Nabble - Tcl

Question on coroutines

2009-12-05T14:25:15Z

I am working on fixing some of the missing things in coros.

For instance, as of today, HEAD can tailcall out of a coro.

The missing thing is to nre-enable coros: if you call
coroutine B $body {*}$arglist
from within a coroutine A, it would be good ig coroutine A would remove
itself from the callstacks before running coroutine B.

I now have a good idea as to how to do this ... sorry it took so long.
It should be in HEAD in the next few days.

The same tech makes possible a new command [yieldTo]: essentially
[tailcall] but yielding instead of returning - that is, saving the state
of the current coro, so that it can be reawakened later on. I am
planning to put this into ::tcl::unsupported.

I think that this command could find very interesting applications. The
big Q is: shall I try to TIP it as an amendment to TIP #328? I DO know
it is rather late ...

Cheers
Miguel

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Core mailing list
Tcl-Core@...
https://lists.sourceforge.net/lists/listinfo/tcl-core

From forum: tcl-core

tcl-cvs: "tcl/tests tailcall.test" modified

2009-12-05T14:05:30Z

Update of /cvsroot/tcl/tcl/tests
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv21719/tests

Modified Files:
tailcall.test
Log Message:
* tests/tailcall.test: remove some old unused crud; improved the
stack depth tests.

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

tcl-cvs: "tcl/tests tailcall.test" modified

2009-12-05T13:30:06Z

Update of /cvsroot/tcl/tcl/tests
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv15632/tests

Modified Files:
tailcall.test
Log Message:
* generic/tclBasic.c: Fixed things so that you can tailcall
* generic/tclNamesp.c: properly out of a coroutine.
* tests/tailcall.test:

* generic/tclInterp.c: Fixed tailcalls for same-interp aliases (no
test)

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

tcl-cvs: "tcl/generic tclBasic.c tclInterp.c tclNamesp.c" modified

2009-12-05T13:30:06Z

Update of /cvsroot/tcl/tcl/generic
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv15632/generic

Modified Files:
tclBasic.c tclInterp.c tclNamesp.c
Log Message:
* generic/tclBasic.c: Fixed things so that you can tailcall
* generic/tclNamesp.c: properly out of a coroutine.
* tests/tailcall.test:

* generic/tclInterp.c: Fixed tailcalls for same-interp aliases (no
test)

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

tcl-cvs: "tcl ChangeLog" modified

2009-12-05T13:30:05Z

Update of /cvsroot/tcl/tcl
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv15632

Modified Files:
ChangeLog
Log Message:
* generic/tclBasic.c: Fixed things so that you can tailcall
* generic/tclNamesp.c: properly out of a coroutine.
* tests/tailcall.test:

* generic/tclInterp.c: Fixed tailcalls for same-interp aliases (no
test)

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

[ tktoolkit-Bugs-2864685 ] Compiz animates each menu item as a new window

2009-12-05T09:21:14Z

Bugs item #2864685, was opened at 2009-09-22 21:06
Message generated for change (Comment added) made by egavilan
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=112997&aid=2864685&group_id=12997

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: 12. Unix Menus
Group: obsolete: 8.5.7
Status: Open
Resolution: None
Priority: 9
Private: No
Submitted By: Andras Muranyi (muranyia)
Assigned to: Donal K. Fellows (dkf)
Summary: Compiz animates each menu item as a new window

Initial Comment:
Compiz somehow thinks that the menus of these apps are separate windows and applies window open and close animation to them. Thus each submenu "pulses" like ordinary window when opened and navigating a menu (specially one with zero delay) looks like a firework.

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

Comment By: Emiliano (egavilan)
Date: 2009-12-05 15:21

Message:
Apparently new window managers expects additional information to be set
about what the function of the window is.
More precisely, it expects the _NET_WM_WINDOW_TYPE property to be set to
_NET_WM_WINDOW_TYPE_MENU for menu windows. See
http://standards.freedesktop.org/wm-spec/1.3/ar01s05.html .
A patch against HEAD (5 dec 2009) adds this information to menu windows on
x11.
As a result, the new property is set for menu windows as shown by xprop
output:

_NET_WM_WINDOW_TYPE(ATOM) = _NET_WM_WINDOW_TYPE_MENU

For the patch, see http://paste.tclers.tk/1880

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-11-29 16:43

Message:
The menu code is very scattered over different files, but the key part
appears to be in tkUnixWm.c in the function to create the actual X window
for the menu (TkpMakeMenuWindow). If I'm reading that function right, the
override_redirect flag is not being set on the menu until after the window
is created, and maybe compiz is not noticing that. Can't tell, as I'm not a
user of the right platform...

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

You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=112997&aid=2864685&group_id=12997

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

[ tktoolkit-Bugs-2768586 ] Menu posting problem on dual monitors

2009-12-05T05:30:11Z

Bugs item #2768586, was opened at 2009-04-16 14:23
Message generated for change (Comment added) made by patthoyts
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=112997&aid=2768586&group_id=12997

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: 13. Win Menus
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Koen Danckaert (danckaert)
>Assigned to: Pat Thoyts (patthoyts)
Summary: Menu posting problem on dual monitors

Initial Comment:
This bug happens on Windows in dual monitor mode, when the second monitor is placed above the primary monitor. Note that this means that the second monitor has negative y coordinates.
When posting a menu on the second monitor (either from a menubutton or directly with tk_popup), the menu is actually displayed on the primary monitor.

The reason can be found in the procedures ::tk::MbPost and ::tk::PostOverPoint. These contains checks like:

if {$y < 0} {
set y 0
}

Which makes it impossible to post menus at negative y coordinates.

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

>Comment By: Pat Thoyts (patthoyts)
Date: 2009-12-05 13:30

Message:
I thought I'd fixed all the multiple-monitor placement issues years ago
with fixes to tk::PlaceWindow. Evidently this got missed. I'll look into
this from work where I've a dual monitor setup.
Wrt to finding the coord space - post any suggestions. If you are running
into this then you at least have some idea of the usage for such a feature.
As I've never run into this as a problem I'm in a worse position to design
the feature. There is an (ancient) bug raised about finding out about
multiple monitors but the developer for that area is clearly not going to
work it so anyone with an interest needs to get involved.

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

Comment By: Tim Davis (aspenlogic)
Date: 2009-11-03 16:02

Message:
Can't Tk assume that if the upper left corner of the application window "."
is in negative coordinates that the menus can be in negative coordinates?
Also, don't post the menu any higher then the top edge of the "." window as
a workaround to the lack of coordinate space API.

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

Comment By: Koen Danckaert (danckaert)
Date: 2009-11-02 09:05

Message:
I've a workaround (see attached file) which works by always posting the
menu at the mouse pointer. Note that this means that "-direction flush" for
option-menus does not work as intended anymore. But at least the menu gets
posted on the right monitor!
For a real fix, we need to know the available coordinate space, and
apparently there is no API in Tk for doing that.

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

Comment By: Tim Davis (aspenlogic)
Date: 2009-10-30 19:09

Message:
I just ran into this problem as well. Is there a work around? I put a large
monitor, connected to my docking station, above my laptop display. I
suspect this is a common arrangement (based on walking around cubes) so
whoever deploys Tk apps to work across multiple monitors will get dinged
for this bug.

