You can subscribe to this list here.
2001 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(2) |
Dec
|
---|---|---|---|---|---|---|---|---|---|---|---|---|
2002 |
Jan
(4) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2003 |
Jan
(1) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(31) |
Dec
(80) |
2004 |
Jan
(30) |
Feb
(31) |
Mar
(46) |
Apr
(31) |
May
(48) |
Jun
(16) |
Jul
|
Aug
|
Sep
(20) |
Oct
(31) |
Nov
(13) |
Dec
(1) |
2005 |
Jan
(4) |
Feb
(7) |
Mar
|
Apr
(3) |
May
(1) |
Jun
(37) |
Jul
(39) |
Aug
(22) |
Sep
(3) |
Oct
(48) |
Nov
(24) |
Dec
(31) |
2006 |
Jan
(4) |
Feb
(6) |
Mar
(19) |
Apr
(17) |
May
(39) |
Jun
(62) |
Jul
(11) |
Aug
(21) |
Sep
(10) |
Oct
(26) |
Nov
(8) |
Dec
|
2007 |
Jan
(7) |
Feb
(6) |
Mar
(2) |
Apr
|
May
|
Jun
(4) |
Jul
(10) |
Aug
(1) |
Sep
(2) |
Oct
|
Nov
(1) |
Dec
(2) |
2008 |
Jan
(19) |
Feb
(24) |
Mar
|
Apr
(4) |
May
(1) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
From: Jeremy W. <jez...@ho...> - 2006-06-27 07:56:40
|
>As Rob said, this is primarily going to impact on an application that opens >and closes many windows. At least the memory is more or less returned >instantly on exit. Yeah - and as it looks like it's XP only we can assume that it's a windows thing rather than being a perl/win32-gui issue. >Is there a "Known Issues" or limitations section somewhere in the >documentation until this is eventually tracked down? Perhaps we should just update the tracker item, but leave it open for future reference? Cheers, jez. |
From: Jeremy W. <jez...@ho...> - 2006-06-27 07:53:10
|
>If appreciate it if you could give this a good thrashing, as, hopefully, >it'll become one of the first applications new users get to see. I've >given up on my grand designs to deal gracefully with the console windows > (at least for the next release), as although I have everything needed >working, there are a few areas that I want to research more fully. The help window, causes problems for me. Menu->Help->Help - Leaves a window that can't be closed, and on top main window. Other than that, it seems fine. Cheers, jez. |
From: Reini U. <ru...@x-...> - 2006-06-27 06:06:04
|
Robert May schrieb: > Robert May wrote: >> Reini Urban wrote: >>> why not Win32::GUI::Region->ExtCreateRegion() and >>> Win32::GUI::Region->GetRegionData() ? >>> >>> Rgn != Region >> >> You're quite correct. I thought I was being consistent with all other >> region functions, but we should be being consistent with the Win32 API. >> >> I'll change it. > > Committed. Thanks -- Reini |
From: Chris W. <ch...@vi...> - 2006-06-26 23:56:25
|
Hi, >>>> Can someone run this with a fixed number of iterations, and tell me how >>>> much memory is leaked per iteration? >>> 56 - 60 K each iteration >> Hummm - I seem to leak about 108 bytes per itteration. >I find the 108 byte figure more believable. At 56-60k per loop >iteration you'd run out of memory pretty quickly, and there's nothing we >allocate from the perl side that would be causing that much memory to be >allocated. Sorry... what's a few decimal places between friends... As Rob said, this is primarily going to impact on an application that opens and closes many windows. At least the memory is more or less returned instantly on exit. Is there a "Known Issues" or limitations section somewhere in the documentation until this is eventually tracked down? Regards Chris -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.1.394 / Virus Database: 268.9.4/375 - Release Date: 25/06/2006 |
From: Robert M. <rm...@po...> - 2006-06-26 19:05:35
|
Jeremy White wrote: >>> Does it still leak if we only create a window (without the button)? >> YES! > > For me, I get no leak when just the window. I also played with a few other > controls, and they leak too (a listview seemed to leak more than a button). > Adding a child window to the parent window almost doesn't leak - a 4k leak > every 30 seconds or so. I can confirm that I don't see this on w2k either, so I afraid that I'm not going to be much use in helping to track this down. >>> Can someone run this with a fixed number of iterations, and tell me how >>> much memory is leaked per iteration? >> 56 - 60 K each iteration > > Hummm - I seem to leak about 108 bytes