RE: Can't obtain write permissions on my file...
> So it seems that is linked to emacs; can you > read/write your file under > another editor (like vi)? just to confirm it's > really an Emacs problem. I haven't got vi, but I tried a simple notepad under Windows, and I modified my file without problem. > Yes, this is an environment variable that can be set > either in the > workstation properties dialog, or in th ecygwin.bat > script (but not after: > it must be set BEFORE the first cygwin program > starts). Okay I've set that environment variable. > Not sure it is CVS, but you may create a file > manually on Unix, then try to > edit it under Emacs from Windows; If I try to create a file on Unix, then there's no problem I can read & write it on Windows. Uh... maybe there's a problem with Emacs detecting my "CVS" file. I mean: the file is checked-out under Unix for user "axelle" (Unix ID). Under Windows, I'm another user - say "axl". It's still me of course, but how can he know (and how to tell him)... > if it works, then > maybe CVS has set-up the > file properties on UNIX strangely. BTW IIRC Solaris > handles not only > traditional Unix attributes (the usual -rw--r--r > stuff) but also ACLs > (access control lists). Yes I know, I've set ACLs on my CVS repository. But the files I access do not have ACLs. I wouldn't think this is a problem. > If these are enabled on your > Solaris box, you should > definitely check the rights set by CVS for the > problematic file and compare > with one you can edit from Emacs on Windows. I can't see any difference. Actually, I tried the following trick : if I commit my file (under Unix) and do not touch it under Unix, then I can read/write it on Emacs on Windows. Whenever I touch the file on Unix, it gets not writable on Windows. But there's absolutely no change in my file permissions... Really strange, no ?! > PS: Please keep these kind of discussion on the > list; the various mails may > trigger ideas from others on the list that may > finally discover the (perhaps > totally unrelated) cause to your problem, or to a > problem of someone else > :-) Yes, you're perfectly right. Actually, I forgot to follow up my last answer to the group... Axelle. ___ Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français ! Yahoo! Mail : http://fr.mail.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: Can't obtain write permissions on my file...
Hi, Actually initially the problem was I did not know who's causing the problem: Cygwin, Emacs or CVS... Now, well, it looks more like an Emacs problem. I'll probably follow up to some group regarding Emacs. Bye. Axelle. --- "Robinow, David" <[EMAIL PROTECTED]> a écrit : > > > If I log on the Unix machine, check out my file, > then > > > go to my NT machine and try to modify it using > Emacs, > > > Emacs complains it is unwritable (though Cygwin > shows > > > I have rw- access, and if I right click on the > file on > > > Win2000, the read-only check box is not > checked). > > > As a matter of fact, I can write on the file: > tried > > > the touch command or an echo >> my file. > Would somebody explain what this problem has to do > with cygwin? I must be > missing something. ___ Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français ! Yahoo! Mail : http://fr.mail.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/
Problem compiling - unix compatible but using winsock2
Hello everybody!
To clarify my problem reported last week:
My program lists
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
which works with Linux and AIX.
If I try this with cygwin it compiles, but upon call it says
setsockopt(IP_MULTICAST_LOOP)
errno=14 (Bad address)
The offending part is
char c;
c=0;
if (setsockopt(sock, IPPROTO_IP,
IP_MULTICAST_LOOP, (void*)&c, sizeof(c)))
ERROR("setsockopt(IP_MULTICAST_LOOP)");
After changing the "char" to an "int" I get
setsockopt(IP_MULTICAST_LOOP)
errno=109 (This option is unsupported)
Which I think after reading in some .h files
is because it uses winsock 1.1.
As I found ws2tcpip.h I try to include that via
#ifdef __WIN32__
#include
#endif
either before or after the above #include lines.
But I get a lot of "conflicting types for xxx"
in netdb.h, sys/socket.h, asm/byteorder.h, and even
cygwin/in.h:25: parse error before '0'
Any help how to use the definitions from ws2tcpip.h?
Thank you!
Regards
Phil
-
This message is RSA-encrypted: n=33389, e=257
--
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: Can't obtain write permissions on my file...
Just wanted to let those who have followed this thread that I found the solution. You were right, this is no Cygwin problem, but an Emacs bug. It doesn't recognize correctly the "w" right for RCS/CVS files. The fix is to put (setq vc-ignore-vc-files t) in the .emacs file. Thanks for your help locating the problem. I really thought this had something to do with my Cygwin config. Cheers, Axelle. ___ Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français ! Yahoo! Mail : http://fr.mail.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/
Order of -l on the command line, compared to linux..
Hello! What is the diff between linker on linux and cygwin linker?? Basically, the problem is that gcc -lX11 foo.o works on linux, but not cygwin (missing symbols), on cygwin I have to do it the order: gcc foo.o -lX11 Is it a new feature in the gcc 2.96-version on my linux-machine, or something else?? Just being curious, myself I preferr to throw the -l last, but I have seen more and more people having a problem wanting to throw -lX11 at the beginning, and wonder where they get the order from.. /Andy p.s. Yeah I know, X11 doesn't really belong here, but I assumed that I have the same problem with any lib I try to link with.. -- The eye of the beholder rests on the beauty! -- 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/
foobaz-unsubscribe-bp=ispelunker.comu@cygwin.com
B.P. -- 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[2]: Core dumped just only with strcat!
Hello Dani, Sorry, but I don't know nothing about mysql programming and I can't help. Still I think you have some kind of programming error, not a bug in mysql or cygwin themselfs, though I don't have any time to look at your code - you should find out yourself. Still, for the record I am forwarding your mail to the cygwin mailing list where possibly someone else would be kind to help you or give you an advise, but I don't really think that this mail belongs there too :( Monday, February 18, 2002, 9:45:51 PM, you wrote: DP> I've trying lots of things and after 1 day with it, I've achieved to work DP> part of the code. What I did? I just declared my array before a MYSQL_RES DP> variable. I had: DP> MYSQL_RES * result; DP> char petition[500]; DP> And I changed to: DP> char petition[500]; DP> MYSQL_RES * result; DP> And also I did all the string work before the mysql stuff (connection call, DP> query call,...). And, don't ask me why, it worked! (now it crashes in other DP> function, in a mysql_query call). DP> I have not a reasonable answer for this behaviour... I'm quite lost. I've DP> asked this same question to the mysql-win32 list, and it seems not to be a DP> mysql problem (according to this list). DP> I have attached the code (it's in mysql_procs.c, but I have attached all the DP> files involved-- In order to compile and run, you need Gtk-win32 library), DP> but you must pardon me because it has been written in Spanish and so, DP> variables and comments may bore you. The problem I related you was among DP> the first lines of the code, but the last problem I told you was among the DP> last lines. DP> Finally, could I ask you for forward this message to the cygwin list? After DP> the answers which claim that the error could be caused by mysql, I DP> unsuscribed from cywgin list, and by now, I'm not suscribed (yet). DP> Thanks, DP> Dani P. DP> - Original Message - DP> From: Pavel Tsekov <[EMAIL PROTECTED]> DP> To: Dani P. <[EMAIL PROTECTED]> DP> Cc: <[EMAIL PROTECTED]> DP> Sent: Monday, February 18, 2002 9:57 AM DP> Subject: Re: Core dumped just only with strcat! >> Just tried it - it doesn't dump core here. Is this snippet taken >> from a larger source ? If so maybe it would be usefull to paste it >> all. Maybe the core happens somewhere else... >> >> Still you can try to debug this by yourself - compile with -g >> option and launch gdb. Then come back to the list if you suspect >> a bug in cygwin and not your code :) >> >> Dani P. wrote: >> >> > Sorry, in my last posting there's an error in the piece of code: >> > >> > >> >>char request[500]; >> >>request[0]=3D'\0'; <-- "3D" should not be here >> >>strcat(request,"SELECT * FROM data"); >> >> >> > >> > I meant that this piece of code dumps a core, not the last posting. >> > >> > Thanks, >> > >> > Dani >> > >> > >> > - Original Message - >> > From: Dani P. <[EMAIL PROTECTED]> >> > To: <[EMAIL PROTECTED]> >> > Sent: Saturday, February 16, 2002 10:57 AM >> > Subject: Core dumped just only with strcat! >> > >> > >> > >> >>I have installed recently cygwin in order to compile a piece of code >> >>which compiled an d worked fine in Linux, using Gtk and MySQL libraries. >> >>It compiles fine, but it dumps a core just only with a "strcat". I do DP> the >> >>following: >> >> >> >>char request[500]; >> >>request[0]=3D'\0'; >> >>strcat(request,"SELECT * FROM data"); >> >> >> >>And in that last instruction, it dumps a core. -- 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/
Fwd: Re: Core dumped just only with strcat!
