is no reliable way to do this. Podcasts rely on extremely frail
> technology. Importing an existing Banshee database is your best bet.

+1

> Longer, more depressing:
> Each Show/Network publishes a feed (typically RSS) which lists the
> episodes they currently have available.  .... The podcast feed does
> have useful information that could be applied towards this task, e.g.
> if there was a published checksum we could calculate that for files
> found in the Podcast folder or specified by the user and match them
> that way. After all checksum must be how we determine if the file was
> downloaded correctly, right?

Technically I believe the Etag on the podcast blob [the MP3 file]
should be able to serve the purpose of change detection with a simple
HEAD request... but, of course, nobody implements this [server side]
correctly.  So there is no point in a client making the request.

> Bingo you are thinking, Banshee sucks for not doing this and you’d be
> right.. was it not for the depressing state of Podcast feeds.

Yep.

--
Adam Tauno Williams <mailto:[hidden email]> GPG D95ED383
Systems Administrator, Python Developer, LPI / NCLA

_______________________________________________
banshee-list mailing list
[hidden email]
https://mail.gnome.org/mailman/listinfo/banshee-list  (unsubscribe here)
Reply | Threaded
Open this post in threaded view
|

Re: Why won't podcast notice already downloaded podcasts?

SnerkRabbledauber
In reply to this post by David Nielsen
Thanks for the great answer.

I restored the old banshee database. After a couple of complete scans I
had my music and my podcast list back.

On 07/28/2015 08:15 AM, [hidden email] wrote:

> Getting Banshee to do that is harder than it sounds. Banshee doesn’t even attempt to do this, no media player with podcast support I know of does this, it is largely down to how Podcasts work today.
>
>
>
>
> tl;dr:
>
> There is no reliable way to do this. Podcasts rely on extremely frail technology. Importing an existing Banshee database is your best bet.
>
>
>
>
> Longer, more depressing:
>
>
>
>
> Each Show/Network publishes a feed (typically RSS) which lists the episodes they currently have available. The norm is to have a history that goes back to the very first episode, but in some cases e.g. the TWiT network (hosting popular shows like Security Now!, This Week in Tech, and friends) limits their feed to the most recent X episodes. This is also the case for feeds which bundle up previously aired seasons to sell, in which case these will be removed from the published feed.
>
>
>
>
> Hence if the feed no longer lists the episodes, they cannot be matched to an existing feed and you are out of luck.
>
>
>
>
>
> We could attach helpful data to the downloaded files, but this would cause problems for episodes that change e.g. to correct an audio problem, publishers will occasionally replace the file and clients then redownload it. Alternatively we could keep such records separately, but that would be no different than you importing an existing Banshee database which should retain your subscriptions.
>
>
>
>
> The podcast feed does have useful information that could be applied towards this task, e.g. if there was a published checksum we could calculate that for files found in the Podcast folder or specified by the user and match them that way. After all checksum must be how we determine if the file was downloaded correctly, right?
>
>
>
>
> Bingo you are thinking, Banshee sucks for not doing this and you’d be right.. was it not for the depressing state of Podcast feeds. Sadly no podcast publisher actually fills out the fields in their feeds correctly and the data gotten from a feed is reliable outside of providing a valid download URL, only in a few cases (forget things like checksums, unique entry IDs, contact emails.. you know, useful stuff). Feel free to confirm this by running your favorite feeds through Feed Validator (http://feedvalidator.org) and watch the horror.
>
>
>
>
> My expressed opinion for quite a while has been that Podcasters need better tools for publishing their work and manage their feeds. Podcasts are simply, sadly, broken at this point in time, and parsing currently published feeds have been known to summon the Great Old Ones with greater reliability than provide the information Podcast clients need.
>
>
>
>
> - David
>
>
> —
> Sent from Mailbox
>
> On Mon, Jul 27, 2015 at 6:48 PM, Kevin Bowersox <[hidden email]> wrote:
>
>> On a new installation of Linux Mint 17.2 I configured Banshee to use a
>> podcast directory in which I had already downloaded many podcasts while
>> running Banshee on my old computer. Now when I subscribe to a podcast it
>> shows no archived shows, even though they are there in the podcast
>> directory.
>> Is there any way to make Banshee see those existing podcasts instead of
>> downloading copies of all of them?
>> Thanks.
>> _______________________________________________
>> banshee-list mailing list
>> [hidden email]
>> https://mail.gnome.org/mailman/listinfo/banshee-list  (unsubscribe here)
>>
>>
>> _______________________________________________
>> banshee-list mailing list
>> [hidden email]
>> https://mail.gnome.org/mailman/listinfo/banshee-list  (unsubscribe here)
_______________________________________________
banshee-list mailing list
[hidden email]
https://mail.gnome.org/mailman/listinfo/banshee-list  (unsubscribe here)