u3zY8V8TiHY9szu6C84u wrote:attempting
(Closed Filehost) http://skyve.io/cf0g4iiiwr9ifrom
viewtopic.php?f=124&t=5578856The Premium Downloader successfully downloads a file, HOWEVER the downloaded file is much smaller than either the Mobilism posting or skyve.io (who agree) say it should be: 27MB vs 426MB:
Code: Select all~/Downloads
$ ls -lh Release_Point.._Drop_Trooper,_Book_8.m4b
-rw-r--r-- 1 xz 197121 27M Apr 4 20:06 Release_Point.._Drop_Trooper,_Book_8.m4b
~/Downloads
$ sha256sum Release_Point.._Drop_Trooper,_Book_8.m4b
075097ab95810763b169b5ccba477e77d00550178244303e4cd7a56a79ced059 *Release_Point.._Drop_Trooper,_Book_8.m4b
I've downloaded the same 27MB file twice using PD.
I am currently attempting a download (of the 426MB version) of this file from skyve.io, but it (and its peer fikper.com) is an unpleasant site to interact with, and the free skyve.io download is much slower than the Premium Downloader.
Help please?
Hi there,
I'm not part of the mobilism team, but I can confirm the issue for this link and can share what I have observed.
When the premium downloader is used for a link, it connects to the filehost, retrieves the name and checks if the name is in the cache. If it's not found, it downloads the file from the remote filehost, stores a copy of the downloaded file in a cache, and then sends the file to the browser of the user. You see a progress bar while the premium downloader retrieves the file from the filehost.
If the premium downloader finds that the filename is contained in the cache, it will return the cached copy, which is why the download is really fast, independent of .the used filehost. You will see no progress bar in this case.
It can happen, that an error occurs while the premium downloader retrieves a file from the filehost. Unfortunately, the premium downloader is currently not able to detect this error, and stores whatever it has retrieved from the filehost, before the error occurred in the cache. Thus any subsequent press on the blue arrow will return the truncated/corrupted file from the cache.
It can also happen that the premium downloader stores the "wrong" file in the cache. If, for example, the site contains two postings, one for and epub and one for an audiobook, and both links happen to have the same filename, e.g. sample.rar, then the file from whichever link is used first with the premium downloader will be stored in the cache and returned for both postings. This happens very rarely.
There are only two solutions for a user in this case. Either manually download the file from one of the filehosts, which is what you are doing. The other solution is to wait until the corrupt/truncated file is removed from the cache. Not sure, how long this takes, as it probably depends on many factors. You might try again in a week.or two. The downside of this approach is, that the file might have already been deleted from the filehost, and the premium downloader will not be able to retrieve it again.
I have asked the mobilism team a few years back to implement a check if a file has been downloaded correctly from a filehost and only store those files in the cache. This would prevent truncated, corrupted files to be stored in the cache, but this seems to be a very hard task, and has, so far, not been implemented.
As a second best alternative, a button to allow a user to remove a particular file from the cache could be implemented, but this also has not been done.
All in all, the premium downloader works great in most cases and saves a lot of trouble, but care must be taken to deal with the few issues of truncated files.
Take care