u/mdw/putty
17 years agoNew logging mode, which records the exact bytes sent over the wire
simon [Tue, 29 Aug 2006 19:07:11 +0000 (19:07 +0000)]
New logging mode, which records the exact bytes sent over the wire
in an SSH connection _in addition_ to the decrypted packets. This
will hopefully come in useful for debugging wire data corruption
issues: you can strace the server, enable this mode in the client,
and compare the sent and received data.

I'd _like_ to have this mode also log Diffie-Hellman private
exponents, session IDs, encryption and MAC keys, so that the
resulting log file could be used to independently verify the
correctness of all cryptographic operations performed by PuTTY.
However, I haven't been able to convince myself that the security
implications are acceptable. (It doesn't matter that this
information would permit an attacker to decrypt the session, because
the _already_ decrypted session is stored alongside it in the log
file. And I'm not planning, under any circumstances, to log users'
private keys. But gaining access to the log file while the session
was still running would permit an attacker to _hijack_ the session,
and that's the iffy bit.)

git-svn-id: svn://svn.tartarus.org/sgt/putty@6835 cda61777-01e9-0310-a592-d414129be87e

17 years agoExplicitly closing logctx on various kinds of error exit means that
simon [Tue, 29 Aug 2006 18:50:07 +0000 (18:50 +0000)]
Explicitly closing logctx on various kinds of error exit means that
the log file gets fclosed properly and the critical last few
messages might be recoverable from the log file more often...

git-svn-id: svn://svn.tartarus.org/sgt/putty@6834 cda61777-01e9-0310-a592-d414129be87e

17 years agoThe Windows HANDLE type, despite being a `void *', does not actually
simon [Tue, 29 Aug 2006 18:32:44 +0000 (18:32 +0000)]
The Windows HANDLE type, despite being a `void *', does not actually
behave like a pointer. In particular, the right thing to set a
HANDLE to to indicate that it's invalid is INVALID_HANDLE_VALUE, not
NULL. Crack down on sloppy use of NULL HANDLEs across all Windows
code.

(There is one oddity, which is that {Create,Open}FileMapping are
documented to return a NULL HANDLE instead of INVALID_HANDLE_VALUE
on failure. Shrug. If MS want to be inconsistent, I suppose I have
to live with it.)

git-svn-id: svn://svn.tartarus.org/sgt/putty@6833 cda61777-01e9-0310-a592-d414129be87e

17 years agoSupport for sending serial breaks, in both the Windows and Unix
simon [Tue, 29 Aug 2006 18:20:57 +0000 (18:20 +0000)]
Support for sending serial breaks, in both the Windows and Unix
serial backends.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6832 cda61777-01e9-0310-a592-d414129be87e

17 years agoInhibit the Serial configuration panel in mid-session if the session
simon [Tue, 29 Aug 2006 09:18:09 +0000 (09:18 +0000)]
Inhibit the Serial configuration panel in mid-session if the session
isn't a serial one. In particular, this causes pterm not to fail an
assertion if you select `Change Settings'. Ahem.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6831 cda61777-01e9-0310-a592-d414129be87e

17 years agoApparently MsgWaitForMultipleObjects doesn't always return the
simon [Mon, 28 Aug 2006 19:09:28 +0000 (19:09 +0000)]
Apparently MsgWaitForMultipleObjects doesn't always return the
values one might expect, which means that GetMessage() was
occasionally blocking the process. That appears to be the last of
the annoying data loss issues, so I think the Windows serial back
end actually looks vaguely reliable now. Phew.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6830 cda61777-01e9-0310-a592-d414129be87e

17 years agoThat OVERLAPPED fix seems to have made flow control useful as well
simon [Mon, 28 Aug 2006 18:36:11 +0000 (18:36 +0000)]
That OVERLAPPED fix seems to have made flow control useful as well
(presumably Windows's serial buffer is actually _filling up_,
causing an XOFF to be sent, now that my dodgy I/O code isn't causing
it to leak). So I think I'll switch the default flow control to
XON/XOFF, since it actually seems to do something now.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6829 cda61777-01e9-0310-a592-d414129be87e

17 years agoEep! Next bit flag after 1 and 2 is _4_, not 3. Perhaps it's time I
simon [Mon, 28 Aug 2006 18:27:54 +0000 (18:27 +0000)]
Eep! Next bit flag after 1 and 2 is _4_, not 3. Perhaps it's time I
stopped coding and went and sat down quietly and tried not to touch
anything for a while.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6828 cda61777-01e9-0310-a592-d414129be87e

17 years agoReading 4K at a time from a serial port turns out to be a bit
simon [Mon, 28 Aug 2006 18:26:50 +0000 (18:26 +0000)]
Reading 4K at a time from a serial port turns out to be a bit
unfriendly in an interactive session, because at 19200 baud it takes
nearly two seconds to receive that much data, and as long as the
data is flowing continuously Windows waits until it has a full
buffer. So here's another annoying flag in the winhandl API, which
restricts reads to length 1 so that serial output shows up as it
appears.