per itteration. I find the 108 byte figure more believable. At 56-60k per loop iteration you'd run out of memory pretty quickly, and there's nothing we allocate from the perl side that would be causing that much memory to be allocated. >> Start code and figures vary but average 40% CPU to Perl.exe and 30% CPU to >> Explorer.exe, the figures change between these two executables by as much >> as >> 10%, however the total between the two stays at 70% total CPU. >> >> When terminating with Ctrl-C, note that perl.exe died straight away however >> Explorer.exe took a few moments more to die and return the memory. Sorry, but for now you're on your own with this one. Regards, Rob. |
From: Robert M. <rm...@po...> - 2006-06-26 18:58:31
|
All, Although I have not completed the full set of things that I had planned for a 1.04 release, I think it is time. So I have bumped everything that remains on my list to 1.05. Specifically this includes: - Reini's CustomDraw changes - Reini's other changes - Thread support (Win32::GUI::ThreadUtils) - dealing with console windows (Win32::GUI::Console) I have a lot more planned for 1.05, but will wait to share that until we have 1.04 out of the door. Now is the time to shout if you think there's anything you were expecting in 1.04 that's not there. I'll be dealing with the documentation next: specifically doHTMLdocs.pl needs some work, and some other bits and pieces will need re-organising to accommodate the new modules. I also intend to introduce a new package of documentation: Win32::GUI::ReleaseNotes, which will contain detailed release notes for each release we make. (it is my hope that we can update this, along with the CHANGELOG, whenever we commit to save this being a single large task for each release) Once I have completed this (in the next few days I hope) I will make a beta PPM (probably for perl 5.8 only), and open it up to the users list for some sanity testing. Hoping this meets everyone's approval. Regards, Rob. -- Robert May Win32::GUI, a perl extension for native Win32 applications http://perl-win32-gui.sourceforge.net/ |
From: Robert M. <rm...@po...> - 2006-06-26 18:34:54
|
All, I just committed a script (win32-gui-demos.pl) in a new scripts directory, along with the Makefile.PL changes necessary to get this script installed in the perl bin directory. It's a script that lists the sample/demo code distributed with Win32::GUI, and allows viewing of the sample source, and running the demos. If appreciate it if you could give this a good thrashing, as, hopefully, it'll become one of the first applications new users get to see. I've given up on my grand designs to deal gracefully with the console windows (at least for the next release), as although I have everything needed working, there are a few areas that I want to research more fully. Apart from documentation, I hope that the code-base as it stands now will make a 1.04 release. (more on this in a separate mail). Regards, Rob. -- Robert May Win32::GUI, a perl extension for native Win32 applications http://perl-win32-gui.sourceforge.net/ |
From: Robert M. <rm...@po...> - 2006-06-26 18:23:54
|
Robert May wrote: > Reini Urban wrote: >> why not Win32::GUI::Region->ExtCreateRegion() and >> Win32::GUI::Region->GetRegionData() ? >> >> Rgn != Region > > You're quite correct. I thought I was being consistent with all other > region functions, but we should be being consistent with the Win32 API. > > I'll change it. Committed. Rob. |
From: Robert M. <rm...@po...> - 2006-06-26 18:02:00
|
Reini Urban wrote: > why not Win32::GUI::Region->ExtCreateRegion() and > Win32::GUI::Region->GetRegionData() ? > > Rgn != Region You're quite correct. I thought I was being consistent with all other region functions, but we should be being consistent with the Win32 API. I'll change it. Rob. |
From: Reini U. <ru...@x-...> - 2006-06-26 13:53:13
|
