| Summary: | Drag & Drop file from Ark to mpv | ||
|---|---|---|---|
| Product: | [Applications] ark | Reporter: | Pierre L. <pihug12+kde> |
| Component: | general | Assignee: | Elvis Angelaccio <elvis.angelaccio> |
| Status: | RESOLVED DUPLICATE | ||
| Severity: | normal | CC: | kdebugs, nate, rthomsen6 |
| Priority: | NOR | ||
| Version: | 20.04.3 | ||
| Target Milestone: | --- | ||
| Platform: | Other | ||
| OS: | Linux | ||
| Latest Commit: | Version Fixed In: | ||
| Sentry Crash Report: | |||
|
Description
Pierre L.
2020-08-29 12:13:35 UTC
This needs support in mpv. Ark provides a dbus API that mpv can call to trigger files extraction. That dbus API is bad. It requires the application to implement Ark-specific handling for drag-and-drop events by making the application tell Ark to extract the file instead of having Ark just extract the file to a temporary location and pass that temporary location to the application through the normal drag and drop mechanism with no involvement of a secondary dbus API to initiate this process. Is there literally any reason as to why it was misengineered this way? (In reply to Nicolas Frattaroli from comment #2) > That dbus API is bad. It requires the application to implement Ark-specific > handling for drag-and-drop events by making the application tell Ark to > extract the file I agree it's not good having to write ark-specific code for each app. Ideally there should be a freedesktop specification for this task. There is https://www.freedesktop.org/wiki/Specifications/XDS/ but I'm not sure if it's what we need. > instead of having Ark just extract the file to a temporary > location and pass that temporary location to the application through the > normal drag and drop mechanism with no involvement of a secondary dbus API > to initiate this process. > > Is there literally any reason as to why it was misengineered this way? I can think of one: extraction takes a long time. Doing it like you say would make the implementation much harder. Can be done of course, but someone needs to work on it. > I can think of one: extraction takes a long time.
One possible solution is to have Ark expose a temporary fuse mount of the archive. It mounts the archive, passes a filename to the application of that path (i.e. no Ark-specific code is needed) and then keeps around a fuse daemon process while there are still some file handles to it open, plus some grace period for reopening and such. That way, ark only really has to extract as the application reads.
*** This bug has been marked as a duplicate of bug 408233 *** |