(I tried this yesterday, but without the OVERLAPPED fix in r6826 it
behaved very erratically. It now seems solid.)

git-svn-id: svn://svn.tartarus.org/sgt/putty@6827 cda61777-01e9-0310-a592-d414129be87e

17 years agoApparently it helps for an OVERLAPPED structure to contain a valid
simon [Mon, 28 Aug 2006 18:16:49 +0000 (18:16 +0000)]
Apparently it helps for an OVERLAPPED structure to contain a valid
event handle. This seems to have fixed _some_, but not all, of the
curious data loss issues in the Windows serial backend.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6826 cda61777-01e9-0310-a592-d414129be87e

17 years agoMinor tweaks to -nc:
jacob [Mon, 28 Aug 2006 17:47:43 +0000 (17:47 +0000)]
Minor tweaks to -nc:
 - log host:port in event log
 - add -nc to Plink usage message

git-svn-id: svn://svn.tartarus.org/sgt/putty@6825 cda61777-01e9-0310-a592-d414129be87e

17 years agoMissed a couple of instances of cfg_launchable().
simon [Mon, 28 Aug 2006 17:41:02 +0000 (17:41 +0000)]
Missed a couple of instances of cfg_launchable().

git-svn-id: svn://svn.tartarus.org/sgt/putty@6824 cda61777-01e9-0310-a592-d414129be87e

17 years agoNew command-line option in Plink (and PuTTY, though it's less useful
simon [Mon, 28 Aug 2006 15:12:37 +0000 (15:12 +0000)]
New command-line option in Plink (and PuTTY, though it's less useful
there): `plink host -nc host2:port' causes the SSH connection's main
channel to be replaced with a direct-tcpip connection to the
specified destination. This feature is mainly designed for use as a
local proxy: setting your local proxy command to `plink %proxyhost
-nc %host:%port' lets you tunnel SSH over SSH with a minimum of
fuss. Works on all platforms.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6823 cda61777-01e9-0310-a592-d414129be87e

17 years agoSerial back end for Unix. Due to hardware limitations (no Linux box
simon [Mon, 28 Aug 2006 14:29:02 +0000 (14:29 +0000)]
Serial back end for Unix. Due to hardware limitations (no Linux box
I own has both an X display and a working serial port) I have been
unable to give this the full testing it deserves; I've managed to
demonstrate the basic functionality of Unix Plink talking to a
serial port, but I haven't been able to test the GTK front end. I
have no reason to think it will fail, but I'll be more comfortable
once somebody has actually tested it.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6822 cda61777-01e9-0310-a592-d414129be87e

17 years ago... and here's the rest of r6820. Ahem.
simon [Mon, 28 Aug 2006 13:08:50 +0000 (13:08 +0000)]
... and here's the rest of r6820. Ahem.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6821 cda61777-01e9-0310-a592-d414129be87e

17 years agoBecause not all OSes will support the same set of serial port
simon [Mon, 28 Aug 2006 13:08:15 +0000 (13:08 +0000)]
Because not all OSes will support the same set of serial port
options, here's a slight change to the API of ser_setup_config_box()
to make it filter its parity and flow control options using
platform-supplied bit masks.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6820 cda61777-01e9-0310-a592-d414129be87e

17 years agoFix small event log bug.
simon [Mon, 28 Aug 2006 11:33:51 +0000 (11:33 +0000)]
Fix small event log bug.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6819 cda61777-01e9-0310-a592-d414129be87e

17 years agoForgot to initialise serial->bufsize to zero.
simon [Mon, 28 Aug 2006 11:32:12 +0000 (11:32 +0000)]
Forgot to initialise serial->bufsize to zero.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6818 cda61777-01e9-0310-a592-d414129be87e

17 years agoFix line endings (svn:eol-style properties and actual CRs).
simon [Mon, 28 Aug 2006 11:13:56 +0000 (11:13 +0000)]
Fix line endings (svn:eol-style properties and actual CRs).

git-svn-id: svn://svn.tartarus.org/sgt/putty@6817 cda61777-01e9-0310-a592-d414129be87e

17 years agoSupport for Windows PuTTY connecting straight to a local serial port
simon [Mon, 28 Aug 2006 10:35:12 +0000 (10:35 +0000)]
Support for Windows PuTTY connecting straight to a local serial port
in place of making a network connection. This has involved a couple
of minor infrastructure changes:
 - New dlg_label_change() function in the dialog.h interface, which
   alters the label on a control. Only used, at present, to switch
   the Host Name and Port boxes into Serial Line and Speed, which
   means that any platform not implementing serial connections (i.e.
   currently all but Windows) does not need to actually do anything
   in this function. Yet.
 - New small piece of infrastructure: cfg_launchable() determines
   whether a Config structure describes a session ready to be
   launched. This was previously determined by seeing if it had a
   non-empty host name, but it has to check the serial line as well
   so there's a centralised function for it. I haven't gone through
   all front ends and arranged for this function to be used
   everywhere it needs to be; so far I've only checked Windows.
 - Similarly, cfg_dest() returns the destination of a connection
   (host name or serial line) in a text format suitable for putting
   into messages such as `Unable to connect to %s'.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6815 cda61777-01e9-0310-a592-d414129be87e

17 years agoNow that Local proxy type exists on both Unix and Windows, document it and
jacob [Sun, 27 Aug 2006 11:55:30 +0000 (11:55 +0000)]
Now that Local proxy type exists on both Unix and Windows, document it and
associated stuff.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6814 cda61777-01e9-0310-a592-d414129be87e

