- Find or set the active system partition only once, when we start the
formatting stage. If the system partition is to be created in some
non-partitioned space, we create it.
- A file-system check is always scheduled for both the system and the
installation partitions.
- If the system partition was already formatted (which usually means
that it already existed on the system), don't ask for formatting it.
CORE-13205
- For each partition to be formatted, the file-system menu list is
re-created, in order to display only the FSes that are compatible
with the selected partition, and allow to give the user the choice
to keep the existing file-system only if the selected partition was
already formatted.
- Check whether the user attempts to delete the partition on which
the installation source is present, only if it is not "new" and is
formatted.
- Check first whether the current partition is the system partition,
before displaying the CONFIRM_DELETE_SYSTEM_PARTITION_PAGE.
- Make some partitioning functions not dependent on the selected "CurrentPartition".
- Add some sanity checks.
- Improve some of the "is-partitioned" checks.
Similarly to FMIFS this structure should be private. Instead file-system
names are passed to the helper functions, allowing to use the names
returned by the FS drivers. The names are then internally mapped to the
corresponding FS providers.
In particular this allows to handle the "RAW" file-system and to assign
the 'Unformatted' flag to partitions having this FS.
Finally this helps us refining the checks performed to see whether the
current "active" system partition uses a supported file-system.
- Within the function's body code, check the status values returned by the called functions.
- Change the BuildInstallPaths's function type to NTSTATUS instead of VOID (and check the status of InitDestinationPaths() as well.
By default, we still fallback to FAT if nothing asked, or if there
is an invalid input.
0 is FAT, 1 is BtrFS. This can be grown as soon as we add more IFS.
- Add checks for GPT disks and either fail or ignore the disk,
depending on the operation being executed.
[USETUP][REACTOS] Display the disk style more accurately.
Always perform the partition validity checks when creating new
primary/extended/logical partitions, and also when a new partition
is automatically created when unpartitioned space is selected for
ReactOS installation.
CORE-12246
Simplified implementation of the proposed solution by George Bișoc
and revised by Victor Perevertkin from PR #952.
NOTE: For whatever reason ERROR_INSTALL_BOOTCODE and ERROR_WRITE_BOOT
seem to be redundant with each other. To be investigated...
Most of Linux filesystems have the MBR ID assigned as 0x83 and when the user chooses BTRFS upon partitions page wizard, we're thinking of the partition as having Ext2 filesystem which is wrong and misleading.
- SETUPLIB and USETUP: Convert PARTENTRY::DriveLetter to WCHAR.
- SETUPLIB: Retrieve volume label.
- SETUPLIB and USETUP: Move the partition types (IDs) table from
USETUP to SETUPLIB so that they can share it with the 1st-stage
GUI installer too. Also take the opportunity to update the table
with information from http://www.win.tue.nl/~aeb/partitions/partition_types-1.html
and cross-checked with the list from Paragon Hard-Disk Manager.
- USETUP: Simplify PrintPartitionData().
- Add some code comments.
- Use adequate access flag when opening symbolic links.
- Simplify the prototype of UpdateRegistry() since now both Setup INF
handle and settings lists are inside the USETUP_DATA structure.
This means, call in the correct order the user callback with the correct
parameters (in particular the correct paths for file copy operations),
and check also for the callback returned value to know whether or not to
continue the file operations.
This allows using some of the SetupApi.dll functions when SETUPLIB is
used in the (Win32) GUI 1st-stage installer "REACTOS", while using the
custom implemented NT-aware functions in "USETUP".
- Add support for delete and move/rename operations, which are needed
for implementing ReactOS upgrading support.
- Use cabinet contexts.
- Use standard LIST_ENTRY structures for implementing the lists.
- Move the path-building hack code in SetupCommitFileQueueW() that had
been introduced in r66604 (97bb83f) out of the file-queues code.
- Make the function prototypes compatible with win32's setupapi functions.
- Fix the format of the data passed to the custom notification handler.
- Adjust the file-copy callback to correctly use its arguments (setupapi-compatible).
- Move a great deal of global variables into the USETUP_DATA structure
(the SetupInf, the SetupFileQueue, the generic lists...).
- Place the common setup initialization code into an InitializeSetup()
routine, and the cleanup code into FinishSetup().
- Implement the setup-code part support for the TXTSETUP.SIF setup
source path override variables "SetupSourceDevice" and "SetupSourcePath"
(see CORE-9023); support for them in SETUPLDR will be added later.
- It can therefore be also used by the 1st-stage GUI setup.
- Rename some function parameters to clarify what they should be.