This is a forwarded message From: Dani P. <[EMAIL PROTECTED]> To: "Pavel Tsekov" <[EMAIL PROTECTED]> Date: Monday, February 18, 2002, 9:45:51 PM Subject: Core dumped just only with strcat! ===8<==Original message text=== I've trying lots of things and after 1 day with it, I've achieved to work part of the code. What I did? I just declared my array before a MYSQL_RES variable. I had: MYSQL_RES * result; char petition[500]; And I changed to: char petition[500]; MYSQL_RES * result; And also I did all the string work before the mysql stuff (connection call, query call,...). And, don't ask me why, it worked! (now it crashes in other function, in a mysql_query call). I have not a reasonable answer for this behaviour... I'm quite lost. I've asked this same question to the mysql-win32 list, and it seems not to be a mysql problem (according to this list). I have attached the code (it's in mysql_procs.c, but I have attached all the files involved-- In order to compile and run, you need Gtk-win32 library), but you must pardon me because it has been written in Spanish and so, variables and comments may bore you. The problem I related you was among the first lines of the code, but the last problem I told you was among the last lines. Finally, could I ask you for forward this message to the cygwin list? After the answers which claim that the error could be caused by mysql, I unsuscribed from cywgin list, and by now, I'm not suscribed (yet). Thanks, Dani P. - Original Message - From: Pavel Tsekov <[EMAIL PROTECTED]> To: Dani P. <[EMAIL PROTECTED]> Cc: <[EMAIL PROTECTED]> Sent: Monday, February 18, 2002 9:57 AM Subject: Re: Core dumped just only with strcat! > Just tried it - it doesn't dump core here. Is this snippet taken > from a larger source ? If so maybe it would be usefull to paste it > all. Maybe the core happens somewhere else... > > Still you can try to debug this by yourself - compile with -g > option and launch gdb. Then come back to the list if you suspect > a bug in cygwin and not your code :) > > Dani P. wrote: > > > Sorry, in my last posting there's an error in the piece of code: > > > > > >>char request[500]; > >>request[0]=3D'\0'; <-- "3D" should not be here > >>strcat(request,"SELECT * FROM data"); > >> > > > > I meant that this piece of code dumps a core, not the last posting. > > > > Thanks, > > > > Dani > > > > > > - Original Message - > > From: Dani P. <[EMAIL PROTECTED]> > > To: <[EMAIL PROTECTED]> > > Sent: Saturday, February 16, 2002 10:57 AM > > Subject: Core dumped just only with strcat! > > > > > > > >>I have installed recently cygwin in order to compile a piece of code > >>which compiled an d worked fine in Linux, using Gtk and MySQL libraries. > >>It compiles fine, but it dumps a core just only with a "strcat". I do the > >>following: > >> > >>char request[500]; > >>request[0]=3D'\0'; > >>strcat(request,"SELECT * FROM data"); > >> > >>And in that last instruction, it dumps a core. > >> > >>Could you help me? > >> > >>Thanks, > >> > >>Dani P. > >> > >> > >>-- > >>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/ > > > > > > > > > > > ===8<===End of original message text=== gui.h Description: Binary data gui.c Description: Binary data mysql_procs.c Description: Binary data mysql_procs.h 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: print command setting in config.ps for cygwin
Gene, On Mon, Feb 18, 2002 at 04:22:11PM -0800, Gene C. Ruzicka wrote: > But the online User's Guide > for Cygwin says that the lp, lpr commands don't > work in Cygwin, and that the MS print command > should be used to print from a cygwin command > line. See attached for my latest PostScript version of lpr. See the following for PCL versions: http://sources.redhat.com/ml/cygwin/2001-04/msg00657.html HTH, Jason #!/bin/bash # $Id: lpr,v 1.4 2002/02/07 15:38:52 jatis Exp $ Server=i3125lp3 Printer=p44sb007 PostscriptOptions='-o l' ProgramName=$(basename $0) WinDir=$(cygpath -u $WINDIR) Lpr=$WinDir/system32/lpr TempFile=/tmp/$ProgramName.$$ WinTempFile=$(cygpath -w $TempFile) cat >$TempFile if [ "$(head -c 2 $TempFile)" = %! ] then LprOptions=$PostscriptOptions fi $Lpr -S $Server -P $Printer $LprOptions "$WinTempFile" Status=$? rm -fr $TempFile exit $Status -- 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: proftpd on cygwin?
I have had success with many ftp daemons on cygwin. ProFTPd was one of them... some minor tweaks needed, but nothing much to it... perhaps these were only due to my peculiar configuration. Stephano Mariani > -Original Message- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of dave > Sent: Tuesday, 19 February 2002 3 51 > To: [EMAIL PROTECTED] > Subject: proftpd on cygwin? > > Hello, > Has anyone ever tried to compile proftpd on cygwin? If so, what all is > involved? > Thanks. > Dave. > > > > -- > 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: Your Cygwin question / cygreadline
Wayne, Please post to [EMAIL PROTECTED] instead of sending private email. On Mon, Feb 18, 2002 at 10:41:03PM -0500, [EMAIL PROTECTED] wrote: > Did you ever arrive at an answer as to how to address the missing > cygreadline.dll? Why are you asking me? > It cropped up when I was working/playing with Octave. > > I tried searching the archives, got 160 some odd hits. No obvious > titles, so I was left with searching that many files. Hmm... > Heres the funny, when I have tried to email the mailing list, my email is > rejected, both from here, and my airbase email account. So? > Anyway, thanks, even if you haven't gotten it resolved. See the following: http://cygwin.com/ml/cygwin-announce/2002/msg00011.html Jason -- 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: Fwd: Re: Core dumped just only with strcat!
On Tue, Feb 19, 2002 at 12:45:35PM +0100, Pavel Tsekov wrote: > I've trying lots of things and after 1 day with it, I've achieved to work > part of the code. What I did? I just declared my array before a MYSQL_RES > variable. I had: > > MYSQL_RES * result; > char petition[500]; This is basic C knowledge. Here petition is a local variable, so it's not filled with '\0'. You either want to place a '\0' in petition[0], or use strcpy() instead of strcat(). strcat() will add at the end of the string, and that might be anywhere in memory since you didn't initialize the string and just contains garbage. Kurt -- 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: Fwd: Re: Core dumped just only with strcat!
On Tue, Feb 19, 2002 at 02:14:38PM +0100, Kurt Roeckx wrote: > On Tue, Feb 19, 2002 at 12:45:35PM +0100, Pavel Tsekov wrote: > > I've trying lots of things and after 1 day with it, I've achieved to work > > part of the code. What I did? I just declared my array before a MYSQL_RES > > variable. I had: > > > > MYSQL_RES * result; > > char petition[500]; > > This is basic C knowledge. Here petition is a local variable, so > it's not filled with '\0'. You either want to place a '\0' in > petition[0], or use strcpy() instead of strcat(). Oops, you did the first one. I should have looked at the code closer. Kurt -- 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: Fwd: Re: Core dumped just only with strcat!