17 years agoGrow some nasty warts on the side of winhandl.c, in preparation for
simon [Sun, 27 Aug 2006 10:00:36 +0000 (10:00 +0000)]
Grow some nasty warts on the side of winhandl.c, in preparation for
a serial port backend:
 - In order to do simultaneous reading and writing on the same
   HANDLE, you must enable overlapped access and pass an OVERLAPPED
   structure to each ReadFile and WriteFile call. This would make
   sense if it were an optional thing I could do if I wanted to do
   the reading and writing in the same thread, but making it
   mandatory even if I'm doing them in _different_ threads is just
   annoying and arbitrary.
 - Serial ports occasionally return length 0 from ReadFile, for no
   particularly good reason. Fortunately serial ports also don't
   have a real EOF condition to speak of, so ignoring EOFs is
   actually a viable response in spite of sounding utterly gross.
Hence, handle_{input,output}_new() now accept a flags parameter,
which includes a flag to enable the OVERLAPPED bureaucracy and a
flag to cause EOFs to be ignored on input handles. The current
clients of winhandl.c do not use either of these.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6813 cda61777-01e9-0310-a592-d414129be87e

17 years agoRemove spurious #include.
simon [Sun, 27 Aug 2006 09:53:34 +0000 (09:53 +0000)]
Remove spurious #include.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6812 cda61777-01e9-0310-a592-d414129be87e

17 years agoCall console_provide_logctx _before_ initialising the back end, so
simon [Sun, 27 Aug 2006 08:34:04 +0000 (08:34 +0000)]
Call console_provide_logctx _before_ initialising the back end, so
that logevent() will go to stderr in -v mode even during the back
end init function.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6811 cda61777-01e9-0310-a592-d414129be87e

17 years agoThe `socket' function in the backends is only ever checked to see if
simon [Sun, 27 Aug 2006 08:03:19 +0000 (08:03 +0000)]
The `socket' function in the backends is only ever checked to see if
it's NULL. Since we already have one back end (uxpty) which doesn't
in fact talk to a network socket, and may well have more soon, I'm
replacing this TCP/IP-centric function with a nice neutral
`connected' function returning a boolean. Nothing else about its
semantics has currently changed.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6810 cda61777-01e9-0310-a592-d414129be87e

17 years agoIt's critically important that the local proxy process should not
simon [Sat, 26 Aug 2006 10:59:09 +0000 (10:59 +0000)]
It's critically important that the local proxy process should not
inherit _our_ ends of its I/O pipes! Otherwise, closing our copy of
those handles does not cause it to see EOF on its stdin, because
it's holding the pipe open itself.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6808 cda61777-01e9-0310-a592-d414129be87e

17 years agoWe _can_ have handle_throttle() called on defunct handles after all,
simon [Sat, 26 Aug 2006 10:58:13 +0000 (10:58 +0000)]
We _can_ have handle_throttle() called on defunct handles after all,
so it should just do nothing rather than failing an assertion.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6807 cda61777-01e9-0310-a592-d414129be87e

17 years agoProxyCommand support for Windows, using the new winhandl.c API.
simon [Sat, 26 Aug 2006 10:20:16 +0000 (10:20 +0000)]
ProxyCommand support for Windows, using the new winhandl.c API.
Seems a bit clunky when I actually try to use it - not sure why -
but I think all the actual functionality is there.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6806 cda61777-01e9-0310-a592-d414129be87e

17 years agoAnother bug fix: always set the busy flag when telling a subthread
simon [Sat, 26 Aug 2006 10:19:23 +0000 (10:19 +0000)]
Another bug fix: always set the busy flag when telling a subthread
to do something, otherwise handle_get_events will forget to tell the
front end to check for that subthread finishing. This applies even
when we're only setting `busy' to tell the subthread to terminate!

git-svn-id: svn://svn.tartarus.org/sgt/putty@6805 cda61777-01e9-0310-a592-d414129be87e

17 years agoCleanups to reduce dependency on Windows SFTP tools always having a
simon [Sat, 26 Aug 2006 10:18:31 +0000 (10:18 +0000)]
Cleanups to reduce dependency on Windows SFTP tools always having a
real network socket.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6804 cda61777-01e9-0310-a592-d414129be87e

17 years agoOn cleanup, PSCP and PSFTP should explicitly check that the back end
simon [Sat, 26 Aug 2006 10:17:39 +0000 (10:17 +0000)]
On cleanup, PSCP and PSFTP should explicitly check that the back end
is still running rather than relying on ssh_sftp_loop_iteration() to
return a bogus value.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6803 cda61777-01e9-0310-a592-d414129be87e

17 years agoStart using notify_remote_exit() in the Windows front end, in place
simon [Sat, 26 Aug 2006 10:04:46 +0000 (10:04 +0000)]
Start using notify_remote_exit() in the Windows front end, in place
of the previous ad-hockery which depended on the return value from
select_result() and hence which will not adapt sensibly to a world
in which the primary session is something local rather than a
network connection.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6802 cda61777-01e9-0310-a592-d414129be87e

17 years agoReturn a non-zero exit code in the event of a fatal-error SSH
simon [Sat, 26 Aug 2006 09:21:52 +0000 (09:21 +0000)]
Return a non-zero exit code in the event of a fatal-error SSH
session termination. `Close window only on clean exit' was not
working properly on Unix in the absence of this:
notify_remote_exit() was being called and ssh_return_exitcode was
returning zero, causing gtk_main_quit() to be called, _before_
connection_fatal() happened.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6801 cda61777-01e9-0310-a592-d414129be87e

