been written before I wrote a never-failing snew.


git-svn-id: svn://svn.tartarus.org/sgt/putty@9985 cda61777-01e9-0310-a592-d414129be87e

half-filled with null pointers.


git-svn-id: svn://svn.tartarus.org/sgt/putty@9986 cda61777-01e9-0310-a592-d414129be87e

now check that all the modular functions (modpow, modinv, modmul,
bigdivmod) have zero moduli, and that modinv also has a nonzero thing
to try to invert.


git-svn-id: svn://svn.tartarus.org/sgt/putty@9987 cda61777-01e9-0310-a592-d414129be87e

git-svn-id: svn://svn.tartarus.org/sgt/putty@9988 cda61777-01e9-0310-a592-d414129be87e

sufficiently silly DSA signature.


git-svn-id: svn://svn.tartarus.org/sgt/putty@9989 cda61777-01e9-0310-a592-d414129be87e

check for that return value everywhere it is used.


git-svn-id: svn://svn.tartarus.org/sgt/putty@9990 cda61777-01e9-0310-a592-d414129be87e

public blob, and return a proper error in that situation rather than a
struct with unhelpful NULLs in.


git-svn-id: svn://svn.tartarus.org/sgt/putty@9991 cda61777-01e9-0310-a592-d414129be87e

git-svn-id: svn://svn.tartarus.org/sgt/putty@9992 cda61777-01e9-0310-a592-d414129be87e

git-svn-id: svn://svn.tartarus.org/sgt/putty@9993 cda61777-01e9-0310-a592-d414129be87e

out of range manages to get past other recent fixes.


git-svn-id: svn://svn.tartarus.org/sgt/putty@9995 cda61777-01e9-0310-a592-d414129be87e

bn_restore_invariant (and the many loops that duplicate it) leaves a
single zero word in a bignum representing 0, whereas the constant
'Zero' does not have any data words at all. Cope with this in
bignum_cmp.

(It would be a better plan to decide on one representation and stick
with it, but this is the less disruptive fix for the moment.)


git-svn-id: svn://svn.tartarus.org/sgt/putty@9996 cda61777-01e9-0310-a592-d414129be87e

git-svn-id: svn://svn.tartarus.org/sgt/putty@9997 cda61777-01e9-0310-a592-d414129be87e

git-svn-id: svn://svn.tartarus.org/sgt/putty@9998 cda61777-01e9-0310-a592-d414129be87e

than fonts. I broke this in r9559 when I added the option for 'both',
because the internal representation got offset by one so as to change
from a boolean to two bitfields and I must have confused myself about
what the default should be.


git-svn-id: svn://svn.tartarus.org/sgt/putty@10008 cda61777-01e9-0310-a592-d414129be87e

For detailed explanation, see here:
http://stackoverflow.com/questions/2520610/detecting-cleartype-optimized-fonts

To reproduce, you need Configuration -> Window -> Colours -> Bolded text is a different colour disabled

…t even if we're not minimised at the time

When the "Adb" connection type was added the radio buttons got
extremely squashed together and it was no longer possible to read
their labels completely. Therefore splitting them into two rows
seemed like a good idea to me.

Since the space was quite limited already I also increased the
entire height of the config dialog box by 10. This appeared to be
the better option instead of making the "Sessions" control smaller
in its height.

To connect a specific device, input device ID in host name
To connect the default device, input a colon(:) in host name

…detection boxes.

…ligned on wide-drawn characters.

actually start the adb shell

previous one only fixed it for device id being given

This is mainly to ease testing; I have no idea what effects
it will have in the real world, but we believe it to be minimal.

For some reason, the icon in the system tray does always not get removed
when the utility closes. This patch closes it.

IPTOS, if defined, contains the ToS for whatever is being compiled..
PuTTY & PuTTYTel = 0x10 = IPTOS_LOWDELAY
pscp & sftp = 0x08 = IPTOS_THROUGHPUT
plink = 0x04 = IPTOS_RELIABILITY

Notes:
Windows XP/2003 users must set HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\DisableUserTOSSetting (DWORD) to 0 to be able to set the ToS.
See http://support.microsoft.com/default.aspx?scid=kb;EN-US;q248611

git-svn-id: http://futty.googlecode.com/svn/trunk@24 8b5b7008-2324-11df-a154-f73f31f273f6

Conflicts:

	windows/MSVC/plink/plink.vcxproj
	windows/MSVC/pscp/pscp.vcxproj
	windows/MSVC/psftp/psftp.vcxproj
	windows/MSVC/putty.suo
	windows/MSVC/putty/putty.vcxproj
	windows/MSVC/puttytel/puttytel.vcxproj

Note, there's no way to disable utmp support, so if you don't want this, use the cthelper.exe from puttycyg instead.

git-svn-id: http://futty.googlecode.com/svn/trunk@44 8b5b7008-2324-11df-a154-f73f31f273f6

When the "Adb" connection type was added the radio buttons got
extremely squashed together and it was no longer possible to read
their labels completely. Therefore splitting them into two rows
seemed like a good idea to me.

Since the space was quite limited already I also increased the
entire height of the config dialog box by 10. This appeared to be
the better option instead of making the "Sessions" control smaller
in its height.

…ygwin/MinGW even when it does have IS_HIGH_SURROGATE defined.

…e font sizes stay the same.

It's the "Window" option called "Change rows and columns only when maximised".

…G was defined).

Now you can scrollback more or less than half a screen at a time!

Thanks goes to [email protected]
http://sourceforge.net/projects/putty-nd/
https://github.com/noodle1983/putty-nd

Code taken from gdi-20120130.diff

http://ice.hotmint.com/putty/
https://code.google.com/p/puttychs/
https://raw.github.com/emonkak/putty/4880c62e3cde6931dc6cb6e663176dca0a2adae3/patch/gdi-20120211.diff

This code isn't great; it effectively drives the UI by clicking
buttons on it... but, then, if you're going to put all the code
directly into WndProc, what do you expect?

This causes startup hangs for some users.

I've been unable to determine a way to ask Windows for
a reasonable guess of the username that is not going to
go to the network.

This can only be enabled when no keys are loaded.  This sucks a bit,
but it doesn't have the paths of the keys available so it can't do it
directly.  Really should make the paths available.