Basically this problem smells like you have managed to mess up the memory/stack before this, and well, memory problems is often 'detected' by the program crashes on some other location.. So start looking around at the code, what is done before these calls. /Andy (with detected I mean that the user notices that the program crashes or similar) / Pavel Tsekov <[EMAIL PROTECTED]> wrote: | This is a forwarded message | From: Dani P. <[EMAIL PROTECTED]> | To: "Pavel Tsekov" <[EMAIL PROTECTED]> | X-Sent: 32 years, 7 weeks, 13 hours, 35 minutes, 10 seconds ago | Subject: Core dumped just only with strcat! | | ===8<==Original message text=== | I've trying lots of things and after 1 day with it, I've achieved to work | part of the code. What I did? I just declared my array before a MYSQL_RES | variable. I had: | | MYSQL_RES * result; | char petition[500]; | | And I changed to: | | char petition[500]; | MYSQL_RES * result; | | And also I did all the string work before the mysql stuff (connection call, | query call,...). And, don't ask me why, it worked! (now it crashes in other | function, in a mysql_query call). -- The eye of the beholder rests on the beauty! -- 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: od
Hi Mark, Thank you for your reply to my message about the od utility: I wrote: >> I created at test file with 4 characters in it: >> >> HTTP >> >> Then, I run od -bcx and I get: >> >> 110 124 124 120 012 000 >> H T T P \n \0 >> 5448 5054 000a You wrote: > That is correct in this little-endian platform, see > http://info.astrian.net/jargon/terms/l/little-endian.html > > >You can see that the hex values in the last line are reversed. > >I think they should be 4854 5450 000a. > > Only if you were on a big-endian platform. I can understand what you are saying if I interpret it to mean that the hex storage values on my machine WindowsNT - which I assume from you message is a big-endian platform - are faithfully represented by the output from the od -bcx display. What then surprises me is that the octal representation of this same storage is 110 124 124 120 which is what I would expect. It seems to me that you are suggesting that the only correct representation of the hex storage values would require od to ouput an ascii value of THPT. I reject, and od itself contradicts, this interpretation of the storage values as being meaningful. I urge you to reconsider your opinion and to modify od to output 4854 5450. David -- 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: Proposed patch for bash (completion spec problem related)
> Possible patch for bash to fix the problem noticed by Jason Tishler in msg: > http://sources.redhat.com/ml/cygwin/2002-02/msg00581.html > > > --- subst.c.old Mon Nov 5 07:19:50 2001 > +++ subst.c Mon Feb 18 16:58:50 2002 > @@ -1339,7 +1339,7 @@ > break; > >i = te /* + member (string[te], d) */; > - while (member (string[i], d) && whitespace(string[i])) > + while (member (string[i], d)/* && whitespace(string[i])*/) > i++; The right fix is to replace the call to whitespace() with one to spctabnl(). There's another fix needed for skip_to_delim(), but my current sources look sufficiently different from bash-2.05a (multibyte character support) that I'm not going to release a patch. Chet -- ``The lyf so short, the craft so long to lerne.'' - Chaucer ( ``Discere est Dolere'' -- chet) Chet Ramey, CWRU[EMAIL PROTECTED]http://cnswww.cns.cwru.edu/~chet/ -- 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/
Using windows graphics routines
I am trying to use windows library specifically graphics library for plotting from my cygwin program. I need to use windows routines like: GETWINDOWCONFIG, DFLIB, SETCOLORRGB, $GCLEARSCREEN. I have a folder wich is called visual. How to go about linking the required windows VISUAL FORTRAN libraries which will serve my purpose? What kind of link comands I need in my makefile? Thanks, Dilip K. Paul mailto:[EMAIL PROTECTED] -- 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: Updating the DLL build instructions
If you will use a piece of code as an "example" for your DLL building docs, I have a suggestion: use tftp... I'd like to obtain TFTP code in the form of a .DLL instead of an .EXE. I need to ultimately give it COM interfaces for use from within Visual Basic 6. It's already part of inetutils, and there aren't a lot of files, so I'm assuming it would be pretty straightforward. Perhaps you could even take this the extra step of building the COM interfaces; I suspect other folks would benefit from this. Thanks, --Kevin -Original Message- From: Charles Wilson [mailto:[EMAIL PROTECTED]] Sent: Monday, February 18, 2002 4:58 PM If you are talking about generic "How do I build a DLL using cygwin tools" - I think I volunteered for that at some point in the past, but I haven't had a chance to get round 'tuit. If somebody else beats me to it, that would be great -- they could even use some of the documentation and examples from dllhelpers (0.4.0 for the lastest spiffy features, 0.2.6 for the "old-style" __declspec() stuff: both techniques need to be explained). -- 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/
Status of Inventor on CygWin
Anyone has an idea of inventor on Cygwin without running X? The same question about Qt? Is there hope for getting Qt / Inventor in the form of OpenGL and GLUT on cygwin, where one could just link a project with -lxxx and it would run on windows? Thanks for your comments, --Piyush -- 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: od
- Original Message - From: "David" <[EMAIL PROTECTED]> To: "Mark Himsley" <[EMAIL PROTECTED]> Cc: <[EMAIL PROTECTED]> Sent: Tuesday, February 19, 2002 06:03 Subject: Re: od > > >You can see that the hex values in the last line are reversed. > > >I think they should be 4854 5450 000a. > > > > Only if you were on a big-endian platform. > > I can understand what you are saying if I interpret it to mean that > the hex storage values on my machine WindowsNT - which I assume from > you message is a big-endian platform - are faithfully represented > by the output from the od -bcx display. > > What then surprises me is that the octal representation of this > same storage is 110 124 124 120 which is what I would expect. > > It seems to me that you are suggesting that the only correct > representation of the hex storage values would require od to > ouput an ascii value of THPT. I reject, and od itself contradicts, > this interpretation of the storage values as being meaningful. > > I urge you to reconsider your opinion and to modify od to > output 4854 5450. No-one here can modify how od works. If it is really important to you, take it up with the appropriate GNU group. Don't expect a lot of sympathy from them though, this behavior has been consistent for decades. -- Mac :}) ** I normally forward private questions to the appropriate mail list. ** Ask Smarter: http://www.tuxedo.org/~esr/faqs/smart-questions.htm 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: od
Sir, You have got to be kidding. For the record, the x86 architecture is little-endian, PowerPC, e.g., is bid-endian. On the off chance that you're running Windows NT (and Cygwin???) on an Alpha, I have to admit I don't know which byte ordering it uses. You will get the 16-bit output you want from od (any radix) if you run Darwin PPC, MacOS X or LinuxPPC, among other systems. Otherwise, what you request ain't gonna happen. If you want od to interpret and display integers with sizes greater than 1 byte in a way that differs from that of the local processor, you could create a hacked version of "od" for your personal use. Randall Schulz Mountain View, CA USA At 06:03 2002-02-19, David wrote: >Hi Mark, > >Thank you for your reply to my message about the od utility: I wrote: > > >> I created at test file with 4 characters in it: > >> > >> HTTP > >> > >> Then, I run od -bcx and I get: > >> > >> 110 124 124 120 012 000 > >> H T T P \n \0 > >> 5448 5054 000a > >You wrote: > > > That is correct in this little-endian platform, see > > http://info.astrian.net/jargon/terms/l/little-endian.html > > > > >You can see that the hex values in the last line are reversed. > > >I think they should be 4854 5450 000a. > > > > Only if you were on a big-endian platform. > >I can understand what you are saying if I interpret it to mean that the >hex storage values on my machine WindowsNT - which I assume from you >message is a big-endian platform - are faithfully represented by the >output from the od -bcx display. > >What then surprises me is that the octal representation of this same >storage is 110 124 124 120 which is what I would expect. > >It seems to me that you are suggesting that the only correct >representation of the hex storage values would require od to ouput an >ascii value of THPT. I reject, and od itself contradicts, this >interpretation of the storage values as being meaningful. > >I urge you to reconsider your opinion and to modify od to output 4854 5450. > >David -- 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: od
At 09:03 AM 2/19/2002, David wrote: >Hi Mark, > >Thank you for your reply to my message about the od utility: I wrote: > > >> I created at test file with 4 characters in it: > >> > >> HTTP > >> > >> Then, I run od -bcx and I get: > >> > >> 110 124 124 120 012 000 > >> H T T P \n \0 > >> 5448 5054 000a > >You wrote: > > > That is correct in this little-endian platform, see > > http://info.astrian.net/jargon/terms/l/little-endian.html > > > > >You can see that the hex values in the last line are reversed. > > >I think they should be 4854 5450 000a. > > > > Only if you were on a big-endian platform. > >I can understand what you are saying if I interpret it to mean that >the hex storage values on my machine WindowsNT - which I assume from >you message is a big-endian platform - are faithfully represented >by the output from the od -bcx display. No, PC's are little-endian. >What then surprises me is that the octal representation of this >same storage is 110 124 124 120 which is what I would expect. > >It seems to me that you are suggesting that the only correct >representation of the hex storage values would require od to >ouput an ascii value of THPT. I reject, and od itself contradicts, >this interpretation of the storage values as being meaningful. od formats the character representation to be in a "human readable" format. It's a convenience. The hex still represents the data in the machine-stored order though. >I urge you to reconsider your opinion and to modify od to >output 4854 5450. I doubt you'll get your way. od has always worked this way and a change in behavior is likely to be considered a bug by others. You'll probably need to adjust your thinking when using od, find something else that fits your way of thinking, or modify a local version to your liking. Still, I'm not the official word on od maintenance so perhaps you can convince the maintainers that you're right and everybody else is wrong! ;-) Larry Hall [EMAIL PROTECTED] RFK Partners, Inc. http://www.rfk.com 838 Washington Street (508) 893-9779 - RFK Office Holliston, MA 01746 (508) 893-9889 - FAX -- 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: Proposed patch for bash (completion spec problem related)
--- Chet Ramey <[EMAIL PROTECTED]> wrote: > > Possible patch for bash to fix the problem noticed by Jason Tishler in msg: > > http://sources.redhat.com/ml/cygwin/2002-02/msg00581.html > > > > > > --- subst.c.old Mon Nov 5 07:19:50 2001 > > +++ subst.c Mon Feb 18 16:58:50 2002 > > @@ -1339,7 +1339,7 @@ > > break; > > > >i = te /* + member (string[te], d) */; > > - while (member (string[i], d) && whitespace(string[i])) > > + while (member (string[i], d)/* && whitespace(string[i])*/) > > i++; > > The right fix is to replace the call to whitespace() with one to spctabnl(). > There's another fix needed for skip_to_delim(), but my current sources look > sufficiently different from bash-2.05a (multibyte character support) that > I'm not going to release a patch. > Ok, let's try that again. --- subst.c.old Mon Nov 5 07:19:50 2001 +++ subst.c Tue Feb 19 11:16:17 2002 @@ -1339,7 +1339,7 @@ break; i = te /* + member (string[te], d) */; - while (member (string[i], d) && whitespace(string[i])) + while (member (string[i], d) && spctabnl(string[i])) i++; if (string[i]) __ Do You Yahoo!? Yahoo! Sports - Coverage of the 2002 Olympic Games http://sports.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/
how to remap backspace and delete keys for 'info'?
