Blockers...

[Alex Ionescu]

Hi,

Here is a current update on the 0.2.7 blockers...if you can help, please DO!

BUG:
- Green icon in run dialog

REGRESSION IDENTIFIED:
Yes. By WaxDragon. Regression due to BUILD.

CAUSE:
Incorrect icon was added to resource file.

FIXED:
Yes. Fix needs to be merged into 0.2.7 branch.

-------------------

BUG:
- Common Dialog Control Broken for File/Save.

REGRESSION IDENTIFIED:
Yes. By WaxDragon. Regression due to 15669.

CAUSE:
Building shell32 as ANSI causes this problem. Building it as UNICODE 
however breaks the My computer dialog. Shell32 should be built as 
Unicode, and the My computer dialog should appear when double-clicked. 
So this bug needs to be fixed.

----------------------

BUG:
- If you use the arrow keys in the explorer start menu to expand an 
entry, you cannot use them anymore to go up and down.

No further information has been researched yet. Martin, explorer is your 
baby, and I'm sure is an easy fix...it doesnt' happen with the 
right-click context menu on the desktop, so maybe this is localized to 
some bug in your start menu code?

---------------------

BUG:
- If you right-click on the desktop to show the context menu, then 
right-click again, then left-click (or other similar combinations), your 
next right-click is not registered anymore.

No further information is available, although I belive hpoussin said the 
problem is above win32k. I haven't heard from tinus in a long time, but 
i8042prt might be where this problem resides.

-------------------

BUG:
- Several applications have regressed due to "memory corruption" and a 
bugcheck related to NtGdiRealizePallette.

More information is on the Bugzilla DB, but it's down atm. I have heard 
reports that the error always happened before, but wasn't seen because 
red zone pool protection was disabled. In any case, it breaks a large 
number of apps.

--------------------

BUG:
 - AllocConsole failure in 1st-stage setup on Dell Laptops (and maybe 
others).

REGRESSION IDENTIFIED:
Yes. By BrandonTurner. A regression range has been identified and 3 key 
builds are being tested to find the actual regression.

Best Regards,
Alex Ionescu

[Robert Köpferl]

Thanks, I just wanted to look for he status of bhe blockers.

Hmmm, I'll wait anoher days for RC2...
There was again just one file changed. Neither did I somehong...so :-/

Alex Ionescu wrote:
> Hi,
>
> Here is a current update on the 0.2.7 blockers...if you can help, please 
> DO!
>
> BUG:
> Green icon in run dialog
>
> REGRESSION IDENTIFIED:
> Yes. By WaxDragon. Regression due to BUILD.
>
> CAUSE:
> Incorrect icon was added to resource file.
>
> FIXED:
> Yes. Fix needs to be merged into 0.2.7 branch.
>
> BUG:
> Common Dialog Control Broken for File/Save.
>
> REGRESSION IDENTIFIED:
> Yes. By WaxDragon. Regression due to 15669.
>
> CAUSE:
> Building shell32 as ANSI causes this problem. Building it as UNICODE 
> however breaks the My computer dialog. Shell32 should be built as 
> Unicode, and the My computer dialog should appear when double-clicked. 
> So this bug needs to be fixed.
>
> BUG:
> If you use the arrow keys in the explorer start menu to expand an
>
> entry, you cannot use them anymore to go up and down.
>
> No further information has been researched yet. Martin, explorer is your 
> baby, and I'm sure is an easy fix...it doesnt' happen with the 
> right-click context menu on the desktop, so maybe this is localized to 
> some bug in your start menu code?
>
> BUG:
> If you right-click on the desktop to show the context menu, then
>
> right-click again, then left-click (or other similar combinations), your 
> next right-click is not registered anymore.
>
> No further information is available, although I belive hpoussin said the 
> problem is above win32k. I haven't heard from tinus in a long time, but 
> i8042prt might be where this problem resides.
>
> BUG:
> Several applications have regressed due to "memory corruption" and a
>
> bugcheck related to NtGdiRealizePallette.
>
> More information is on the Bugzilla DB, but it's down atm. I have heard 
> reports that the error always happened before, but wasn't seen because 
> red zone pool protection was disabled. In any case, it breaks a large 
> number of apps.
>
> BUG:
> AllocConsole failure in 1st-stage setup on Dell Laptops (and maybe
>
> others).
>
> REGRESSION IDENTIFIED:
> Yes. By BrandonTurner. A regression range has been identified and 3 key 
> builds are being tested to find the actual regression.
>
> Best Regards,
> Alex Ionescu
> _______________________________________________
> Ros-dev mailing list
> [e-mail]
> http://reactos.com:8080/mailman/listinfo/ros-dev

[Hervé Poussineau]

BUG:
- If you right-click on the desktop to show the context menu, then 
right-click again, then left-click (or other similar combinations), your 
next right-click is not registered anymore.

