summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Make all Gtk calls from inside of it's main loop (and thread)threadingMartin Sivak2012-10-167-106/+109
| | | | | | | | | | | | | | | | | | | | | | | | | | | | - Marks all methods containing mostly gtk calls as gtk_thread_wait or gtk_thread_nowait - Uses gtk_call_once instead of GLib.idle_add to make sure the method is called only once (returns False) - Removes some code from the threading locks, because it wasn´t touching Gtk at all This change was discussed in the mailinglist here: https://www.redhat.com/archives/anaconda-devel-list/2012-October/msg00030.html The main point was: According to the Gtk team, the gdk_threads_enter/leave pair should not be used at all (and they have apparently discouraged usage of it since early releases of Gtk2). Moreover in the current Gdk docs (http://developer.gnome.org/gdk3/stable/gdk3-Threads.html) those functions are now marked as deprecated. The preferred way (and now the only way) is to use g_idle_add (GLib.idle_add) with a callback method to schedule GUI changes. The callback method will then get called by the Gtk main loop so no locking is needed (and GLib.idle_add performs none). But that is also the reason why everything Gtk related must be done from the mainloop thread either directly or via idle_add.
* Remove Gdk thread initialization, introduce new helper functions and make ↵Martin Sivak2012-10-163-11/+62
| | | | exception handler be called by Gtk only once
* Fix threading initializationMartin Sivak2012-10-161-2/+4
|
* Do not remove the layout if it was added back (#865830)Vratislav Podzimek2012-10-161-1/+2
| | | | | | When replacing the last layout in the list, user may mark the same layout (among the others) for adding. In such cases we shouldn't remove the layout from the list and just leave it where it was.
* Release Gdk lock in exception handlingVratislav Podzimek2012-10-161-1/+5
| | | | | Since we add actions to main loop, not releasing lock may lead to deadlock.
* Configure new-kernel-pkg to keep tboot configuration on updates (#742885)Peter Jones2012-10-151-0/+4
| | | | | | | | | | | If we want tboot to keep getting used, we need to configure new-kernel-pkg to tell grubby to do that. When grubby finds a multiboot stanza as the default template, it won't use the multiboot parts of the template unless specifically configured to. This patch adds a configuration option to tell grubby that we always want to use the tboot multiboot kernel by default when we add a new kernel. Signed-off-by: Peter Jones <pjones@redhat.com>
* Honor the nompath option.David Lehman2012-10-152-7/+14
|
* Validate lv names.David Lehman2012-10-151-0/+9
| | | | Resolves: rhbz#747278
* Add support for specifying encryption cipher mode via kickstart.David Lehman2012-10-151-0/+6
| | | | Related: rhbz#838736
* Acquire the yum lock before accessing YumBase.repos. (#858993)David Lehman2012-10-151-1/+5
|
* Remove the entry on the resize dialog's combo boxes.Chris Lumens2012-10-151-0/+1
| | | | | There's no need for these to be editable, and setting this off also helps a little with the packing.
* disks_free -> disks_size (#863647).Chris Lumens2012-10-151-1/+1
|
* Fix one more reference to bits (#859932).Chris Lumens2012-10-151-1/+1
|
* Fix a traceback in media check (#865897).Chris Lumens2012-10-151-4/+3
|
* Add support for deleting an entire root via the existing ConfirmDeleteDialog.Chris Lumens2012-10-152-20/+47
| | | | | | You've already bought into doing something destructive, so why not have the chance to really tear things up? Also, make the cancel button the default action.
* Don't traceback when removing a mountpoint with no expanded selector (#862746).Chris Lumens2012-10-151-32/+35
| | | | | Most of this patch is just un-indenting stuff. The rest is replacing references to currentPage with references to selector._root. Same data, really.
* Remove the code for removing an entire Root all at once.Chris Lumens2012-10-151-24/+0
| | | | | | | | | First, it simply does not work right now. Secondly, it's not at all obvious in the UI how you would go about doing this. Click on the expander and then hit remove? Yes, but then the expander collapses and you still have a mountpoint displayed on the right. Which thing are you deleting then? We need a way to do this, but I don't think this code is it.
* Yet more TODO list updates.Chris Lumens2012-10-151-3/+8
|
* Don't display "None" in the name of a root.Chris Lumens2012-10-151-2/+8
| | | | | | It's possible to end up in a situation where the root is "None Linux None for None" which just looks silly. This happens in my testing if you kill an install after it's made partitions, but before fedora-release gets installed.
* Fix configuration of protected wireless connections (#855526)Radek Vykydal2012-10-131-7/+12
|
* Fix graphical kickstart with %packages dataJesse Keating2012-10-121-24/+30
| | | | | | | | Without this change we were automatically selecting the first environment from the list, and adding it to the existing package data. This means no matter what you had in %packages you also got the GNOME environment. Obviously this is wrong, so we don't autoselect if we're doing an automated install and have package data.
* Add password validation to text password spokeJesse Keating2012-10-121-9/+24
| | | | | | This brings it up to feature parity with the gui spoke, mostly. Still don't have a way to completely disable the password here, but that's a good thing for now.
