Re: Rootless mode
Kensuke, Harold, I was just wondering whether it would be possible to create a test build of the X server with this multi-window mode. Obviously there are issues with it but I presume they only come up in multi-window mode? It would just be really nice to see it beginning to work - this is something lots of people have wanted for a long time, congratulations for getting it going! David Harold L Hunt II wrote: JS, Kensuke titled his message ``rootless mode'' when it should really have been titled ``multi-window mode''. The goal of multi-window mode is to create a Windows-window for each top-level X Window, rather than creating one huge window for your entire X desktop. Harold J S wrote: Kensuke, The new patch is an architectual improvement. However, when I run it, if I move a window it gets moved, then it jumps back to its original position and retraces the move path that I took it on, over and over again until I feel like I will throw up. :) I am not sure what is causing this problem, but obviously the queue of messages is not being cleared and is instead being looped through repeatedly. My log file from this session is on the web: http://www.msu.edu/~huntharo/xwin/XWinrl.log.bz2 (33 KiB) The unpacked log file is 800+ KiB. Harold Kensuke Matsuzaki wrote: Harold, It remains only for debugging. Am I correct that the root window is still being drawn, even thought it is not really usable? Is that something that remains to be fixed, or did I have something go wrong with my patching? By the way, this is new patch that integrates XWin and the window manager. Kensuke Matsuzaki How does this patch change what Rootless mode already does? I thought that rootless mode already integrated the window manager? I have windowmaker running, and when I'm in rootless mode the xterms come up with windowmaker frames. Can you integrate with the Windows windows manager? _ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=features/junkmail
Re: Cygwin GNOME 1.4 binary packages
Ian Harcombe wrote: Just to voice a quick vote of thanks to all involved in getting this working. I am now happily running Gnome 1.4 (with all the extra bits n bobs you've provided) under Cygwin on a Windows XP Professional laptop as my normal development environment at work. It is clean and responsive, and now that sawfish is running properly, very pretty indeed. My vote of thanks too! But I can't get it to work, would appreciate any comments. It seems to be a problem with the goad_server which seems to be something to do with CORBA/COMM. Is there any way I can debug it? I have attached the startup log below. BTW, I installed it using the cygwin setup utility, not the installpkg script, could this be a problem? Thanks David starting gnome on display :0 _IceTransmkdir: Owner of /tmp/.ICE-unix should be set to root _IceTransmkdir: Mode of /tmp/.ICE-unix should be set to 1777 SESSION_MANAGER=local/auga:/tmp/.ICE-unix/1868 env: gnome-upgrade.py: No such file or directory /bin/bash: line 1: aumix-minimal: command not found can't load fixed font during initialisationxscreensaver-command: not found xscreensaver: not found Gnome-Message: gnome_execute_async_with_env_fds: returning -1 ** WARNING **: goad_server_activate: goad.c 673: unexpected exception IDL:CORBA/ COMM_FAILURE:1.0: ** WARNING **: sysex: IDL:CORBA/COMM_FAILURE:1.0. ** WARNING **: usrex: IDL:CORBA/COMM_FAILURE:1.0.
Re: Rootless mode
* David Fraser [EMAIL PROTECTED] [2002-12-12 10:03:53 +0200]: Kensuke, Harold, I was just wondering whether it would be possible to create a test build of the X server with this multi-window mode. Obviously there are issues with it but I presume they only come up in multi-window mode? It would just be really nice to see it beginning to work - this is something lots of people have wanted for a long time, congratulations for getting it going! I wholeheartedly second that; either a test build, or perhaps some instructions on how to take the patch(es) you supplied and build a modified version using the existing Cygwin tools. Is this just a matter of building a modified version of the XWin.exe binary and replacing my current XWin -rootless line with XWin2 -multiwindow in the startx script? I think a lot of us non-programmers would be willing to become beta testers and/or bug reporters if we could get our foot in the door. :) Also, someone else suggested making the actual Cygwin/XFree86 control window (is that the right term?) minimize to the system tray rather than the taskbar; I think this would be a good idea too... -- John Buttery (Web page temporarily unavailable) msg04456/pgp0.pgp Description: PGP signature
RE: Cygwin GNOME 1.4 binary packages
At most it only seems to be a configuration problem, and not a bug. Like if it can't find your fonts? I'm no expert. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of David Fraser Sent: Thursday, December 12, 2002 3:29 AM To: [EMAIL PROTECTED] Subject: Re: Cygwin GNOME 1.4 binary packages Ian Harcombe wrote: Just to voice a quick vote of thanks to all involved in getting this working. I am now happily running Gnome 1.4 (with all the extra bits n bobs you've provided) under Cygwin on a Windows XP Professional laptop as my normal development environment at work. It is clean and responsive, and now that sawfish is running properly, very pretty indeed. My vote of thanks too! But I can't get it to work, would appreciate any comments. It seems to be a problem with the goad_server which seems to be something to do with CORBA/COMM. Is there any way I can debug it? I have attached the startup log below. BTW, I installed it using the cygwin setup utility, not the installpkg script, could this be a problem? Thanks David starting gnome on display :0 _IceTransmkdir: Owner of /tmp/.ICE-unix should be set to root _IceTransmkdir: Mode of /tmp/.ICE-unix should be set to 1777 SESSION_MANAGER=local/auga:/tmp/.ICE-unix/1868 env: gnome-upgrade.py: No such file or directory /bin/bash: line 1: aumix-minimal: command not found can't load fixed font during initialisationxscreensaver-command: not found xscreensaver: not found Gnome-Message: gnome_execute_async_with_env_fds: returning -1 ** WARNING **: goad_server_activate: goad.c 673: unexpected exception IDL:CORBA/ COMM_FAILURE:1.0: ** WARNING **: sysex: IDL:CORBA/COMM_FAILURE:1.0. ** WARNING **: usrex: IDL:CORBA/COMM_FAILURE:1.0.
Re: xkbcomp.exe segfaults
Alexander Gottwald wrote: Compile xkbcomp from sources and run it in gdb. Ok. Should I download the X420src- .tgz files from ftp://ftp.xfree86.org/pub/XFree86/4.2.0/source/ and go on from there, or how should I proceed? Alexander Skwar -- How to quote: http://learn.to/quote (german) http://quote.6x.to (en) Homepage: http://www.iso-top.biz | Jabber: [EMAIL PROTECTED] iso-top.biz - Die günstige Art an Linux Distributionen zu kommen
Re: Host Chooser Program?
Tim Thomson wrote: Yep, There are lots of them :) I even wrote one :) In the meantime, I've also written one. Mine is written in Python with wxPython. You can get it from http://message-center.info/stuff/xhc/ Alexander Skwar -- How to quote: http://learn.to/quote (german) http://quote.6x.to (en) Homepage: http://www.iso-top.biz | Jabber: [EMAIL PROTECTED] iso-top.biz - Die günstige Art an Linux Distributionen zu kommen
Re: xkbcomp.exe segfaults
On Thu, 12 Dec 2002, Alexander Skwar wrote: Compile xkbcomp from sources and run it in gdb. Ok. Should I download the X420src- .tgz files from ftp://ftp.xfree86.org/pub/XFree86/4.2.0/source/ and go on from there, or how should I proceed? I think the contributor's guide covers this topic very well http://xfree86.cygwin.com/docs/cg/ HTH ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
Re: Rootless mode
* Kensuke Matsuzaki [EMAIL PROTECTED] [2002-12-12 21:10:54 +0900]: To build XWin.exe, please read http://xfree86.cygwin.com/docs/cg/cygwin-xfree-cg.html I should have figured that out without being told, sorry. I'm now getting the source via CVS. There sure is a lot of it; I hope x86 is still a standard architecture when it finishes. :p The patch can be applied as shown in the following command $pwd ~/x-devel/xc/programs/Xserver/hw/xwin $bunzip2 rootless.patch.bz2 $patch -p1 rootless.patch I edited xc/config/cf/cygwin.rules to disable version controlled .dll, but is it right? I don't understand this question. If you're asking if a change you made causes your instructions to break, I'll let you know when I try them. When I get a modified version of XWin.exe out of this process, can other people just download it and drop it in to their own Cygwin installs? (presumably they would back up their current XWin.exe first) Let me know...if this is the case, I'll put my modified version online somewhere for people who don't have the patience/bandwidth/disk space to build their own. After the patch is applied, do I just go back up to the root directory and run make? Then find the built XWin.exe and copy it over to the directory where the current one lives? -- John Buttery (Web page temporarily unavailable) msg04463/pgp0.pgp Description: PGP signature
X server window size
With the current version of XFree86 (XFree86-bin-4.2.0-2; XFree86-xserv-4.2.0-16) I am getting the following: I start a (xdm based) session on another (linux) X server, using a bat which does: start /B XWin -indirect linux This creates a window as large as if being maximized. Ok. If either I click on Quit button (from the xdm window) or choose a host, login and then logout from it, the Cygwin/XFree86 window is closed and opened again but shrinked by some 4-5 pixels in width and about the height of the taskbar in height. Repeating the procedure (clicking Quit or along several sessions on the linux machine) the Cygwin/XFree86 window gets smaller and smaller. This behaviour dissapears when I specify -screen 0 1152 844 to XWin command. Is this a known issue? I unsuccessfuly searched in the archives for it. Am I missing something? Another problem (which I did not noticed in *some* previous version) was redraw not being performed properly for eg when moving windows. I work only on another computer (linux) with DISPLAY set to Cygwin machine. Sorry I can't be more specific. Best regards, Corneliu Rudeanu
Re: Rootless mode
-multiwindow is not stable, so I thought binary is not needed. XWin.exe is here. http://peppermint.jp/products/asis/XWin.2002_12_09.exe.bz2 And libfreetype 2.1.1 is here. http://peppermint.jp/temp/libfreetype.tar.bz2 Kensuke Matsuzaki
Re: Host Chooser Program?