More information:
MOUSE_RIGHT_BUTTON_DOWN/MOUSE_RIGHT_BUTTON_UP events are correctly sent 
from i8042prt driver to mouclass driver, which sends them correctly to 
win32k. So the problem is in win32k or above

Here is a patch to see mouse buttons events received by win32k:
Index: subsys/win32k/ntuser/input.c
===================================================================
--- subsys/win32k/ntuser/input.c	(revision 16704)
+++ subsys/win32k/ntuser/input.c	(working copy)
@@ -77,6 +77,19 @@
    for(i = 0; i < InputCount; i++)
    {
      mid = (Data + i);
+    if (mid->ButtonFlags & (MOUSE_LEFT_BUTTON_DOWN | 
MOUSE_LEFT_BUTTON_UP | MOUSE_RIGHT_BUTTON_DOWN | MOUSE_RIGHT_BUTTON_UP))
+    {
+      DPRINT1("{ ");
+      if (mid->ButtonFlags & MOUSE_LEFT_BUTTON_DOWN)
+        DbgPrint("MOUSE_LEFT_BUTTON_DOWN ");
+      if (mid->ButtonFlags & MOUSE_LEFT_BUTTON_UP)
+        DbgPrint("MOUSE_LEFT_BUTTON_UP ");
+      if (mid->ButtonFlags & MOUSE_RIGHT_BUTTON_DOWN)
+        DbgPrint("MOUSE_RIGHT_BUTTON_DOWN ");
+      if (mid->ButtonFlags & MOUSE_RIGHT_BUTTON_UP)
+        DbgPrint("MOUSE_RIGHT_BUTTON_UP ");
+      DbgPrint("}\n");
+    }
      mi.dx += mid->LastX;
      mi.dy += mid->LastY;

Hervé

[Martin Fuchs]

2005/7/23, Alex Ionescu [e-mail]:
> BUG:
> If you use the arrow keys in the explorer start menu to expand an
>
> entry, you cannot use them anymore to go up and down.
>
> No further information has been researched yet. Martin, explorer is your
> baby, and I'm sure is an easy fix...it doesnt' happen with the
> right-click context menu on the desktop, so maybe this is localized to
> some bug in your start menu code?

Sorry, I can't reproduce this problem on (MS) Windows. As I currently
don't have a working ROS installation - and don't know how to debug
such in reasonable time, I don't think I can help you here very much.
However I can say, the Explorer start menu and the context menus are
not related in any way. I guess this might be a focus handling problem
somewhere in the USER32 code handling window creation/keyboard message
processing.

Regards,

Martin

[Steven Edwards]

Hi Martin,

--- Martin Fuchs [e-mail] wrote:
> Sorry, I can't reproduce this problem on (MS) Windows. As I currently
> don't have a working ROS installation - and don't know how to debug
> such in reasonable time, I don't think I can help you here very much.
> However I can say, the Explorer start menu and the context menus are
> not related in any way. I guess this might be a focus handling problem
> somewhere in the USER32 code handling window creation/keyboard message
> processing.

Right. We should have thought of this. In the future we can save time tracking down bugs in
explorer and isolating them by testing on Windows and comparing the results to ros.

Thanks
Steven

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com

[Alex Ionescu]

Martin Fuchs wrote:
> Sorry, I can't reproduce this problem on (MS) Windows. As I currently
> don't have a working ROS installation - and don't know how to debug
> such in reasonable time, I don't think I can help you here very much.
> However I can say, the Explorer start menu and the context menus are
> not related in any way. I guess this might be a focus handling problem
> somewhere in the USER32 code handling window creation/keyboard message
> processing.
>
> Regards,
>
> Martin

You're right, it works fine on Windows. But I'm not going to let you go 
so easily ;). Here's something which is in explorer, has been there 
since 0.2.0 and really bugs me, it's on Windows too... you create the 
buttons on the task bar too low... they aren't centered, so they are 
much lower then they should be...and it looks really bad, especially on 
ReactOS (but on Windows too).

Can you fix that please ? :)

Best regards,
Alex Ionescu

[Gedi]

Alex Ionescu wrote:
> Martin Fuchs wrote:
> > Sorry, I can't reproduce this problem on (MS) Windows. As I currently
> > don't have a working ROS installation - and don't know how to debug
> > such in reasonable time, I don't think I can help you here very much.
> > However I can say, the Explorer start menu and the context menus are
> > not related in any way. I guess this might be a focus handling problem
> > somewhere in the USER32 code handling window creation/keyboard message
> > processing.
> >
> > Regards,
> >
> > Martin
>
> You're right, it works fine on Windows. But I'm not going to let you 
> go so easily ;). Here's something which is in explorer, has been there 
> since 0.2.0 and really bugs me, it's on Windows too... you create the 
> buttons on the task bar too low... they aren't centered, so they are 
> much lower then they should be...and it looks really bad, especially 
> on ReactOS (but on Windows too).
>
> Can you fix that please ? :)
>
> Best regards,
> Alex Ionescu

I'm going to piggyback this with another request  :)

When browsing through the start menu and pointing at an item which has 
further sub-items, it is expanded straight away. This makes the start 
menu expand everything as the mouse quickly moves across it.
It would be nicer to have it as it is in Windows or KDE, etc, whereby 
the pointer needs to hover over a particular item for around 0.5-1.0 
secs before it expands.