17 years agoAlways initialise the `addresses' field of a SockAddr to NULL,
simon [Sat, 26 Aug 2006 08:37:42 +0000 (08:37 +0000)]
Always initialise the `addresses' field of a SockAddr to NULL,
because it gets unconditionally sfree()d in sk_addr_free(). This
just bit me when running under the MSVC debugger; not sure how it
hasn't bitten anyone until now!

git-svn-id: svn://svn.tartarus.org/sgt/putty@6800 cda61777-01e9-0310-a592-d414129be87e

17 years agoBug fix: since the input thread does not wait for the event object
simon [Sat, 26 Aug 2006 08:15:53 +0000 (08:15 +0000)]
Bug fix: since the input thread does not wait for the event object
until _after_ its first read, we should not start by signalling that
object in order to trigger the first read. Ahem.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6799 cda61777-01e9-0310-a592-d414129be87e

17 years agoSmall tweak to the new handle API: provide a `privdata' field in
simon [Sat, 26 Aug 2006 07:41:15 +0000 (07:41 +0000)]
Small tweak to the new handle API: provide a `privdata' field in
each handle structure, set on initialisation and readable by an API
call.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6798 cda61777-01e9-0310-a592-d414129be87e

17 years agoNew piece of Windows infrastructure: winhandl.c takes Plink's
simon [Fri, 25 Aug 2006 22:10:16 +0000 (22:10 +0000)]
New piece of Windows infrastructure: winhandl.c takes Plink's
thread-based approach to stdin and stdout, wraps it in a halfway
sensible API, and makes it a globally available service across all
network tools.

There is no direct functionality enhancement from this checkin:
winplink.c now talks to the new API instead of doing it all
internally, but does nothing different as a result.

However, this should lay the groundwork for several diverse pieces
of work in future: pipe-based ProxyCommand on Windows, a serial port
back end, and (hopefully) a pipe-based means of communicating with
Pageant, which should have sensible blocking behaviour and hence
permit asynchronous agent requests and decrypt-on-demand.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6797 cda61777-01e9-0310-a592-d414129be87e

17 years agoEr, ahem. Other half of r6788. :-/
simon [Tue, 15 Aug 2006 22:48:01 +0000 (22:48 +0000)]
Er, ahem. Other half of r6788. :-/

git-svn-id: svn://svn.tartarus.org/sgt/putty@6791 cda61777-01e9-0310-a592-d414129be87e

17 years agoShifts left and right by 32 were tripping a gcc warning (fatal with
simon [Tue, 15 Aug 2006 20:29:02 +0000 (20:29 +0000)]
Shifts left and right by 32 were tripping a gcc warning (fatal with
-Werror, of course) about shifting by more than the range of a data
type. They only appeared in `if' statements testing sizeof(off_t),
but gcc warns even when the code is unreachable. I've removed the
conditional code (the general case should still work even on 32-bit
machines), and hacked each shift by 32 into a pair of shifts by 16.

Note that the gcc warning is not just a helpful indication that you
may be using the wrong data type; it's actually pointing out ANSI-
undefined behaviour in shifting a signed integer beyond the size of
its type.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6789 cda61777-01e9-0310-a592-d414129be87e

17 years agoAha! At long last I've managed to reproduce the intermittent problem
simon [Tue, 15 Aug 2006 12:45:21 +0000 (12:45 +0000)]
Aha! At long last I've managed to reproduce the intermittent problem
I've been having with the cursor sometimes restoring to the wrong
place when screen(1) terminates. The offending sequence of escape
sequences goes ESC 7 (save cursor), ESC [?47h (switch to alternate
screen), ESC 7 (save cursor _again_), do some stuff, ESC 8 (restore
cursor), run screen session for a bit, ESC [?47l (return to main
screen), ESC 8 (restore cursor). The final ESC 8 is expected to
restore the cursor to where it was saved by the initial ESC 7.

Translation: the ESC 7 saved cursor state is part of the state we
must swap out when switching to the alternate screen. In other
words, we need to track _four_ cursor positions: active and saved,
on each of main and alternate screen. Previously we were tracking
only three.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6788 cda61777-01e9-0310-a592-d414129be87e

17 years agoLarge file support for psftp and pscp on both Windows and Unix. On Unix
owen [Sat, 12 Aug 2006 15:20:19 +0000 (15:20 +0000)]
Large file support for psftp and pscp on both Windows and Unix.  On Unix
we set _FILE_OFFSET_BITS to 64 on the compiler command line (via mkfiles.pl),
and on Windows we use SetFilePointer and GetFileSize to cope with 64-bit sizes
where possible.  Not tested on Win9x.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6783 cda61777-01e9-0310-a592-d414129be87e

17 years agoSome extra int64 functions.
owen [Sat, 5 Aug 2006 13:48:53 +0000 (13:48 +0000)]
Some extra int64 functions.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6778 cda61777-01e9-0310-a592-d414129be87e

17 years agoSpell out more explicitly what needs to be done with vanilla VC6, since it
jacob [Fri, 7 Jul 2006 20:56:22 +0000 (20:56 +0000)]
Spell out more explicitly what needs to be done with vanilla VC6, since it
seems to be becoming a FAQ.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6752 cda61777-01e9-0310-a592-d414129be87e

17 years agoRandom Unix puttygen improvements highlighted by a post to comp.security.ssh:
jacob [Fri, 7 Jul 2006 14:18:47 +0000 (14:18 +0000)]
Random Unix puttygen improvements highlighted by a post to comp.security.ssh:
 - fix diagnostic if keyfile and '-t' both specified
 - add diagnostic for generating a key but discarding the private part
 - document '-q' option

git-svn-id: svn://svn.tartarus.org/sgt/putty@6750 cda61777-01e9-0310-a592-d414129be87e