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

You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=112997&aid=2768586&group_id=12997

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

Re: NSCFString error in Tk-Cocoa

2009-12-05T01:48:52Z

Hi,

> I have a 10.5.8 user reporting an uncaught Objective-C exception in one
> of my programs that runs on Tk-Cocoa 8.5.7. The error log causes my
> application to crash, with the messages below logged to Console:

> The error occurs when the user is typing a password into a ttk::entry
> field. The relevant code is below:
>
> ttk::entry .password.top.entry -show \u2022 -textvariable
> authorize::password

This sounds like what I experienced when entering a circumflex (^) in an entry on Tk 8.6b1.1.
It works on 8.5.8 though. I have filed a bug about this:

http://sourceforge.net/tracker/?func=detail&atid=112997&aid=2907388&group_id=12997

Torsten
------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-mac mailing list
tcl-mac@...
https://lists.sourceforge.net/lists/listinfo/tcl-mac

From forum: tcl-mac

[ tcl-Bugs-2906841 ] calling [glob ../*] in an safe interp returns weird things

2009-12-04T08:37:55Z

Bugs item #2906841, was opened at 2009-12-01 14:21
Message generated for change (Comment added) made by dkf
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=110894&aid=2906841&group_id=10894

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: 33. Safe Base
Group: development: 8.6b1.1
Status: Open
Resolution: Fixed
Priority: 9
Private: No
Submitted By: Johannes13 (john13a)
Assigned to: Donal K. Fellows (dkf)
Summary: calling [glob ../*] in an safe interp returns weird things

Initial Comment:
8.5.7 - Windows
% safe::interpCreate test
test
% test eval glob ../*
{../All Users} ../Johannes

8.6b1.1 - Windows
% safe::interpCreate test
test
% test eval glob ../*
can't read "msg": no such variable
% set errorInfo
can't read "msg": no such variable
while executing
"Log $slave "GLOB @ $msg" NOTICE"
(procedure "::safe::AliasGlob" line 65)
invoked from within
"glob ../*"
invoked from within
"test eval glob ../*"

Some results from others
<dkf> $ tclsh8.5
<dkf> % safe::interpCreate test
<dkf> test
<dkf> % test eval glob ../*
<dkf> invalid command name "glob"

<emiliano> emiliano@merlot:~$ tclsh8.6
<emiliano> % safe::interpCreate test
<emiliano> test
<emiliano> % test eval glob ../*
<emiliano> ../ftp ../emiliano
<emiliano> HEAD (two or three days ago), linux

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

>Comment By: Donal K. Fellows (dkf)
Date: 2009-12-04 16:37

Message:
Correction, we pass -nocomplain through, but if we've got it, we fail
noisily in some cases where we shouldn't (i.e., with paths that are
non-readable by policy).

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-04 09:27

Message:
Notable issues:
1. we check the second -directory option
2. we don't handle -nocomplain
3. we get confused by glob metachars in the path part of a pattern

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-04 09:26

Message:
More work required. Currently we handle many things wrongly anyway

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-03 15:51

Message:
Fixed for 8.6 HEAD I think, but needs review and backport (can't just use
patch; 8.6 version uses [try]).

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-03 14:09

Message:
OK, with digging around I find that this security bug was introduced with
the fix for [Bug 1999119] (enabling Tcl Module support) which corresponds
to the 1.16->1.17 or 1.16->1.16.4.1 change.

The issue is that AliasGlob does not check for directory components in the
pattern arguments to [glob], let alone work out whether those components
refer to to locations that are "bad". :-(

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-02 15:52

Message:
Correction, that's with 1.16.4.2

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-02 15:48

Message:
With safe.tcl 1.33 in the tip of the 8.5 branch, I get globbing where I
wouldn't expect it:

bash$ make -C unix shell
DYLD_LIBRARY_PATH="`pwd`:${DYLD_LIBRARY_PATH}"; export DYLD_LIBRARY_PATH;
\
TCL_LIBRARY="/Users/dkf/Documents/software/tcl8.5/library"; export
TCL_LIBRARY; \
./tclsh
% safe::interpCreate test
test
% test eval glob ../*
../ChangeLog ../ChangeLog.1999 ../ChangeLog.2000 ../ChangeLog.2001
../ChangeLog.2002 ../ChangeLog.2003 ../ChangeLog.2004 ../ChangeLog.2005
../ChangeLog.~1.3975.2.269.~ ../changes ../compat ../CVS ../doc ../generic
../library ../libtommath ../license.terms ../macosx ../README ../tests
../tools ../unix ../win

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

Comment By: Andreas Kupries (andreas_kupries)
Date: 2009-12-01 17:52

Message:
Definitely the old buggy code I fixed on 2009-10-05 (The first fixed
revision was rev 1.19).

http://tcl.cvs.sourceforge.net/viewvc/tcl/tcl/library/safe.tcl?r1=1.18&r2=1.19

You need a newer Tcl to get the fix.

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

Comment By: Johannes13 (john13a)
Date: 2009-12-01 17:31

Message:
See uploaded files. :)

8.5.7: # RCS: @(#) $Id: safe.tcl,v 1.16.4.1 2008/06/25 16:42:05
andreas_kupries Exp $
8.6b1.1 # RCS: @(#) $Id: safe.tcl,v 1.18 2009/07/26 11:40:24 dkf Exp $

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

Comment By: Andreas Kupries (andreas_kupries)
Date: 2009-12-01 17:15

Message:
Hm. What is your revision of safe.tcl ?
(Installed under [info library], see the RCS comment line at the top).

Mine is CVS head as of the time of this comment, revision 1.33, and I do
not see the problem/error.

The relevant code in my file safe.tcl is

try {
::interp invokehidden $slave glob {*}$cmd
* } on ok msg {
* # Nothing to be done, just capture the 'msg' for later.
} on error msg {
Log $slave $msg
return -code error "script error"
}

Log $slave "GLOB @ $msg" NOTICE

and the lines I marked with * are presumably missing in your revision ...
Checking the ChangeLog I also find

2009-10-05 Andreas Kupries <andreask@...>

* library/safe.tcl (AliasGlob): Fixed conversion of catch to
try/finally, it had an 'on ok msg' branch missing, causing a
silent error immediately, and bogus glob results, breaking
search for Tcl modules.

which looks to me is the problem you are running into. MY guess is that
your Tcl is older than that entry.

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

Comment By: Andreas Kupries (andreas_kupries)
Date: 2009-12-01 17:00

Message:
I believe that I introduced this problem when I reworked the safebase to be
more uptodate in its use of Tcl's features. Reassigning to myself.