Hope that makes sense.

Ged.

[Royce Mitchell III]

Gedi wrote:
> Alex Ionescu wrote:
> > Martin Fuchs wrote:
> > > Sorry, I can't reproduce this problem on (MS) Windows. As I currently
> > > don't have a working ROS installation - and don't know how to debug
> > > such in reasonable time, I don't think I can help you here very much.
> > > However I can say, the Explorer start menu and the context menus are
> > > not related in any way. I guess this might be a focus handling problem
> > > somewhere in the USER32 code handling window creation/keyboard message
> > > processing.
> > >
> > > Regards,
> > >
> > > Martin
> >
> > You're right, it works fine on Windows. But I'm not going to let you 
> > go so easily ;). Here's something which is in explorer, has been 
> > there since 0.2.0 and really bugs me, it's on Windows too... you 
> > create the buttons on the task bar too low... they aren't centered, 
> > so they are much lower then they should be...and it looks really bad, 
> > especially on ReactOS (but on Windows too).
> >
> > Can you fix that please ? :)
> >
> > Best regards,
> > Alex Ionescu
>
> I'm going to piggyback this with another request  :)
>
> When browsing through the start menu and pointing at an item which has 
> further sub-items, it is expanded straight away. This makes the start 
> menu expand everything as the mouse quickly moves across it.
> It would be nicer to have it as it is in Windows or KDE, etc, whereby 
> the pointer needs to hover over a particular item for around 0.5-1.0 
> secs before it expands.
>
> Hope that makes sense.

I prefer the instant reaction. If we're going to change it, can we make 
it configurable, please?

[Thomas Weidenmueller]

Royce Mitchell III wrote:
> I prefer the instant reaction. If we're going to change it, can we make
> it configurable, please?

The menu behavior (delay of popup menus) can be configured in windows
and even disabled, so explorer should use these settings imo. Use
SystemParametersInfo with SPI_GETMENUSHOWDELAY to determine the
behavior. There are more menu settings you can and should query using
the same method.

Best Regards,
Thomas

[Rick Langschultz]

I though ReactOS was supposed to be a Windows-like environment; having said
that I think that there should be an option obviously a home user would like
those flashy things included, whether they choose to turn them on or off is
their choice. Now a developer's machine is about the programming aspects,
and cpu cycles would be better spent on code, services, and performance. The
way I see it is Windows XP implemented a great feature in letting the user
choose using the Advanced Performance Options. Also, let me say, that if a
developer chooses to use a lot of services, such as high-end databases used
on an e-commerce site the process host should be allocated more memory than
say the GUI. Optimization of services is an important factor in choosing a
server environment.

I think BeOS also touched on a key feature in isolating each background
service. If ReactOS has a bug or a server or service crashed the system
should not BSOD instead there should be an auto-restart option that will
effect/affect only that process or service. This will allow little downtime
especially in a mission critical environment.

[Nick Journals]

Hi guys,

Whats up with all our websites, they all seem to be down? :-/

Grtz,
Nick

[Andrew Murphy]

Yeah, it was hacked, you missed it :)

GvG is on holiday, he'll be back in about a week to fix it

On 27/07/05, Nick Journals [e-mail] wrote:
> Hi guys,
>
> Whats up with all our websites, they all seem to be down? :-/
>
> Grtz,
> Nick
>
> Ros-dev mailing list
> [e-mail]
> http://reactos.com:8080/mailman/listinfo/ros-dev

[Steven Edwards]

Hi,

--- Royce Mitchell III [e-mail] wrote:
> > When browsing through the start menu and pointing at an item which has 
> > further sub-items, it is expanded straight away. This makes the start 
> > menu expand everything as the mouse quickly moves across it.
> > It would be nicer to have it as it is in Windows or KDE, etc, whereby 
> > the pointer needs to hover over a particular item for around 0.5-1.0 
> > secs before it expands.
> >
> > Hope that makes sense.
>
> I prefer the instant reaction. If we're going to change it, can we make 
> it configurable, please?

Yes I prefer instant reaction as well. Microsoft added that stupid fade effect for no other reason
than to eat CPU cycles and look flashy. If you don't want the menu to expand.....don't move the
mouse there =P.

Or at the very least make it a configurable option so I don't have to waste 0.5-1.0 secs of my
life everytime I use the menu. You know those secs from all this eye candy add up over the
years.....

