Summary: | webdav ioslave does not implement random access, forcing kio-fuse-accessed files to be read in their entirety, rather than streamed | ||
---|---|---|---|
Product: | [Frameworks and Libraries] frameworks-kio | Reporter: | ihipop <ihipop> |
Component: | WebDAV | Assignee: | KIO Bugs <kio-bugs-null> |
Status: | CONFIRMED --- | ||
Severity: | normal | CC: | a.saoutkin, fabian, kdelibs-bugs, kio-bugs-null, mapatrapa, micraft.b, nate, qydwhotmail |
Priority: | NOR | ||
Version: | git master | ||
Target Milestone: | --- | ||
Platform: | Manjaro | ||
OS: | Linux | ||
Latest Commit: | Version Fixed In: | ||
Sentry Crash Report: | |||
Attachments: |
GVFS network speed chart
KIO-FUSE network speed chart |
Description
ihipop
2021-09-10 06:13:02 UTC
Created attachment 141436 [details]
KIO-FUSE network speed chart
The kioslave for webdav:// (http.so) does not implement random access, so kio-fuse has no choice but to download the file in a single chunk. It should start playing during the download already though, unless the video file isn't directly streamable and contains necessary data (like an index) at the end. @Fabian Vogt It not start playing during the download ,as you said ,a video file often contains necessary data (like an index) at the end. It will make the video player to `JAM` and Keeps Not Responding I have observed the same behavior with sftp and fish. For a few months, the entire file is downloaded first and then the video plays (In reply to Jesus from comment #4) > I have observed the same behavior with sftp and fish. For a few months, the > entire file is downloaded first and then the video plays I also have this problem with sftp in SMPlayer, which did not exist a few months back. Now the player opens but can't even draw its interface before the entire file has been downloaded by kio Fabian is correct, for WebDav unfortunately this is the expected behaviour until the HTTP slave implementation gains random access support. For the SFTP protocol, this is most likely that the file is opened by KIOExec, not KIO FUSE because of this regression tracked here: https://bugs.kde.org/show_bug.cgi?id=436553 any update on this issue? |