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

You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=110894&aid=2906841&group_id=10894

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

[ tcl-Bugs-2906841 ] calling [glob ../*] in an safe interp returns weird things

2009-12-04T01:27:39Z

Bugs item #2906841, was opened at 2009-12-01 14:21
Message generated for change (Comment added) made by dkf
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=110894&aid=2906841&group_id=10894

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: 33. Safe Base
Group: development: 8.6b1.1
Status: Open
Resolution: Fixed
Priority: 9
Private: No
Submitted By: Johannes13 (john13a)
Assigned to: Donal K. Fellows (dkf)
Summary: calling [glob ../*] in an safe interp returns weird things

Initial Comment:
8.5.7 - Windows
% safe::interpCreate test
test
% test eval glob ../*
{../All Users} ../Johannes

8.6b1.1 - Windows
% safe::interpCreate test
test
% test eval glob ../*
can't read "msg": no such variable
% set errorInfo
can't read "msg": no such variable
while executing
"Log $slave "GLOB @ $msg" NOTICE"
(procedure "::safe::AliasGlob" line 65)
invoked from within
"glob ../*"
invoked from within
"test eval glob ../*"

Some results from others
<dkf> $ tclsh8.5
<dkf> % safe::interpCreate test
<dkf> test
<dkf> % test eval glob ../*
<dkf> invalid command name "glob"

<emiliano> emiliano@merlot:~$ tclsh8.6
<emiliano> % safe::interpCreate test
<emiliano> test
<emiliano> % test eval glob ../*
<emiliano> ../ftp ../emiliano
<emiliano> HEAD (two or three days ago), linux

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

>Comment By: Donal K. Fellows (dkf)
Date: 2009-12-04 09:27

Message:
Notable issues:
1. we check the second -directory option
2. we don't handle -nocomplain
3. we get confused by glob metachars in the path part of a pattern

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-04 09:26

Message:
More work required. Currently we handle many things wrongly anyway

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-03 15:51

Message:
Fixed for 8.6 HEAD I think, but needs review and backport (can't just use
patch; 8.6 version uses [try]).

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-03 14:09

Message:
OK, with digging around I find that this security bug was introduced with
the fix for [Bug 1999119] (enabling Tcl Module support) which corresponds
to the 1.16->1.17 or 1.16->1.16.4.1 change.

The issue is that AliasGlob does not check for directory components in the
pattern arguments to [glob], let alone work out whether those components
refer to to locations that are "bad". :-(

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-02 15:52

Message:
Correction, that's with 1.16.4.2

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-02 15:48

Message:
With safe.tcl 1.33 in the tip of the 8.5 branch, I get globbing where I
wouldn't expect it:

bash$ make -C unix shell
DYLD_LIBRARY_PATH="`pwd`:${DYLD_LIBRARY_PATH}"; export DYLD_LIBRARY_PATH;
\
TCL_LIBRARY="/Users/dkf/Documents/software/tcl8.5/library"; export
TCL_LIBRARY; \
./tclsh
% safe::interpCreate test
test
% test eval glob ../*
../ChangeLog ../ChangeLog.1999 ../ChangeLog.2000 ../ChangeLog.2001
../ChangeLog.2002 ../ChangeLog.2003 ../ChangeLog.2004 ../ChangeLog.2005
../ChangeLog.~1.3975.2.269.~ ../changes ../compat ../CVS ../doc ../generic
../library ../libtommath ../license.terms ../macosx ../README ../tests
../tools ../unix ../win

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

Comment By: Andreas Kupries (andreas_kupries)
Date: 2009-12-01 17:52

Message:
Definitely the old buggy code I fixed on 2009-10-05 (The first fixed
revision was rev 1.19).

http://tcl.cvs.sourceforge.net/viewvc/tcl/tcl/library/safe.tcl?r1=1.18&r2=1.19

You need a newer Tcl to get the fix.

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

Comment By: Johannes13 (john13a)
Date: 2009-12-01 17:31

Message:
See uploaded files. :)

8.5.7: # RCS: @(#) $Id: safe.tcl,v 1.16.4.1 2008/06/25 16:42:05
andreas_kupries Exp $
8.6b1.1 # RCS: @(#) $Id: safe.tcl,v 1.18 2009/07/26 11:40:24 dkf Exp $

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

Comment By: Andreas Kupries (andreas_kupries)
Date: 2009-12-01 17:15

Message:
Hm. What is your revision of safe.tcl ?
(Installed under [info library], see the RCS comment line at the top).

Mine is CVS head as of the time of this comment, revision 1.33, and I do
not see the problem/error.

The relevant code in my file safe.tcl is

try {
::interp invokehidden $slave glob {*}$cmd
* } on ok msg {
* # Nothing to be done, just capture the 'msg' for later.
} on error msg {
Log $slave $msg
return -code error "script error"
}

Log $slave "GLOB @ $msg" NOTICE

and the lines I marked with * are presumably missing in your revision ...
Checking the ChangeLog I also find

2009-10-05 Andreas Kupries <andreask@...>

* library/safe.tcl (AliasGlob): Fixed conversion of catch to
try/finally, it had an 'on ok msg' branch missing, causing a
silent error immediately, and bogus glob results, breaking
search for Tcl modules.

which looks to me is the problem you are running into. MY guess is that
your Tcl is older than that entry.

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

Comment By: Andreas Kupries (andreas_kupries)
Date: 2009-12-01 17:00

Message:
I believe that I introduced this problem when I reworked the safebase to be
more uptodate in its use of Tcl's features. Reassigning to myself.

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

You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=110894&aid=2906841&group_id=10894

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

[ tcl-Bugs-2906841 ] calling [glob ../*] in an safe interp returns weird things

2009-12-04T01:26:01Z

Bugs item #2906841, was opened at 2009-12-01 14:21
Message generated for change (Comment added) made by dkf
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=110894&aid=2906841&group_id=10894

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: 33. Safe Base
Group: development: 8.6b1.1
Status: Open
Resolution: Fixed
Priority: 9
Private: No
Submitted By: Johannes13 (john13a)
Assigned to: Donal K. Fellows (dkf)
Summary: calling [glob ../*] in an safe interp returns weird things

Initial Comment:
8.5.7 - Windows
% safe::interpCreate test
test
% test eval glob ../*
{../All Users} ../Johannes

8.6b1.1 - Windows
% safe::interpCreate test
test
% test eval glob ../*
can't read "msg": no such variable
% set errorInfo
can't read "msg": no such variable
while executing
"Log $slave "GLOB @ $msg" NOTICE"
(procedure "::safe::AliasGlob" line 65)
invoked from within
"glob ../*"
invoked from within
"test eval glob ../*"

Some results from others
<dkf> $ tclsh8.5
<dkf> % safe::interpCreate test
<dkf> test
<dkf> % test eval glob ../*
<dkf> invalid command name "glob"

<emiliano> emiliano@merlot:~$ tclsh8.6
<emiliano> % safe::interpCreate test
<emiliano> test
<emiliano> % test eval glob ../*
<emiliano> ../ftp ../emiliano
<emiliano> HEAD (two or three days ago), linux

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

>Comment By: Donal K. Fellows (dkf)
Date: 2009-12-04 09:26

Message:
More work required. Currently we handle many things wrongly anyway

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-03 15:51

Message:
Fixed for 8.6 HEAD I think, but needs review and backport (can't just use
patch; 8.6 version uses [try]).

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-03 14:09

Message:
OK, with digging around I find that this security bug was introduced with
the fix for [Bug 1999119] (enabling Tcl Module support) which corresponds
to the 1.16->1.17 or 1.16->1.16.4.1 change.