why not Win32::GUI::Region->ExtCreateRegion() and Win32::GUI::Region->GetRegionData() ? Rgn != Region 2006/6/24, Robert May <rm...@po...>: > I just committed 2 new functions for handling windows regions: > > New region method: > my $data = $rgn->GetRgnData(); > wraps the win32 API GetRegionData() function > > New region constructor: > my $newrgn = Win32::GUI::Region->ExtCreateRgn($data); > wraps the win32 API ExtCreateRegion() function > > Tracker 1469648: > https://sourceforge.net/tracker/index.php?func=detail&aid=1469648&group_id=16572&atid=366572 > > As a quick test I did the following: > > #!perl -w > use strict; > use warnings; > > use Win32::GUI(); > > my $mw = Win32::GUI::Window->new( > -title => "Region test", > -size => [400,300], > ); > > $mw->Show(); > my $DC = $mw->GetDC(); > > my $rgn = Win32::GUI::Region->CreateRoundRectRgn(0,0,100,100,50,50); > $DC->InvertRgn($rgn); > > my $rgndata = $rgn->GetRgnData(); > my $newrgn = Win32::GUI::Region->ExtCreateRgn($rgndata); > $newrgn->OffsetRgn(150,150); > $DC->InvertRgn($newrgn); > > > Win32::GUI::Dialog(); > $mw->Hide(); > exit(0); > __END__ > > Regards, > Rob. -- Reini Urban http://phpwiki.org/ http://murbreak.at/ http://spacemovie.mur.at/ http://helsinki.at/ |
From: Jeremy W. <jez...@ho...> - 2006-06-26 08:09:52
|
> >Does it still leak if we only create a window (without the button)? > >YES! For me, I get no leak when just the window. I also played with a few other controls, and they leak too (a listview seemed to leak more than a button). Adding a child window to the parent window almost doesn't leak - a 4k leak every 30 seconds or so. > >Can someone run this with a fixed number of iterations, and tell me how > >much memory is leaked per iteration? > >56 - 60 K each iteration Hummm - I seem to leak about 108 bytes per itteration. >Start code and figures vary but average 40% CPU to Perl.exe and 30% CPU to >Explorer.exe, the figures change between these two executables by as much >as >10%, however the total between the two stays at 70% total CPU. > >When terminating with Ctrl-C, note that perl.exe died straight away however >Explorer.exe took a few moments more to die and return the memory. Yeah, I see the same thing. Cheers, jez. |
From: Chris W. <ch...@vi...> - 2006-06-26 00:41:08
|
Hi, >This is looking like an XP thing at the moment. Can either of you >reduce the problem to something that doesn't need Win32::GUI? I.e. can >we find out if it's a window creation/deletion thing, or a perl thing? >Does it still leak if we only create a window (without the button)? YES! >Can someone run this with a fixed number of iterations, and tell me how >much memory is leaked per iteration? 56 - 60 K each iteration >Is all the memory returned when the app exits? YES! Also Jez asked if only 50% of the CPU was taken up... When running Task Manager Processes before commencement I have 99% System Idles processes. Only things open are a command prompt and Task Manager. Start code and figures vary but average 40% CPU to Perl.exe and 30% CPU to Explorer.exe, the figures change between these two executables by as much as 10%, however the total between the two stays at 70% total CPU. When terminating with Ctrl-C, note that perl.exe died straight away however Explorer.exe took a few moments more to die and return the memory. Regards, Chris > use strict; > use Win32::GUI; > > my ($W,$but); > while (1) { > $W = new Win32::GUI::Window( > #-name => "TestWindow", > -pos => [ 0, 0], > -size => [210, 200], > -text => "TestWindow", > ); > > $but=$W->AddButton( > #-name => "test", > -text => "Button 1", > -size => [ 70, 22 ], > -pos => [ 20, 20 ], > ); > } -- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.1.394 / Virus Database: 268.9.3/374 - Release Date: 23/06/2006 -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.1.394 / Virus Database: 268.9.4/375 - Release Date: 25/06/2006 |
From: Robert M. <rm...@po...> - 2006-06-25 13:24:40
|