When I'm in 'info' and try to scroll backward a page using the backspace key, I get *Info Help* instead. Since I also get help when I hit C-h, I suspect my backspace maps to C-h instead of DEL. (FWIW using the delete key () gets the response > Unknown command (~). ) Note also that I get this behavior running 'info' from either bash or cmd. How can I remap my backspace key in/for 'info'? Alternatively, how can I otherwise get the desired DEL behavior? Apologies if this is a FAQ (I know it is for Emacs, where I know how to fix it), but I didn't see anything relevant in info.info, the Cygwin FAQ, the Cygwin User's Guide, or the 'cygwin' list archives. Also, please reply to me as well as the list: I tried to subscribe, and got the happy page saying "your request has been received, and we'll send you a confirmation post shortly" but I haven't received the email yet. FWIW info --version > info (GNU texinfo) 4.0 cygcheck -s info > Found: C:\cygwin\bin\info.exe > C:\cygwin\bin\info.exe > C:\cygwin\bin\cygwin1.dll > C:\WINNT\System32\KERNEL32.dll > C:\WINNT\System32\NTDLL.DLL > Cygwin Win95/NT Configuration Diagnostics > Current System Time: Tue Feb 19 11:46:38 2002 > Windows 2000 Professional Ver 5.0 Build 2195 Service Pack 1 > Path: C:\cygwin\usr\local\bin > C:\cygwin\bin > C:\cygwin\bin > d:\bin\CVS\cvsnt > C:\cygwin\bin > c:\tlroche\bin > c:\WINNT\system32 > c:\WINNT > SysDir: C:\WINNT\System32 > WinDir: C:\WINNT > c: /c usertextmode > d: /d usertextmode > t: /t usertextmode > . /cygdrive userbinmode,noumount > C:/cygwin / system binmode > C:/cygwin/bin /usr/bin system binmode > C:/cygwin/lib /usr/lib system binmode > Found: C:\cygwin\bin\bash.exe > Found: C:\cygwin\bin\cat.exe > Not Found: cpp (good!) > Found: C:\cygwin\bin\find.exe > Not Found: gcc > Not Found: gdb > Not Found: ld > Found: C:\cygwin\bin\ls.exe > Found: c:\tlroche\bin\ls.exe > Found: C:\cygwin\bin\make.exe > Found: C:\cygwin\bin\sh.exe > 119k 2002/02/09 C:\cygwin\bin\cygjpeg6b.dll >35k 2002/01/09 C:\cygwin\bin\cygform6.dll >20k 2002/01/09 C:\cygwin\bin\cygmenu6.dll > 175k 2002/01/09 C:\cygwin\bin\cygncurses++6.dll >12k 2002/01/09 C:\cygwin\bin\cygpanel6.dll >18k 2000/10/23 C:\cygwin\bin\cyggdbm.dll >45k 2001/04/25 C:\cygwin\bin\cygform5.dll >26k 2001/04/25 C:\cygwin\bin\cygmenu5.dll > 156k 2001/04/25 C:\cygwin\bin\cygncurses++5.dll > 226k 2001/04/25 C:\cygwin\bin\cygncurses5.dll >15k 2001/04/25 C:\cygwin\bin\cygpanel5.dll > 202k 2002/01/09 C:\cygwin\bin\cygncurses6.dll > 170k 2002/01/21 C:\cygwin\bin\cygpng2.dll >41k 2002/01/20 C:\cygwin\bin\cygXpm-noX4.dll >46k 2002/01/20 C:\cygwin\bin\cygXpm-X4.dll >50k 2002/01/20 C:\cygwin\bin\cygz.dll >22k 2001/12/13 C:\cygwin\bin\cygintl-1.dll > 621k 2002/01/16 C:\cygwin\bin\cygcrypto.dll > 156k 2002/01/16 C:\cygwin\bin\cygssl.dll >21k 2001/06/20 C:\cygwin\bin\cygintl.dll > 253k 2002/02/10 C:\cygwin\bin\cygtiff3.dll > 751k 2002/01/21 C:\cygwin\bin\cygwin1.dll > Cygwin DLL version info: > DLL version: 1.3.9 > DLL epoch: 19 > DLL bad signal mask: 19005 > DLL old termios: 5 > DLL malloc env: 28 > API major: 0 > API minor: 51 > Shared data: 3 > DLL identifier: cygwin1 > Mount registry: 2 > Cygnus registry name: Cygnus Solutions > Cygwin registry name: Cygwin > Program options name: Program Options > Cygwin mount registry name: mounts v2 > Cygdrive flags: cygdrive flags > Cygdrive prefix: cygdrive prefix > Cygdrive default prefix: > Build date: Mon Jan 21 12:48:41 EST 2002 > Shared id: cygwin1S3 > Cygwin Package Information > Package Version > ash 20020131-1 > bash2.05a-2 > cygwin 1.3.9-1 > diff0.0 > fileutils 4.1-1 > findutils 4.1 > gawk3.0.4-1 > gdbm1.8.0-3 > grep2.4.2-1 > groff 1.17.2-1 > gzip1.3.2-1 > jpeg6b-7 > less358-3 > libintl 0.10.38-3 > libintl10.10.40-1 > libncurses5 5.2-1 > libncurses6 5.2-8 > libpng 1.0.12-1 > libpng2 1.0.12-1 > libreadline44.1-2 > libreadline54.2a-1 > login 1.4-3 > lynx2.8.4-1 > make3.79.1-5 > man 1.5g-2 > mutt1.2.5i-6 > ncurses 5.2-8 > openssl 0.9.6c-3 > patch 2.5-2 > readline4.2a-1 > sed 3.02-1 > sh-utils2.0-2 > ssmtp 2.38.7-3 > tar 1.13.19-1 > termcap 20010825-1 > terminfo5.2-1 > texinfo
Re: Re[2]: Core dumped just only with strcat!
Dani, You should really give us more info about the fault like say, the actual line number. Compile with -g option, run in gbd, it should tell you where it crashes, send us as much info as possible. I do not have experience in mysql but I did have a look at the code. If you are crashing in teh mysql call maybe you not allocating and initialising the MYSQL *base_de_datos structure? Are you calling mysql_init() as per http://www.mysql.com/documentation/mysql/bychapter/manual_Clients.html#mysql _init? Another thing: mysql expects you to put ';' at the end of every SQL query, thats in their docs too. Eugene. - Original Message - From: "Pavel Tsekov" <[EMAIL PROTECTED]> To: "Dani P." <[EMAIL PROTECTED]> Cc: <[EMAIL PROTECTED]> Sent: Tuesday, February 19, 2002 10:44 PM Subject: Re[2]: Core dumped just only with strcat! Lots of stuff omitted... -- 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: od
On Tue, 19 Feb 2002 14:03:55 GMT you wrote: > > > Then, I run od -bcx and I get: > > > > > > 110 124 124 120 012 000 > > > H T T P \n \0 > > > 5448 5054 000a > > > > That is correct in this little-endian platform, see > > http://info.astrian.net/jargon/terms/l/little-endian.html > > I can understand what you are saying if I interpret it to mean that > the hex storage values on my machine WindowsNT - which I assume from > you message is a big-endian platform - are faithfully represented > by the output from the od -bcx display. In case you misinterpreted what I said and did not read the URL I listed, your Intel machine is little-endian. If you don't know what a 'short' is or do not understand the difference between a 'short' and a 'byte' or you don't understand the difference between a big and a little endian platform is then I suggest you do some research. > > >I think they should be 4854 5450 000a. > > > > Only if you were on a big-endian platform. > >What then surprises me is that the octal representation of this >same storage is 110 124 124 120 which is what I would expect. Yes, because the octal and the ascii been split into the byte values therefore they must be in the correct order. What you are looking for is a byte sized hex representation (as opposed to the shorts which are obviously confusing you). >It seems to me that you are suggesting that the only correct >representation of the hex storage values would require od to >ouput an ascii value of THPT. Where exactly did I say that? >I urge you to reconsider your opinion and to modify od to >output 4854 5450. If you want to modify your copy of od then do so. If you feel like it then send a patch to the maintainer (not me). You can urge me to reconsider what every you like but since I understand the output of od and I did not write it your urges will have no affect on me what-so-ever. More over, this whole discussion has no relevance that I can see to cygwin and therefore I urge you to discuss this issue with the appropriate people through the appropriate channels. -- Mark Himsley In Acton -- 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: Updating the DLL build instructions
On Tue, Feb 19, 2002 at 10:38:29AM -0500, Roth, Kevin P. wrote: >If you will use a piece of code as an "example" for your DLL building >docs, I have a suggestion: use tftp... > >I'd like to obtain TFTP code in the form of a .DLL instead of an .EXE. I >need to ultimately give it COM interfaces for use from within Visual >Basic 6. It's already part of inetutils, and there aren't a lot of >files, so I'm assuming it would be pretty straightforward. Sure. Send your words + example here for review. That would be best. 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/
How to create a windows link in script?