17 years agoIn the config dialog, clip RGB values to [0..255] rather than reducing them
jacob [Tue, 20 Jun 2006 21:10:33 +0000 (21:10 +0000)]
In the config dialog, clip RGB values to [0..255] rather than reducing them
mod 256.
Document that the RGB values can be edited, and their range.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6734 cda61777-01e9-0310-a592-d414129be87e

17 years agoWhen Unix PuTTYgen gives brief usage information, it should mention "--help"!
jacob [Sat, 17 Jun 2006 13:01:04 +0000 (13:01 +0000)]
When Unix PuTTYgen gives brief usage information, it should mention "--help"!

git-svn-id: svn://svn.tartarus.org/sgt/putty@6732 cda61777-01e9-0310-a592-d414129be87e

17 years agoRobert Evans spotted that bignum_decimal() failed to cope with being given
jacob [Sat, 17 Jun 2006 12:02:03 +0000 (12:02 +0000)]
Robert Evans spotted that bignum_decimal() failed to cope with being given
a zero input.
This shouldn't matter for PuTTY, as these routines are only used in PuTTYgen,
to output SSH-1 format public key exponents/moduli, which should be nonzero.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6731 cda61777-01e9-0310-a592-d414129be87e

17 years agoUpdate status of our PocketPC port.
jacob [Fri, 16 Jun 2006 13:09:47 +0000 (13:09 +0000)]
Update status of our PocketPC port.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6730 cda61777-01e9-0310-a592-d414129be87e

17 years agoDo not send raw mouse events in the middle of a selection operation, even if
jacob [Sun, 11 Jun 2006 12:56:52 +0000 (12:56 +0000)]
Do not send raw mouse events in the middle of a selection operation, even if
we otherwise would (for instance, if Shift is released before the mouse button
being used for selection).

git-svn-id: svn://svn.tartarus.org/sgt/putty@6727 cda61777-01e9-0310-a592-d414129be87e

17 years agoLionel Fourquaux offers this very simple patch to speed up SFTP,
simon [Fri, 2 Jun 2006 08:46:34 +0000 (08:46 +0000)]
Lionel Fourquaux offers this very simple patch to speed up SFTP,
simply by upping the packet sizes and maximum in-flight packet
count. Got to be worth a try, I think!

git-svn-id: svn://svn.tartarus.org/sgt/putty@6722 cda61777-01e9-0310-a592-d414129be87e

17 years agoMake it clearer that `psftp -b' doesn't return to the interactive
simon [Fri, 26 May 2006 12:45:21 +0000 (12:45 +0000)]
Make it clearer that `psftp -b' doesn't return to the interactive
prompt if the script doesn't end with an explicit `quit'.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6717 cda61777-01e9-0310-a592-d414129be87e

17 years agoPort r6710 from puzzles:
jacob [Sun, 21 May 2006 12:20:42 +0000 (12:20 +0000)]
Port r6710 from puzzles:
Patch from Ben Hutchings to prevent an ugly special case in
&splitline in which a line is `split' into a line ending in a
backslash followed by a completely blank line.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6714 cda61777-01e9-0310-a592-d414129be87e

18 years agoLink to new `key-formats-natively' wishlist item from appropriate FAQ.
jacob [Wed, 17 May 2006 12:06:08 +0000 (12:06 +0000)]
Link to new `key-formats-natively' wishlist item from appropriate FAQ.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6693 cda61777-01e9-0310-a592-d414129be87e

18 years agopty_init should put _something_ into realhost
owen [Fri, 12 May 2006 11:02:28 +0000 (11:02 +0000)]
pty_init should put _something_ into realhost

git-svn-id: svn://svn.tartarus.org/sgt/putty@6679 cda61777-01e9-0310-a592-d414129be87e

18 years agosbcsgen.pl was giving different results on different machines in the case
jacob [Wed, 26 Apr 2006 23:01:06 +0000 (23:01 +0000)]
sbcsgen.pl was giving different results on different machines in the case
where two SBCS code points mapped to a single Unicode point.
Changed so that by default it favours the lower SBCS code point.
On ixion, this highlighted ambiguities in CS_MAC_THAI, CS_MAC_SYMBOL, and
CS_VISCII. Guessed at a preference for the first two and added "sortpriority"
directives. (No idea about VISCII.)

git-svn-id: svn://svn.tartarus.org/sgt/putty@6641 cda61777-01e9-0310-a592-d414129be87e

18 years agoSprinkle some header comments in various files in an attempt to explain what
jacob [Sun, 23 Apr 2006 18:26:03 +0000 (18:26 +0000)]
Sprinkle some header comments in various files in an attempt to explain what
they're for.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6639 cda61777-01e9-0310-a592-d414129be87e

18 years agoPocketPuTTY website moved.
jacob [Sun, 16 Apr 2006 11:15:20 +0000 (11:15 +0000)]
PocketPuTTY website moved.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6638 cda61777-01e9-0310-a592-d414129be87e

18 years agoEverywhere we print an fxp_error(), try to make it clear what we were trying
jacob [Thu, 13 Apr 2006 21:18:09 +0000 (21:18 +0000)]
Everywhere we print an fxp_error(), try to make it clear what we were trying
to do at the time.
(A lot of these say just "canonify:". This isn't a nice thing to show to a
user, but I don't believe canonify() will ever return failure due to a server
error, so users shouldn't actually see it, and it means we have a chance of
tracing it if reported.)