* Make use of the validatePassword routine from users.pyJesse Keating2012-10-121-39/+15
| | | | | | Errors bring returned is a simple mismatch or invalid chars, which we can't allow. Raised items are actual quality issues which the user could use anyway.
* Add a password verification method to users.pyJesse Keating2012-10-121-0/+38
| | | | | | This moves it to a more central spot so that it can be used by both the gui and the text spoke, or any other thing that wants to validate passwords.
* Always honor the 'nokill' flagVratislav Podzimek2012-10-121-6/+6
| | | | | | There may be cases when we don't get exit code. Especially in those nonstandard cases we should honor the 'nokill' flag and don't reboot the machine.
* Fall back to lvm autopart if the default fails. (#864708)David Lehman2012-10-121-5/+29
| | | | | This will only help if the disks and/or their free regions are very small.
* Special boot devices are handled the same whether they exist or not.David Lehman2012-10-121-7/+4
| | | | | | They won't show up in /etc/fstab and they don't have mountpoints, so it really doesn't matter if they exist or not. We want to include them in the new root regardless.
* Fix a bug allocating fixed-size partitions.David Lehman2012-10-121-1/+1
| | | | | This was only triggered if none of the partitions being allocated were growable, which has become a very rare case.
* Clean up size sets immediately after allocation run. (#864771)David Lehman2012-10-121-0/+3
|
* Make sure partition base sizes are adequate for their formatting.David Lehman2012-10-121-3/+6
| | | | | There is no guarantee of growth, so be sure to start with a size that can hold the specified formatting.
* Don't fail to account for all set members' growth.David Lehman2012-10-121-3/+0
| | | | | | | If we skip accounting for any requests, the set's data will no longer match the state of its devices. We need to know how much every device has grown. This bug was leading to severely overgrown sets in some cases.
* Remove some extra calls to show_first_mountpoint.David Lehman2012-10-121-5/+0
| | | | | It already gets called from _do_refresh and multiple calls to save_right_side are both wasteful and potentially dangerous.
* Show the correct raid options for btrfs.David Lehman2012-10-121-1/+5
|
* Support change of raid level in custom spoke.David Lehman2012-10-121-6/+26
|
* Use devicetree as partition list source instead of parted. (#864718)David Lehman2012-10-121-3/+1
| | | | | | The parted list should be pretty much in sync, but this way saves some trouble with resolving the parted partition paths to devices in the devicetree.
* Use Storage convenience methods to schedule reclaim actions.David Lehman2012-10-121-7/+3
| | | | | This also uses Storage.recursiveRemove to remove dependent devices before trying to remove the partition.
* Pass disk list when trying to recover from device type change failure.David Lehman2012-10-121-0/+1
|
* Fill in missing parts of the disabled raid features dict.David Lehman2012-10-121-1/+5
| | | | | This is used to prevent combinations of features that do not reduce to a valid raid level.
* Clear errors when entering or leaving the custom spoke.David Lehman2012-10-121-0/+2
|
* Hook up signal handler for raid feature checkbuttons.David Lehman2012-10-121-0/+9
|
* Raise MDRaidError instead of ValueError from devicelibs.mdraid.David Lehman2012-10-121-7/+7
| | | | | ValueError makes exception handling difficult since you can't know if the error is something expected as opposed to a programming error.
* Minimum we have to do with HW clockVratislav Podzimek2012-10-123-3/+34
|
* Check X layouts specified in kickstart for validityVratislav Podzimek2012-10-122-0/+13
| | | | | | If we are given invalid X layouts (not supported), we have to remove them from the list. If none are left in the list, we should act like if there was none specified at all and add our defaults.
* Work with VConsole keymap and X layouts separatelyVratislav Podzimek2012-10-126-62/+112
| | | | | | | | | | | | | | | | | We need to set VConsole keymap and X layouts separately. We may get them all from kickstart, but if not, we have to convert the given one to the missing one and save the configuration. If X11 configuration is written before the server is started, it is loaded. This way we can get rid of using XklWrapper in main anaconda script. Also we need to write one more configuration file. And it's better to try to write all configuration files ignoring errors and in case some error(s) appeared, raise exception at the end of the procedure. Resolves: rhbz#853877 Resolves: rhbz#856362 Resolves: rhbz#859867
* Add class wrapping systemd-localed functionalityVratislav Podzimek2012-10-121-0/+112
|
* Don't write XkbVariants if none are specifiedVratislav Podzimek2012-10-121-2/+4
|
* Add comment to the begining of generated xorg.conf fileVratislav Podzimek2012-10-121-3/+5
| | | | | We should indicate where does this file come from. Moreover, without this comment vim doesn't turn on the syntax highlighting automatically.
* Don't display "None" for NIC vendors and products NM can't identify (#859540)Radek Vykydal2012-10-121-2/+2
|
* New version.Brian C. Lane2012-10-113-42/+87
|
* Don't try to load ifcfg files for wifi devices (#865355)Vratislav Podzimek2012-10-111-1/+2
| | | | | | | Wifi connections have separate ifcfg files, not specified by the device name. Signed-off-by: Vratislav Podzimek <vpodzime@redhat.com>
generated by cgit (git 2.34.1) at 2024-05-01 16:56:34 +0000