Thanks
Steven

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com

[Gedi]

Steven Edwards wrote:
> Hi,
>
> --- Royce Mitchell III [e-mail] wrote:
> > > When browsing through the start menu and pointing at an item which has 
> > > further sub-items, it is expanded straight away. This makes the start 
> > > menu expand everything as the mouse quickly moves across it.
> > > It would be nicer to have it as it is in Windows or KDE, etc, whereby 
> > > the pointer needs to hover over a particular item for around 0.5-1.0 
> > > secs before it expands.
> > >
> > > Hope that makes sense.
> >
> > I prefer the instant reaction. If we're going to change it, can we make 
> > it configurable, please?
>
> Yes I prefer instant reaction as well. Microsoft added that stupid fade effect for no other reason
> than to eat CPU cycles and look flashy. If you don't want the menu to expand.....don't move the
> mouse there =P.
>
> Or at the very least make it a configurable option so I don't have to waste 0.5-1.0 secs of my
> life everytime I use the menu. You know those secs from all this eye candy add up over the
> years.....
>
> Thanks
> Steven

I'm not into all the eye candy stuff either. I'm not sure if we're 
taking about the same thing here.
What I meant was, when you move your mouse up the start menu quickly, 
everything tries to expand immedietly. This turns the menu into a mad 
flashing strobe like something out of a prodigy video.
Also, when wanting to follow an expanded item, if you drift off with the 
cursor there is no leeway and you have to start over again.  It's sorta 
like that game at the fair where you have to move the loop around 
without touching the wire, else you have to start over.

Ged.

[Richard]

FYI the Windows version of explorer uses a button bar, MF's explorer 
doesn't.  Using a button bar would fix a couple of annoyances as far as 
the taskbar goes.

Alex Ionescu wrote:
> Martin Fuchs wrote:
> > Sorry, I can't reproduce this problem on (MS) Windows. As I currently
> > don't have a working ROS installation - and don't know how to debug
> > such in reasonable time, I don't think I can help you here very much.
> > However I can say, the Explorer start menu and the context menus are
> > not related in any way. I guess this might be a focus handling problem
> > somewhere in the USER32 code handling window creation/keyboard message
> > processing.
> >
> > Regards,
> >
> > Martin
>
> You're right, it works fine on Windows. But I'm not going to let you 
> go so easily ;). Here's something which is in explorer, has been there 
> since 0.2.0 and really bugs me, it's on Windows too... you create the 
> buttons on the task bar too low... they aren't centered, so they are 
> much lower then they should be...and it looks really bad, especially 
> on ReactOS (but on Windows too).
>
> Can you fix that please ? :)
>
> Best regards,
> Alex Ionescu
>
> Ros-dev mailing list
> [e-mail]
> http://reactos.com:8080/mailman/listinfo/ros-dev

[Alex Ionescu]

Alex Ionescu wrote:
> BUG:
> Common Dialog Control Broken for File/Save.
>
> REGRESSION IDENTIFIED:
> Yes. By WaxDragon. Regression due to 15669.
>
> CAUSE:
> Building shell32 as ANSI causes this problem. Building it as UNICODE 
> however breaks the My computer dialog. Shell32 should be built as 
> Unicode, and the My computer dialog should appear when double-clicked. 
> So this bug needs to be fixed.

UPDATE:
FIXED. Resolution found by Filip Navara. Awaiting patch.

Best Regards,
Alex Ionescu

[WaxDragon]

On 7/23/05, Alex Ionescu [e-mail] wrote:
> Hi,
>
> Here is a current update on the 0.2.7 blockers...if you can help, please DO!

The keyboard input bug was a blocker too.  Luckly i had the email from bugzilla.
-----
http://reactos.com/bugzilla/show_bug.cgi?id=657

Summary: Shift sticks when selecting text with the arrow keys.
           Product: ReactOS
           Version: 0.3.0-SVN
          Platform: VMWare 4
        OS/Version: ReactOS
            Status: NEW
          Severity: major
          Priority: P3
         Component: Drivers
        AssignedTo: [e-mail]
        ReportedBy: [e-mail]

VMWare 4.5.2
ReactOS r16251 (this is not a recent bug, I think 15xxx has it)

1. Boot ROS
2. Open notepad
3. Type "This sucks"
4. Press and hold Shift and use the arrow keys to select the previous text (not
the numberic keypad*)
5. Type "This sucks" again, notice it is in UPPERCASE and backspace no
longer works

* Numlock seems to affect this.  I boot vmware with numlock off, and this bug
occurs.  If turn num lock on before typing, imput works as expected.  It's Shift
that gets stuck because capslock Shifts them back to lowercase, but the numbers
come out as ! @ # $ still.  The capslock - arrow keys interaction is also
discussed in bug 617
-----

There were some other  comments on this bug, basically keycode dumps
showing how the arrow keys are emulated my qemu and vmware.

Has anyone seen tinus?

WD
-- 
#irc.freenode.net #reactos
01:03PM <filip2307> i don't know about any bug
01:04PM <filip2307> none exist
01:04PM <filip2307> ReactOS is prefect

[Steven Edwards]

Hi,

--- WaxDragon [e-mail] wrote:
> The keyboard input bug was a blocker too.  Luckly i had the email from bugzilla.

This is a good time to mention this. I would like to propose that we have a regression or (QA)
coordinator position in the Constitution. This position is designed with WaxDragon in mind. For
those of you that don't know he hangs out on IRC all the time and has been doing a incredible job
tracking down regressions and which revisions caused them using ISOs from each build.

The position would entail certain rights and responsibilities:

Finial Authority on blocker bugs for a release (Could be overridden by a vote)
Maintains the documentation on testing, hunting regressions, etc
Would be the person in charge of directing any testing team that develops.
Maintain bugzilla
Anything else that we would need a QA person for.