git-svn-id: svn://svn.tartarus.org/sgt/putty@6636 cda61777-01e9-0310-a592-d414129be87e

18 years agoRemove login name prompt from PSFTP.
jacob [Fri, 7 Apr 2006 21:42:30 +0000 (21:42 +0000)]
Remove login name prompt from PSFTP.
ssh.c will prompt for a login name as required, and doing so in psftp.c
before we've even made a connection is incorrect wrt `bypass-ssh2-userauth'.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6634 cda61777-01e9-0310-a592-d414129be87e

18 years agoECCN / FIPS - we can't be arsed.
owen [Fri, 31 Mar 2006 09:26:59 +0000 (09:26 +0000)]
ECCN / FIPS - we can't be arsed.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6619 cda61777-01e9-0310-a592-d414129be87e

18 years agoUpdate size when going from maximised to full screen.
owen [Wed, 29 Mar 2006 17:55:40 +0000 (17:55 +0000)]
Update size when going from maximised to full screen.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6618 cda61777-01e9-0310-a592-d414129be87e

18 years agoExplicitly add linking to our website to the "asking permission for things"
jacob [Wed, 15 Mar 2006 14:31:39 +0000 (14:31 +0000)]
Explicitly add linking to our website to the "asking permission for things"
section.
Replace the FAQ text with a summary + link to feedback.but.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6613 cda61777-01e9-0310-a592-d414129be87e

18 years agoFix minor memory leak.
jacob [Tue, 14 Mar 2006 22:01:27 +0000 (22:01 +0000)]
Fix minor memory leak.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6611 cda61777-01e9-0310-a592-d414129be87e

18 years agoPageant docs didn't mention that you could load keys into an existing
jacob [Tue, 14 Mar 2006 11:21:59 +0000 (11:21 +0000)]
Pageant docs didn't mention that you could load keys into an existing
Pageant from the command line.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6610 cda61777-01e9-0310-a592-d414129be87e

18 years agoEquivalent of r6583 window-border palette-change fix for Gtk.
jacob [Sun, 12 Mar 2006 22:17:46 +0000 (22:17 +0000)]
Equivalent of r6583 window-border palette-change fix for Gtk.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6609 cda61777-01e9-0310-a592-d414129be87e

18 years agoDimitry Andric spotted that DH gex with SHA-256 was overflowing a buffer.
jacob [Sun, 12 Mar 2006 19:24:05 +0000 (19:24 +0000)]
Dimitry Andric spotted that DH gex with SHA-256 was overflowing a buffer.
Fixed, and added paranoia so that this shouldn't happen again.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6606 cda61777-01e9-0310-a592-d414129be87e

18 years agoLog the hash used for DH kex (now there's a choice).
jacob [Sun, 12 Mar 2006 15:39:19 +0000 (15:39 +0000)]
Log the hash used for DH kex (now there's a choice).

git-svn-id: svn://svn.tartarus.org/sgt/putty@6605 cda61777-01e9-0310-a592-d414129be87e

18 years agoFix inadvertent O(N^2) loop within do_paint() which I just
simon [Sun, 12 Mar 2006 14:53:18 +0000 (14:53 +0000)]
Fix inadvertent O(N^2) loop within do_paint() which I just
discovered when profiling IPBT.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6602 cda61777-01e9-0310-a592-d414129be87e

18 years agoFlesh out `-m' caveats slightly.
jacob [Wed, 8 Mar 2006 23:16:52 +0000 (23:16 +0000)]
Flesh out `-m' caveats slightly.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6595 cda61777-01e9-0310-a592-d414129be87e

18 years agoOh, and update the docs for NetHack keypad mode too.
simon [Wed, 8 Mar 2006 18:15:16 +0000 (18:15 +0000)]
Oh, and update the docs for NetHack keypad mode too.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6594 cda61777-01e9-0310-a592-d414129be87e

18 years agoDavid Damerell tells me I should be using Ctrl-hjklyubn rather than
simon [Wed, 8 Mar 2006 18:10:12 +0000 (18:10 +0000)]
David Damerell tells me I should be using Ctrl-hjklyubn rather than
Shift-hjklyubn for batch movement in NetHack, because they have
subtly different behaviour within the game and the Ctrl-moves are
more useful. Unfortunately, PuTTY's NetHack keypad mode doesn't
support Ctrl-moves. Therefore, it does now :-)

git-svn-id: svn://svn.tartarus.org/sgt/putty@6593 cda61777-01e9-0310-a592-d414129be87e

18 years agoFix a memory leak in key generation.
jacob [Mon, 27 Feb 2006 23:55:07 +0000 (23:55 +0000)]
Fix a memory leak in key generation.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6587 cda61777-01e9-0310-a592-d414129be87e

18 years agoChange XXX-REMOVE-BEFORE-RELEASE to XXX-REVIEW-BEFORE-RELEASE to allow
jacob [Mon, 27 Feb 2006 22:30:38 +0000 (22:30 +0000)]
Change XXX-REMOVE-BEFORE-RELEASE to XXX-REVIEW-BEFORE-RELEASE to allow
more general changes to be queued up.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6586 cda61777-01e9-0310-a592-d414129be87e