The issue is that AliasGlob does not check for directory components in the
pattern arguments to [glob], let alone work out whether those components
refer to to locations that are "bad". :-(

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-02 15:52

Message:
Correction, that's with 1.16.4.2

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-02 15:48

Message:
With safe.tcl 1.33 in the tip of the 8.5 branch, I get globbing where I
wouldn't expect it:

bash$ make -C unix shell
DYLD_LIBRARY_PATH="`pwd`:${DYLD_LIBRARY_PATH}"; export DYLD_LIBRARY_PATH;
\
TCL_LIBRARY="/Users/dkf/Documents/software/tcl8.5/library"; export
TCL_LIBRARY; \
./tclsh
% safe::interpCreate test
test
% test eval glob ../*
../ChangeLog ../ChangeLog.1999 ../ChangeLog.2000 ../ChangeLog.2001
../ChangeLog.2002 ../ChangeLog.2003 ../ChangeLog.2004 ../ChangeLog.2005
../ChangeLog.~1.3975.2.269.~ ../changes ../compat ../CVS ../doc ../generic
../library ../libtommath ../license.terms ../macosx ../README ../tests
../tools ../unix ../win

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

Comment By: Andreas Kupries (andreas_kupries)
Date: 2009-12-01 17:52

Message:
Definitely the old buggy code I fixed on 2009-10-05 (The first fixed
revision was rev 1.19).

http://tcl.cvs.sourceforge.net/viewvc/tcl/tcl/library/safe.tcl?r1=1.18&r2=1.19

You need a newer Tcl to get the fix.

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

Comment By: Johannes13 (john13a)
Date: 2009-12-01 17:31

Message:
See uploaded files. :)

8.5.7: # RCS: @(#) $Id: safe.tcl,v 1.16.4.1 2008/06/25 16:42:05
andreas_kupries Exp $
8.6b1.1 # RCS: @(#) $Id: safe.tcl,v 1.18 2009/07/26 11:40:24 dkf Exp $

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

Comment By: Andreas Kupries (andreas_kupries)
Date: 2009-12-01 17:15

Message:
Hm. What is your revision of safe.tcl ?
(Installed under [info library], see the RCS comment line at the top).

Mine is CVS head as of the time of this comment, revision 1.33, and I do
not see the problem/error.

The relevant code in my file safe.tcl is

try {
::interp invokehidden $slave glob {*}$cmd
* } on ok msg {
* # Nothing to be done, just capture the 'msg' for later.
} on error msg {
Log $slave $msg
return -code error "script error"
}

Log $slave "GLOB @ $msg" NOTICE

and the lines I marked with * are presumably missing in your revision ...
Checking the ChangeLog I also find

2009-10-05 Andreas Kupries <andreask@...>

* library/safe.tcl (AliasGlob): Fixed conversion of catch to
try/finally, it had an 'on ok msg' branch missing, causing a
silent error immediately, and bogus glob results, breaking
search for Tcl modules.

which looks to me is the problem you are running into. MY guess is that
your Tcl is older than that entry.

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

Comment By: Andreas Kupries (andreas_kupries)
Date: 2009-12-01 17:00

Message:
I believe that I introduced this problem when I reworked the safebase to be
more uptodate in its use of Tcl's features. Reassigning to myself.

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

You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=110894&aid=2906841&group_id=10894

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

Re: Inconsistent buffered I/O

2009-12-03T18:57:03Z

Alex,

Okay, I did the additional testing. Instead of logging to stderr, I
just put a select timer for 100,000 microseconds into WillRead &
WillWrite.

Here are the results if anyone is interested:

http://rmadilo.com/files/tcl-io/testio2/TIMINGS.TXT

The maximum number of calls to the two helpers is 10, so the max
sleepy time is 1 sec.

The question to be answered by the test was if the Seek or Flush
executed at any other time than when necessary.

Answer: never happens. Perfect score.

The patch fixes the problem exactly and should not have any detectable
effect on existing "bug-free" code which uses [read], [puts] or
[gets].

Code which uses the workarounds will probably run slightly faster if
the workarounds are removed.

tom jackson

On Thu, Dec 3, 2009 at 10:24 AM, Alexandre Ferrieux
<alexandre.ferrieux@...> wrote:

> On 12/3/09, Tom Jackson <tom.rmadilo@...> wrote:
>> Alex,
>>
>> I just tested the patch. All 144 configurations tested give the
>> expected result.
>
> Ha !
>
> So, is the patch "OK for you" now ?
>
> -Alex
>

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Core mailing list
Tcl-Core@...
https://lists.sourceforge.net/lists/listinfo/tcl-core

From forum: tcl-core

[ tktoolkit-Bugs-2908052 ] TkAqua crashes on scrolling event

2009-12-03T18:45:51Z

Bugs item #2908052, was opened at 2009-12-03 02:52
Message generated for change (Comment added) made by das
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=112997&aid=2908052&group_id=12997

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: 66. Aqua Window Operations
Group: development: 8.6b1.1
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Torsten Berg (treincke)
Assigned to: Daniel A. Steffen (das)
Summary: TkAqua crashes on scrolling event

Initial Comment:
The 8.6b1.1 version of TkAqua crashes when scrolling in a certain way. Holding the mouse button pressed on a scrolling arrow will cause wish to crash with the following message:

wish8.6-aqua(1797,0xb0103000) malloc: free_garbage: garbage ptr = 0x1418c80, has non-zero refcount = 1
2009-12-03 11:32:32.995 wish8.6-aqua[1797:10b] *** Assertion failure in +[NSEvent startPeriodicEventsAfterDelay:withPeriod:], /SourceCache/AppKit/AppKit-949.54/AppKit.subproj/NSEvent.m:1765
2009-12-03 11:32:32.995 wish8.6-aqua[1797:10b] An uncaught exception was raised
2009-12-03 11:32:32.997 wish8.6-aqua[1797:10b] Periodic events are already being generated
2009-12-03 11:32:32.997 wish8.6-aqua[1797:10b] *** Terminating app due to uncaught exception 'NSInternalInconsistencyException', reason: 'Periodic events are already being generated'
2009-12-03 11:32:32.998 wish8.6-aqua[1797:10b] Stack: (
2470170603,
2514165307,
2470170059,
2478157508,
2525052112,
2526164841,
2524622328,
2523486967,
2523276965,
2195129,
2196449,
3235242,
1389874,
2705113,
2707855,
3011097,
2707855,
2737327,
2738193,
2198337,
2523500175,
2523499980,
2524627019,
2526164928,
2524622328,
2523486967,
2523276965,
2195129,
2196449,
3235242,
1421705,
1483183,
25904,
25689
)
Trace/BPT trap

The same is true, when scrolling by holding the mouse pressed on the slider trough (tested on MacOS X 10.5 and 10.6).

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

>Comment By: Daniel A. Steffen (das)
Date: 2009-12-03 18:45

Message:
I cannot reproduce this with 8.5.8 or HEAD on 10.6.2, please verify that
you still see this there.

If yes, please provide more details or a sample script, maybe this depends
on how your scrollbars were created?

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

You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=112997&aid=2908052&group_id=12997

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

Re: Inconsistent buffered I/O

2009-12-03T16:41:17Z

The tests included 40 configurations which could be considered
work-arounds: they worked! There were 36 which included at least flush
before read and seek before write. There were four where
buffering=none which only included a seek prior to write.

All passed.

Only thing I would like to test is to add a log message to the two new
helper API to see when they fire in relation to the different cases.

I'm hoping they don't fire at all for the work-around cases. The only
thing this would prove is that performance would be at a maximum.
Although I guess it would also prove is that no code which use
read/write will perform _any_ differently as viewed from outside, no
new flush or seek calls.

But the patch looks very good.