I would like to be able to create a .lnk file to a batch file (and ideally associate an icon with it) from a bash script on a Windows 2000 machine. Does anyone have any suggestions? I think "ln -s" used to work, but has since been improved? Anyway, this does not provide the functionality for the icon. -rgm -- 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/
openssh publickey authentication problem
Hi, We are trying to run shell scripts and binaries on remote machine (NT and Unix) by using ssh (openssh 3.02). The NT (2k and 4) network has PDC. The Cygwin version is 1.3.9. From NT to Unix the public key authentication is working good (RSA and DSA both). From Unix to NT it doesn't work. The output of ssh -v and the sshd_config (both Unix and NT) are attached. Thanks for any assistance. Guy Amir Unix system administrator Mysticom L.t.d mailto:[EMAIL PROTECTED] sshd_config.NT Description: Binary data sshd_config.unix Description: Binary data nt2unix.log Description: Binary data unix2nt.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: od
David - I agree with you that the little-endian representation can be somewhat confusing; however, as Mark pointed out, it's not likely to ever change... I've had my own "struggles" with od's idea of output formats, and have come up with a few options along the way. Here's some that will help, if you're looking to get hex characters output in the order you're expecting: $ od -toC -tc -tx1 filename 110 124 124 120 012 H T T P \n 48 54 54 50 0a $ od -tcx1 -w24 filename | sed '/^[0-9]/s/ \( .\)/\1/g'; H T T P \n 48 54 54 50 0a $ od -txlz filename 48 54 54 50 0a >HTTP.< --Kevin -- 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/
inputstream!
i'm sorry, but i just had this question and i thought you might be able to help me. if i write a simple java program that uses a BufferedReader reader object wrapped from System.in and then try to read a line from user and then echo it back, cygwin for some reason already has stuff on the stream and so without waiting for the user to type something in the program fetches that and prints it out (it is usually) just a newline. why does this happen! i know it's not the program because it works in djgpp and other platforms. any help would be greatly appreciated. thanks. -- 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/
od commands
Hi Kevin I used od -txz and it's exactly what I was looking for. I couldn't get the z argument to work before you commented on my request. Thanks very much to you and everyone. David -- 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/
1.4.5 gnu sort 2.0.16 extremely slow under Win98 (tiny default buffer)
I looked at the temp files sort creates and noticed that they were around 700 bytes (bytes, not kilobytes) in size. The normal behavior of sort is to create temp files of around 16kB in size. When I use the --buffer-size option to force a large buffer, like --buffer-size=20 (200MB) sort completes in seconds. I'm running sort like this: cat <47k line file> | perl -ne "
Re: TCSH startup scripts failure
On Sat, Feb 09, 2002 at 06:23:54PM -0800, Michael wrote: > Hi > > As of the latest version of Cygwin, there is a flaw in the tcsh startup > scripts in /etc and /etc/profile.d that causes tcsh to fail if $HOME > contains a space (eg "/home/John Smith") > > Lines such as > if ( -r $HOME/.cshrc ) ... > will cause errors unless changed to > if ( -r "$HOME/.cshrc" ) ... I changed /etc/csh.cshrc slightly The above is fixed, too. A new tcsh is available in a few hours. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc. -- 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: od
Hi Randall, no need to hack od because of endianess, we are on *nix after all, so pipe it thru: dd conv=swab If needed often, put it in a wrapper shell script or an alias! Testing portably if you want it swapped or not is left as an excercise for the reader 8-) Bye, Heribert ([EMAIL PROTECTED]) > -Original Message- > From: Randall R Schulz [SMTP:[EMAIL PROTECTED]] > Sent: Tuesday, February 19, 2002 17:56 > To: David; Mark Himsley > Cc: [EMAIL PROTECTED] > Subject: Re: od > > Sir, > > You have got to be kidding. > > For the record, the x86 architecture is little-endian, PowerPC, e.g., is > bid-endian. On the off chance that you're running Windows NT (and > Cygwin???) on an Alpha, I have to admit I don't know which byte ordering > it > uses. > > You will get the 16-bit output you want from od (any radix) if you run > Darwin PPC, MacOS X or LinuxPPC, among other systems. > > Otherwise, what you request ain't gonna happen. If you want od to > interpret > and display integers with sizes greater than 1 byte in a way that differs > from that of the local processor, you could create a hacked version of > "od" > for your personal use. > > Randall Schulz > Mountain View, CA USA > [Heribert] [snip] -- 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: How to create a windows link in script?
Roland,
Well, I was going to say this:
If this is a one-off sort of thing, just use Windows Explorer (i.e., the
Properties dialog) to set the icon on the ".lnk" file (shortcut) that
results from invoking "ln -s ".
However, I find that the resulting .lnk file won't allow the icon to be
changed. I get an error alert when I try to "Apply" the changed icon
("Unable to save changes to gvi.lnk"; "Access is denied."). This is true
even after I go to the Security pane of the link/shortcut's Properties
dialog and enable "Full Control" and "Modify" permissions, apply them,
close the dialog, come back in and try all over again. The permissions
change "takes" but I still cannot set the icon.
I wonder what's up here? This is an up-to-date Cygwin (1.3.9), NTFS file
system and CYGWIN=ntsec. My BASH "umask" is 2, but I did a chmod 777
afterward (as well as interactively changing the permissions as I mentioned
above).
Can anybody clue me in on what's happening with this?
Thanks.
Randall Schulz
Mountain View, CA USA
At 10:07 2002-02-19, Roland Glenn McIntosh wrote:
>I would like to be able to create a .lnk file to a batch file (and ideally
>associate an icon with it) from a bash script on a Windows 2000 machine.
>
>Does anyone have any suggestions?
>
>I think "ln -s" used to work, but has since been improved? Anyway, this
>does not provide the functionality for the icon.