Hmm, weird. Haven't had that combination before :) Could you please post me (not the list) your Xwin.log, and your launchx.bat file that XLauncher creates from one of these 95 machines, after running XLauncher on it? Hopefully that will turn up something. This is not a problem of XLauncher nor a direct problem of XWin. The cygwin layer can not handle the dynamic IP address and does always report 0.0.0.0 as address for the network interface. Even the -from parameter will not help here. The XServer has no possibility to get the local IP address to send it to the xdmcp server. To add this to the cygwin dll is with high priority on my todo list. bye ago Alex, this problem with Xlauncher happened when I used a fixed IP address, (not DHCP). I'll check the xwin.log and post it tomorrow 'cos I'm working from home today. Cheers, JS. _ MSN 8 with e-mail virus protection service: 2 months FREE* http://join.msn.com/?page=features/virus
Re: Multiwindow mode
[EMAIL PROTECTED] wrote: -multiwindow is not stable, so I thought binary is not needed. XWin.exe is here. http://peppermint.jp/products/asis/XWin.2002_12_09.exe.bz2 And libfreetype 2.1.1 is here. http://peppermint.jp/temp/libfreetype.tar.bz2 Kensuke Matsuzaki Thank you Kensuke! I have konsole running by itself in a separate window! This is great! Much appreciated! By the way I have no stability issues other than moving windows as Harold mentioned. I can resize a window as long as I don't move it. David -=-=- ... Micro$oft is not an answer. It is a question. The answer is 'no'. * TagZilla 0.030
Re: Rootless mode
Kensuke, Did you ever look into the bug I reported where a dragged window just keeps on repeating the dragging and never stops moving? Seems to be a problem with the message queue not being cleared properly. My email is in the list archives if you missed it. Harold Kensuke Matsuzaki wrote: I wholeheartedly second that; either a test build, or perhaps some instructions on how to take the patch(es) you supplied and build a modified version using the existing Cygwin tools. To build XWin.exe, please read http://xfree86.cygwin.com/docs/cg/cygwin-xfree-cg.html The patch can be applied as shown in the following command $pwd ~/x-devel/xc/programs/Xserver/hw/xwin $bunzip2 rootless.patch.bz2 $patch -p1 rootless.patch I edited xc/config/cf/cygwin.rules to disable version controlled .dll, but is it right? Is this just a matter of building a modified version of the XWin.exe binary and replacing my current XWin -rootless line with XWin2 -multiwindow in the startx script? Yes. Kensuke Matsuzaki
Re: Rootless mode
David, I have not released a test version because the version that Kensuke sent me didn't work at all. It just kept repeating the dragging of a window if you ever moved one. I sent a report to the list but he never responded. I didn't want to bug him about it for awhile in case he had seen it... but it seems now that he never saw it. I will release a test version once there are no obvious bugs. I mean, are we really going to benefit from 100 people saying ``multiwindow mode doesn't work, it keeps doing foo''? I don't think so :) Harold David Fraser wrote: Kensuke, Harold, I was just wondering whether it would be possible to create a test build of the X server with this multi-window mode. Obviously there are issues with it but I presume they only come up in multi-window mode? It would just be really nice to see it beginning to work - this is something lots of people have wanted for a long time, congratulations for getting it going! David Harold L Hunt II wrote: JS, Kensuke titled his message ``rootless mode'' when it should really have been titled ``multi-window mode''. The goal of multi-window mode is to create a Windows-window for each top-level X Window, rather than creating one huge window for your entire X desktop. Harold J S wrote: Kensuke, The new patch is an architectual improvement. However, when I run it, if I move a window it gets moved, then it jumps back to its original position and retraces the move path that I took it on, over and over again until I feel like I will throw up. :) I am not sure what is causing this problem, but obviously the queue of messages is not being cleared and is instead being looped through repeatedly. My log file from this session is on the web: http://www.msu.edu/~huntharo/xwin/XWinrl.log.bz2 (33 KiB) The unpacked log file is 800+ KiB. Harold Kensuke Matsuzaki wrote: Harold, It remains only for debugging. Am I correct that the root window is still being drawn, even thought it is not really usable? Is that something that remains to be fixed, or did I have something go wrong with my patching? By the way, this is new patch that integrates XWin and the window manager. Kensuke Matsuzaki How does this patch change what Rootless mode already does? I thought that rootless mode already integrated the window manager? I have windowmaker running, and when I'm in rootless mode the xterms come up with windowmaker frames. Can you integrate with the Windows windows manager? _ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=features/junkmail
Re: Rootless mode
Please do not release your own test version. As I said, I am waiting for the repeating window movement to be looked into and possibly fixed. I do not wish to get lots of duplicate bug reports for such an obvious bug. If Kensuke says it doesn't happen with his version, then I will rebuild mine and try it again. Harold John Buttery wrote: * Kensuke Matsuzaki [EMAIL PROTECTED] [2002-12-12 21:10:54 +0900]: To build XWin.exe, please read http://xfree86.cygwin.com/docs/cg/cygwin-xfree-cg.html I should have figured that out without being told, sorry. I'm now getting the source via CVS. There sure is a lot of it; I hope x86 is still a standard architecture when it finishes. :p The patch can be applied as shown in the following command $pwd ~/x-devel/xc/programs/Xserver/hw/xwin $bunzip2 rootless.patch.bz2 $patch -p1 rootless.patch I edited xc/config/cf/cygwin.rules to disable version controlled .dll, but is it right? I don't understand this question. If you're asking if a change you made causes your instructions to break, I'll let you know when I try them. When I get a modified version of XWin.exe out of this process, can other people just download it and drop it in to their own Cygwin installs? (presumably they would back up their current XWin.exe first) Let me know...if this is the case, I'll put my modified version online somewhere for people who don't have the patience/bandwidth/disk space to build their own. After the patch is applied, do I just go back up to the root directory and run make? Then find the built XWin.exe and copy it over to the directory where the current one lives?
Re: X server window size
Corneliu, Yes, I have seen a similar problem before. I have not yet had a chance to look into it and not that many people are complaining about it :) Harold Corneliu Rudeanu wrote: With the current version of XFree86 (XFree86-bin-4.2.0-2; XFree86-xserv-4.2.0-16) I am getting the following: I start a (xdm based) session on another (linux) X server, using a bat which does: start /B XWin -indirect linux This creates a window as large as if being maximized. Ok. If either I click on Quit button (from the xdm window) or choose a host, login and then logout from it, the Cygwin/XFree86 window is closed and opened again but shrinked by some 4-5 pixels in width and about the height of the taskbar in height. Repeating the procedure (clicking Quit or along several sessions on the linux machine) the Cygwin/XFree86 window gets smaller and smaller. This behaviour dissapears when I specify -screen 0 1152 844 to XWin command. Is this a known issue? I unsuccessfuly searched in the archives for it. Am I missing something? Another problem (which I did not noticed in *some* previous version) was redraw not being performed properly for eg when moving windows. I work only on another computer (linux) with DISPLAY set to Cygwin machine. Sorry I can't be more specific. Best regards, Corneliu Rudeanu
Re: Rootless mode
On Thu, 12 Dec 2002 11:22:45 -0500, Harold L Hunt II [EMAIL PROTECTED] wrote: David, I have not released a test version because the version that Kensuke sent me didn't work at all. It just kept repeating the dragging of a window if you ever moved one. I sent a report to the list but he never responded. I didn't want to bug him about it for awhile in case he had seen it... but it seems now that he never saw it. I will release a test version once there are no obvious bugs. I mean, are we really going to benefit from 100 people saying ``multiwindow mode doesn't work, it keeps doing foo''? I don't think so :) I'll add to the description of the bug. If I try to launch any application (xterm, for instance) the window just slowly moves diagonally across the screen regardless of if I try to move it. Eventually it disappears, never to be seen again. If I intercept it and move it, it behanves as you reported. -- Jonathan Fosburgh Software Systems Specialist III AIX/SAN Administrator University of Texas MD Anderson Cancer Center Houston, TX
Re: Rootless mode
* Harold L Hunt II [EMAIL PROTECTED] [2002-12-12 11:24:25 -0500]: Please do not release your own test version. As I said, I am waiting for the repeating window movement to be looked into and possibly fixed. I do not wish to get lots of duplicate bug reports for such an obvious bug. If Kensuke says it doesn't happen with his version, then I will rebuild mine and try it again. OK; sorry, I guess I got a little ahead of myself. I'm one of those people that's been waiting for a Windows-as-window-manager X server for a long time, so I was excited. :) -- John Buttery (Web page temporarily unavailable) msg04473/pgp0.pgp Description: PGP signature
Re: Rootless mode
Harold, I could not reproduce that bug, but now I found that this bug occur when Show window contents while dragging enabled. Maximizing a window never stop too. Matsuzaki Kensuke
Re: Rootless mode
Kensuke, Hmm... that sounds about right. Windows 2000 and Windows XP have that option on by default, I believe. Are you going to try to debug this? Harold Matsuzaki Kensuke [EMAIL PROTECTED] said: Harold, I could not reproduce that bug, but now I found that this bug occur when Show window contents while dragging enabled. Maximizing a window never stop too. Matsuzaki Kensuke
RE: Rootless mode
This may be obvious, but the difference between the two is that Windows sends you a bunch of messages when you're dragging the window when fulldrag is in effect... So it might be some message handler repositioning the window during one of those notifications... Maybe a WM_SIZE or WM_WINDOWPOSCHANGING... Anyhow, I think running spy and just looking in on the stream of messages that a window doing fulldrag receives could be a way to figure out which message. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Harold L Hunt Sent: Thursday, December 12, 2002 1:33 PM To: [EMAIL PROTECTED] Subject: Re: Rootless mode Kensuke, Hmm... that sounds about right. Windows 2000 and Windows XP have that option on by default, I believe. Are you going to try to debug this? Harold Matsuzaki Kensuke [EMAIL PROTECTED] said: Harold, I could not reproduce that bug, but now I found that this bug occur when Show window contents while dragging enabled. Maximizing a window never stop too. Matsuzaki Kensuke
Re: Rootless mode
Harold L Hunt II wrote: David, I have not released a test version because the version that Kensuke sent me didn't work at all. It just kept repeating the dragging of a window if you ever moved one. I sent a report to the list but he never responded. I didn't want to bug him about it for awhile in case he had seen it... but it seems now that he never saw it. I will release a test version once there are no obvious bugs. I mean, are we really going to benefit from 100 people saying ``multiwindow mode doesn't work, it keeps doing foo''? I don't think so :) Harold OK, I see the reason now, thanks for explaining. In the mean time has been quite fun getting it to run :-) I'm quite happy with not moving windows as long as I know that it's to be avoided. David -- Micro$oft is not an answer. It is a question. The answer is 'no'.
Re: Rootless mode -multiwindow bug
I was able to reproduce the -multiwindow moving bug on two Win2k machines, and Show window contents while dragging does not seem to have any effect on those machines. (I already had it switched off) Neither does disabling ActiveDesktop help.