If anyone has example code for logging to stderr from a tcl library
function, please send me an email.

tom jackson

On Thu, Dec 3, 2009 at 10:17 AM, Barry Skidmore <atri@...> wrote:

> Just out of curiosity of the expected result.
>
> Your tests have included the workaround code against the patched source yes?
>
> Effectively proving that any code written to work around this issue will continue to function correctly with the patched version?
>
> On Dec 3, 2009, at 10:13 AM, Tom Jackson wrote:
>
>> Alex,
>>
>> I just tested the patch. All 144 configurations tested give the
>> expected result.
>>
>> Here is a link to the test results (in html table form):
>>
>> http://rmadilo.com/files/tcl-io/testio2/RESULTS.html
>>
>> Note that these tests include 16 tests with no flush/seek prior to
>> read or write:
>> (r+, w+) x (none, full) x (buffsiz 10, 20, 1024, and no -buffersize option)
>>
>> tom jackson
>>
>> On Wed, Dec 2, 2009 at 11:44 PM, Alexandre Ferrieux
>> <alexandre.ferrieux@...> wrote:
>>> On 12/3/09, Tom Jackson <tom.rmadilo@...> wrote:
>>>>
>>>> Is the patch against the HEAD on sf?
>>>
>>> Yaeh of course. Why don't you go fetch it and test at last ?
>>>
>>> -Alex
>>>
>>
>> ------------------------------------------------------------------------------
>> Join us December 9, 2009 for the Red Hat Virtual Experience,
>> a free event focused on virtualization and cloud computing.
>> Attend in-depth sessions from your desk. Your couch. Anywhere.
>> http://p.sf.net/sfu/redhat-sfdev2dev
>> _______________________________________________
>> Tcl-Core mailing list
>> Tcl-Core@...
>> https://lists.sourceforge.net/lists/listinfo/tcl-core
>
>

From forum: tcl-core

tk-cvs: "tk/win ttkWinXPTheme.c" modified

2009-12-03T15:58:35Z

Update of /cvsroot/tktoolkit/tk/win
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv18178/win

Modified Files:
Tag: core-8-5-branch
ttkWinXPTheme.c
Log Message:
Fix the selection of treeview rows on Windows XP and Vista

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

tk-cvs: "tk/library/ttk vistaTheme.tcl xpTheme.tcl" modified

2009-12-03T15:58:35Z

Update of /cvsroot/tktoolkit/tk/library/ttk
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv18178/library/ttk

Modified Files:
Tag: core-8-5-branch
vistaTheme.tcl xpTheme.tcl
Log Message:
Fix the selection of treeview rows on Windows XP and Vista

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

tk-cvs: "tk ChangeLog" modified

2009-12-03T15:58:35Z

Update of /cvsroot/tktoolkit/tk
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv18178

Modified Files:
Tag: core-8-5-branch
ChangeLog
Log Message:
Fix the selection of treeview rows on Windows XP and Vista

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

[ tktoolkit-Bugs-2908379 ] ttk:* widgets - selection text illegible on Mac OS X

2009-12-03T13:29:41Z

Bugs item #2908379, was opened at 2009-12-03 12:55
Message generated for change (Settings changed) made by das
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=112997&aid=2908379&group_id=12997

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: 88. Themed Tk
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Joe English (jenglish)
>Assigned to: Joe English (jenglish)
Summary: ttk:* widgets - selection text illegible on Mac OS X

Initial Comment:
peterc reports in #2908368:

Selected text on Mac needs have the font as a light colour (white?) rather
than black, otherwise it's unreadable: eg,

http://pc.autons.net/stuff/ttk_treeview_unreadable_macosx.png

and reports on the chat that ttk::entry widgets also have the same problem.

I can't tell which (if any) of the kThemeTextColor* constants in Appearance.h is the right one to use. das?

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

>Comment By: Daniel A. Steffen (das)
Date: 2009-12-03 13:29

Message:
there doesn't seem to be an appearance text color for the foreground, so
you probably need to use NSColor (and do the reverse conversion from what
TkMacOSXGetNSColor() is doing if you need a CGColor)
i.e. use [NSColor selectedTextColor] and [NSColor
selectedTextBackgroundColor]

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

You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=112997&aid=2908379&group_id=12997

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

Re: TCL 8.5 and MacOS 10.3.9

2009-12-03T12:56:05Z

Dear Jasper,

Jasper Taylor wrote:
> You need to give the compiler switch:
> -mmacosx-version-min=10.3

Thanks Jasper. I'll give that a shot. I'm still working on 10.4, so
should be able to compile for all.

Yours, Kevan

--
Kevan Hashemi, Electrical Engineer
Physics Department, Brandeis University
http://alignment.hep.brandeis.edu/

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-mac mailing list
tcl-mac@...
https://lists.sourceforge.net/lists/listinfo/tcl-mac

From forum: tcl-mac

[ tktoolkit-Bugs-2908379 ] ttk:* widgets - selection text illegible on Mac OS X

2009-12-03T12:55:53Z

Bugs item #2908379, was opened at 2009-12-03 12:55
Message generated for change (Tracker Item Submitted) made by jenglish
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=112997&aid=2908379&group_id=12997

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: 88. Themed Tk
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Joe English (jenglish)
Assigned to: Daniel A. Steffen (das)
Summary: ttk:* widgets - selection text illegible on Mac OS X

Initial Comment:
peterc reports in #2908368:

Selected text on Mac needs have the font as a light colour (white?) rather
than black, otherwise it's unreadable: eg,

http://pc.autons.net/stuff/ttk_treeview_unreadable_macosx.png

and reports on the chat that ttk::entry widgets also have the same problem.

I can't tell which (if any) of the kThemeTextColor* constants in Appearance.h is the right one to use. das?

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

You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=112997&aid=2908379&group_id=12997

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

Re: Inconsistent buffered I/O

2009-12-03T10:24:28Z

On 12/3/09, Tom Jackson <tom.rmadilo@...> wrote:
> Alex,
>
> I just tested the patch. All 144 configurations tested give the
> expected result.

Ha !

So, is the patch "OK for you" now ?

-Alex

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Core mailing list
Tcl-Core@...
https://lists.sourceforge.net/lists/listinfo/tcl-core

From forum: tcl-core

Re: Inconsistent buffered I/O

2009-12-03T10:17:36Z

Just out of curiosity of the expected result.

Your tests have included the workaround code against the patched source yes?

Effectively proving that any code written to work around this issue will continue to function correctly with the patched version?

On Dec 3, 2009, at 10:13 AM, Tom Jackson wrote:

> Alex,
>
> I just tested the patch. All 144 configurations tested give the
> expected result.
>
> Here is a link to the test results (in html table form):
>
> http://rmadilo.com/files/tcl-io/testio2/RESULTS.html
>
> Note that these tests include 16 tests with no flush/seek prior to
> read or write:
> (r+, w+) x (none, full) x (buffsiz 10, 20, 1024, and no -buffersize option)
>
> tom jackson
>
> On Wed, Dec 2, 2009 at 11:44 PM, Alexandre Ferrieux
> <alexandre.ferrieux@...> wrote:
>> On 12/3/09, Tom Jackson <tom.rmadilo@...> wrote:
>>>
>>> Is the patch against the HEAD on sf?
>>
>> Yaeh of course. Why don't you go fetch it and test at last ?
>>
>> -Alex
>>
>
> ------------------------------------------------------------------------------
> Join us December 9, 2009 for the Red Hat Virtual Experience,
> a free event focused on virtualization and cloud computing.
> Attend in-depth sessions from your desk. Your couch. Anywhere.
> http://p.sf.net/sfu/redhat-sfdev2dev
> _______________________________________________
> Tcl-Core mailing list
> Tcl-Core@...
> https://lists.sourceforge.net/lists/listinfo/tcl-core

From forum: tcl-core

Re: Inconsistent buffered I/O

2009-12-03T10:13:30Z

Alex,

I just tested the patch. All 144 configurations tested give the
expected result.