18 years agoBatch of miscellaneous tweaks to the Windows installer:
jacob [Mon, 27 Feb 2006 22:27:49 +0000 (22:27 +0000)]
Batch of miscellaneous tweaks to the Windows installer:
 - Now we've fixed `win-versioninfo', choose some sensible outcomes from
   the installer's comparisons of binary version numbers. Also, give the
   installer _itself_ a matching binary version.
   In particular, without this change, it would not have been possible
   to downgrade PuTTY -- it would have silently left the "newer" files in
   place. Now it will make some fuss, but permit it.
 - Also remove descriptions from shortcuts, on the grounds that the
   binaries have embedded descriptions now. (Although I've not checked
   whether those are actually visible in the Start Menu.)
 - At the request of various people (e.g., PJB), add flags so that if
   files are in use at the time the (un)installer is run, replacement is
   deferred to the next restart. (The user may be prompted to restart,
   which isn't ideal; see comments).
   This is supposed to make centrally-pushed silent upgrades more robust.
 - Note some limitations of the installer.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6585 cda61777-01e9-0310-a592-d414129be87e

18 years agoAlain Guibert points out that palette changes weren't causing the space
jacob [Sat, 25 Feb 2006 14:13:46 +0000 (14:13 +0000)]
Alain Guibert points out that palette changes weren't causing the space
between the text area and the window border to be refreshed.
Fixed on Windows. Gtk still has a similar problem.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6583 cda61777-01e9-0310-a592-d414129be87e

18 years agoDo proper select-for-write on ptys. Currently, pasting a
simon [Thu, 23 Feb 2006 13:38:44 +0000 (13:38 +0000)]
Do proper select-for-write on ptys. Currently, pasting a
sufficiently large string into pterm in any circumstances in which
it's echoed back to the terminal will cause a deadlock once the
pty's write buffer fills up.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6582 cda61777-01e9-0310-a592-d414129be87e

18 years agoAlain Guibert points out that ESC]P sequences were erroneously
simon [Mon, 20 Feb 2006 19:54:17 +0000 (19:54 +0000)]
Alain Guibert points out that ESC]P sequences were erroneously
accepting 'G' as a hex digit. (The _first_ digit of the sequence
intentionally goes up further than F, but the remaining ones
shouldn't have.)

git-svn-id: svn://svn.tartarus.org/sgt/putty@6581 cda61777-01e9-0310-a592-d414129be87e

18 years agoDaniel Meidlinger points out a redundant test :-)
simon [Sun, 19 Feb 2006 22:37:31 +0000 (22:37 +0000)]
Daniel Meidlinger points out a redundant test :-)

git-svn-id: svn://svn.tartarus.org/sgt/putty@6579 cda61777-01e9-0310-a592-d414129be87e

18 years agoFix `restart-reset-terminal': terminal now restored to a sensible state when
jacob [Sun, 19 Feb 2006 14:59:48 +0000 (14:59 +0000)]
Fix `restart-reset-terminal': terminal now restored to a sensible state when
reusing a window to restart a session.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6577 cda61777-01e9-0310-a592-d414129be87e

18 years agoAdd/tweak a couple of comments.
jacob [Sun, 19 Feb 2006 14:10:02 +0000 (14:10 +0000)]
Add/tweak a couple of comments.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6576 cda61777-01e9-0310-a592-d414129be87e

18 years agoFix up documentation/usage messages for r6572.
jacob [Sun, 19 Feb 2006 12:52:28 +0000 (12:52 +0000)]
Fix up documentation/usage messages for r6572.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6574 cda61777-01e9-0310-a592-d414129be87e

18 years agoIntroduce a new checkbox and command-line option to inhibit use of
simon [Sun, 19 Feb 2006 12:05:12 +0000 (12:05 +0000)]
Introduce a new checkbox and command-line option to inhibit use of
Pageant for local authentication. (This is a `don't use Pageant for
authentication at session startup' button rather than a `pretend
Pageant doesn't exist' button: that is, agent forwarding is
independent of this option.)

git-svn-id: svn://svn.tartarus.org/sgt/putty@6572 cda61777-01e9-0310-a592-d414129be87e

18 years agoAt last, a fix for `large-clipboard-crash'.
jacob [Sun, 19 Feb 2006 01:37:45 +0000 (01:37 +0000)]
At last, a fix for `large-clipboard-crash'.
A growable buffer was only being grown for actual text, not for newlines or
trailing NULs. A large run of empty lines could lead to newlines overflowing
the buffer (> 100 should be enough to guarantee this on all platforms, after
the initial 5k size of the buffer).

Also fix some valgrind in the same area (was probably harmless), and a memory
leak introduced by the RTF attribute pasting.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6570 cda61777-01e9-0310-a592-d414129be87e

18 years agoPointer type correction in term_clrsb().
jacob [Sat, 18 Feb 2006 22:30:10 +0000 (22:30 +0000)]
Pointer type correction in term_clrsb().
(Harmless, as it happens.)

git-svn-id: svn://svn.tartarus.org/sgt/putty@6568 cda61777-01e9-0310-a592-d414129be87e

18 years agoTypo in r6555 spotted by Daniel Meidlinger.
jacob [Tue, 14 Feb 2006 13:14:08 +0000 (13:14 +0000)]
Typo in r6555 spotted by Daniel Meidlinger.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6558 cda61777-01e9-0310-a592-d414129be87e