>
>-rgm
--
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: Proposed patch for bash (completion spec problem related)
On Tue, Feb 19, 2002 at 12:27:54PM -0500, Jason Tishler wrote: > On Tue, Feb 19, 2002 at 09:05:01AM -0500, Chet Ramey wrote: > > On Mon, Feb 18, 2002 at 05:06:36PM -0800, AJ Reins wrote: > > > Possible patch for bash to fix the problem noticed by Jason Tishler in msg: > > > http://sources.redhat.com/ml/cygwin/2002-02/msg00581.html > > > > > > [snip] > > AJ, thanks for tracking down this problem. > > > The right fix is to replace the call to whitespace() with one to spctabnl(). > > There's another fix needed for skip_to_delim(), ... > > Chet, thanks for reviewing AJ's patch. > > Corinna, attached is AJ's patch redone with a call to spctabnl() but > without the unknown fix to skip_to_delim(). Nevertheless, Cygwin bash > 2.05a, with this patch, can now handle completion specs. Would you be > willing to release a new bash 2.05a with this patch? Yeah, and thanks to you all. I'm just uploading the new version of bash. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc. -- 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: od
Hi, Heribert, Ahh, yes, my old friend dd and his funky command syntax (inspired by some old IBM command language, if I'm not mistaken). The Swiss Army Knife of Unix data transfer. Thanks for reminding me. However, I don't see a 4-byte equivalent to "conv=swab" in the manual page, so this will not extend to 32-bit entities. Randy At 11:21 2002-02-19, Heribert Dahms wrote: >Hi Randall, > >no need to hack od because of endianess, we are on *nix after all, so pipe >it thru: dd conv=swab If needed often, put it in a wrapper shell script or >an alias! Testing portably if you want it swapped or not is left as an >excercise for the reader8-) > >Bye, Heribert ([EMAIL PROTECTED]) -- 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: Proposed patch for bash (completion spec problem related)
On Tue, Feb 19, 2002 at 09:05:01AM -0500, Chet Ramey wrote: > On Mon, Feb 18, 2002 at 05:06:36PM -0800, AJ Reins wrote: > > Possible patch for bash to fix the problem noticed by Jason Tishler in msg: > > http://sources.redhat.com/ml/cygwin/2002-02/msg00581.html > > > > [snip] AJ, thanks for tracking down this problem. > The right fix is to replace the call to whitespace() with one to spctabnl(). > There's another fix needed for skip_to_delim(), ... Chet, thanks for reviewing AJ's patch. Corinna, attached is AJ's patch redone with a call to spctabnl() but without the unknown fix to skip_to_delim(). Nevertheless, Cygwin bash 2.05a, with this patch, can now handle completion specs. Would you be willing to release a new bash 2.05a with this patch? Thanks, Jason --- subst.c.origTue Feb 19 09:29:06 2002 +++ subst.c Tue Feb 19 09:30:01 2002 @@ -1339,7 +1339,7 @@ split_at_delims (string, slen, delims, s break; i = te /* + member (string[te], d) */; - while (member (string[i], d) && whitespace(string[i])) + while (member (string[i], d) && spctabnl (string[i])) i++; if (string[i]) -- 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/
cygreadline5.dll redux
This file appears to be missing. Before someone jumps on my case and says "Search the archives" I'd like to point out that a) The archives are highly polluted with messages containing cygcheck output that lists cygreadline5.dll. A search for just "cygreadline5.dll" returns hundreds of hits most of which have nothing to do with the problem. b) The archive search engine is too stupid to understand quoted strings meaning adjacency, so that searching for 'cygrealine5.dll "not found"' is equivalent to searching for the individual words. Since 'found' appears in cygcheck output, this does not reduce the number of hits at all. Could someone please either answer the question as to why some packages still require this library even though it does not appear to be included in the distribution, and then add the answer to the FAQ? TIA -- James GarrisonAthens Group, Inc. mailto:[EMAIL PROTECTED]5608 Parkcrest Dr http://www.athensgroup.comAustin, TX 78731 PGP: RSA=0x92E90A3B DH/DSS=0x498D331C (512) 345-0600 x150 -- 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: cygreadline5.dll redux - resolved
I am using the most current installer. However, I just solved the problem by forcing it to reinstall the readline package. For whatever reason, the library was missing even though the installer thought the package was installed. Andrew Markebo wrote: > Just a quick note, which installer did you use.. what I mean is that > the modern setup.exe should include this if it is needed, > > But otoh someone might have forgotten to mention readline in the > config of his package, how have you missed?? > > How I get it.. I run the installer and select it :-) -- James GarrisonAthens Group, Inc. mailto:[EMAIL PROTECTED]5608 Parkcrest Dr http://www.athensgroup.comAustin, TX 78731 PGP: RSA=0x92E90A3B DH/DSS=0x498D331C (512) 345-0600 x150 -- 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: cywin!
Well, a quick search of the cygwin mailing list about Java would have unfortunately told you that Java is not "cygwin-aware". I run javac from cmd.exe for this reason. You might try wrapping your Java program in a script like: CMD.EXE /c "javaprog.exe" Hope it helps. --- "pramod m." <[EMAIL PROTECTED]> wrote: > if i write a simple java program that uses a > BufferedReader reader object wrapped from System.in > and then try to read a line from user and then echo > it back, cygwin for some reason already has stuff on > the stream and so without waiting for the user to > type something in the program fetches that and > prints it out (it is usually) just a newline. why > does this happen! > > i know it's not the program because it works in > djgpp and other platforms. __ Do You Yahoo!? Yahoo! Sports - Coverage of the 2002 Olympic Games http://sports.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: cygreadline5.dll redux
http://cygwin.com/ml/cygwin-announce/2002/msg00011.html http://cygwin.com/ml/cygwin/2002-01/msg00855.html http://cygwin.com/cgi-bin2/package-grep.cgi?grep=cygreadline5.dll http://cygwin.com/ml/cygwin/2002-01/msg01619.html http://cygwin.com/faq/faq.html#SEC32 Sigh. --Chuck James Garrison wrote: > This file appears to be missing. > > Before someone jumps on my case and says "Search the archives" > I'd like to point out that > > a) The archives are highly polluted with messages containing >cygcheck output that lists cygreadline5.dll. A search for >just "cygreadline5.dll" returns hundreds of hits most of which >have nothing to do with the problem. > > b) The archive search engine is too stupid to understand >quoted strings meaning adjacency, so that searching for >'cygrealine5.dll "not found"' is equivalent to searching >for the individual words. Since 'found' appears in cygcheck >output, this does not reduce the number of hits at all. > > Could someone please either answer the question as to why > some packages still require this library even though it does > not appear to be included in the distribution, and then > add the answer to the FAQ? > > TIA > -- 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: cygreadline5.dll redux
> b) The archive search engine is too stupid to understand > quoted strings meaning adjacency, so that searching for > 'cygrealine5.dll "not found"' is equivalent to searching > for the individual words. Since 'found' appears in cygcheck > output, this does not reduce the number of hits at all. Try using google to search. It's quite a bit smarter. Rob -- 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: cygreadline5.dll redux
On Tue, Feb 19, 2002 at 11:24:51AM -0600, James Garrison wrote: >This file appears to be missing. > >Before someone jumps on my case and says "Search the archives" >I'd like to point out that > >a) The archives are highly polluted with messages containing > cygcheck output that lists cygreadline5.dll. A search for > just "cygreadline5.dll" returns hundreds of hits most of which > have nothing to do with the problem. > >b) The archive search engine is too stupid to understand > quoted strings meaning adjacency, so that searching for > 'cygrealine5.dll "not found"' is equivalent to searching > for the individual words. Since 'found' appears in cygcheck > output, this does not reduce the number of hits at all. > >Could someone please either answer the question as to why >some packages still require this library even though it does >not appear to be included in the distribution, and then >add the answer to the FAQ? Hmm. A mailing list search of the stupid search engine with "cygreadline5.dll not found" unearthed this within a few clicks: http://cygwin.com/ml/cygwin/2002-01/msg01619.html If this doesn't answer the question, then I don't know what the question is. 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: Registry freedom
> >The Cygwin + custom GNU tools > >installer I created installs this tools for system wide use by various > >product installers. The default location is %SystemDrive%\wrtools to > >avoid cluttering an existing C:\cygwin. > > It would hardly be cluttering to produce an /opt/wrtools. That's what > you'd do on linux or something similar. You wouldn't create your own > bin directory somewhere off the root on UNIX. For UNIX tools in a UNIX environment yes. My need is to control a separate, 'stable' cygwin.dll + GNU tools installation. I can't put it in RH distro /. It can cause both of us support problems. It is also kind of installing the 'stable' distro in the /opt dir of an existing one. It got to be separate. > >Some users might have various versions and is not a good practice to > >mess up what they have. I really don't want folks to be creative and > >make changes to the stuff I rely on. They can play with c:\cygwin if > >they want to. I also advocate the slogan "you package it, you maintain > >it". > > That's fine, but none of this is really a goal for cygwin as I see it. I can see your point. I thought that other users might need this too. The lack of replies from other folks tell the story. I will put this on my to-do list. I still believe it is worth looking into. > >When the user starts a WR product installer he/she will be prompted for > >a location to install for that media. The product install dir has no > >connection whatsoever with the location of the Unix emulator providing > >the POSIX API. > > Ok. Then the mount table doesn't really matter at all. Just ignore it. > Use /cygdrive/whatever for everything. That is what I thought in the beginning. What I discover later was that yes I can start my own BASH with my own cygwin1.dll version but the DLL will read the existing mount table. One need to make sure there are no mount points in the BASH path to avoid using stuff from the existing installation. > I'm always open to inspecting actual > source code patches. Until then, I think my stance on "good ideas" > should be well known in this mailing list. > > cgf Time allowing I will try to take a closer look and see what is the magnitude of the change. You will definitely see a patch if I will be able to make it work. Thank you Christopher. Doru Carastan -- 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: cygreadline5.dll redux
Christopher Faylor wrote: > On Tue, Feb 19, 2002 at 11:24:51AM -0600, James Garrison wrote: > > Hmm. A mailing list search of the stupid search engine with > "cygreadline5.dll not found" unearthed this within a few clicks: > > http://cygwin.com/ml/cygwin/2002-01/msg01619.html > > If this doesn't answer the question, then I don't know what the > question is. Sorry, but I don't usually expect files in installed packages to simply disappear without warning. What exactly causes files to disappear such that the FAQ answer is "just reinstall the package"? -- James GarrisonAthens Group, Inc. mailto:[EMAIL PROTECTED]5608 Parkcrest Dr http://www.athensgroup.comAustin, TX 78731 PGP: RSA=0x92E90A3B DH/DSS=0x498D331C (512) 345-0600 x150 -- 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/
recvfrom and timeout signal
Hi All,
I try to implement timeout for recvfrom:
struct sigaction action;
action.sa_handler = alarm_signal;
action.sa_flags = 0;
sigemptyset(&(action.sa_mask)); /* ignore all known signals */
sigaction(SIGALRM,&action,NULL);/* ensures that SA_RESTART is NOT set
*/
alarm(TIMEOUT);
int st=recvfrom(sockfd,buff,sizeof(buff),0,(sockaddr*)&srv_addr,&sz);
if(st == -1){
if(errno==EINTR){
/* timeout */
}else{
alarm(0); /* reset alarm */
}
}
but the process is never woken by SIGALRM when no packet is received.