Installation
I have got cygwin on a cd-rom because I use a 56K modem. I have read the FAQ s that are on the cd and tried every solution that is there but it will not install. Every time when the packeges should come in the sceen it says that there is nothing availeble. Should I first unzip every folder that is on the cd or is there an other way of doing this. I know that I am a dumb Windows user but I can not get cygwin working this way. I would be gratefull for any help. Dirk Dorst
Windows Manager
Hello! For default, when I execute the startxwin.bat file, the twm windows manager is loaded... How I can configure X to use the windows of my Windows system? I have any problems of fonts visualisation... How I can configure X so that use the server of fonts of the remote system? Thanks! Martín.
Email status flags
John, Whatever you are doing to your email to cause it to be flagged as important and needing followup, please stop. I sort the Cygwin/XFree86 email by the importance flag and I have a list of emails that I need to follow up on for the Cygwin/XFree86 project. Some flags in your email are causing your emails to appear at the top of my email list, where they do not belong because I have not flagged them for followup. Now, this is probably just some default setting in your mail client, so please try to find a way to disable it. Or, if you like this setting, please refrain from using it when sending emails to the Cygwin/XFree86 mailing list. Thanks so much, Harold
RE: Rootless mode
Looking at Kensuke's code, he uses an internal queue to store messages that need to be processed by child windows. For some reason this queue is not being properly reset when the messages are processed. Or, it could be as simple as returning the wrong value after processing these messages, causing Windows to send them again. Harold -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Jean-Claude Gervais Sent: Thursday, December 12, 2002 1:52 PM To: [EMAIL PROTECTED] Subject: RE: Rootless mode This may be obvious, but the difference between the two is that Windows sends you a bunch of messages when you're dragging the window when fulldrag is in effect... So it might be some message handler repositioning the window during one of those notifications... Maybe a WM_SIZE or WM_WINDOWPOSCHANGING... Anyhow, I think running spy and just looking in on the stream of messages that a window doing fulldrag receives could be a way to figure out which message. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Harold L Hunt Sent: Thursday, December 12, 2002 1:33 PM To: [EMAIL PROTECTED] Subject: Re: Rootless mode Kensuke, Hmm... that sounds about right. Windows 2000 and Windows XP have that option on by default, I believe. Are you going to try to debug this? Harold Matsuzaki Kensuke [EMAIL PROTECTED] said: Harold, I could not reproduce that bug, but now I found that this bug occur when Show window contents while dragging enabled. Maximizing a window never stop too. Matsuzaki Kensuke
RE: Rootless mode
Other than the WM_PAINT message, I don't think Windows ever sends the same message again. On a WM_PAINT, if you do not mark validate the region by calling BeginPaint and EndPaint and return from processing the WM_PAINT, you WILL receive a WM_PAINT again. On the other hand, if you call GetMessage without PM_REMOVE, the message retrieved is not removed from the queue. But if that were the case, the SAME message would keep being retrieved because the Windows message queue is more or less a FIFO structure. There might be some recursion at work here though, because if the child windows notify the parent and in response the parent posts or sends the same message again, things could get hairy. Probably not though, because they flood of messages does stop, right? I'm sure running Spy++ would yield interesting information. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Harold L Hunt II Sent: Thursday, December 12, 2002 5:55 PM To: [EMAIL PROTECTED] Subject: RE: Rootless mode Looking at Kensuke's code, he uses an internal queue to store messages that need to be processed by child windows. For some reason this queue is not being properly reset when the messages are processed. Or, it could be as simple as returning the wrong value after processing these messages, causing Windows to send them again. Harold -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Jean-Claude Gervais Sent: Thursday, December 12, 2002 1:52 PM To: [EMAIL PROTECTED] Subject: RE: Rootless mode This may be obvious, but the difference between the two is that Windows sends you a bunch of messages when you're dragging the window when fulldrag is in effect... So it might be some message handler repositioning the window during one of those notifications... Maybe a WM_SIZE or WM_WINDOWPOSCHANGING... Anyhow, I think running spy and just looking in on the stream of messages that a window doing fulldrag receives could be a way to figure out which message. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Harold L Hunt Sent: Thursday, December 12, 2002 1:33 PM To: [EMAIL PROTECTED] Subject: Re: Rootless mode Kensuke, Hmm... that sounds about right. Windows 2000 and Windows XP have that option on by default, I believe. Are you going to try to debug this? Harold Matsuzaki Kensuke [EMAIL PROTECTED] said: Harold, I could not reproduce that bug, but now I found that this bug occur when Show window contents while dragging enabled. Maximizing a window never stop too. Matsuzaki Kensuke
Re: quandary with pthreads
I have attached a bunch of mutex testcases borrowed from the pthreads-win32 project. Thomas On Wed, 11 Dec 2002, Thomas Pfaff wrote: Rob, thank you that you are willing to add my patches. FYI, the logic of the mutex stuff is running in pthreads-win32 about 1 1/2 years now unchanged (guess who has contributed). I will try to add some of the testcases from pthreads-win32. On Wed, 11 Dec 2002, Robert Collins wrote: I *really* want to get Thomas's contribution properly reviewed in time for 1.3.18. Unfortunately my cygwin development machine's hard drive is failing, and it will be off the air for ~ 1 week. So, I'd like to suggest the following: Thomas.. If you can write test cases for all the new capabilities you introduce - see the ones I added based on your previous work, and the README in the test-suite dir. Then I will be happy to have your changes committed if the test suite shows no pthreads regressions on a win9x machine, and on an NT kernel based machine. pthreads-tests.tar.gz Description: GNU Zip compressed data
Re: Small security patches
Christopher Faylor wrote: Actually, if you can get away without using a constructor that would be best. Constructors are a noticeable part of cygwin's startup cost. - Is there a C++ way to initialize a constant class and have it in the .text section, as const int i = 1; would be? - If not, I can get the desired effect by using gcc Asm Labels, like int foo asm (myfoo) = 2; Would that be acceptable in Cygwin? Pierre
What bash --login -i has done for my project?
My project runs well in cygwin's bash. But it fails in cmd.exe with the following error: Exception: STATUS_ACCESS_VIOLATION at eip=00558706 eax= ebx=00691E64 ecx= edx= esi=00692140 edi=005BF220 ebp=0022F430 esp=0022F388 program=D:\cygwin\home\Administrator\win9.1.4\rcssserver.exe cs=001B ds=0023 es=0023 fs=0038 gs= ss=0023 Stack trace: Frame Function Args 0022F430 00558706 (0022F880, 00691E64, 0022F890, 00555C18) ... End of stack trace What cygwin's init script bash --login -i has done that seems necessary for my project? Thanks in advance _ ÓëÁª»úµÄÅóÓѽøÐн»Á÷£¬ÇëʹÓà MSN Messenger: http://messenger.msn.com/cn -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: ntsec and remote copy
Hi Pierre, thanks for your reply! That indeed turned out to be the case. The owner of the remote file has a SID that does not appear in your passwd file and thus it cannot be mapped to a uid. Cygwin then uses uid = -1 This also explains what you describe in your follow up message. Perhaps the Windows security gui or the cacls program will reveal the identity of the owner. Pierre What happened was the following: - I had trimmed down my /etc/passwd file (I think I read somewhere it was advisable to remove users from there for which you don't want access. For example when running telnetd, this seems to be the only way to restrict access). In the process, I also removed the Administrators group from /etc/passwd. (I don't really want somebody to break into my system as Administrator via telnet) - I am member of the Administrators group, so as the ntsec doc mentions, all files I create are owned by Administrators, but that wasn't in /etc/passwd, so it's an unrecognised user. Adding Administrators to /etc/passwd solved my problems. Remaining questions: - is there another way to prevent specific users access to telnet or ftp ? (or ssh when I get round to installing sshd) ? - It seems cygwin does not use the above mentioned behaviour that files that I create are owned by Administrators, instead they are owned by myself. This is fine for me, but I guess worth mentioning in the doc. - I still don't understand why certain cygwin programs could read the file and others not (see mail below). It can't be NT, because I could obviously read/write all files I created myself using NT programs. So, I have the impression that some cygwin programs use ntsec in different ways. For example, cat or vi could read the file with an unrecognised owner, but test -r couldn't. (I recently reinstalled and upgraded the whole of cygwin, so I don't think it's because I have old versions lying around). Thanks for the help Kris -Original Message- From: Kris Thielemans [mailto:[EMAIL PROTECTED]] Sent: 10 December 2002 16:26 To: Gnuwin Subject: RE: ntsec and remote copy Oh yes, a related remark. On the file with the funny UID, some things work, some don't. For instance, I can vi it, but I cannot write it. This seems to say that some cygwin programs call this file readable, but others don't... Bearing in mind the premissions that it is said to, I would actually claim that I should NOT be able to read it. Here's an illustration of 'test' that cannot read it. $ ls -l test.txt -rwx--+ 1 65535None 1003 Dec 10 00:24 test.txt $ if [ -r test.txt ]; then echo 'I can read it'; else echo 'I cannot read it'; fi I cannot read it $ chown kris test.txt $ ls -l test.txt -rwx--1 kris None 1003 Dec 10 00:24 test.txt $ if [ -r test.txt ]; then echo 'I can read it'; else echo 'I cannot read it'; fi I can read it I also see now that not only files copied remotely have this strang eUID, but others as well (maybe older files?) Kris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: postgresql postmaster and LC_MESSAGES EN_US problem
