Summary: | Support for file type ARJ | ||
---|---|---|---|
Product: | [Applications] ark | Reporter: | Thomas Winischhofer <thomas> |
Component: | plugins | Assignee: | Ragnar Thomsen <rthomsen6> |
Status: | RESOLVED FIXED | ||
Severity: | wishlist | CC: | elvis.angelaccio, nate, psychonaut, rakuco |
Priority: | NOR | ||
Version: | unspecified | ||
Target Milestone: | --- | ||
Platform: | Unlisted Binaries | ||
OS: | Linux | ||
Latest Commit: | https://commits.kde.org/ark/01d0a2dd13fd888dd75caa5844ecb69f40e6ae00 | Version Fixed In: | 22.12 |
Sentry Crash Report: | |||
Bug Depends on: | |||
Bug Blocks: | 192630 | ||
Attachments: | Test ARJ archive |
Description
Thomas Winischhofer
2005-04-23 00:58:37 UTC
Changing the default assignee for currently open Ark bug reports to me. Created attachment 98827 [details]
Test ARJ archive
Since p7zip can extract ARJ archives, should we add application/x-arj to the supported MIME types of the cli7zplugin? The problem is that the cli7zplugin is marked as "readwrite", so it should also be able to create ARJ archives. Can p7zip do that? Oh, it seems p7zip can read but cannot write ARJ archives. As an alternative, how about using the cliunarchiverplugin? Unarchiver can open many formats, including ARJ and LHA (Bug 321951). (In reply to 2wxsy58236r3 from comment #5) > Oh, it seems p7zip can read but cannot write ARJ archives. > > As an alternative, how about using the cliunarchiverplugin? Unarchiver can > open many formats, including ARJ and LHA (Bug 321951). That should be doable, yes. (In reply to Elvis Angelaccio from comment #6) The Unarchiver plugin claims that it is capable of opening the following file types: application/x-msi application/x-stuffit application/x-arj (This bug) application/x-arc application/x-pak application/x-ace application/x-zoo application/x-lha (Bug 321951, a.k.a. LZH) application/x-amiga-disk-format application/x-alz application/x-nintendo-ds-rom Reference: https://code.google.com/archive/p/theunarchiver/wikis/SupportedFormats.wiki (A more readable version: https://github.com/mietek/theunarchiver/wiki/SupportedFormats) A possibly relevant merge request was started @ https://invent.kde.org/utilities/ark/-/merge_requests/23 Implemented with https://invent.kde.org/utilities/ark/-/commit/89832ce1dc272c1b7ff5d1d8476071a48e1e08f1 Unarchiver has a bug [1] which prevents correct decompression of arj archives which contain filenames with slashes, so I have submitted a request to revert my commit earlier. Sorry for that. [1] https://github.com/MacPaw/XADMaster/issues/69 See also: https://phabricator.kde.org/T2443#78198 (In reply to Higa from comment #10) > Unarchiver has a bug [1] which prevents correct decompression of arj > archives which contain filenames with slashes, so I have submitted a request > to revert my commit earlier. So why reopen this bug report instead of opening a new one for that particular edge case? Surely the commit you submitted will add functionality that works with 99.9+% of ARJ archives, and so it would be better to keep it in place pending a fix for the bug. Yep, that's what should happen. > So why reopen this bug report instead of opening a new one for that particular edge case? Surely the commit you submitted will add functionality that works with 99.9+% of ARJ archives If an ARJ archive has folders, it is likely affected by the unarchiver bug. The autotest file [1] is also affected - the contents are listed properly, but if one attempts to extract it by dragging the contents to Dolphin, an error will be thrown. If it is extracted in Dolphin by right-click -> Extract archive, the folder hierarchy is lost and filenames become something like data_A_B_C_test2.txt. Is there a way to workaround this bug in Ark? If so, I can open a new bug report in KDE Bugtracker. From T2443, Ark developers are waiting for upstream fix, and additionally I am not sure how many ARJ archives are affected by the bug, so I thought I should revert my commit. I apologize for the troubles I have caused. I should have done more testing. For the time being, affected ARJ archives can be extracted by p7zip [2] in command line. [3] [1] https://invent.kde.org/utilities/ark/-/blob/master/autotests/plugins/cliunarchiverplugin/data/multiple_toplevel_entries.arj [2] Or 7zz (the binary is called "7zz" instead of "7z"), provided by the upstream 7-Zip 21.01 alpha version for Linux, released by Igor Pavlov (source code is not yet available though) [3] Although Ark has a p7zip plugin, if I remember correctly, it is currently a read-write plugin, so read-only formats like ARJ cannot be added to cli7zplugin for the time being. (In reply to Higa from comment #13) > If an ARJ archive has folders, it is likely affected by the unarchiver bug. Then I think you ought to have said that the bug affects paths containing a slash, not filenames containing a slash. The former is exceedingly common and the latter exceedingly uncommon. https://invent.kde.org/utilities/ark/-/commit/01d0a2dd13fd888dd75caa5844ecb69f40e6ae00 Ark master (future 22.12 KDE Gear release) now has a proper backend to handle the ARJ format. Thanks to Ilya Pominov for the contribution. |