The recvfrom waits forever.
Has anybody an idea what can be wrong ?
--
Piotr Stepien
--
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: cygreadline5.dll redux
On Tue, Feb 19, 2002 at 02:57:43PM -0600, James Garrison wrote: >Christopher Faylor wrote: >>On Tue, Feb 19, 2002 at 11:24:51AM -0600, James Garrison wrote: >> >>Hmm. A mailing list search of the stupid search engine with >>"cygreadline5.dll not found" unearthed this within a few clicks: >> >>http://cygwin.com/ml/cygwin/2002-01/msg01619.html >> >>If this doesn't answer the question, then I don't know what the >>question is. > >Sorry, but I don't usually expect files in installed packages to >simply disappear without warning. What exactly causes files to >disappear such that the FAQ answer is "just reinstall the package"? I'm not sure what this has to do with an inability to find things in the mailing list archives. If someone points you to the archives and the definitive answer from the package maintainer is to reinstall, then reinstall. However, to answer your question, the current version of setup.exe is simple-minded in the way it uninstalls a package prior to installing. If new 'package a' installs foo.dll and old 'package b' no longer contains foo.dll, then setup.exe will happily install 'package a' with it's new foo.dll *but* then uninstall 'package b' (including foo.dll) and reinstall 'package b'. The result is that foo.dll is missing. "We" are working on fixing this in setup.exe but, until then, we'll be answering this indignant question on a weekly basis, it seems. 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: recvfrom and timeout signal
On Tue, Feb 19, 2002 at 10:28:02PM +0100, Piotr St?pie? wrote:
>Hi All,
>
>I try to implement timeout for recvfrom:
>
> struct sigaction action;
> action.sa_handler = alarm_signal;
> action.sa_flags = 0;
> sigemptyset(&(action.sa_mask)); /* ignore all known signals */
> sigaction(SIGALRM,&action,NULL);/* ensures that SA_RESTART is NOT set
>*/
>
> alarm(TIMEOUT);
> int st=recvfrom(sockfd,buff,sizeof(buff),0,(sockaddr*)&srv_addr,&sz);
> if(st == -1){
> if(errno==EINTR){
> /* timeout */
> }else{
> alarm(0); /* reset alarm */
> }
> }
>
>but the process is never woken by SIGALRM when no packet is received.
>The recvfrom waits forever.
>
>Has anybody an idea what can be wrong ?
Nope. Send an actual small working test case and someone will probably
investigate it.
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/
Problems compiling latest snapshot - cygwin-src-20020219.tar.bz2
Hey there! ;) The latest snapshot as in the subject fails to compile. The error occures when the make commands tries to execute in winsup/w32api, which is empty. The snapshot itself does contain a win32 directory but its empty. I check the source of the stock 1.3.9-1 and it doesn't have any files in the w32api directory. This makes me think that this question maybe have an answer and the post may be considred RTFM or stupid or something else :) Still I'd like to ask it - are there any other steps that should be done to build the cygwin dll except downloading the snapshot and configuring in separate dir ? Thanks! :) P.S. I read the mailing list but haven't noticed anything on this topic :( -- 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/
[ANNOUNCEMENT] Updated: ghostscript-6.51-3
The Ghostscript package has been updated to fix a bug reported by Greg Bond related to operation on text-mode mounts: http://sources.redhat.com/ml/cygwin/2001-12/msg00400.html This release of Ghostscript is an "X-less" version (i.e. one that doesn't require X11), and so can be used without the cygwin-xfree packages. The GNU 'standard' and 'other' Ghostscript font collections are included for convenience. Since this release is built using the unix-gcc.mak makefile, Windows GUI support is not included. This means that Postscript files must be previewed indirectly (they must be rasterized to an output file first, e.g. JPEG or PDF, and then viewed with a suitable viewer.) For build instructions, please read: /usr/doc/Cygwin/ghostscript-6.51.README The instructions have also been included at the end of this message. I want to thank Greg Bond for submitting the text-mode mount patch, and Chuck Wilson for help uploading the package. Release Notes -- GNU Ghostscript for Cygwin Version 6.51, Release 3 (6.51-3) (9 January, 2002) This Cygwin version of GNU Ghostscript has been built without support for X. This version fixes problems seen with release 2 of 6.51 (7/5/01) when running on text-mode mounts. (Thanks to Greg Bond, <[EMAIL PROTECTED]>, for submitting a patch to fix this problem.) REQUIREMENTS: Previous releases required shared libraries for zlib, libpng, and jpeg. However, due to the very specific requirements Ghostscript has regarding these libraries, these libraries are instead statically linked. As a result, Ghostscript no longer requires the use of shared libraries. The libraries are included in the source package, ready to be built with Ghostscript. BUILDING: At a minimum, you must have all the tools needed to build Cygwin programs, i.e. make, gcc, and binutils. You also need to fetch the source archive, ghostscript-6.51-src.tar.gz from a Cygwin mirror site. Once you have everything required for building, follow these steps: 1. Unpack ghostscript-6.51-src.tar.gz (the source is pre-patched to build on Cygwin): $ tar xzf ghostscript-6.51-src.tar.gz 2. Build from the top-level source directory: $ cd ghostscript-6.51 ; make -f src/unix-gcc.mak 3. Install the entire package: $ make -f src/unix-gcc.mak install 4. Grab the following font compressed tar files: ftp://mirror.cs.wisc.edu/pub/mirrors/ghost/gnu/fonts: gnu-gs-fonts-std-6.0.tar.gz (required) gnu-gs-fonts-other-6.0.tar.gz (optional, recommended) Unpack them into /usr/share/ghostscript/fonts (font directory contained within tar file): $ tar xzCf /usr/share/ghostscript gnu-gs-fonts-*-6.0.tar.gz 5. Once you install it, you can test it: $ gs -sDEVICE=jpeg -dBATCH -sOutputFile=tiger.jpg \ /usr/share/ghostscript/6.51/examples/tiger.ps The Cygwin patches are in CYGWIN-PATCHES, along with this file. Have fun! -- Dario Alcocer -- Sr. Software Developer, Helix Digital Inc. [EMAIL PROTECTED] -- http://www.helixdigital.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: od
Randall R Schulz <[EMAIL PROTECTED]> writes: > Sir, > > You have got to be kidding. 22:53:33 fred@appel:~$ uname -a Linux appel.lilypond.org 2.4.18-pre9-benh #1 Sat Feb 16 20:51:15 CET 2002 ppc unknown 22:53:37 fred@appel:~$ echo http > http 22:53:43 fred@appel:~$ od -bcx http 000 150 164 164 160 012 000 h t t p \n \0 6874 7470 0a00 005 22:54:01 jan@verdon:~$ uname -a Linux verdon 2.4.17-k7 #2 Sat Dec 22 22:03:49 EST 2001 i686 unknown 22:54:05 jan@verdon:~$ echo http > http 22:54:15 jan@verdon:~$ od -bcx http 000 150 164 164 160 012 000 h t t p \n \0 7468 7074 000a 005 > For the record, the x86 architecture is little-endian, PowerPC, e.g., is > bid-endian. Jep. Jan. -- Jan Nieuwenhuizen <[EMAIL PROTECTED]> | GNU LilyPond - The music typesetter http://www.xs4all.nl/~jantien | http://www.lilypond.org -- 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 compiling latest snapshot - cygwin-src-20020219.tar.bz2
> The latest snapshot as in the subject fails to compile. The error > occures when the make commands tries to execute in winsup/w32api, > which is empty. The snapshot itself does contain a win32 directory > but its empty. I check the source of the stock 1.3.9-1 and it doesn't > have any files in the w32api directory. This makes me think that this > question maybe have an answer and the post may be considred RTFM or > stupid or something else :) Still I'd like to ask it - are there > any other steps that should be done to build the cygwin dll except > downloading the snapshot and configuring in separate dir ? Download the sources to w32api and make a symlink. IIRC the currently released w32api package has a missing #define which needs to go in winnt.h: #defineINVALID_FILE_ATTRIBUTES (-1) I believe this has been in CVS for a while and therefore will show up in the next package (if it has not already). Regards Chris -- 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: recvfrom and timeout signal
On Tue, Feb 19, 2002 at 11:51:59PM +0100, Piotr Stepien wrote:
>Christopher,
>
>Hope this working test case is small enough. I removed all unecessary code.