Here is a link to the test results (in html table form):

http://rmadilo.com/files/tcl-io/testio2/RESULTS.html

Note that these tests include 16 tests with no flush/seek prior to
read or write:
(r+, w+) x (none, full) x (buffsiz 10, 20, 1024, and no -buffersize option)

tom jackson

On Wed, Dec 2, 2009 at 11:44 PM, Alexandre Ferrieux
<alexandre.ferrieux@...> wrote:
> On 12/3/09, Tom Jackson <tom.rmadilo@...> wrote:
>>
>> Is the patch against the HEAD on sf?
>
> Yaeh of course. Why don't you go fetch it and test at last ?
>
> -Alex
>

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Core mailing list
Tcl-Core@...
https://lists.sourceforge.net/lists/listinfo/tcl-core

From forum: tcl-core

[ tcl-Bugs-2906841 ] calling [glob ../*] in an safe interp returns weird things

2009-12-03T07:51:40Z

Bugs item #2906841, was opened at 2009-12-01 14:21
Message generated for change (Comment added) made by dkf
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=110894&aid=2906841&group_id=10894

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: 33. Safe Base
Group: development: 8.6b1.1
Status: Open
>Resolution: Fixed
Priority: 9
Private: No
Submitted By: Johannes13 (john13a)
>Assigned to: Donal K. Fellows (dkf)
Summary: calling [glob ../*] in an safe interp returns weird things

Initial Comment:
8.5.7 - Windows
% safe::interpCreate test
test
% test eval glob ../*
{../All Users} ../Johannes

8.6b1.1 - Windows
% safe::interpCreate test
test
% test eval glob ../*
can't read "msg": no such variable
% set errorInfo
can't read "msg": no such variable
while executing
"Log $slave "GLOB @ $msg" NOTICE"
(procedure "::safe::AliasGlob" line 65)
invoked from within
"glob ../*"
invoked from within
"test eval glob ../*"

Some results from others
<dkf> $ tclsh8.5
<dkf> % safe::interpCreate test
<dkf> test
<dkf> % test eval glob ../*
<dkf> invalid command name "glob"

<emiliano> emiliano@merlot:~$ tclsh8.6
<emiliano> % safe::interpCreate test
<emiliano> test
<emiliano> % test eval glob ../*
<emiliano> ../ftp ../emiliano
<emiliano> HEAD (two or three days ago), linux

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

>Comment By: Donal K. Fellows (dkf)
Date: 2009-12-03 15:51

Message:
Fixed for 8.6 HEAD I think, but needs review and backport (can't just use
patch; 8.6 version uses [try]).

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-03 14:09

Message:
OK, with digging around I find that this security bug was introduced with
the fix for [Bug 1999119] (enabling Tcl Module support) which corresponds
to the 1.16->1.17 or 1.16->1.16.4.1 change.

The issue is that AliasGlob does not check for directory components in the
pattern arguments to [glob], let alone work out whether those components
refer to to locations that are "bad". :-(

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-02 15:52

Message:
Correction, that's with 1.16.4.2

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-02 15:48

Message:
With safe.tcl 1.33 in the tip of the 8.5 branch, I get globbing where I
wouldn't expect it:

bash$ make -C unix shell
DYLD_LIBRARY_PATH="`pwd`:${DYLD_LIBRARY_PATH}"; export DYLD_LIBRARY_PATH;
\
TCL_LIBRARY="/Users/dkf/Documents/software/tcl8.5/library"; export
TCL_LIBRARY; \
./tclsh
% safe::interpCreate test
test
% test eval glob ../*
../ChangeLog ../ChangeLog.1999 ../ChangeLog.2000 ../ChangeLog.2001
../ChangeLog.2002 ../ChangeLog.2003 ../ChangeLog.2004 ../ChangeLog.2005
../ChangeLog.~1.3975.2.269.~ ../changes ../compat ../CVS ../doc ../generic
../library ../libtommath ../license.terms ../macosx ../README ../tests
../tools ../unix ../win

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

Comment By: Andreas Kupries (andreas_kupries)
Date: 2009-12-01 17:52

Message:
Definitely the old buggy code I fixed on 2009-10-05 (The first fixed
revision was rev 1.19).

http://tcl.cvs.sourceforge.net/viewvc/tcl/tcl/library/safe.tcl?r1=1.18&r2=1.19

You need a newer Tcl to get the fix.

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

Comment By: Johannes13 (john13a)
Date: 2009-12-01 17:31

Message:
See uploaded files. :)

8.5.7: # RCS: @(#) $Id: safe.tcl,v 1.16.4.1 2008/06/25 16:42:05
andreas_kupries Exp $
8.6b1.1 # RCS: @(#) $Id: safe.tcl,v 1.18 2009/07/26 11:40:24 dkf Exp $

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

Comment By: Andreas Kupries (andreas_kupries)
Date: 2009-12-01 17:15

Message:
Hm. What is your revision of safe.tcl ?
(Installed under [info library], see the RCS comment line at the top).

Mine is CVS head as of the time of this comment, revision 1.33, and I do
not see the problem/error.

The relevant code in my file safe.tcl is

try {
::interp invokehidden $slave glob {*}$cmd
* } on ok msg {
* # Nothing to be done, just capture the 'msg' for later.
} on error msg {
Log $slave $msg
return -code error "script error"
}

Log $slave "GLOB @ $msg" NOTICE

and the lines I marked with * are presumably missing in your revision ...
Checking the ChangeLog I also find

2009-10-05 Andreas Kupries <andreask@...>

* library/safe.tcl (AliasGlob): Fixed conversion of catch to
try/finally, it had an 'on ok msg' branch missing, causing a
silent error immediately, and bogus glob results, breaking
search for Tcl modules.

which looks to me is the problem you are running into. MY guess is that
your Tcl is older than that entry.

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

Comment By: Andreas Kupries (andreas_kupries)
Date: 2009-12-01 17:00

Message:
I believe that I introduced this problem when I reworked the safebase to be
more uptodate in its use of Tcl's features. Reassigning to myself.

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

You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=110894&aid=2906841&group_id=10894

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

tcl-cvs: "tcl/tests safe.test" modified

2009-12-03T07:49:22Z

Update of /cvsroot/tcl/tcl/tests
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv19562/tests

Modified Files:
safe.test
Log Message:
Fix [Bug 2906841] and a few other smaller issues.

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

tcl-cvs: "tcl ChangeLog" modified

2009-12-03T07:49:22Z

Update of /cvsroot/tcl/tcl
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv19562

Modified Files:
ChangeLog
Log Message:
Fix [Bug 2906841] and a few other smaller issues.

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

tcl-cvs: "tcl/library safe.tcl" modified

2009-12-03T07:49:22Z

Update of /cvsroot/tcl/tcl/library
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv19562/library

Modified Files:
safe.tcl
Log Message:
Fix [Bug 2906841] and a few other smaller issues.