Based on:
svn path=/branches/setup_improvements/; revision=75750
[USETUP] Improve usage of ERROR_TXTSETUP_SECTION and ERROR_CABINET_SECTION error codes, correctly display associated error message, and remove a deprecated string.
- Apart from allowing a UI cache variable that may be used when
displaying GENERIC_LIST_ENTRY-ies, do not store any display strings
associated to these list entries. They should be instead computed only
when initializing a list UI (or a combo-box or list control if the
code is used in Win32 environment).
For this matter a callback is provided to InitGenericListUi() that
does the job of computing the displayed string corresponding to a
given GENERIC_LIST_ENTRY.
- Simplify the calls to InitGenericListUi(), and refactor the
RestoreGenericListUiState() function.
- Use for-loops for iterating over GENERIC_LIST items.
- Adapt the storage data format for lists of settings items.
- The txtsetup.sif INF format specified in LoadSetupInf() should not be
INF_STYLE_WIN4 (to be investigated...).
- Use OBJ_CASE_INSENSITIVE when initializing object attributes
(no actual reason why to keep case sensitivity there).
- Check the success of a RtlStringCchPrintfW call in EnumerateReactOSEntries().
- Explicitly check for returned STATUS_NOT_SUPPORTED from ChkdskPartition()
or FormatPartition(), and display an appropriate error message.
- Remove some left-over comments but also explain why I kept some
commented code (mainly for future reference).
Only convert to ANSI when needed (e.g. in the display code for usetup).
The 1st-stage GUI setup will however use the UNICODE strings directly.
svn path=/branches/setup_improvements/; revision=75753
[SETUPLIB] Copy the mui* files from usetup into the setuplib for later usage, and add mui.c to build.
svn path=/branches/setup_improvements/; revision=75711
[SETUPLIB][USETUP] Start massaging the MUI files and split up what concerns only the usetup "resources",
and what concerns general language parameters (in the setuplib).
It may be interesting to retrieve the language parameters
from INF file (e.g. intl.inf) (suggested by Giannis).
svn path=/branches/setup_improvements/; revision=75715
[SETUPLIB] Add the necessary headers to mui.c to be able to compile.
[USETUP] Comment out the languages that don't have any resources.
svn path=/branches/setup_improvements/; revision=75716
[SETUPLIB][USETUP] Adjust MUI and settings code.
- Re-enable settings code concerning language, keyboards & layout
in the setuplib, and remove the old code in usetup.
- Remove useless code in setuplib's mui.c.
- Rename usetup's MUI "LanguageList" into "ResourceList" so as to
avoid colliding with the "LanguageList" of setuplib.
- Add the magic CMakeLists line "add_definitions(${I18N_DEFS})" that
I forgot previously, that caused the "LanguageList" of setuplib to be empty!
The code compiles and works during my tests.
svn path=/branches/setup_improvements/; revision=75717
[SETUPLIB] Change some string pointer types & function prototypes so that the usetup-specific
"SelectedLanguageId" variable is not used in the library.
svn path=/branches/setup_improvements/; revision=75719
[SETUPLIB] Change some function prototypes so that the usetup-specific "SelectedLanguageId" variable is not used in the library.
Also, make AddEntriesFromInfSection(), DefaultProcessEntry() and the typedef PPROCESS_ENTRY_ROUTINE private again.
svn path=/branches/setup_improvements/; revision=75720
svn path=/branches/setup_improvements/; revision=75724
[USETUP] Code adaptations.
- Transform some do{}while() loops into while(){} ones, since the lists on which we work may only contain one single {NULL} element.
- Modify MUIGetOEMCodePage() call after r75719.
- Use PCWSTR where needed.
svn path=/branches/setup_improvements/; revision=75722
[USETUP] Adjust some MUI & settings calls, following the commits r75719, r75720 and r75721.
svn path=/branches/setup_improvements/; revision=75723
[INPUT.CPL] Adjust inclusion of "muifonts.h" (commit 3dbd44f) due to changes introduced in r75711 and r75715.