Thoughts?

Thanks
Steven

____________________________________________________
Start your day with Yahoo! - make it your home page 
http://www.yahoo.com/r/hs

[James Tabor]

Steven Edwards wrote:
> Hi,
>
> --- WaxDragon [e-mail] wrote:
> > The keyboard input bug was a blocker too.  Luckly i had the email from bugzilla.
>
> This is a good time to mention this. I would like to propose that we have a regression or (QA)
> coordinator position in the Constitution. This position is designed with WaxDragon in mind. For
> those of you that don't know he hangs out on IRC all the time and has been doing a incredible job
> tracking down regressions and which revisions caused them using ISOs from each build.
>
> The position would entail certain rights and responsibilities:
>
> Finial Authority on blocker bugs for a release (Could be overridden by a vote)
> Maintains the documentation on testing, hunting regressions, etc
> Would be the person in charge of directing any testing team that develops.
> Maintain bugzilla
> Anything else that we would need a QA person for.
>
> Thoughts?
>
> Thanks
> Steven

WaxDragon, Has my vote for Top Test and Release Coordinator.
Very Good!
James

[Andrew Murphy]

Same here,

Go Wax

Andrew

On 7/24/05, James Tabor [e-mail] 
wrote:
> Steven Edwards wrote:
> > Hi,
> >
> > --- WaxDragon [e-mail] wrote:
> > > The keyboard input bug was a blocker too. Luckly i had the email from
>
> bugzilla.
> > This is a good time to mention this. I would like to propose that we
>
> have a regression or (QA)
> > coordinator position in the Constitution. This position is designed with
>
> WaxDragon in mind. For
> > those of you that don't know he hangs out on IRC all the time and has
>
> been doing a incredible job
> > tracking down regressions and which revisions caused them using ISOs
>
> from each build.
> > The position would entail certain rights and responsibilities:
> >
> > Finial Authority on blocker bugs for a release (Could be overridden by a
>
> vote)
> > Maintains the documentation on testing, hunting regressions, etc
> > Would be the person in charge of directing any testing team that
>
> develops.
> > Maintain bugzilla
> > Anything else that we would need a QA person for.
> >
> > Thoughts?
> >
> > Thanks
> > Steven
>
> WaxDragon, Has my vote for Top Test and Release Coordinator.
> Very Good!
> James
> _______________________________________________
> Ros-dev mailing list
> [e-mail]
> http://reactos.com:8080/mailman/listinfo/ros-dev

[Casper Hornstrup]

> -----Original Message-----
> From: [e-mail] [mailto:[e-mail]] On Behalf Of Steven Edwards
> Sent: 24. juli 2005 17:50
> To: WaxDragon; ReactOS Development List
> Subject: Re: [ros-dev] Blockers...
>
> Hi,
>
> --- WaxDragon [e-mail] wrote:
> > The keyboard input bug was a blocker too.  Luckly i had the email from bugzilla.
>
> This is a good time to mention this. I would like to propose that we have a regression or (QA)
> coordinator position in the Constitution. This position is designed with WaxDragon in mind. For
> those of you that don't know he hangs out on IRC all the time and has been doing a incredible job
> tracking down regressions and which revisions caused them using ISOs from each build.
>
> The position would entail certain rights and responsibilities:
>
> Finial Authority on blocker bugs for a release (Could be overridden by a vote)
> Maintains the documentation on testing, hunting regressions, etc
> Would be the person in charge of directing any testing team that develops.
> Maintain bugzilla
> Anything else that we would need a QA person for.
>
> Thoughts?
>
> Thanks
> Steven

The constitution is not about team coordinator roles. It's about the decision making process of the project. A description of the
organizational structure of ReactOS belong in a separate document that is easier to change.

Casper

[Steven Edwards]

Hi Casper,

--- Casper Hornstrup [e-mail] wrote:
> The constitution is not about team coordinator roles. It's about the decision making process of
> the project. A description of the
> organizational structure of ReactOS belong in a separate document that is easier to change.

Works for me. How about WaxDragon in such a role?

Thanks
Steven

____________________________________________________
Start your day with Yahoo! - make it your home page 
http://www.yahoo.com/r/hs

[James Tabor]

WaxDragon wrote:
> On 7/23/05, Alex Ionescu [e-mail] wrote:
> > Hi,
> >
> > Here is a current update on the 0.2.7 blockers...if you can help, please DO!
> The keyboard input bug was a blocker too.  Luckly i had the email from bugzilla.
>
> http://reactos.com/bugzilla/show_bug.cgi?id=657
>        Summary: Shift sticks when selecting text with the arrow keys.
>        Product: ReactOS
>        Version: 0.3.0-SVN
>       Platform: VMWare 4
>     OS/Version: ReactOS
>         Status: NEW
>       Severity: major
>       Priority: P3
>      Component: Drivers
>     AssignedTo: [e-mail]
>     ReportedBy: [e-mail]
>
>
> VMWare 4.5.2
> ReactOS r16251 (this is not a recent bug, I think 15xxx has it)
> Boot ROS
> Open notepad
> Type "This sucks"
> Press and hold Shift and use the arrow keys to select the previous text (not
>
> the numberic keypad*)
> 5. Type "This sucks" again, notice it is in UPPERCASE and backspace no
> longer works
> Numlock seems to affect this.  I boot vmware with numlock off, and this bug
>
> occurs.  If turn num lock on before typing, imput works as expected.  It's Shift
> that gets stuck because capslock Shifts them back to lowercase, but the numbers
> come out as ! @ # $ still.  The capslock - arrow keys interaction is also
> discussed in bug 617
>
> There were some other  comments on this bug, basically keycode dumps
> showing how the arrow keys are emulated my qemu and vmware.
>
> Has anyone seen tinus?
>
> WD