Heitzso, On Wed, Dec 11, 2002 at 12:48:20PM -0500, Heitzso wrote: I just updated to current postgresql and got past the ipc-daemon upgrade problem (yes it's noted in the README) but then got snagged on the LC_MESSAGES EN_US setting in postgresql.conf. [snip] I would try the [EMAIL PROTECTED] list -- you may have better luck there. Jason -- PGP/GPG Key: http://www.tishler.net/jason/pubkey.asc or key servers Fingerprint: 7A73 1405 7F2B E669 C19D 8784 1AFD E4CC ECF4 8EF6 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: pipe improvements in snapshot
Christopher Faylor writes: In the most recent cygwin snapshot (uploading now) I've attempted to work around the 10ms delay in pipe reads. snip Please check out the latest snapshot and report here if there are problems. I haven't yet tried this on Windows 9x class systems so it's entirely possible that there is a problem there. With a Cygwin CVS tree I built this AM (2) below rxvt has problems when configuring a project from a build directory ie $TOP_DIR src build cd $TOP_DIR/build ../configure The problems vary from run to run ranging from 'error parsing uname' to program hang recoverable with 'ctrl-C' making me suspect a possible data loss Using the gcc option -pipe seems to allow the configure process Using the -pipe flag I have not seen the uname error but it does eventually hang when actually creating the Makefiles This project uses libtool With the CVS tree build (1) below I do not experience this (1) 10-Dec-2002 7:10:38a6,683,847 cygwin-20021210.tar.gz (2) 12-Dec-2002 6:09:42a6,688,318 cygwin-20021212.tar.gz The times are EST and the tarballs are created by a script that does a CVS update, make, make tarball which takes ~20 minutes so the CVS update time should be approximately 20 minutes earlier Any 'tips' as to how to best debug this appreciated Norman -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: postgresql postmaster and LC_MESSAGES EN_US problem
Heitzso, On Wed, Dec 11, 2002 at 12:48:20PM -0500, Heitzso wrote: I got around this problem (??) by commenting out the 4 localization lines at the end of postgresql.conf in the data directory, but would greatly appreciate any suggestions as to what's south with my setup that would cause this. It appears that having LANG=en_us set before you ran initdb caused this problem. My recommendation it to either unset LANG or use LANG=C before you rerun initdb. IIRC, Cygwin does not fully support locale. Jason -- PGP/GPG Key: http://www.tishler.net/jason/pubkey.asc or key servers Fingerprint: 7A73 1405 7F2B E669 C19D 8784 1AFD E4CC ECF4 8EF6 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: postgres 7.3 postmaster: invalid value for option 'LC_MESSAGES':'en_US'
Jari, On Wed, Dec 11, 2002 at 10:27:43PM +0200, Jari Aalto wrote: $ env | grep LC_ LC_ALL=en_US See the following: http://cygwin.com/ml/cygwin/2002-12/msg00666.html Jason -- PGP/GPG Key: http://www.tishler.net/jason/pubkey.asc or key servers Fingerprint: 7A73 1405 7F2B E669 C19D 8784 1AFD E4CC ECF4 8EF6 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: pipe improvements in snapshot
I have similar experience. Examining the processes when the ./configure hangs reveals that hanging leaf seems to be always 'sed', running as a part of config.status script creating Makefiles etc. I'm running XPPro, the same results were achieved when running configure inside xterm or inside FSFemacs (X version). Attached is cygcheck output, but it was already running *after* I reverted to 20021209 snapshot. If someone needs cygcheck output with offending snapshot, pls let me know. Pavel P.S.: Sorry for 'me too' type of post, butI think that this can be important.. [EMAIL PROTECTED] wrote: With a Cygwin CVS tree I built this AM (2) below rxvt has problems when configuring a project from a build directory ie $TOP_DIR src build cd $TOP_DIR/build ../configure The problems vary from run to run ranging from 'error parsing uname' to program hang recoverable with 'ctrl-C' making me suspect a possible data loss Using the gcc option -pipe seems to allow the configure process Using the -pipe flag I have not seen the uname error but it does eventually hang when actually creating the Makefiles This project uses libtool With the CVS tree build (1) below I do not experience this (1) 10-Dec-2002 7:10:38a6,683,847 cygwin-20021210.tar.gz (2) 12-Dec-2002 6:09:42a6,688,318 cygwin-20021212.tar.gz The times are EST and the tarballs are created by a script that does a CVS update, make, make tarball which takes ~20 minutes so the CVS update time should be approximately 20 minutes earlier Any 'tips' as to how to best debug this appreciated Norman -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ Cygwin Win95/NT Configuration Diagnostics Current System Time: Thu Dec 12 13:42:19 2002 Windows XP Professional Ver 5.1 Build 2600 Service Pack 1 Path: f:\cygwin\usr\local\bin f:\cygwin\bin f:\cygwin\usr\X11R6\bin c:\cygwin-mounts\opt\gnome2\bin f:\bin f:\Program Files\Microsoft Platform SDK\Bin f:\Program Files\Microsoft Visual Studio\Common\MSDev98\Bin f:\Program Files\Microsoft Visual Studio\Common\Tools f:\Program Files\Wise for Windows Installer f:\Program Files\Microsoft Visual Studio\VC98\bin f:\jdk1.3.1\bin f:\MetaDeveloper\hcarc\bin f:\src\psuite\main\src\buildengine f:\bin SysDir: F:\WINDOWS\System32 WinDir: F:\WINDOWS CYGWIN = `nowinsymlinks tty error_start=f:\cygwin\bin\dumper.exe' HOME = `c:\cygwin-mounts\home\pholejs' MAKE_MODE = `unix' PWD = `/home/pholejs' USER = `pholejs' ALLUSERSPROFILE = `F:\Documents and Settings\All Users' APPDATA = `F:\Documents and Settings\PHolejs\Application Data' BASEMAKE = `F:\Program Files\Microsoft Platform SDK\Include\BKOffice.Mak' BASHRC_INITED = `BASHRC_INITED' BKOFFICE = `F:\Program Files\Microsoft Platform SDK\.' CLASSPATH = `f:\Gemplus\GemXpresso.rad3\lib\gse\gse_gxp211_pk.jar;f:\Gemplus\GemXpresso.rad3\lib\cryptix-gemxpresso.jar;f:\Gemplus\GemXpresso.rad3\lib\cryptix-jce-api.jar' COMMONPROGRAMFILES = `F:\Program Files\Common Files' COMPUTERNAME = `PRGPC009' COMSPEC = `F:\WINDOWS\system32\cmd.exe' DISPLAY = `:0.0' DXSDKROOT = `F:\Program Files\Microsoft Platform SDK\.' EDITOR = `nano' HOMEDRIVE = `F:' HOMEPATH = `\Documents and Settings\PHolejs' IMNINST = `help' IMNINSTSRV = `F:\IMNnq_NT' INCLUDE = `F:\Program Files\Microsoft Platform SDK\Include;F:\Program Files\Microsoft Platform SDK\Src\WTL\Include;F:\Program Files\Microsoft Visual Studio\VC98\atl\include;F:\Program Files\Microsoft Visual Studio\VC98\mfc\include;F:\Program Files\Microsoft Visual Studio\VC98\include' INETSDK = `F:\Program Files\Microsoft Platform SDK\.' JAVA_HOME = `F:\jdk1.3.1' JC21_HOME = `C:\Store\Chipcards\javacard\jc211' JDKHOME = `F:\jdk1.3.1' LIB = `F:\Program Files\Intel\Compiler50\ia32\lib;F:\Program Files\Microsoft Platform SDK\Lib;F:\Program Files\Microsoft Visual Studio\VC98\mfc\lib;F:\Program Files\Microsoft Visual Studio\VC98\lib' LOGNAME = `pholejs' LOGONSERVER = `\\PRGOP02' LS_COLORS = `no=00:fi=00:di=01;34:ln=01;36:pi=47;33:so=01;35:bd=47;32;01:cd=47;32;01:or=47;31;01:ex=01;31:*.tar=31:*.tgz=31:*.taz=31:*.zip=31:*.z=31:*.Z=31:*.gz=31:*.bz2=31:*.rpm=31:*.jpg=01;35:*.png=01;35:*.gif=01;35:*.bmp=01;35:*.ppm=01;35:*.tga=01;35:*.xbm=01;35:*.xpm=01;35:*.tif=01;35:*.mpg=01;35:*.avi=01;35:*.fli=01;35:*.gl=01;35:*.dl=01;35:' MANPATH = `:/usr/ssl/man' MSDEVDIR = `F:\Program Files\Microsoft Visual Studio\Common\MSDev98;F:\Program Files\Debugging Tools for Windows' MSSDK = `F:\Program Files\Microsoft Platform SDK\.' MSTOOLS = `F:\Program Files\Microsoft Platform SDK\.' NUMBER_OF_PROCESSORS = `1' OS = `Windows_NT' PATHEXT = `.COM;.EXE;.BAT;.CMD;.VBS;.VBE;.JS;.JSE;.WSF;.WSH' PROCESSOR_ARCHITECTURE = `x86' PROCESSOR_IDENTIFIER = `x86 Family 6 Model 4 Stepping 4, AuthenticAMD
Problems with GCC and C_INCLUDE_PATH
I'm having problems with GCC and the C_INCLUDE_PATH, if i add the path to the command line with -Iw:\source then all is ok, but if I add the same path to the environment variable and remove the -I from the command line it just will not see the header file, whats wrong?? I get the following error: System__Public.h: No such file or directory this file is in w:\source, Please help -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: postgresql postmaster and LC_MESSAGES EN_US problem
Just want to note that my LANG was not set when I ran initdb which left me with LC_MESSAGES=EN_US. I tried setting LANG=C in my environment and wiped my postgres data directory and re initdb'ed and still ended up with LC_MESSAGES=EN_US. I'm guessing there's another environment variable that initdb may be working with? In any case, just commenting out the last 4 lines in postgresql.conf seems to take care of the problem for now. Heitzso -Original Message- From: Jason Tishler [mailto:[EMAIL PROTECTED]] Sent: Thursday, December 12, 2002 7:41 AM To: [EMAIL PROTECTED] Subject: Re: postgresql postmaster and LC_MESSAGES EN_US problem Heitzso, On Wed, Dec 11, 2002 at 12:48:20PM -0500, Heitzso wrote: I got around this problem (??) by commenting out the 4 localization lines at the end of postgresql.conf in the data directory, but would greatly appreciate any suggestions as to what's south with my setup that would cause this. It appears that having LANG=en_us set before you ran initdb caused this problem. My recommendation it to either unset LANG or use LANG=C before you rerun initdb. IIRC, Cygwin does not fully support locale. Jason -- PGP/GPG Key: http://www.tishler.net/jason/pubkey.asc or key servers Fingerprint: 7A73 1405 7F2B E669 C19D 8784 1AFD E4CC ECF4 8EF6 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: sftp ssh chroot
I solved my problem regarding jailing users in cygwin. If anyone is interested, the patch for linux works fine for cygwin. Here are the steps: 1) Download http://chrootssh.sourceforge.net/patches/osshChroot-3.5.diff 2) Go to http://www.openssh.org/portable.html and get the sourcefile openssh-3.5p1.tar.gz. 3) tar xzf openssh-3.5p1.tar.gz patch -p0 osshChroot-3.5.diff cd openssh-3.5p1.tar.gz ./configure --prefix=/usr --sysconfdir=/etc --libexecdir=/usr/sbin (Be sure to check that configure script, I can't remember the exact command.) make make install 4) Now run ssh-host-config -y or however you would normally set it up. 5) Create a user with the shell script I wrote, or write your own. #!/bin/sh # add a caged ssh user # usage sshuser username # make a directory according to the username if [ $1 ] then mkdir /home/$1 mkdir /home/$1/bin mkdir /home/$1/home mkdir /home/$1/usr mkdir /home/$1/usr/sbin mkdir /home/$1/tmp cp /bin/bash /home/$1/bin cp /bin/cp /home/$1/bin cp /bin/cygwin1.dll /home/$1/bin cp /bin/rm /home/$1/bin cp /bin/mv /home/$1/bin cp /bin/ls /home/$1/bin cp /bin/cygcrypto.dll /home/$1/bin cp /usr/sbin/sftp-server /home/$1/usr/sbin echo Remember to change /home/$1 to echo /home/$1/./home in /etc/passwd else echo Usage sshuser username fi 6) Modify the users directory in /etc/passwd file to be home/username/./home 7) Hope this helps, and sorry if someone already posted this although I wish I had seen it if they had cause it would have saved me a lot of trouble, although trouble can be a good thing when it's a learning experience! =) -- Nothing would please me more than being able to hire ten programmers and deluge the hobby market with good software. -- Bill Gates 1976 We are still waiting -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: using MFC with cygwin