Chris Wearn wrote: > Checked out latest (25/06/2006 about 1:40AM GMT)... compiled, tested, > installed all okay. Thanks for the report. > Ran code below and watched memory in task manager diminish... > Running on Windows XP Professional SP2, Intel CPU (P4 2.4GHz), 1GB RAM... > Installed: ActivePerl 5.8.7 build 815, VC6++ Robert May wrote previously: >>But I can't duplicate the memory leak reported (perl 5.8.7 813, >>Win32::GUI 1.03, win98 and win2k). Jez reports this as still happening >>on WinXP - can anyone else duplicate a memory leak running this code: This is looking like an XP thing at the moment. Can either of you reduce the problem to something that doesn't need Win32::GUI? I.e. can we find out if it's a window creation/deletion thing, or a perl thing? Does it still leak if we only create a window (without the button)? Can someone run this with a fixed number of iterations, and tell me how much memory is leaked per iteration? Is all the memory returned when the app exits? I can confirm that I don't see this problem in win98, and will re-test win2k tomorrow. Regards, Rob. > use strict; > use Win32::GUI; > > my ($W,$but); > while (1) { > $W = new Win32::GUI::Window( > #-name => "TestWindow", > -pos => [ 0, 0], > -size => [210, 200], > -text => "TestWindow", > ); > > $but=$W->AddButton( > #-name => "test", > -text => "Button 1", > -size => [ 70, 22 ], > -pos => [ 20, 20 ], > ); > } |
From: Jeremy W. <jez...@ho...> - 2006-06-25 12:42:55
|
Hi Chris, >Ran code below and watched memory in task manager diminish... > >Running on Windows XP Professional SP2, Intel CPU (P4 2.4GHz), 1GB RAM... > When you run that script do you find that Perl only take about 50% of CPU while some other process takes up the rest? Looks like this could be an issue with XP. Cheers, jez. |
From: Chris W. <ch...@vi...> - 2006-06-25 10:27:04
|
Hi Rob, Checked out latest (25/06/2006 about 1:40AM GMT)... compiled, tested, installed all okay. Ran code below and watched memory in task manager diminish... Running on Windows XP Professional SP2, Intel CPU (P4 2.4GHz), 1GB RAM... Installed: ActivePerl 5.8.7 build 815 VC6++ Rgds Chris Wearn >But I can't duplicate the memory leak reported (perl 5.8.7 813, >Win32::GUI 1.03, win98 and win2k). Jez reports this as still happening >on WinXP - can anyone else duplicate a memory leak running this code: use strict; use Win32::GUI; my ($W,$but); while (1) { $W = new Win32::GUI::Window( #-name => "TestWindow", -pos => [ 0, 0], -size => [210, 200], -text => "TestWindow", ); $but=$W->AddButton( #-name => "test", -text => "Button 1", -size => [ 70, 22 ], -pos => [ 20, 20 ], ); } -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.1.394 / Virus Database: 268.9.3/374 - Release Date: 23/06/2006 |
From: Robert M. <rm...@po...> - 2006-06-24 14:54:05
|
I just committed 2 new functions for handling windows regions: New region method: my $data = $rgn->GetRgnData(); wraps the win32 API GetRegionData() function New region constructor: my $newrgn = Win32::GUI::Region->ExtCreateRgn($data); wraps the win32 API ExtCreateRegion() function Tracker 1469648: https://sourceforge.net/tracker/index.php?func=detail&aid=1469648&group_id=16572&atid=366572 As a quick test I did the following: #!perl -w use strict; use warnings; use Win32::GUI(); my $mw = Win32::GUI::Window->new( -title => "Region test", -size => [400,300], ); $mw->Show(); my $DC = $mw->GetDC(); my $rgn = Win32::GUI::Region->CreateRoundRectRgn(0,0,100,100,50,50); $DC->InvertRgn($rgn); my $rgndata = $rgn->GetRgnData(); my $newrgn = Win32::GUI::Region->ExtCreateRgn($rgndata); $newrgn->OffsetRgn(150,150); $DC->InvertRgn($newrgn); Win32::GUI::Dialog(); $mw->Hide(); exit(0); __END__ Regards, Rob. -- Robert May Win32::GUI, a perl extension for native Win32 applications http://perl-win32-gui.sourceforge.net/ |
From: Robert M. <rm...@po...> - 2006-06-24 13:30:45
|
Jeremy White wrote: >> I've added support for the Win32 AnimateWindow() API, as requested by >> tracker 1266930: >> https://sourceforge.net/tracker/index.php?func=detail&aid=1266930&group_id=16572&atid=366572 >> >> Now, as well as $win->Show(), $win->Hide() you can use $win->Animate(). > > I had problems building under mingw - I had to set: > > #define WINVER 0x0501 > > in windef.h for it to build - after that, all tests pass, and the > example runs (nice BTW!). The MS headers set WINVER if it's not already set (in the case of the headers I'm using it gets set to Ox501). MinGW headers don't do this. It's probably a good idea for us to set it anyway, so I'll add #define WINVER 0x0501 to the start of GUI.h Thanks for the feedback. Rob. |
From: Jeremy W. <jez...@ho...> - 2006-06-24 11:04:17
|