After looking into this I found in win32k/ntuser/keyboard.c. All controls keys in wParam
return left key as in VK_LCONTROL, VK_LMENU and VK_LSHIFT. I can not down menu with VK_MENU
because wParam is VK_LMENU.

The numlock keys are also translated into,
static WORD NumpadConversion[][2] =
     { { VK_DELETE, VK_DECIMAL },
       { VK_INSERT, VK_NUMPAD0 },
       { VK_END,    VK_NUMPAD1 },
       { VK_DOWN,   VK_NUMPAD2 },
       { VK_NEXT,   VK_NUMPAD3 },
       { VK_LEFT,   VK_NUMPAD4 },
       { VK_CLEAR,  VK_NUMPAD5 },
       { VK_RIGHT,  VK_NUMPAD6 },
       { VK_HOME,   VK_NUMPAD7 },
       { VK_UP,     VK_NUMPAD8 },
       { VK_PRIOR,  VK_NUMPAD9 },
this is in W32kKeyProcessMessage. Maybe the bug is in keyboard.c or input.c.
Thanks,
James

[Maarten Bosma]

Hi,

the mailing list might not be the best place for this, but blight was so 
nice to provide his wiki.

http://blight.eu.org/reactos/wiki/index.php/0.2.7

Maarten Bosma

I wanted to add details about te memory corrution bug, but I can't
register, the  "Create an account"-fields are missing
> Hi,
>
> the mailing list might not be the best place for this, but blight was so
> nice to provide his wiki.
>
> http://blight.eu.org/reactos/wiki/index.php/0.2.7
>
> Maarten Bosma
> _______________________________________________
> Ros-dev mailing list
> [e-mail]
> http://reactos.com:8080/mailman/listinfo/ros-dev

[Maarten Bosma]

[e-mail] wrote:
> I wanted to add details about te memory corrution bug, but I can't
> register, the  "Create an account"-fields are missing

I emailed you the passwort for the 'rosdev' accout blight gave us.

[Brandon Mark Turner]

Sorry for the delay, i will narrow it down father, which of the 3 rev it is, 
later today once i get back home from my small vacation.

Brandon

Alex Ionescu writes:
> Hi,
>
> Here is a current update on the 0.2.7 blockers...if you can help, please 
> DO!
>
> BUG:
> Green icon in run dialog
>
> REGRESSION IDENTIFIED:
> Yes. By WaxDragon. Regression due to BUILD.
>
> CAUSE:
> Incorrect icon was added to resource file.
>
> FIXED:
> Yes. Fix needs to be merged into 0.2.7 branch.
>
> BUG:
> Common Dialog Control Broken for File/Save.
>
> REGRESSION IDENTIFIED:
> Yes. By WaxDragon. Regression due to 15669.
>
> CAUSE:
> Building shell32 as ANSI causes this problem. Building it as UNICODE 
> however breaks the My computer dialog. Shell32 should be built as Unicode, 
> and the My computer dialog should appear when double-clicked. So this bug 
> needs to be fixed.
>
> BUG:
> If you use the arrow keys in the explorer start menu to expand an entry,
>
> you cannot use them anymore to go up and down.
>
> No further information has been researched yet. Martin, explorer is your 
> baby, and I'm sure is an easy fix...it doesnt' happen with the right-click 
> context menu on the desktop, so maybe this is localized to some bug in 
> your start menu code?
>
> BUG:
> If you right-click on the desktop to show the context menu, then
>
> right-click again, then left-click (or other similar combinations), your 
> next right-click is not registered anymore.
>
> No further information is available, although I belive hpoussin said the 
> problem is above win32k. I haven't heard from tinus in a long time, but 
> i8042prt might be where this problem resides.
>
> BUG:
> Several applications have regressed due to "memory corruption" and a
>
> bugcheck related to NtGdiRealizePallette.
>
> More information is on the Bugzilla DB, but it's down atm. I have heard 
> reports that the error always happened before, but wasn't seen because red 
> zone pool protection was disabled. In any case, it breaks a large number 
> of apps.
>
> BUG:
> AllocConsole failure in 1st-stage setup on Dell Laptops (and maybe
>
> others).
>
> REGRESSION IDENTIFIED:
> Yes. By BrandonTurner. A regression range has been identified and 3 key 
> builds are being tested to find the actual regression.
>
> Best Regards,
> Alex Ionescu
> _______________________________________________
> Ros-dev mailing list
> [e-mail]
> http://reactos.com:8080/mailman/listinfo/ros-dev

[Alex Ionescu]

Alex Ionescu wrote:

As I don't have the wiki password yet, I'm updating this here:
> BUG:
> Several applications have regressed due to "memory corruption" and a
>
> bugcheck related to NtGdiRealizePallette.
>
> More information is on the Bugzilla DB, but it's down atm. I have 
> heard reports that the error always happened before, but wasn't seen 
> because red zone pool protection was disabled. In any case, it breaks 
> a large number of apps.