Hm, someone's been digging around in the email archives! ;-) Actually, that's good to see. While I won't argue the slight semantic differences pointed out here between what I said and this response, I believe we both agree it makes the same point, at least in the context of Cygwin. Anyone that wants to can try to port MFC to any Win32-based compiler they want but they will end up with something restricted by the Microsoft license. Larry Original Message: - From: Samuel [EMAIL PROTECTED] Date: Wed, 11 Dec 2002 20:13:52 -0800 To: [EMAIL PROTECTED] Subject: Re: using MFC with cygwin - Original Message - From: [EMAIL PROTECTED] To: [EMAIL PROTECTED]; [EMAIL PROTECTED] Sent: Thursday, August 22, 2002 6:59 AM Subject: RE: using MFC with cygwin No one has ported MFC to any compiler besides VC++. And while it could be done, the benefits are minimal since the Microsoft license would prevent you from distributing the result. MFC was not ported to VC; it did not exist prior to VC. I am 95% sure that (some) Borland compilers license (include) MFC. You might be correct that it has not been ported, but Microsoft licenses it. The VC compiler includes the MFC source code so it would not be too much work to port (unlicensed). I assume that if source code existed that did the same thing as MFC then it could be entirely legal. I think the original question was about a console program that used MFC and I would probably suggest not using MFC for that. --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.419 / Virus Database: 235 - Release Date: 11/14/2002 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ mail2web - Check your email from the web at http://mail2web.com/ . -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: postgresql postmaster and LC_MESSAGES EN_US problem
Heitzso, On Thu, Dec 12, 2002 at 09:34:37AM -0500, Heitzso wrote: I'm guessing there's another environment variable that initdb may be working with? Grepping initdb yields: $ fgrep LANG /usr/bin/initdb for var in PGLC_$arg PGLOCALE LC_ALL LC_$arg LANG; do Did you set any of the above? Jason -- PGP/GPG Key: http://www.tishler.net/jason/pubkey.asc or key servers Fingerprint: 7A73 1405 7F2B E669 C19D 8784 1AFD E4CC ECF4 8EF6 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: using MFC with cygwin
I believe Metrowerks CodeWarrior for Windows comes with MFC. http://www.metrowerks.com/ CodeWarrior is a much more standards-compliant compiler than VC++ and I think even more than g++, but it has a compatibility mode that allows it to compile a lot of broken VC++ code. Maybe if you ported MFC, you could distribute a patch instead of your port. Alternatively, maybe you could help me port ZooLib to Cygwin. If you succeeded (or helped me to) the changes could be folded into the ZooLib distribution and everyone would benefit. ZooLib is open source under the MIT license and is a lot nicer API than MFC. http://zoolib.sourceforge.net/ some documentation can be had at http://www.goingware.com/zoolib/cookbook/ Best, Michael D. Crawford GoingWare Inc. - Expert Software Development and Consulting http://www.goingware.com/ [EMAIL PROTECTED] Tilting at Windmills for a Better Tomorrow. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: problems with CPAN and cygwin 1.3.12
Hallo JoNO, At first, make sure that in your Cygwin environment are no paths in PATH with spaces like this: /cygdrive/c/Program Files/SecureCRT 3.0 Then check the permissions on your executables: $ ls -l /bin/perl should look like this: -rwxr-xr-x1 gerrit Domänen-14848 Dec 9 17:56 /bin/perl* ^ ^^ ^^ ^ including at least the 'r-x' for all groups. Also try to update the MakeMaker module to the latest version, like Soren suggested. Gerrit -- =^..^=
Re: Problems with GCC and C_INCLUDE_PATH
On Thu, 12 Dec 2002, Allan Crook wrote: I'm having problems with GCC and the C_INCLUDE_PATH, if i add the path to the command line with -Iw:\source then all is ok, but if I add the same path to the environment variable and remove the -I from the command line it just will not see the header file, whats wrong?? I get the following error: System__Public.h: No such file or directory this file is in w:\source, Please help Hmm, let me get this straight: you're adding a Windows-style path to the include path of the Cygwin gcc? One obvious solution would be to change the path to POSIX, something like /cygdrive/w/source. If you insist on using Windows-style paths, this may be a quoting issue (you may have to quote your backslashes twice). Igor -- http://cs.nyu.edu/~pechtcha/ |\ _,,,---,,_[EMAIL PROTECTED] ZZZzz /,`.-'`'-. ;-;;,_[EMAIL PROTECTED] |,4- ) )-,_. ,\ ( `'-' Igor Pechtchanski '---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow! Oh, boy, virtual memory! Now I'm gonna make myself a really *big* RAMdisk! -- /usr/games/fortune -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
URL correction
The correct URL for the; Cygwin GNOME 1.4 binary packages under NEWS is; http://sourceforge.net/projects/cygnome/ Have Fun Sterling -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Why No such file ... in path ? 1.3.17
Kun, I need to correct myself (since no one else did!)... At 19:30 2002-12-11, Randall R Schulz wrote: Kun, ... Include files required during C or C++ compilation are located via a separate search path maintained by the compiler and not driven by any environment variable, at least not in the Cygwin compiler, which is GCC. There _are_ environment variables to tell GCC where to find files. This is an excerpt from the (rather long) man page for gcc: -==--==--==--==--==--==--==--==--==--==--==--==--==--==--==- Some additional environments variables affect the behavior of the preprocessor. CPATH C_INCLUDE_PATH CPLUS_INCLUDE_PATH OBJC_INCLUDE_PATH Each variable's value is a list of directories separated by a special character, much like PATH, in which to look for header files. The special character, PATH_SEPARATOR, is target-dependent and determined at GCC build time. For Windows-based targets it is a semicolon, and for almost all other targets it is a colon. CPATH specifies a list of directories to be searched as if specified with -I, but after any paths given with -I options on the command line. The environment variable is used regardless of which language is being preprocessed. The remaining environment variables apply only when preprocessing the particular language indicated. Each specifies a list of directories to be searched as if specified with -isystem, but after any paths given with -isystem options on the command line. -==--==--==--==--==--==--==--==--==--==--==--==--==--==--==- I believe the rest of what I wrote was accurate. Randall Schulz Mountain View, CA USA ... Randall Schulz Mountain View, CA USA At 19:08 2002-12-11, Wu Kun wrote: Hi, all: When I use CygWin 3.17.1 to build Sun's CLDC 1.0.x, an error occurs: ../../src/check_class.c(30) : fatal error C1083: Cannot open include file: 'ctype.h': No such file or directory But the file ctype.h is right in the /usr/include, which has been set in the $PATH. Can some body help me out ? Best Regards Wu Kun -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: pipe improvements in snapshot
On Thu, Dec 12, 2002 at 07:32:16AM -0500, Norman Vine wrote: Christopher Faylor writes: In the most recent cygwin snapshot (uploading now) I've attempted to work around the 10ms delay in pipe reads. snip Please check out the latest snapshot and report here if there are problems. I haven't yet tried this on Windows 9x class systems so it's entirely possible that there is a problem there. With a Cygwin CVS tree I built this AM (2) below rxvt has problems when configuring a project from a build directory ie $TOP_DIR src build cd $TOP_DIR/build ../configure The problems vary from run to run ranging from 'error parsing uname' to program hang recoverable with 'ctrl-C' making me suspect a possible data loss Using the gcc option -pipe seems to allow the configure process Using the -pipe flag I have not seen the uname error but it does eventually hang when actually creating the Makefiles This project uses libtool With the CVS tree build (1) below I do not experience this (1) 10-Dec-2002 7:10:38a6,683,847 cygwin-20021210.tar.gz (2) 12-Dec-2002 6:09:42a6,688,318 cygwin-20021212.tar.gz The times are EST and the tarballs are created by a script that does a CVS update, make, make tarball which takes ~20 minutes so the CVS update time should be approximately 20 minutes earlier Any 'tips' as to how to best debug this appreciated - Attach to the hung process with gdb and see where it is hung. - Provide cygcheck output. - Run under strace and see if you can infer where hangs or problems are occurring. cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Script Doesn't Work Correctly with Cron
I have a script that works when I run it from the command line, but when it is executed with cron, it produces a null file. I am sure that cron is executing it because it creates the file. It is just not populated. Here is the script. I want to query an Oracle database every couple of minutes, capture some data, and append it to a flat file. I want to keep 2000 lines in the flat file. Can someone spot the error? Again, it works run from the command line. #!/bin/bash sqlplus -s scott@mydb up @ss.sql | grep -v Enter password: monsql.lst tail -2000 monsql.lst montemp.lst rm -f monsql.lst mv montemp.lst monsql.lst -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
GCC Include Paths
Help, We're trying to make GCC automatically search for required header files, unforunately if we use the C_INCLUDE_PATH environment variable or -I you need to enter every single search directory. For our current project this results in a line over 2000chars long (too long for windows or GCC to handle. Can we somehow tell GCC to search subfolders or is there some other way to do this??? thanks, Allan. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Script Doesn't Work Correctly with Cron