>I've added support for the Win32 AnimateWindow() API, as requested by >tracker 1266930: >https://sourceforge.net/tracker/index.php?func=detail&aid=1266930&group_id=16572&atid=366572 > >Now, as well as $win->Show(), $win->Hide() you can use $win->Animate(). I had problems building under mingw - I had to set: #define WINVER 0x0501 in windef.h for it to build - after that, all tests pass, and the example runs (nice BTW!). Cheers, jez. |
From: Robert M. <rm...@po...> - 2006-06-23 23:12:18
|
More commits :-) How do I get <enter> key strokes into a Textfield/RichEdit control when I'm using -dialogui option if a frequently asked question on the users list, so I've added a -wantreturn option to Textfield/RichEdit to make applying the ES_WANTRETURN style easier. I've added support for the Win32 AnimateWindow() API, as requested by tracker 1266930: https://sourceforge.net/tracker/index.php?func=detail&aid=1266930&group_id=16572&atid=366572 Now, as well as $win->Show(), $win->Hide() you can use $win->Animate(). Enjoy. Rob. -- Robert May Win32::GUI, a perl extension for native Win32 applications http://perl-win32-gui.sourceforge.net/ |
From: Jeremy W. <jez...@ho...> - 2006-06-23 19:28:06
|
>But I can't duplicate the memory leak reported (perl 5.8.7 813, >Win32::GUI 1.03, win98 and win2k). Jez reports this as still happening >on WinXP - can anyone else duplicate a memory leak running this code: She builds fine with all tests ok - I still see that leak:) Cheers, jez. |
From: Robert M. <rm...@po...> - 2006-06-23 18:44:27
|
All, I've been working with Jez to add a ClassData() method to Win32::GUI, to give us somewhere to store class instance data, without having to push it into the blessed object. I just committed an implementation of what was discussed previously on the list (I'd add a url to the message, but the SF mail archives are down at the moment). Whilst doing so we identified a source of memory leaks that I have also corrected accross the codebase - the worst of which would only have been seen by someone creating and destroying a lot of windows (as we were leaving an empty HV around during window destruction ~50 bytes per window). We were hoping to close down this bug report too: https://sourceforge.net/tracker/index.php?func=detail&aid=1417288&group_id=16572&atid=116572 But I can't duplicate the memory leak reported (perl 5.8.7 813, Win32::GUI 1.03, win98 and win2k). Jez reports this as still happening on WinXP - can anyone else duplicate a memory leak running this code: use strict; use Win32::GUI; my ($W,$but); while (1) { $W = new Win32::GUI::Window( #-name => "TestWindow", -pos => [ 0, 0], -size => [210, 200], -text => "TestWindow", ); $but=$W->AddButton( #-name => "test", -text => "Button 1", -size => [ 70, 22 ], -pos => [ 20, 20 ], ); } Thanks for your help. Rob. -- Robert May Win32::GUI, a perl extension for native Win32 applications http://perl-win32-gui.sourceforge.net/ |
From: Reini U. <ru...@x-...> - 2006-06-23 05:37:19
|
2006/6/22, Robert May <rm...@po...>: > Reini Urban wrote: > > 2006/6/18, Robert May <rm...@po...>: > >>>>> 3. gnu make, make test, make install doesn't recurse into the subdirs > >>>>> Win32-GUI-AxWindow, Win32-GUI-DIBitmap, Win32-GUI-Grid > >>>> That's by design. None of those 3 modules can be built (as far as I > >>>> know) with gcc, as they rely on one or other of the MFC or ATL > >>>> frameworks, which only come with MS compilers. > >>> Ok, I'll port them then. Shouldn't be that hard to produce > > [snip] > > > Indeed, the real problem are the header files, which are not available > > in /usr/include/w32api > > I simply copy the original MSVC (6) MFC headers to a private subdir > > and patch these there. > > The stack manipulation #pragmas and some more MS specifics need to be addressed. > > If I have a working Stdafx.o the rest is easy. > > That's OK as an approach for doing a private build, and a binary > distribution, but AFAIK those headers don't have a re-distributable > licence, so we won't be able to add them to the project source. sure. ask user for the path of the MSVC path cp the local MFC headers to a private dir, patch them. -- Reini |
From: Robert M. <rm...@po...> - 2006-06-22 20:53:08
|