Possible patch posted by Hartmut on the ML. Does anyone remember which 
apps were broken? I'd like to test it.
> BUG:
> AllocConsole failure in 1st-stage setup on Dell Laptops (and maybe
>
> others).
>
> REGRESSION IDENTIFIED:
> Yes. By BrandonTurner. A regression range has been identified and 3 
> key builds are being tested to find the actual regression.

Regression found by BrandonTurner. Caused by commit 14906:
Author: hbirr
Fixed the calculation of timeout values.

However, I believe that Hartmut's patch is correct...

Best regards,
Alex Ionescu

[James Tabor]

Alex Ionescu wrote:
> Alex Ionescu wrote:
>
> As I don't have the wiki password yet, I'm updating this here:
> > BUG:
> > Several applications have regressed due to "memory corruption" and a
> >
> > bugcheck related to NtGdiRealizePallette.
> >
> > More information is on the Bugzilla DB, but it's down atm. I have 
> > heard reports that the error always happened before, but wasn't seen 
> > because red zone pool protection was disabled. In any case, it breaks 
> > a large number of apps.
>
> Possible patch posted by Hartmut on the ML. Does anyone remember which 
> apps were broken? I'd like to test it.
> > BUG:
> > AllocConsole failure in 1st-stage setup on Dell Laptops (and maybe
> >
> > others).
> >
> > REGRESSION IDENTIFIED:
> > Yes. By BrandonTurner. A regression range has been identified and 3 
> > key builds are being tested to find the actual regression.
>
> Regression found by BrandonTurner. Caused by commit 14906:
> Author: hbirr
> Fixed the calculation of timeout values.
>
> However, I believe that Hartmut's patch is correct...
>
> Best regards,
> Alex Ionescu

Hi!
Do you mean NtGdiResizePalette? I do not have a listing in Undoc W2k of a
NtGdiRealizePallette.
Thanks,
James

BTW NtGdiUnrealizeObject is finished, it does nothing more than what I committed
to the patch.

[James Tabor]

James Tabor wrote:
> Hi!
> Do you mean NtGdiResizePalette? I do not have a listing in Undoc W2k of a
> NtGdiRealizePallette.
> Thanks,
> James
>
> BTW NtGdiUnrealizeObject is finished, it does nothing more than what I 
> committed
> to the patch.

UserRealizePalette ping!

[Richard]

http://msdn.microsoft.com/library/default.asp?url=/library/en-us/gdi/colors_...

James Tabor wrote:
> James Tabor wrote:
> > Hi!
> > Do you mean NtGdiResizePalette? I do not have a listing in Undoc W2k 
> > of a
> > NtGdiRealizePallette.
> > Thanks,
> > James
> >
> > BTW NtGdiUnrealizeObject is finished, it does nothing more than what 
> > I committed
> > to the patch.
>
> UserRealizePalette ping!
> _______________________________________________
> Ros-dev mailing list
> [e-mail]
> http://reactos.com:8080/mailman/listinfo/ros-dev

[James Tabor]

Hi!
Richard wrote:
> http://msdn.microsoft.com/library/default.asp?url=/library/en-us/gdi/colors_...

Thanks, I know. 8^D
> James Tabor wrote:
> > James Tabor wrote:
> > > Hi!
> > > Do you mean NtGdiResizePalette? I do not have a listing in Undoc W2k 
> > > of a
> > > NtGdiRealizePallette.
> > > Thanks,
> > > James
> > >
> > > BTW NtGdiUnrealizeObject is finished, it does nothing more than what 
> > > I committed
> > > to the patch.
> >
> > UserRealizePalette ping!

I was in my little way maken a joke.
8^P
James

Hi Winrar is one of two apps that have memory corruption bugs that I know of

Quoting Alex Ionescu [e-mail]:
> Alex Ionescu wrote:
>
> As I don't have the wiki password yet, I'm updating this here:
> > BUG:
> > Several applications have regressed due to "memory corruption" and a
> >
> > bugcheck related to NtGdiRealizePallette.
> >
> > More information is on the Bugzilla DB, but it's down atm. I have
> > heard reports that the error always happened before, but wasn't seen
> > because red zone pool protection was disabled. In any case, it breaks
> > a large number of apps.
>
> Possible patch posted by Hartmut on the ML. Does anyone remember which
> apps were broken? I'd like to test it.
> > BUG:
> > AllocConsole failure in 1st-stage setup on Dell Laptops (and maybe
> >
> > others).
> >
> > REGRESSION IDENTIFIED:
> > Yes. By BrandonTurner. A regression range has been identified and 3
> > key builds are being tested to find the actual regression.
>
> Regression found by BrandonTurner. Caused by commit 14906:
> Author: hbirr
> Fixed the calculation of timeout values.
>
> However, I believe that Hartmut's patch is correct...
>
> Best regards,
> Alex Ionescu
>
> Ros-dev mailing list
> [e-mail]
> http://reactos.com:8080/mailman/listinfo/ros-dev