On Thu, 12 Dec 2002, Buck Turgidson wrote: I have a script that works when I run it from the command line, but when it is executed with cron, it produces a null file. I am sure that cron is executing it because it creates the file. It is just not populated. Here is the script. I want to query an Oracle database every couple of minutes, capture some data, and append it to a flat file. I want to keep 2000 lines in the flat file. Can someone spot the error? Again, it works run from the command line. #!/bin/bash sqlplus -s scott@mydb up @ss.sql | grep -v Enter password: monsql.lst tail -2000 monsql.lst montemp.lst rm -f monsql.lst mv montemp.lst monsql.lst This is probably not cron-specific, but try redirecting stderr to a file (by using 2filename) to see if the script produces any errors. Also, are all the programs you use (sqlplus, grep, tail, rm, mv) in the path for the user that cron runs as (most likely system)? Are they executable by system? Igor -- http://cs.nyu.edu/~pechtcha/ |\ _,,,---,,_[EMAIL PROTECTED] ZZZzz /,`.-'`'-. ;-;;,_[EMAIL PROTECTED] |,4- ) )-,_. ,\ ( `'-' Igor Pechtchanski '---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow! Oh, boy, virtual memory! Now I'm gonna make myself a really *big* RAMdisk! -- /usr/games/fortune -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: GCC Include Paths
Yes, we have faced a similar problem. The problem is with the windows command shell which limits a line to 2048 characters. I don't know how to make that problem go away. But you shouldn't get the same problem from a cygwin shell. Thanks, Vijay -Original Message- From: Allan Crook [mailto:[EMAIL PROTECTED]] Sent: Thursday, December 12, 2002 10:10 AM To: [EMAIL PROTECTED] Subject: GCC Include Paths Help, We're trying to make GCC automatically search for required header files, unforunately if we use the C_INCLUDE_PATH environment variable or -I you need to enter every single search directory. For our current project this results in a line over 2000chars long (too long for windows or GCC to handle. Can we somehow tell GCC to search subfolders or is there some other way to do this??? thanks, Allan. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: GCC Include Paths
Vijay, Allan, Cygwin is similarly limited. All Unix / POSIX systems have such a limit, but Cygwin's limit is much smaller than the typical limit on a Unix (-like) system. I don't know it for a fact, but I'm pretty sure this limit is not imposed by Cygwin itself (why would it?) but is a Windows limitation. Most of the time xargs resolves this, but obviously that's not the case for -I arguments to gcc or in general when the argument overload originates in auxiliary arguments that name file system entities and which must all be present concurrently. I suggest that you create a separate directory containing links (symbolic links or, if feasible (*), hard links) to all the (required) include files in all the include directories. Then you can side-step the argument list limit. (*) Hard links are an option (the preferred option, actually) if the file systems are NTFS and the target of the link is on the same file system volume as the link. If the latter does not hold, Cygwin will copy the files, so this approach will still work, but you won't be using actual links. Lastly, in keeping with the pedantic theme of which I'm so fond, I put target in quotes since hard links are all co-equal and there is no original or target vs. link distinction, just alternate names for the same underlying file. Good luck. Randall Schulz Mountain View, CA USA At 10:29 2002-12-12, Vijay Sampath wrote: Yes, we have faced a similar problem. The problem is with the windows command shell which limits a line to 2048 characters. I don't know how to make that problem go away. But you shouldn't get the same problem from a cygwin shell. Thanks, Vijay -Original Message- From: Allan Crook [mailto:[EMAIL PROTECTED]] Sent: Thursday, December 12, 2002 10:10 AM To: [EMAIL PROTECTED] Subject: GCC Include Paths Help, We're trying to make GCC automatically search for required header files, unforunately if we use the C_INCLUDE_PATH environment variable or -I you need to enter every single search directory. For our current project this results in a line over 2000chars long (too long for windows or GCC to handle. Can we somehow tell GCC to search subfolders or is there some other way to do this??? thanks, Allan. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: GCC Include Paths
Randall, I just tried out a line with 2 characters and it works fine on bash as an input to GCC. At this point I gave up trying to increase the length to find out the limit. Whatever the bash limit is, it is definitely greater than a windows shell. Of course, if you are invoking a bat file or some such thing from a cygwin shell then you will be bounded by the windows limit. Therefore Allan, I suggest trying to move to a pure cygwin enviroment, if there is such a thing. Thanks, Vijay -Original Message- From: Randall R Schulz [mailto:[EMAIL PROTECTED]] Sent: Thursday, December 12, 2002 10:43 AM To: [EMAIL PROTECTED] Subject: RE: GCC Include Paths Vijay, Allan, Cygwin is similarly limited. All Unix / POSIX systems have such a limit, but Cygwin's limit is much smaller than the typical limit on a Unix (-like) system. I don't know it for a fact, but I'm pretty sure this limit is not imposed by Cygwin itself (why would it?) but is a Windows limitation. Most of the time xargs resolves this, but obviously that's not the case for -I arguments to gcc or in general when the argument overload originates in auxiliary arguments that name file system entities and which must all be present concurrently. I suggest that you create a separate directory containing links (symbolic links or, if feasible (*), hard links) to all the (required) include files in all the include directories. Then you can side-step the argument list limit. (*) Hard links are an option (the preferred option, actually) if the file systems are NTFS and the target of the link is on the same file system volume as the link. If the latter does not hold, Cygwin will copy the files, so this approach will still work, but you won't be using actual links. Lastly, in keeping with the pedantic theme of which I'm so fond, I put target in quotes since hard links are all co-equal and there is no original or target vs. link distinction, just alternate names for the same underlying file. Good luck. Randall Schulz Mountain View, CA USA At 10:29 2002-12-12, Vijay Sampath wrote: Yes, we have faced a similar problem. The problem is with the windows command shell which limits a line to 2048 characters. I don't know how to make that problem go away. But you shouldn't get the same problem from a cygwin shell. Thanks, Vijay -Original Message- From: Allan Crook [mailto:[EMAIL PROTECTED]] Sent: Thursday, December 12, 2002 10:10 AM To: [EMAIL PROTECTED] Subject: GCC Include Paths Help, We're trying to make GCC automatically search for required header files, unforunately if we use the C_INCLUDE_PATH environment variable or -I you need to enter every single search directory. For our current project this results in a line over 2000chars long (too long for windows or GCC to handle. Can we somehow tell GCC to search subfolders or is there some other way to do this??? thanks, Allan. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
The application failed to initialize properly (0xc0000022) -- cause and cure
I think this is unrelated to previous messages on this list with a similar subject. Instead, it pertains to a problem that I encountered and solved myself; I thought it might benefit others to hear about it, so here it is. The symptom is that I was trying to start a non-Cygwin program that I'd just built, and saw the above-mentioned popup. It didn't occur to me for a long time that the problem had anything to do with Cygwin, since neither the program -- nor the DLLs on which it depended -- used Cygwin. However, I *had* installed those DLLs by using Cygwin's unzip to unpack a ZIP file. I then noticed that if I simply did chmod +x *.dll on those DLLs, the problem went away. I figure unzip set the permissions on those DLLs to 0640, which seems reasonable from Cygwin's point of view ... unfortunately, though, NT requires those DLLs to be executable; hence the infuriatingly obtuse error popup. -- PGP Fingerprint: 3E7B A3F3 96CA 8958 ACC5 C8BD 6337 0041 C01C 5276 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: GCC Include Paths
Vijay, I guess I was misled by this: /usr/include/limits.h:#define _POSIX_ARG_MAX 4096 /usr/include/sys/syslimits.h:#define ARG_MAX 65536 /* max bytes for an exec function */ Furthermore, /usr/include/limits.h bears a Red Hat copyright and is specifically marked as a part of Cygwin, while /usr/include/sys/syslimits.h bears a Regents of the University of California copyright. Thus I assumed the former was more authoritative. Sorry (again) for the misinformation. Randall Schulz Mountain View, CA USA At 10:53 2002-12-12, Vijay Sampath wrote: Randall, I just tried out a line with 2 characters and it works fine on bash as an input to GCC. At this point I gave up trying to increase the length to find out the limit. Whatever the bash limit is, it is definitely greater than a windows shell. Of course, if you are invoking a bat file or some such thing from a cygwin shell then you will be bounded by the windows limit. Therefore Allan, I suggest trying to move to a pure cygwin enviroment, if there is such a thing. Thanks, Vijay -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: /tmp
Soren A wrote: [...] you can nenver tell another person what they should like in a shell. Well (sorry for OT drift), the main thing was the case-insensitive filename completion, which is invaluable on Windows. I need to look at the tcsh source and see if it is implemented within an #ifdef WIN32 like block. It would be nice if the feature was automatically also available in the cygwin build of tcsh .. -- Shankar. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: GCC Include Paths
Vijay Sampath wrote: I just tried out a line with 2 characters and it works fine on bash as an input to GCC. I think that direct Cygwin-to-Cygwin invocation has a higher limit. If you're calling a Cygwin program from a non-Cygwin program (e.g. CMD.EXE), you're still stuck with Windows limits. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Rsync performance increase through buffering (cygwin/ssh solution?)