Move the bootsup.c code into the setuplib, as the code is generic
enough to be used by both USETUP and the 1st-stage GUI installer.
svn path=/branches/setup_improvements/; revision=75674
- Move several global setup variables into a structure "USETUP_DATA",
similar to the syssetup structure "SETUPDATA" (or the WIP 1st-stage
installer structure of the same name), so that these variables can be
set easily by different helper setup functions;
- Move CheckUnattendedSetup() and GetSourcePaths() to setuplib and make
CheckUnattendedSetup() use the USETUP_DATA structure;
- Add a LoadSetupInf() function that loads the txtsetup.sif file
(factoring out the corresponding code in USETUP);
- Add a InstallSetupInfFile() function (that I'll probably rename later on)
whose purpose is to create a valid "$winnt$.inf" setup installation file
in the ReactOS\system32 directory, which should help the 2nd-stage installer
to correctly retrieve the source installation media we used during 1st-stage,
and contain the unattended setup lines copied from unattend.inf. This is
done in a Windows-compatible way.
svn path=/branches/setup_improvements/; revision=75518
[USETUP] Close the txtsetup.sif file at the end of the operations.
svn path=/branches/setup_improvements/; revision=75539
They currently do nothing, since the getter functions don't actually
capture (copy) the strings but merely return pointers to read-only strings.
But the calls are placed here for consistency, because if one day the
getters' implementation is changed so that strings are captured, it would
then be needed to free the allocated buffers.
In addition, fix a buggy call to INF_GetData() -- should be instead
INF_GetDataField() -- in AddSectionToCopyQueue().
svn path=/branches/setup_improvements/; revision=75516
- As GetSourcePaths() is used once in USETUP to initialize global
UNICODE_STRING path strings once, move it out of drivesup.c and
put it in usetup.c. Then remove drivesup.c : 1 file less!
- Move some INF file prototype declarations out of usetup.h and
inside inffile.h where they should better be, as inffile.h and .c
is the glue code for the INF library, defining similar functions
as the ones in setupapi.dll.
- I rename our local SetupOpenInfFileW() into SetupOpenInfFileExW()
because the latter one takes an extra user-provided LCID parameter,
and this is this one that we use in USETUP.
- Make 'UNICODE_STRING SourcePath;' visible only inside usetup.c
(not used elsewhere).
- Implement installation path validity check in case we are either
in repair/update, or unattended setup mode. If the path is detected
as invalid, then we fall back into manual path specification
(for now...; note that we could instead fail the installation too).
svn path=/branches/setup_improvements/; revision=75246
- Modify a bit the page flow so that the upgrade page is inserted before
the Device-settings page, and after the Install-Intro page.
- Insert some extra 'RepairUpdateFlag' checks in SelectPartitionPage() and
InstallDirectoryPage() to take specific actions in case the flag is TRUE.
svn path=/branches/setup_improvements/; revision=74576
svn path=/branches/setup_improvements/; revision=75525
Implement work-in-progress code that, when upgrading an existing
ReactOS installation, verifies whether the (existing) system registry
hives are valid (by loading & unloading them, this allows using the
built-in repair functionality if needed), or not.
If a given hive cannot be repaired successfully, it is backed up
(with a '.brk' extension, "brk" as "broken"), then is marked up for
recreation.
When all hives have been checked, if there are any hive that needs
to be recreated, we recreate its hive file, mount it, but we do *NOT*
mount the other valid existing hives for update. We create instead
dummy registry hives so that we can actually use, as the update code,
the same one as the one used when fully creating the registry hives
for a clean installation (of course, this choice can be improved later on).
The update code (i.e. the same as the registry clean-install one) then
adds the registry keys & values, either putting them in the dummy
registry hives (the ones that we don't want to recreate) or in the
registry hive that is recreated.
At the end, the (re)created registry hives are flushed back to disk,
and a copy of them (under a '.sav' extension) are created, so that they
can be used for restoration purposes if 2nd-stage (and up) goes berserk.
Extra fixes:
- Use the correct structure member field when initializing the
'InstallDir' variable, when performing an upgrade.
- CreateNestedKey() should be better analysed to see whether it correctly
creates the full registry path compatible with volatile/non-volatile keys
(under inspection).
svn path=/branches/setup_improvements/; revision=75010
The rationale is as follows.
We now have the 1st-stage ReactOS setup running with its own registry SYSTEM hive,
similarly to regular ROS running instances (livecd, regular installation...).
The ReactOS-specific SetInstallPathValue() hack, introduced in a76689e9 (r3794)
and 5f973ce0 (r3795), is removed. This hack told the kernel that, during the setup,
it had to "switch" the used registry hives and instead use the ones of the ROS
installation being prepared.
This was really hackish because this means, mixing between registry settings used only
for the setup running instance, that could use different registry settings
than the ones that should be set for the ROS installation being actually performed.
Also, note that in the case of a 1st-stage GUI setup, consisting in running the
LiveCD + the GUI setup program, this situation would be untenable. Note also that
for people willing to use the Setup*** functions exported by setupapi.dll to parse
the registry INF files to initialize the registry of the ROS installation being
prepared, this would be impossible either.
Hence the need to have offline registry modification functionality.
svn path=/branches/setup_improvements/; revision=74766