Other examples are Dr. Hardware and Total Commander (not always!)
> Hi Winrar is one of two apps that have memory corruption bugs that I know
> of
>
> Quoting Alex Ionescu [e-mail]:
> > Alex Ionescu wrote:
> >
> > As I don't have the wiki password yet, I'm updating this here:
> > > BUG:
> > > Several applications have regressed due to "memory corruption" and a
> > >
> > > bugcheck related to NtGdiRealizePallette.
> > >
> > > More information is on the Bugzilla DB, but it's down atm. I have
> > > heard reports that the error always happened before, but wasn't seen
> > > because red zone pool protection was disabled. In any case, it breaks
> > > a large number of apps.
> >
> > Possible patch posted by Hartmut on the ML. Does anyone remember which
> > apps were broken? I'd like to test it.
> > > BUG:
> > > AllocConsole failure in 1st-stage setup on Dell Laptops (and maybe
> > >
> > > others).
> > >
> > > REGRESSION IDENTIFIED:
> > > Yes. By BrandonTurner. A regression range has been identified and 3
> > > key builds are being tested to find the actual regression.
> >
> > Regression found by BrandonTurner. Caused by commit 14906:
> > Author: hbirr
> > Fixed the calculation of timeout values.
> >
> > However, I believe that Hartmut's patch is correct...
> >
> > Best regards,
> > Alex Ionescu
> >
> > Ros-dev mailing list
> > [e-mail]
> > http://reactos.com:8080/mailman/listinfo/ros-dev
>
> Ros-dev mailing list
> [e-mail]
> http://reactos.com:8080/mailman/listinfo/ros-dev

[Hartmut Birr]

Alex Ionescu wrote:
> > BUG:
> > AllocConsole failure in 1st-stage setup on Dell Laptops (and maybe
> >
> > others).
> >
> > REGRESSION IDENTIFIED:
> > Yes. By BrandonTurner. A regression range has been identified and 3
> > key builds are being tested to find the actual regression.
>
> Regression found by BrandonTurner. Caused by commit 14906:
> Author: hbirr
> Fixed the calculation of timeout values.
>
> However, I believe that Hartmut's patch is correct...

If the 1st-stage setup  is the console setup which copies the files from
the cd, it isn't possible that my patch has any effect for this bug.
Usetup.exe does only use ntdll.dll. Kernel32.dll isn't loaded at that time.

- Hartmut

[Brandon Turner]

I am almost certain i made a mistake when doing the reg testing.  I have 
talked with alex, but i think it was after he sent this out.  I dont 
think it broke on this patch.  I think it a few after it.  I will get 
the exact one a little later.  Sorry for the confusion and the delay.

Brandon

Hartmut Birr wrote:
> Alex Ionescu wrote:
> > > BUG:
> > > AllocConsole failure in 1st-stage setup on Dell Laptops (and maybe
> > >
> > > others).
> > >
> > > REGRESSION IDENTIFIED:
> > > Yes. By BrandonTurner. A regression range has been identified and 3
> > > key builds are being tested to find the actual regression.
> >
> > Regression found by BrandonTurner. Caused by commit 14906:
> > Author: hbirr
> > Fixed the calculation of timeout values.
> >
> > However, I believe that Hartmut's patch is correct...
>
> If the 1st-stage setup  is the console setup which copies the files from
> the cd, it isn't possible that my patch has any effect for this bug.
> Usetup.exe does only use ntdll.dll. Kernel32.dll isn't loaded at that time.
> Hartmut
>
> Ros-dev mailing list
> [e-mail]
> http://reactos.com:8080/mailman/listinfo/ros-dev

[Alex Ionescu]

Hartmut Birr wrote:
> Alex Ionescu wrote:
> > > BUG:
> > > AllocConsole failure in 1st-stage setup on Dell Laptops (and maybe
> > >
> > > others).
> > >
> > > REGRESSION IDENTIFIED:
> > > Yes. By BrandonTurner. A regression range has been identified and 3
> > > key builds are being tested to find the actual regression.
> >
> > Regression found by BrandonTurner. Caused by commit 14906:
> > Author: hbirr
> > Fixed the calculation of timeout values.
> >
> > However, I believe that Hartmut's patch is correct...
>
> If the 1st-stage setup  is the console setup which copies the files from
> the cd, it isn't possible that my patch has any effect for this bug.
> Usetup.exe does only use ntdll.dll. Kernel32.dll isn't loaded at that time.
> Hartmut

Hi Hartmut,

This confirms what I thought, and also your patch is correct. Brandon 
told me that he made some mistakes in regression testing, but it was 
after I posted the email. Sorry.

Best regards,
Alex Ionescu