Below is a description of a buffering speed improvement. The original message with complete patch for performance can be found here: http://groups.google.com/groups?hl=enlr=ie=UTF-8oe=UTF-8newwindow=1selm=at1i0e%241tj3%241%40FreeBSD.csie.NCTU.edu.tw This may explain the horrendous performance of rsync with ssh under cygwin. Perhaps the way that cygwin emulates these calls. It is especially bad under the circumstances mentioned here - with a large sync of data where the files are identical (-I). The hang fix has been identified elsewhere in cygwin: Check out these links: http://www.cygwin.com/ml/cygwin/2002-10/msg00308.html http://sources.redhat.com/ml/cygwin/2002-09/msg01155.html patched rsync for hang under cygwin: http:// www.addr.com/~dmitry/rsync.exe Jim [EMAIL PROTECTED] (Craig Barratt) wrote in message news:at1i0e$1tj3$[EMAIL PROTECTED]... I've been studying the read and write buffering in rsync and it turns out most I/O is done just a couple of bytes at a time. This means there are lots of system calls, and also most network traffic comprises lots of small packets. The behavior is most extreme when sending/receiving file deltas of identical files. The main case where I/O is buffered is writes from the server (when io multiplexing is on). These are usually buffered in 4092 byte chunks with a 4 byte header. However, reading of these packets is usually unbuffered, and writes from the client are generally not buffered. For example: when receiving 1st phase checksums (6 bytes per block), 2 reads are done: one of 4 bytes and one of 2 bytes, meaning there are 4 system calls (select/read/select/read) per 6 bytes of checksum data). One cost of this is some performance, but a significant issue is that unbuffered writes generate very short (and very many) ethernet packets, which means the overhead is quite large on slow network connections. ... see this link for the rest: http://groups.google.com/groups?hl=enlr=ie=UTF-8oe=UTF-8newwindow=1selm=at1i0e%241tj3%241%40FreeBSD.csie.NCTU.edu.tw -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: completion-ignore-case (was: /tmp)
On Thu, 12 Dec 2002 11:07:37 -0800 Shankar Unni [EMAIL PROTECTED] wrote: Soren A wrote: [...] you can nenver tell another person what they should like in a shell. Well (sorry for OT drift), the main thing was the case-insensitive filename completion, which is invaluable on Windows. I need to look at the tcsh source and see if it is implemented within an #ifdef WIN32 like block. It would be nice if the feature was automatically also available in the cygwin build of tcsh .. I use case insensitive completion all the time in Cygwin bash. The following lines in ~/.inputrc control that feature for bash and may affect other shells that use the same input library. # Ignore case while completing set completion-ignore-case on -- Mac :}) ** I normally forward private questions to the appropriate mail list. ** Ask Smarter: http://www.tuxedo.org/~esr/faqs/smart-questions.html Give a hobbit a fish and he eats fish for a day. Give a hobbit a ring and he eats fish for an age. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: completion-ignore-case (was: /tmp)
Michael A Chase wrote: # Ignore case while completing set completion-ignore-case on Hmm. Thanks! Now that I RTFM more closely, I also see set complete=enhance. Is there a difference? The latter seems to work as I'd expect it to, so I'm now beginning my Big Move To Cygwin Tcsh :-/. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: completion-ignore-case (was: /tmp)
On Thu, 12 Dec 2002 12:12:56 -0800 Shankar Unni [EMAIL PROTECTED] wrote: Michael A Chase wrote: # Ignore case while completing set completion-ignore-case on Hmm. Thanks! Now that I RTFM more closely, I also see set complete=enhance. Is there a difference? The latter seems to work as I'd expect it to, so I'm now beginning my Big Move To Cygwin Tcsh :-/. I looked in the fine manual for tcsh after I sent my email and found the same thing. I don't know if ~/.inputrc applies to tcsh at all. If you find any other differences between Cygwin and Linux tcsh, they may be of interest since the intention is to make things as much alike as practical. -- Mac :}) ** I normally forward private questions to the appropriate mail list. ** Ask Smarter: http://www.tuxedo.org/~esr/faqs/smart-questions.html Give a hobbit a fish and he eats fish for a day. Give a hobbit a ring and he eats fish for an age. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: problems with CPAN and cygwin 1.3.12
Hallo JoNO, Please keep replies to the list, others may help debugging your problem too! --- Gerrit P. Haase [EMAIL PROTECTED] wrote: Hallo JoNO, At first, make sure that in your Cygwin environment are no paths in PATH with spaces like this: /cygdrive/c/Program Files/SecureCRT 3.0 Then check the permissions on your executables: $ ls -l /bin/perl should look like this: -rwxr-xr-x1 gerrit Domänen-14848 Dec 9 17:56 /bin/perl* ^ ^^ ^^ ^ including at least the 'r-x' for all groups. Also try to update the MakeMaker module to the latest version, like Soren suggested. Nope, it doesn't work. - my enviroment is ok - the perl's permissions are ok - I have updated ExtUtils::MakeMaker to the latest version. Note that i'm using: perl-5.8.0 + cygwin-1.3.17-1 + latest version of CPAN (1.63) if I replace the cygwin1.dll with the older version (1.3.12) ALL is OK! so there must be a problem with this version of cygwin.(?!) Please check again, or try to reproduce my enviroment :) Thanks. again Ok. I'll test the next days. Is the perl-5.8.0 you use from the original Cygwin netrelease mirrors or did you compile it yourself? Gerrit -- =^..^=
Re: Shell script wont run with cron, but it works ok in the command line.
How can I make the mv, and rm commands excecutable by system? where I can add the path to this commands for System??? This is the hint I have been looking for since Dec 07 thanks in advance! Eric. Re: Script Doesn't Work Correctly with Cron From: Igor Pechtchanski pechtcha at cs dot nyu dot edu To: Buck Turgidson jc_va at hotmail dot com Cc: cygwin at cygwin dot com Date: Thu, 12 Dec 2002 13:26:48 -0500 (EST) Subject: Re: Script Doesn't Work Correctly with Cron Reply-to: cygwin at cygwin dot com On Thu, 12 Dec 2002, Buck Turgidson wrote: I have a script that works when I run it from the command line, but when it is executed with cron, it produces a null file. I am sure that cron is executing it because it creates the file. It is just not populated. Here is the script. I want to query an Oracle database every couple of minutes, capture some data, and append it to a flat file. I want to keep 2000 lines in the flat file. Can someone spot the error? Again, it works run from the command line. #!/bin/bash sqlplus -s scott@mydb up @ss.sql | grep -v Enter password: monsql.lst tail -2000 monsql.lst montemp.lst rm -f monsql.lst mv montemp.lst monsql.lst This is probably not cron-specific, but try redirecting stderr to a file (by using 2filename) to see if the script produces any errors. Also, are all the programs you use (sqlplus, grep, tail, rm, mv) in the path for the user that cron runs as (most likely system)? Are they executable by system? __ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Shell script wont run with cron, but it works ok in the command line.
Try adding c:\cygwin\bin to your *system* (not user) PATH environment variable (in the Control Panel System applet select the Advanced tab, then Environment Variables, and change the System PATH variable to contain the above). If you have any other POSIX-compliant toolset installed and in your path, make sure the cygwin bin directory precedes that. As for making the files executable by SYSTEM, chmod a+x /bin/* should do it. As a side note, it's always a good idea to use absolute paths in scripts (i.e. /bin/rm) if you are uncertain what the value of PATH may be when the script executes. Igor On Thu, 12 Dec 2002, Eric De La Cruz Lugo wrote: How can I make the mv, and rm commands excecutable by system? where I can add the path to this commands for System??? This is the hint I have been looking for since Dec 07 thanks in advance! Eric. Re: Script Doesn't Work Correctly with Cron From: Igor Pechtchanski pechtcha at cs dot nyu dot edu To: Buck Turgidson jc_va at hotmail dot com Cc: cygwin at cygwin dot com Date: Thu, 12 Dec 2002 13:26:48 -0500 (EST) Subject: Re: Script Doesn't Work Correctly with Cron Reply-to: cygwin at cygwin dot com On Thu, 12 Dec 2002, Buck Turgidson wrote: I have a script that works when I run it from the command line, but when it is executed with cron, it produces a null file. I am sure that cron is executing it because it creates the file. It is just not populated. Here is the script. I want to query an Oracle database every couple of minutes, capture some data, and append it to a flat file. I want to keep 2000 lines in the flat file. Can someone spot the error? Again, it works run from the command line. #!/bin/bash sqlplus -s scott@mydb up @ss.sql | grep -v Enter password: monsql.lst tail -2000 monsql.lst montemp.lst rm -f monsql.lst mv montemp.lst monsql.lst This is probably not cron-specific, but try redirecting stderr to a file (by using 2filename) to see if the script produces any errors. Also, are all the programs you use (sqlplus, grep, tail, rm, mv) in the path for the user that cron runs as (most likely system)? Are they executable by system? -- http://cs.nyu.edu/~pechtcha/ |\ _,,,---,,_[EMAIL PROTECTED] ZZZzz /,`.-'`'-. ;-;;,_[EMAIL PROTECTED] |,4- ) )-,_. ,\ ( `'-' Igor Pechtchanski '---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow! Oh, boy, virtual memory! Now I'm gonna make myself a really *big* RAMdisk! -- /usr/games/fortune -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Installation
Actually, I'm not sure if there is a proper list for this. This list can't really answer questions about someone else's distribution, assuming that's the source of the problem. This list can help you determine if the CD is the source of the problem (or if it's just user error). However, we would need to know exactly the steps you take and selections you make prior to the problem. For example, one way that this problem could arise is if you picked a local directory which did not contain the any packages. If there is nothing obviously wrong with your process, then the problem would have to be with the CD, it's contents, and/or it's structure. But to resolve that issue, you'd need to contact the provider of the CD. Larry Original Message: - From: Igor Pechtchanski [EMAIL PROTECTED] Date: Thu, 12 Dec 2002 16:41:25 -0500 (EST) To: [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED] Subject: Re: Installation Wrong list, redirected. Igor On Thu, 12 Dec 2002, d. dorst wrote: I have got cygwin on a cd-rom because I use a 56K modem. I have read the FAQ s that are on the cd and tried every solution that is there but it will not install. Every time when the packeges should come in the sceen it says that there is nothing availeble. Should I first unzip every folder that is on the cd or is there an other way of doing this. I know that I am a dumb Windows user but I can not get cygwin working this way. I would be gratefull for any help. Dirk Dorst -- http://cs.nyu.edu/~pechtcha/ |\ _,,,---,,_[EMAIL PROTECTED] ZZZzz /,`.-'`'-. ;-;;,_[EMAIL PROTECTED] |,4- ) )-,_. ,\ ( `'-' Igor Pechtchanski '---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow! Oh, boy, virtual memory! Now I'm gonna make myself a really *big* RAMdisk! -- /usr/games/fortune -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ mail2web - Check your email from the web at http://mail2web.com/ . -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: MS_types and addr_t name polution
Elfyn McBratney wrote: Hi, The addr_t type is used in memory managment (memory addresses??) in /usr/include/sys/mman.h. If you plan on using this (or is already on the code, mman.h i mean) you may have problems. What is the type of your addr_t? is it type char of * (pointer)? Actually, mman.h uses caddr_t, not addr_t, at least in the cygwin 1.3.15 system I have installed. Nothing in /usr/include appears to use addr_t. My typedef of addr_t is for 64-bit int. If the code is open source you can just change the addr_t typedef name (in your source) to something else eg. addr2_t in all of your source files (a simple search and replace). Yes, I renamed all references to addr_t to a different name. It's just a nuisance and appears to be unnecessary name polution. Elfyn [EMAIL PROTECTED] --- Michael Eager [EMAIL PROTECTED] wrote: I'm porting a small program to Cygwin. One of the typedefs in the program is for a symbol named addr_t. Since this is defined in /usr/include/sys/types.h, there is a redefinition conflict. This (and a number of other symbols) are defined when __MS_types_ is defined, which, in turn, is defined whenever __CYGWIN__, _WIN32 or __MSDOS__ is defined. I don't see any use of addr_t in /usr/include. Is there a reason for defining addr_t and the other symbols? -- Michael Eager [EMAIL PROTECTED] 408-328-8426 MontaVista Software, Inc. 1237 E. Arques Ave., Sunnyvale, CA 94085 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ _ www.smokeJet.com - Free UK Internet Services _ Select your own custom email address for FREE! Get [EMAIL PROTECTED] w/No Ads, 6MB, POP more! http://www.everyone.net/selectmail?campaign=tag -- Michael Eager [EMAIL PROTECTED] 408-328-8426 MontaVista Software, Inc. 1237 E. Arques Ave., Sunnyvale, CA 94085 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Emacs: completion of buffer names ignores case
