summaryrefslogtreecommitdiffstats
path: root/pyanaconda
Commit message (Collapse)AuthorAgeFilesLines
...
* Fix a couple pylint errors (#867125).Chris Lumens2012-12-141-21/+8
|
* Fix an undefined variable error (#867129).Chris Lumens2012-12-141-1/+1
|
* The fs type combo should be sensitive when reformat is checked (#887201).Chris Lumens2012-12-141-4/+2
| | | | | | Reformat being sensitive shouldn't have anything to do with it. If we are forcing the user the reformat a filesystem, they should be able to pick what it's being reformatted with.
* Remove idiomatic, hard to translate text (#865598).Chris Lumens2012-12-141-2/+1
|
* Activate default layout when it is changed (#882440)Vratislav Podzimek2012-12-142-2/+23
| | | | | | Users expect to have the layout they marked as the default one (by moving it to the top of the list) activated for the installation. This patch adds such activation.
* Validate and correct vg names as needed.David Lehman2012-12-132-1/+17
| | | | Related: rhbz#747278
* Don't allow resize of devices with no/unrecognized formatting. (#869841)David Lehman2012-12-131-2/+1
|
* Add keyboard dracut setup args (#875567)Vratislav Podzimek2012-12-134-16/+46
| | | | | Without vconsole.keymap on kernel command line, dracut/plymouth uses 'us' as keymap in LUKS password prompt.
* recheck software when source changes (#875599)Brian C. Lane2012-12-122-5/+23
| | | | | | Refresh the environments and groups when the source changes, and allow the software spoke's apply() method to run when the source is different than the last time the spoke was visited.
* Include the new lib directory in the package (#886319, #886470).Chris Lumens2012-12-122-2/+24
|
* Add a gradient background to spoke headers (mizmo, clumens).Chris Lumens2012-12-1211-77/+129
| | | | | | | | | | | The idea here is that by adding a gradient up top, the "Done" button looks more like it belongs with the header than with the widgets down in the spoke. It also subtly makes the spoke look like a subpart of the hub. Adding this required adding an additional internal widget, which means also changing all existing glade files to include this internal widget. Make sure to build new glade support and install locally (or use the proper makefile targets) before editing.
* Only allow changing filesystem type if the reformat combo is checked (#885906).Chris Lumens2012-12-121-0/+1
| | | | | We were doing fine at this except for one case, where the contents of the customize combo are displayed for the first time.
* It's possible for mountpoint to be None (#885279).Chris Lumens2012-12-121-1/+1
|
* Explicitly set True/False in the bootloader setting (#885381).Chris Lumens2012-12-121-2/+2
|
* Make sure software selection is checked against filesystem space. (#853636)David Lehman2012-12-116-7/+120
|
* Update default install size and disk space estimate.David Lehman2012-12-111-2/+3
|
* Add checkbox for setting language default layout (#866887)Vratislav Podzimek2012-12-112-13/+45
| | | | | | | | Some users expect that when they select language Anaconda automatically sets their keyboard layout to the one that is default for the selected language. But this would not be a good implicit behaviour because we would be changing layout in background without letting user now. Adding checkbox for the behaviour some users want seems to be a best solution.
* Change testing area label to something more appropriate (KeyboardSpoke)Vratislav Podzimek2012-12-111-1/+1
| | | | | | | Since we are not switching layouts on selection changes we shouldn't state that in the label above the testing area. Resolves: rhbz#854557
* Fix getting country layout variantsVratislav Podzimek2012-12-111-5/+7
| | | | | | | | | Commit 70e339b86b0d4a000e0e62253703ac4ca8618b74 introduced the bug resulting in almost empty dictionary of country layout variants. Iteration over countries uses the same _get_variant method as iteration over languages, but comes afterwards. The check added in 70e339b86b0d4a000e0e62253703ac4ca8618b74 fails for almost every layout that is processed in iteration over countries.
* Wait for slower dhcp before running vnc server (#868777)Radek Vykydal2012-12-111-0/+1
| | | | Regards NM's default autoconnections.
* Network spoke: fix NMClient signal callback arguments (#885488)Radek Vykydal2012-12-101-2/+2
|
* Add logging for networking and improve logging of ifcfg filesRadek Vykydal2012-12-102-16/+37
|
* Honor user request via UI to not install a bootloader. (#885240)David Lehman2012-12-072-3/+1
|
* Handle partition removals regardless of deepcopy. (#884896)David Lehman2012-12-071-1/+2
| | | | | | | | | | | | | A StorageDevice containing a DiskLabel will have both an originalFormat and a format attribute, both of which are DiskLabel instances. They are copied using a shallow copy, so they share _partedDisk and _origPartedDisk attributes. That means you only have to do partition removals from the parted Disk once. When you make a deep copy of such a StorageDevice, things change. The originalFormat and format no longer share a single _partedDisk or _origPartedDisk. In this case, it is sometimes necessary to remove a partition from both self.disk.format and self.disk.originalFormat.