18 years agoRTF pasting now includes colours and style. Say so.
owen [Mon, 13 Feb 2006 22:25:59 +0000 (22:25 +0000)]
RTF pasting now includes colours and style.  Say so.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6557 cda61777-01e9-0310-a592-d414129be87e

18 years agoPreserve more attributes of text copied as RTF. Thanks to Stephen Balousek.
owen [Mon, 13 Feb 2006 22:18:17 +0000 (22:18 +0000)]
Preserve more attributes of text copied as RTF.  Thanks to Stephen Balousek.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6555 cda61777-01e9-0310-a592-d414129be87e

18 years agoDon Heap spotted that our heuristics for dealing with IPv6 literal addresses
jacob [Sat, 11 Feb 2006 19:10:01 +0000 (19:10 +0000)]
Don Heap spotted that our heuristics for dealing with IPv6 literal addresses
in the PSCP command line were bogus, giving "remote to remote not supported"
errors with filenames like '[].txt'. Made the heuristic less bogus.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6551 cda61777-01e9-0310-a592-d414129be87e

18 years agoFailure to connect to a Unix-domain socket could cause a segfault. Fixed.
jacob [Sat, 11 Feb 2006 18:29:55 +0000 (18:29 +0000)]
Failure to connect to a Unix-domain socket could cause a segfault. Fixed.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6550 cda61777-01e9-0310-a592-d414129be87e

18 years agoA zero-length return from platform_get_x_display() (for instance, a
jacob [Sat, 11 Feb 2006 18:00:23 +0000 (18:00 +0000)]
A zero-length return from platform_get_x_display() (for instance, a
zero-length DISPLAY variable in the environment) caused an assertion
failure when X11 forwarding was attempted. Fixed (now treated the same
as a NULL return, e.g., a non-existent DISPLAY variable).

git-svn-id: svn://svn.tartarus.org/sgt/putty@6549 cda61777-01e9-0310-a592-d414129be87e

18 years agoOops. Since r6546, old "Special Commands" menus weren't being deleted from the
jacob [Fri, 10 Feb 2006 20:57:40 +0000 (20:57 +0000)]
Oops. Since r6546, old "Special Commands" menus weren't being deleted from the
context menu, and they tended to pile up.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6547 cda61777-01e9-0310-a592-d414129be87e

18 years agoUsers of Virtual Dimension are reporting that the "Close" menu item and
jacob [Thu, 9 Feb 2006 23:06:57 +0000 (23:06 +0000)]
Users of Virtual Dimension are reporting that the "Close" menu item and
button tend to get disabled on login.
After a suggestion by "Tkil", change the way we handle the specials menu
to be robust against the window menu being externally modified.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6546 cda61777-01e9-0310-a592-d414129be87e

18 years agoSomewhat gruesome tweak to use SetClassLongPtr where available and degrade
jacob [Fri, 27 Jan 2006 20:49:59 +0000 (20:49 +0000)]
Somewhat gruesome tweak to use SetClassLongPtr where available and degrade
nicely elsewhere, which should fix `win64' _properly_.
Tested on recent-ish MinGW (with GetWindowLongPtr but not GetClassLongPtr),
and VC++ 6.0 with a recent SDK, but not with vanilla VC++.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6535 cda61777-01e9-0310-a592-d414129be87e

18 years agoDon't explicitly open the logfile on startup; it'll automatically be opened
owen [Wed, 25 Jan 2006 22:46:33 +0000 (22:46 +0000)]
Don't explicitly open the logfile on startup; it'll automatically be opened
by logwrite() as necessary.  Should fix win-askappend-multi.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6533 cda61777-01e9-0310-a592-d414129be87e

18 years agos/PuTTY/PuTTYgen/ in two places.
jacob [Tue, 24 Jan 2006 11:08:34 +0000 (11:08 +0000)]
s/PuTTY/PuTTYgen/ in two places.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6528 cda61777-01e9-0310-a592-d414129be87e

18 years agoVC6 doesn't define LONG_PTR
owen [Wed, 11 Jan 2006 23:43:04 +0000 (23:43 +0000)]
VC6 doesn't define LONG_PTR

git-svn-id: svn://svn.tartarus.org/sgt/putty@6520 cda61777-01e9-0310-a592-d414129be87e

18 years agoConfigurable font quality on Windows. (Together with a little bit of
owen [Wed, 11 Jan 2006 23:42:02 +0000 (23:42 +0000)]
Configurable font quality on Windows.  (Together with a little bit of
macro stuff to cope with the inadequacy of VC++ 6 headers.)

git-svn-id: svn://svn.tartarus.org/sgt/putty@6519 cda61777-01e9-0310-a592-d414129be87e

18 years agoFix zero-rows-cols-crash
owen [Tue, 10 Jan 2006 20:02:22 +0000 (20:02 +0000)]
Fix zero-rows-cols-crash

git-svn-id: svn://svn.tartarus.org/sgt/putty@6515 cda61777-01e9-0310-a592-d414129be87e

18 years agoIt's a new year.
jacob [Sun, 8 Jan 2006 18:18:11 +0000 (18:18 +0000)]
It's a new year.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6513 cda61777-01e9-0310-a592-d414129be87e

18 years agoUpdate status of this library wrt other variants.
jacob [Sun, 18 Dec 2005 17:05:21 +0000 (17:05 +0000)]
Update status of this library wrt other variants.

git-svn-id: svn://svn.tartarus.org/sgt/putty@6500 cda61777-01e9-0310-a592-d414129be87e