Reini Urban wrote: > 2006/6/18, Robert May <rm...@po...>: >>>>> 3. gnu make, make test, make install doesn't recurse into the subdirs >>>>> Win32-GUI-AxWindow, Win32-GUI-DIBitmap, Win32-GUI-Grid >>>> That's by design. None of those 3 modules can be built (as far as I >>>> know) with gcc, as they rely on one or other of the MFC or ATL >>>> frameworks, which only come with MS compilers. >>> Ok, I'll port them then. Shouldn't be that hard to produce [snip] > Indeed, the real problem are the header files, which are not available > in /usr/include/w32api > I simply copy the original MSVC (6) MFC headers to a private subdir > and patch these there. > The stack manipulation #pragmas and some more MS specifics need to be addressed. > If I have a working Stdafx.o the rest is easy. That's OK as an approach for doing a private build, and a binary distribution, but AFAIK those headers don't have a re-distributable licence, so we won't be able to add them to the project source. Regards, Rob. |
From: Reini U. <ru...@x-...> - 2006-06-19 08:59:24
|
2006/6/18, Robert May <rm...@po...>: > >>> 3. gnu make, make test, make install doesn't recurse into the subdirs > >>> Win32-GUI-AxWindow, Win32-GUI-DIBitmap, Win32-GUI-Grid > >> That's by design. None of those 3 modules can be built (as far as I > >> know) with gcc, as they rely on one or other of the MFC or ATL > >> frameworks, which only come with MS compilers. > > > > Ok, I'll port them then. Shouldn't be that hard to produce > > import libs on the fly for those MS dll's. > > Just the names will have to be redecorated. Some .def file munging. > > > > http://cygwin.com/faq/faq_3.html#SEC103 > > My suspicion is that it's harder than simply producing a few lib*.lib > files - there's a whole set of header files that I doubt are available > for Cygwin (but I haven't looked in any detail). If you do look at > this, and have success, then I'd be very interested in your notes, so > that I can do something similar for MinGW. Please let me know how you > get on. Indeed, the real problem are the header files, which are not available in /usr/include/w32api I simply copy the original MSVC (6) MFC headers to a private subdir and patch these there. The stack manipulation #pragmas and some more MS specifics need to be addressed. If I have a working Stdafx.o the rest is easy. -- Reini Urban http://phpwiki.org/ http://murbreak.at/ http://spacemovie.mur.at/ http://helsinki.at/ |
From: Robert M. <rm...@po...> - 2006-06-18 21:13:54
|
Reini Urban wrote: >>> 2. In Makefile for gnu make the "-" before cp didn't work for me to copy >>> the demos. So I changed that to >>> >>> demos: >>> $(MKPATH) $(INST_DEMO) >>> $(CP) samples/*.pl $(INST_DEMO) >>> $(CP) samples/*.cur $(INST_DEMO) >>> $(CP) samples/*.bmp $(INST_DEMO) >>> $(CP) samples/*.ico $(INST_DEMO) >>> >>> Otherwise the errorlevel is set: >>> cp samples/* blib/lib/Win32/GUI/demos >>> cp: omitting directory `samples/CVS' >>> make: [demos] Error 1 (ignored) >> That look like it is expected to. The "-" before the cp results in the >> "(ignored)" on the last line: cp is returning a non-zero exit status, >> but make (which would normally abort when this happens), is ignoring it >> and continuing. > > I know that "-" should suppress the make errorlevel breakout. Hmm. > Something really weird. Really? It looks like it is doing what it is supposed to. Are the demo files not copied? >>> 3. gnu make, make test, make install doesn't recurse into the subdirs >>> Win32-GUI-AxWindow, Win32-GUI-DIBitmap, Win32-GUI-Grid >> That's by design. None of those 3 modules can be built (as far as I >> know) with gcc, as they rely on one or other of the MFC or ATL >> frameworks, which only come with MS compilers. > > Ok, I'll port them then. Shouldn't be that hard to produce > import libs on the fly for those MS dll's. > Just the names will have to be redecorated. Some .def file munging. > > http://cygwin.com/faq/faq_3.html#SEC103 My suspicion is that it's harder than simply producing a few lib*.lib files - there's a whole set of header files that I doubt are available for Cygwin (but I haven't looked in any detail). If you do look at this, and have success, then I'd be very interested in your notes, so that I can do something similar for MinGW. Please let me know how you get on. Regards, Rob.11 |