* Default to partitions for /boot* instead of just /boot/efi. (#884606)David Lehman2012-12-071-5/+3
| | | | | | | | There is some argument that grub2 does not do well with non-partition /boot (https://bugzilla.redhat.com/show_bug.cgi?id=737508#c75). Also, this change brings the defaults in line with our own automatic partitioning.
* Fix a logic error in ActionDestroyFormat.obsoletes. (#885004)David Lehman2012-12-071-1/+1
|
* Take device type into account when making the config button sensitive (#885051).Chris Lumens2012-12-071-1/+2
| | | | We were doing this elsewhere, but not when a selector was clicked on.
* Hide VG-related widgets when displaying a non-LV mountpoint first (#885131).Chris Lumens2012-12-071-6/+6
| | | | | | Remember that after the refresh method, a show_all() will be called. Thus it's not good enough to call hide. You must also call set_no_show_all(True) on the widgets you want to be hidden.
* Install default system for %packages --default (#869978)Brian C. Lane2012-12-071-0/+3
|
* Fix a typo in the live cd completion text (#884373).Chris Lumens2012-12-071-1/+1
|
* Do not allow deleting or editing a protected device in custom part (#884599).Chris Lumens2012-12-071-12/+15
|
* If path doesn't exist, don't traceback. Return None.Chris Lumens2012-12-071-1/+5
|
* Add/remove the HDISO source from protectedDevSpecs (#882147).Chris Lumens2012-12-071-0/+4
| | | | | | | | | Should storage ever get reset (which it can in the case of moving back to the hub from reclaim, resulting in an error), any protected attributes will be wiped. protectedDevSpecs is the definitive source for what's protected and what's not. The dev.protected lines are still there in case we do not hit an error.
* Put the bad VG name into the error message (#884359).Chris Lumens2012-12-071-1/+1
|
* Use updated connection settings object for default auto config files (#883383)Radek Vykydal2012-12-071-1/+1
|
* Call udev_settle from inside udev_trigger.David Lehman2012-12-051-0/+1
| | | | You always want to let things settle, so just take care of it.
* Prevent enabling the encryption checkbutton erroneously.David Lehman2012-12-051-0/+11
| | | | | The same complicated rules apply when reformat is toggled as when populating the right side.
* Make sure Storage is initialized before refreshing the custom spoke.David Lehman2012-12-051-3/+4
| | | | Related: rhbz#884270
* Fix initialization of Storage.roots to use a list. (#884270)David Lehman2012-12-051-1/+1
|
* Don't allow reformat without setting a mountpoint. (#883076)David Lehman2012-12-051-2/+7
|
* Fix check for toggled encryption checkbutton. (#882722)David Lehman2012-12-051-1/+1
|
* Make sure FS minSize is never greater than its currentSize. (#876547)David Lehman2012-12-051-4/+10
|
* When considering whether anything can be shrunk, throw out protected devs.Chris Lumens2012-12-051-1/+2
|
* In the UI, mark the HDISO source device as protected (#879610).Chris Lumens2012-12-052-0/+13
| | | | | Then later when we are setting up filesystems, do not tear down protected devices.
* update mdraid superBlock space calculation (#883483)Brian C. Lane2012-12-051-1/+56
| | | | | | | | | | | | mdadm 3.2.4 made a major change in the amount of space used for 1.1 and 1.2 metadata in order to reserve space for reshaping. See commit 508a7f16 in the upstream mdadm repository. This has resulted in a large difference between what anaconda thinks the size of the array should be and what it actually is. We now use 2.0MB for metadata that isn't 1.1 or 1.2 For 1.1 and 1.2 we calculate the reserved space value and use that.
* Remove resetResolver function, we don't need it anymore (#868695)Radek Vykydal2012-12-053-20/+0
| | | | | Also removes urlgrabber.grabber.reset_curl_obj() from getDefaultHostname() where all its job were to cause races occasionally.
* Network spoke: improve logging.Radek Vykydal2012-12-051-0/+7
|
* Unify writeNetworkConf with other modules (rename, put in ks.execute)Radek Vykydal2012-12-054-28/+23
|
* Fix network command --onboot and --activate options.Radek Vykydal2012-12-052-129/+73
| | | | | | In dracut we set ONBOOT=yes for --activate so that NM activates the device upon its start, in anaconda we set the real value of ONBOOT.
* Fix two calls of self.window.set_info (#883632)Vratislav Podzimek2012-12-051-4/+2
| | | | It was replaced by self.set_info, self.set_error and self.set_warning.
generated by cgit (git 2.34.1) at 2025-09-03 14:36:38 +0000