I'm using emacs-21.2-10. I frequently want to switch to the buffer named `*shell*', so I type C-x b * s h e SPC RET expecting that, since `*shell*' is the only buffer whose name begins with `*she', the SPC will cause its complete name (`*shell*') to appear in the minibuffer, and the RET will cause Emacs to switch to it. Instead, when there's a buffer named `*Shell Command Output*', the name doesn't complete fully: after I hit RET, I'm looking at a brand-new buffer named `*shell'. This isn't what I wanted :( It's as if Emacs has decided to ignore case when completing what I've typed. Indeed, if I type TAB instead of RET, I see Click mouse-2 on a completion to select it. In this buffer, type RET to select the completion near point. Possible completions are: *Shell Command Output* *shell* Now, I can understand that Emacs would ignore case when completing file names on Windows, since Windows itself doesn't distinguish file names by case. But since Emacs does distinguish buffer names by case, I expect it, when completing buffer names, to ignore a buffer whose name differs in case from what I've typed. Is this a bug? -- PGP Fingerprint: 3E7B A3F3 96CA 8958 ACC5 C8BD 6337 0041 C01C 5276 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
[ANN] Re: cyg-wrapper.sh
Hello Matt, * On Tue, Dec 10, 2002 at 09:38:02AM -0700, Matt Armstrong [EMAIL PROTECTED] wrote: Luc, you might consider adding an option to unset the PWD environment variable (or convert it to a Windows path). That's done. I haven't done any test with perforce, so, let me know. The two new options are: --cyg-PWD-clear : to unset PWD --cyg-PWD-convert : to convert PWD to its DOS (short) form. I was considering to handle other environment variables, but I didn't see any good candidate -- ie. environment variables specific to bash or cygwin that win32 native applications may also use. -- Luc Hermitte -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
LaPack++ help
Hello! Hope you are doing well. My name is Huma Babak, I have been researching for school about the LaPack++ package and I got your name as someone who has used LaPack++ package. Can you please tell me your experience with this package? Positives and Negative, more focused on the negatives. I am asked to wrap the package in .NET. Can you just pass any size matrix to the library and expect it to perform the general operations on the matrix? Thank you, Huma Babak -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Installation
Wrong list, redirected. Igor On Thu, 12 Dec 2002, d. dorst wrote: I have got cygwin on a cd-rom because I use a 56K modem. I have read the FAQ s that are on the cd and tried every solution that is there but it will not install. Every time when the packeges should come in the sceen it says that there is nothing availeble. Should I first unzip every folder that is on the cd or is there an other way of doing this. I know that I am a dumb Windows user but I can not get cygwin working this way. I would be gratefull for any help. Dirk Dorst -- http://cs.nyu.edu/~pechtcha/ |\ _,,,---,,_[EMAIL PROTECTED] ZZZzz /,`.-'`'-. ;-;;,_[EMAIL PROTECTED] |,4- ) )-,_. ,\ ( `'-' Igor Pechtchanski '---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow! Oh, boy, virtual memory! Now I'm gonna make myself a really *big* RAMdisk! -- /usr/games/fortune -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Broken Link on the ported page.
Hello, The link to the rpm package on http://sources.redhat.com/cygwin/ported.html is broken. Does anyone have that file or know if it's posted on another site somewhere? Thanks in advance. Adam -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Shell script wont run with cron, but it works ok in the command line.
Igor Pechtchanski wrote: Try adding c:\cygwin\bin to your *system* (not user) PATH environment variable (in the Control Panel System applet select the Advanced tab, then Environment Variables, and change the System PATH variable to contain the above). If you have any other POSIX-compliant toolset installed and in your path, make sure the cygwin bin directory precedes that. And don't forget to reboot to make the changes effective! -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Individual vs. list-only replies (Was Re: Force bash to startas administrator)
Igor Pechtchanski wrote: Andrew, I would like to point out that most people on this list are busy people, As am I. and that the most convenient way to reply is the one that takes the least effort on their part. In my software I simply hit reply instead of reply all. Either button is as easy. The message that started this discussion used the default behavior of pine (reply to author, cc to list - I hit reply, write the message, and press send), whereas for this message I actually had to explicitly remove your name from the To: line and move the cygwin list from Cc: to To:. I'm only doing this to prove the point now, and will most probably not do so in the future. Yes and in my software the default way is to either hit the reply button or type Control-R. In order to reply all you need to hit that button or Control-Shift-R (i.e. a little more difficult). But the issue is not only are people who reply busy but so are the readers. With your software's defaults I get a message in my email AND see it and have to skip it later when I read the list. Thus it is more work for me, hence the request not to do this. As for the path of least inconvenience, if your method was the most popular then I would be receiving email whenever somebody response to one of my posts but I don't. Therefore it sounds to me like this is a pine'ism. Otherwise I will simply set up a filter to ignore all email from you :-) ! This is certainly your choice. Ploink! Again, I don't appreciate getting email that's a copy of what I'll be reading on the list shortly anyway. As I mentioned before, the standard way to deal with this (on any mailing list) is to use a Reply-To header. Since your method for reading and posting to this list is non-standard (namely, the gmane newsgroup) and incompatible with Reply-To, It didn't used to be incompatible however the gname maintainer changed that behavior. you would need to seek alternate solutions. Yeah, ploink I guess... :-) P.S. Also note that, even with your current setup, you can get a double copy only if someone replies to a message that you posted. Again, when others respond I do not get an additional email copy of it. Just with some people. Apparently people who use pine and perhaps some other mailers. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Individual vs. list-only replies (Was Re: Force bash to startas administrator)
On Fri, 2002-12-13 at 13:50, Andrew DeFaria wrote: But the issue is not only are people who reply busy but so are the readers. With your software's defaults I get a message in my email AND see it and have to skip it later when I read the list. Thus it is more work for me, hence the request not to do this. Heh, fix your mail reader. Seriously. It's neither the posters, nor the list serve's responsibility to reduce duplicates. And given that there are valid reasons for duplicates (say an issue that affects both kde-cygwin and general-cygwin discussions), it's up to you to solve the problem at your end. As for the path of least inconvenience, if your method was the most popular then I would be receiving email whenever somebody response to one of my posts but I don't. Therefore it sounds to me like this is a pine'ism. Heh. I use Reply-All. You'll probably see two responses, unless you fix your mail reader first :}. Apparently people who use pine and perhaps some other mailers. I use evolution, and occasionally MS Outlook / MS Outlook Express. You'll get the same behaviour from me every time. Rob -- --- GPG key available at: http://users.bigpond.net.au/robertc/keys.txt. --- signature.asc Description: This is a digitally signed message part
Re: pipe improvements in snapshot
Christopher Faylor wrote: On Thu, Dec 12, 2002 at 07:32:16AM -0500, Norman Vine wrote: With a Cygwin CVS tree I built this AM (2) below rxvt has problems when configuring a project from a build directory ie $TOP_DIR src build cd $TOP_DIR/build ../configure The problems vary from run to run ranging from 'error parsing uname' to program hang recoverable with 'ctrl-C' making me suspect a possible data loss Using the gcc option -pipe seems to allow the configure process Using the -pipe flag I have not seen the uname error but it does eventually hang when actually creating the Makefiles This project uses libtool With the CVS tree build (1) below I do not experience this (1) 10-Dec-2002 7:10:38a6,683,847 cygwin-20021210.tar.gz (2) 12-Dec-2002 6:09:42a6,688,318 cygwin-20021212.tar.gz The times are EST and the tarballs are created by a script that does a CVS update, make, make tarball which takes ~20 minutes so the CVS update time should be approximately 20 minutes earlier Any 'tips' as to how to best debug this appreciated - Attach to the hung process with gdb and see where it is hung. - Provide cygcheck output. - Run under strace and see if you can infer where hangs or problems are occurring. Attached find cygcheck output and the config.status output upto where 'sed' gets unhappy and enters a repeating loop This happens when running a bash shell in an rxvt terminal config.status runs to completion from a bash shell in the Win2k cmd.exe terminal I have marked where the output differs from the successful run in make.log This is repeatable Norman cygcheck.log Description: Binary data make.log Description: Binary data -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Broken Link on the ported page.
On Thu, Dec 12, 2002 at 06:25:40PM -0800, Adam Jurevicius wrote: The link to the rpm package on http://sources.redhat.com/cygwin/ported.html is broken. Does anyone have that file or know if it's posted on another site somewhere? It's not broken right now, at least. cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Individual vs. list-only replies (Was Re: Force bash to start as administrator)
On Thu, Dec 12, 2002 at 06:50:13PM -0800, Andrew DeFaria wrote: P.S. Also note that, even with your current setup, you can get a double copy only if someone replies to a message that you posted. Again, when others respond I do not get an additional email copy of it. Just with some people. Apparently people who use pine and perhaps some other mailers. I have to note that your receipt of additional email is really not of any consequence to the cygwin mailing list. I can see one admonishment but a long-running scolding thread is pretty boring. We've stopped talking about this now, right? cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Some inquiries about Cygwin?
Dear Sir: I am interested in Cygwin. May I have the following inquiries? 1. Can I install and use RedHat's RPM in Cygwin in order to install others packages like uClinux? 2. Can I install uClinux in Cygwin to cross-compile it? 3. Can I install cross-compiler like GNU toolchain for ARM7 (like Lineo's product) in Cygwin? Your response is very much appreciated. Regards, MJLIU -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Changing to another user in cygwin -- login?
Hi, I have postgres installed and running as a service under user postgres. Now I want to run the createdb, etc scritps as user postgres to create a new database. I have tried 'login postgres' from a bash shell, but this is the result: $ login postgres Password: Login incorrect login: I know the password is correct, as I logged into windows with user postgres to init the postgres system. I also followed all the instructions in the postgres readme. Do I need to make any special changes to the NT user account or something? (I do not know the Windows environment too well - use it by force, not choice) :-) -- Andre Truter Software Engineer Registered Linux user #185282 ICQ #40935899 AIM: trusoftzaf http://www.trusoft.za.net - The box said: Requires Windows 95 or better... So I installed Linux - -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/