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

[ tcl-Bugs-2906841 ] calling [glob ../*] in an safe interp returns weird things

2009-12-03T06:09:22Z

Bugs item #2906841, was opened at 2009-12-01 14:21
Message generated for change (Comment added) made by dkf
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=110894&aid=2906841&group_id=10894

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: 33. Safe Base
Group: development: 8.6b1.1
Status: Open
Resolution: Out of Date
>Priority: 9
Private: No
Submitted By: Johannes13 (john13a)
Assigned to: Andreas Kupries (andreas_kupries)
Summary: calling [glob ../*] in an safe interp returns weird things

Initial Comment:
8.5.7 - Windows
% safe::interpCreate test
test
% test eval glob ../*
{../All Users} ../Johannes

8.6b1.1 - Windows
% safe::interpCreate test
test
% test eval glob ../*
can't read "msg": no such variable
% set errorInfo
can't read "msg": no such variable
while executing
"Log $slave "GLOB @ $msg" NOTICE"
(procedure "::safe::AliasGlob" line 65)
invoked from within
"glob ../*"
invoked from within
"test eval glob ../*"

Some results from others
<dkf> $ tclsh8.5
<dkf> % safe::interpCreate test
<dkf> test
<dkf> % test eval glob ../*
<dkf> invalid command name "glob"

<emiliano> emiliano@merlot:~$ tclsh8.6
<emiliano> % safe::interpCreate test
<emiliano> test
<emiliano> % test eval glob ../*
<emiliano> ../ftp ../emiliano
<emiliano> HEAD (two or three days ago), linux

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

>Comment By: Donal K. Fellows (dkf)
Date: 2009-12-03 14:09

Message:
OK, with digging around I find that this security bug was introduced with
the fix for [Bug 1999119] (enabling Tcl Module support) which corresponds
to the 1.16->1.17 or 1.16->1.16.4.1 change.

The issue is that AliasGlob does not check for directory components in the
pattern arguments to [glob], let alone work out whether those components
refer to to locations that are "bad". :-(

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-02 15:52

Message:
Correction, that's with 1.16.4.2

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

Comment By: Donal K. Fellows (dkf)
Date: 2009-12-02 15:48

Message:
With safe.tcl 1.33 in the tip of the 8.5 branch, I get globbing where I
wouldn't expect it:

bash$ make -C unix shell
DYLD_LIBRARY_PATH="`pwd`:${DYLD_LIBRARY_PATH}"; export DYLD_LIBRARY_PATH;
\
TCL_LIBRARY="/Users/dkf/Documents/software/tcl8.5/library"; export
TCL_LIBRARY; \
./tclsh
% safe::interpCreate test
test
% test eval glob ../*
../ChangeLog ../ChangeLog.1999 ../ChangeLog.2000 ../ChangeLog.2001
../ChangeLog.2002 ../ChangeLog.2003 ../ChangeLog.2004 ../ChangeLog.2005
../ChangeLog.~1.3975.2.269.~ ../changes ../compat ../CVS ../doc ../generic
../library ../libtommath ../license.terms ../macosx ../README ../tests
../tools ../unix ../win

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

Comment By: Andreas Kupries (andreas_kupries)
Date: 2009-12-01 17:52

Message:
Definitely the old buggy code I fixed on 2009-10-05 (The first fixed
revision was rev 1.19).

http://tcl.cvs.sourceforge.net/viewvc/tcl/tcl/library/safe.tcl?r1=1.18&r2=1.19

You need a newer Tcl to get the fix.

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

Comment By: Johannes13 (john13a)
Date: 2009-12-01 17:31

Message:
See uploaded files. :)

8.5.7: # RCS: @(#) $Id: safe.tcl,v 1.16.4.1 2008/06/25 16:42:05
andreas_kupries Exp $
8.6b1.1 # RCS: @(#) $Id: safe.tcl,v 1.18 2009/07/26 11:40:24 dkf Exp $

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

Comment By: Andreas Kupries (andreas_kupries)
Date: 2009-12-01 17:15

Message:
Hm. What is your revision of safe.tcl ?
(Installed under [info library], see the RCS comment line at the top).

Mine is CVS head as of the time of this comment, revision 1.33, and I do
not see the problem/error.

The relevant code in my file safe.tcl is

try {
::interp invokehidden $slave glob {*}$cmd
* } on ok msg {
* # Nothing to be done, just capture the 'msg' for later.
} on error msg {
Log $slave $msg
return -code error "script error"
}

Log $slave "GLOB @ $msg" NOTICE

and the lines I marked with * are presumably missing in your revision ...
Checking the ChangeLog I also find

2009-10-05 Andreas Kupries <andreask@...>

* library/safe.tcl (AliasGlob): Fixed conversion of catch to
try/finally, it had an 'on ok msg' branch missing, causing a
silent error immediately, and bogus glob results, breaking
search for Tcl modules.

which looks to me is the problem you are running into. MY guess is that
your Tcl is older than that entry.

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

Comment By: Andreas Kupries (andreas_kupries)
Date: 2009-12-01 17:00

Message:
I believe that I introduced this problem when I reworked the safebase to be
more uptodate in its use of Tcl's features. Reassigning to myself.

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

You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=110894&aid=2906841&group_id=10894

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

TCL 8.5 and MacOS 10.3.9

2009-12-03T05:04:30Z

Hi,

One of my users has a 10.3.9 machine and claims that our code, which
runs on top of TCL 8.5.5, won't start up. I don't have any 10.3.9
machines left to confirm or deny her claim. This 8.5.5 is one I compiled
from sources myself.

I doubt any of you have 10.3.9 machines either, but I'm just wondering:
is there a special compiler directive or makefile option I need to
specify during compilation to make the code compatible with 10.3.9?

Yours, Kevan

--
Kevan Hashemi, Electrical Engineer
Physics Department, Brandeis University
http://alignment.hep.brandeis.edu/

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-mac mailing list
tcl-mac@...
https://lists.sourceforge.net/lists/listinfo/tcl-mac

From forum: tcl-mac

[ tktoolkit-Bugs-2908052 ] TkAqua crashes on scrolling event

2009-12-03T02:52:39Z

Bugs item #2908052, was opened at 2009-12-03 11:52
Message generated for change (Tracker Item Submitted) made by treincke
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=112997&aid=2908052&group_id=12997

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: 66. Aqua Window Operations
Group: development: 8.6b1.1
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Torsten Berg (treincke)
Assigned to: Daniel A. Steffen (das)
Summary: TkAqua crashes on scrolling event

Initial Comment:
The 8.6b1.1 version of TkAqua crashes when scrolling in a certain way. Holding the mouse button pressed on a scrolling arrow will cause wish to crash with the following message:

wish8.6-aqua(1797,0xb0103000) malloc: free_garbage: garbage ptr = 0x1418c80, has non-zero refcount = 1
2009-12-03 11:32:32.995 wish8.6-aqua[1797:10b] *** Assertion failure in +[NSEvent startPeriodicEventsAfterDelay:withPeriod:], /SourceCache/AppKit/AppKit-949.54/AppKit.subproj/NSEvent.m:1765
2009-12-03 11:32:32.995 wish8.6-aqua[1797:10b] An uncaught exception was raised
2009-12-03 11:32:32.997 wish8.6-aqua[1797:10b] Periodic events are already being generated
2009-12-03 11:32:32.997 wish8.6-aqua[1797:10b] *** Terminating app due to uncaught exception 'NSInternalInconsistencyException', reason: 'Periodic events are already being generated'
2009-12-03 11:32:32.998 wish8.6-aqua[1797:10b] Stack: (
2470170603,
2514165307,
2470170059,
2478157508,
2525052112,
2526164841,
2524622328,
2523486967,
2523276965,
2195129,
2196449,
3235242,
1389874,
2705113,
2707855,
3011097,
2707855,
2737327,
2738193,
2198337,
2523500175,
2523499980,
2524627019,
2526164928,
2524622328,
2523486967,
2523276965,
2195129,
2196449,
3235242,
1421705,
1483183,
25904,
25689
)
Trace/BPT trap

The same is true, when scrolling by holding the mouse pressed on the slider trough (tested on MacOS X 10.5 and 10.6).

