http://bugs.gpodder.org/show_bug.cgi?id=211





--- Comment #11 from ad <[EMAIL PROTECTED]>  2008-10-01 22:13:10 ---
(In reply to comment #10)
> Ok, reopening then.
> 
> So, here's how we should do it:
> 
> 1.) Rely on the RSS feed to give some sane value (should we hide the size or
> return "unknown" for obviously (say, below 2kb) wrong file sizes?)
> 2.) On download (when it's finished), update the database with the correct
> size, if it differs
> 3.) For updates to the feed for downloaded items - see below
> 
> The good thing with updating metadata for already-downloaded items is for
> things like when some podcast releases the file with "description follows" and
> then updates the feed afterwards. This might be a good thing (i.e. DO update
> the database for downloaded files when the feed changes).
> 
> For the filesize, as soon as we've downloaded a file, we should always display
> the file size of the downloaded file. Does stat'ing the downloaded file have 
> an
> impact on performance? If so, we might just update the file size in the
> database after the download and consider the value in the database "correct"
> after it has been downloaded (and NEVER update the file size for downloaded
> files from the feed).
> 
> An alternative would be to update file sizes after every feed update by
> stat'ing all downloaded files and making sure that the file size is correct in
> the database. Advantage: Also works for already-downloaded episodes;
> disadvantage: performance impact on the "update feeds" function
> 
> Your input?
> 

I agree this is the correct process to use, certainly hiding the filesize or
displaying 'Unknown' would be good for podcast feeds showing episodes at 2B or
less. When a file is downloaded, the size should be updated in the database to
reflect the correct size.

Perhaps isolating the function to check all downloaded files into a Tools menu
function would be preferable - i.e. 'Check downloaded files' or similar. This
could also be expanded at a later date to include a sanity check of other data
stored in the database - whether files exist, etc.

Updating the database with information when a feed changes is also of use, but
leave the filezsize alone - unless perhaps the filesize on the feed is larger
than the downloaded file - in which case the user is alerted, or the episode is
flagged as such. It's possible a podcast is partially posted, then updated with
the full one at a later date.


-- 
Configure bugmail: http://bugs.gpodder.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.
_______________________________________________
gPodder-Bugs mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/gpodder-bugs

Reply via email to