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

You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=112997&aid=2908052&group_id=12997

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

[ tktoolkit-Bugs-220690 ] Linking VTK with Tcl/Tk support

2009-12-03T01:54:46Z

Bugs item #220690, was opened at 2000-11-01 00:48
Message generated for change (Comment added) made by nijtmans
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=112997&aid=220690&group_id=12997

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: 85. Win Build
Group: obsolete: 8.3
Status: Closed
Resolution: Fixed
Priority: 2
Private: No
Submitted By: Brent B. Welch (welch)
Assigned to: Nobody/Anonymous (nobody)
Summary: Linking VTK with Tcl/Tk support

Initial Comment:
OriginalBugID: 4240 Bug
Version: 8.3
SubmitDate: '2000-02-25'
LastModified: '2000-03-27'
Severity: LOW
Status: Released
Submitter: techsupp
ChangedBy: hobbs
OS: Windows NT
FixedDate: '2000-03-27'
FixedInVersion: 8.3.1

Name:
Bernhard Hombitzer

Comments:
Linking VTK (www.kitware.com) with tcl/tk failed with the error
_TkWinChildProc in library not found.
Changing the line
extern LRESULT CALLBACK TkWinChildProc _ANSI_ARGS_((HWND hwnd, UINT
message, WPARAM wParam, LPARAM lParam));
to
EXTERN LRESULT CALLBACK TkWinChildProc _ANSI_ARGS_((HWND hwnd, UINT
message, WPARAM wParam, LPARAM lParam));
in file tk8.3.0/win/tkWinInt.h resolved that problem.

The fix as noted is not correct. It should go into the platform
specific stubs table. For some reason it wasn't done with the
rest in 8.1...
-- 02/25/2000 hobbs
This required a bit more than the above, but the fix is in 8.3.1.
-- 03/27/2000 hobbs

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

>Comment By: Jan Nijtmans (nijtmans)
Date: 2009-12-03 10:54

Message:
Added TkWinChildProc to the internal stub table (HEAD and 8.5)

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

You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=112997&aid=220690&group_id=12997

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

[ tktoolkit-Bugs-220600 ] missing symbol in windows binary tk82.lib

2009-12-03T01:53:06Z

Bugs item #220600, was opened at 2000-11-01 00:46
Message generated for change (Settings changed) made by nijtmans
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=112997&aid=220600&group_id=12997

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: 85. Win Build
Group: obsolete: 8.2
Status: Closed
Resolution: Fixed
Priority: 2
Private: No
Submitted By: Brent B. Welch (welch)
Assigned to: Nobody/Anonymous (nobody)
Summary: missing symbol in windows binary tk82.lib

Initial Comment:
OriginalBugID: 2963 Bug
Version: 8.2
SubmitDate: '1999-09-28'
LastModified: '1999-11-16'
Severity: SER
Status: Released
Submitter: techsupp
ChangedBy: hobbs
OS: Windows NT
OSVersion: 4.0 SP4
Machine: X86
FixedDate: '1999-11-16'
FixedInVersion: 8.2.2

Name:
Christopher Volpe

ReproducibleScript:
I'm building a 3rd party application that links with theTcl/Tk
libraries. This is the first time I have tried it with Tcl/Tk 8.2. Using
the library file tk82.lib that ships with the MS Windows binary
distribution, the linker complains about the missing symbol
"_TkWinChildProc@16". I used the cygwin "nm" command to check for
symbols in tk82.lib and confirmed that this symbol is indeed not in the
library. But then I downloaded the source distributions for Tcl and Tk
8.2, built each of them, and discovered that the symbol
"_TkWinChildProc@16" does appear in the tk82.lib file that I built.
Using this file solves my problem, but I don't think it was intended
that the library you shipped as part of the binary distribution would be
missing this symbol.

8.2.2 shows this symbol, as seen with NT's Explorer QuickView
of tk82.dll.
-- 11/16/1999 hobbs

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

>Comment By: Jan Nijtmans (nijtmans)
Date: 2009-12-03 10:53

Message:
Added TkWinChildProc to the internal stub table (HEAD and 8.5)

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

You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=112997&aid=220600&group_id=12997

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Bugs mailing list
Tcl-Bugs@...
https://lists.sourceforge.net/lists/listinfo/tcl-bugs

From forum: tcl-bugs

Memory leak in try "on" handlers

2009-12-03T01:07:53Z

Hi

There seems to be a memory leak in the result handler code of the C
based try command.

This leaks the result:

try {
string repeat \0 1048576
} on ok {} {}

... this doesn't:

try {
string repeat \0 1048576
}

Adding Tcl_DecrRefCount(resultObj); at line 4461 of generic/tclCmdMZ.c
fixes the leak for me.

Cyan

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Core mailing list
Tcl-Core@...
https://lists.sourceforge.net/lists/listinfo/tcl-core

From forum: tcl-core

Re: Inconsistent buffered I/O

2009-12-02T23:44:47Z

On 12/3/09, Tom Jackson <tom.rmadilo@...> wrote:
>
> Is the patch against the HEAD on sf?

Yaeh of course. Why don't you go fetch it and test at last ?

-Alex

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-Core mailing list
Tcl-Core@...
https://lists.sourceforge.net/lists/listinfo/tcl-core

From forum: tcl-core

NSCFString error in Tk-Cocoa

2009-12-02T17:36:31Z

I have a 10.5.8 user reporting an uncaught Objective-C exception in one
of my programs that runs on Tk-Cocoa 8.5.7. The error log causes my
application to crash, with the messages below logged to Console:

09-12-02 15:16:21.598 PortAuthority[2071:10b] An uncaught exception was
raised
2009-12-02 15:16:21.601 PortAuthority[2071:10b] *** -[NSCFString
characterAtIndex:]: Range or index out of bounds
2009-12-02 15:16:21.602 PortAuthority[2071:10b] *** Terminating app due
to uncaught exception 'NSRangeException', reason: '*** -[NSCFString
characterAtIndex:]: Range or index out of bounds'

The error occurs when the user is typing a password into a ttk::entry
field. The relevant code is below:

ttk::entry .password.top.entry -show \u2022 -textvariable
authorize::password

I can't reproduce this on my 10.5.8 system, and the error does not
appear to be related to any of the other bugs that users have reported
with some of my Tk-Cocoa programs (conflicts with input managers,
conflicts between 64-bit and 32-bit libraries, etc.). I've also had this
code in the application for years, and it's never crashed before (with
Tk-Carbon or Tk-Cocoa).

I've found various references to "-[NSCFString characterAtIndex:]: Range
or index out of bounds" on Google, but there doesn't appear to be any
common pattern to them.

Any insight is appreciated.

--Kevin

--
Kevin Walzer
Code by Kevin
http://www.codebykevin.com

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Tcl-mac mailing list
tcl-mac@...
https://lists.sourceforge.net/lists/listinfo/tcl-mac

From forum: tcl-mac