>Please let me know if you need anything else.
I wasn't asking for personal email. That's precisely why I set the
Reply-To and Mail-Followup-To to go to the cygwin mailing list.
"Someone" does not necessarily mean me.
cgf
>-Original Message-
>From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of
>Christopher Faylor
>Sent: Tuesday, February 19, 2002 10:42 PM
>To: [EMAIL PROTECTED]
>Subject: Re: recvfrom and timeout signal
>
>
>On Tue, Feb 19, 2002 at 10:28:02PM +0100, Piotr St?pie? wrote:
>>Hi All,
>>
>>I try to implement timeout for recvfrom:
>>
>> struct sigaction action;
>> action.sa_handler = alarm_signal;
>> action.sa_flags = 0;
>> sigemptyset(&(action.sa_mask)); /* ignore all known signals */
>> sigaction(SIGALRM,&action,NULL);/* ensures that SA_RESTART is NOT set
>*/
>>
>> alarm(TIMEOUT);
>> int st=recvfrom(sockfd,buff,sizeof(buff),0,(sockaddr*)&srv_addr,&sz);
>> if(st == -1){
>> if(errno==EINTR){
>> /* timeout */
>> }else{
>> alarm(0); /* reset alarm */
>> }
>> }
>>
>>but the process is never woken by SIGALRM when no packet is received.
>>The recvfrom waits forever.
>>
>>Has anybody an idea what can be wrong ?
>
>Nope. Send an actual small working test case and someone will probably
>investigate it.
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#define BADSIGNAL ((void(*)(int))-1)
#define TIMEOUT 5
void err_dump(const char* msg){
cerr << msg << ", errno=" << errno << endl;
exit(1);
}
void alarm_signal(int){
cout << "signal received\n";
cout.flush();
}
int main(){
struct sockaddr_in cli_addr,srv_addr;
bzero( (char*)&srv_addr, sizeof(srv_addr) );
srv_addr.sin_family = AF_INET;
srv_addr.sin_addr.s_addr = inet_addr("255.255.255.255");
srv_addr.sin_port = htons(192);
bzero( (char*)&cli_addr, sizeof(cli_addr) );
cli_addr.sin_family = AF_INET;
cli_addr.sin_addr.s_addr = htonl(INADDR_ANY);
cli_addr.sin_port = htons(0);
int sockfd = socket(AF_INET,SOCK_DGRAM,0);//IPPROTO_UDP);
if(sockfd == -1){
err_dump("can't open socket");
}
{
int on=1;
if(setsockopt(sockfd,SOL_SOCKET,SO_BROADCAST,&on,sizeof(on))==-1){
err_dump("error setsockopt");
}
}
if(bind(sockfd,(sockaddr*)&cli_addr,sizeof(cli_addr))== -1){
err_dump("can't bind to port");
}
char buff[0x100];
/*** BEGIN AREA OF INTEREST ***/
struct sigaction action;
action.sa_handler = alarm_signal;
action.sa_flags = 0;
sigemptyset(&(action.sa_mask)); /* ignore all known signals */
sigaction(SIGALRM,&action,NULL);/* ensures that SA_RESTART is NOT set
*/
int sz=sizeof(srv_addr);
alarm(TIMEOUT);
int st=recvfrom(sockfd,buff,sizeof(buff),0,(sockaddr*)&srv_addr,&sz);
/*** END AREA OF INTEREST ***/
if(st == -1){
if(errno==EINTR){
cout << "timeout occured\n";
}else{
alarm(0);
err_dump("error recvfrom");
}
}
close(sockfd);
return 0;
}
--
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/
creating a root/superuser under cygwin????
Hi: creating a root/superuser under cygwin How to do it? Thanks. -- 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 not a news server?
On Sun, Feb 17, 2002 at 08:57:24PM -0500, Christopher Faylor wrote: >Please go to http://cygwin.com/ and post a news item. Hmm. I guess the excitement about this doesn't extend to actually doing this very minor thing. 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: Problems compiling latest snapshot - cygwin-src-20020219.tar.bz2
On Tue, Feb 19, 2002 at 11:45:37PM +0100, [EMAIL PROTECTED] wrote: >The latest snapshot as in the subject fails to compile. The error >occures when the make commands tries to execute in winsup/w32api, >which is empty. I re-added w32api and mingw to the snapshot sources. cgf -- Please do not send me personal email with cygwin questions. Use the resources at http://cygwin.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/
BUG with 1.3.9 : -mno-cygwin target still depends on cygwin1.dll
Hello, I'm currently working on a cygwin32 port of DevIL (http://openil.sf.net). I get strange results when trying to compile the libs with -mno-cygwin (we prefer to not depend on cygwin unix emulation as we don't need it, we just use standard C functions) The problem is that i get dlls that still depend on cygwin1.dll ! but the compilation is done with -mno-cygwin flag. You have the entire libILU.dll compilation process in attachment (and the cygcheck output). And there, you have the cygwin1.dll import symbols : 100134b0 I ___libs_libimp_cygIL_0_a_iname 1000fba8 T __alloca 1001166c B __bss_end__ 10011000 B __bss_start__ 1000fd5c T __cygwin_crt0_common@8 1000fbfc T __cygwin_dll_entry@12 1000fcac T __cygwin_noncygwin_dll_entry@12 10010194 D __data_end__ 1001 D __data_start__ 0001 A __dll__ 10001000 T __end__ 0200 A __file_alignment__ 10011638 B __fmode 10013000 I __head__libs_libimp_cygIL_0_a 10013014 I __head_cygwin1_dll 1001318c I __imp__atan 10013190 I __imp__calloc 10013194 I __imp__ceil 10013198 I __imp__cygwin_detach_dll 1001319c I __imp__cygwin_internal 100131a0 I __imp__dll_dllcrt0 100131a4 I __imp__floor 100131a8 I __imp__free 100131ac I __imp__malloc 100131b0 I __imp__memcpy 100131b4 I __imp__pow 100131b8 I __imp__rand 100131bc I __imp__realloc 10011630 B __impure_ptr Is this cygwin version known to be buggy with -mno-cygwin targets ? -- Edouard Gomez NB : if you want to test yourself, use the cvs, because latest *official* release doesn't have cygwin support at all. Cygwin Win95/NT Configuration Diagnostics Current System Time: Wed Feb 20 00:07:08 2002 Windows 98 SE Ver 4.10 Build Path: C:\cygwin\usr\local\bin C:\cygwin\bin C:\cygwin\bin c:\WIN c:\WIN\COMMAND SysDir: C:\WIN\SYSTEM WinDir: C:\WIN HOME = `C:\cygwin\usr\src\DevIL\src-ILU\src\GomGom' MAKE_MODE = `unix' PWD = `/usr/src/DevIL/src-ILU/src' USER = `GomGom' BLASTER = `A220 I7 D1 H7 P330 T6' CMDLINE = `bash --login -i' COMSPEC = `C:\WIN\COMMAND.COM' MANPATH = `:/usr/ssl/man' OLDPWD = `/usr/src/DevIL/src-ILU' PROMPT = `$p$g' PS1 = `\[\033]0;\w\007 \033[32m\]\u@\h \[\033[33m\w\033[0m\] $ ' SBPCI = `C:\SBPCI' SHLVL = `1' TEMP = `c:\WIN\TEMP' TERM = `cygwin' TMP = `c:\WIN\TEMP' WINBOOTDIR = `C:\WIN' WINDIR = `C:\WIN' _ = `/usr/bin/cygcheck.exe' HKEY_CURRENT_USER\Software\Cygnus Solutions HKEY_CURRENT_USER\Software\Cygnus Solutions\Cygwin HKEY_CURRENT_USER\Software\Cygnus Solutions\Cygwin\mounts v2 (default) = `/cygdrive' cygdrive flags = 0x0022 HKEY_CURRENT_USER\Software\Cygnus Solutions\Cygwin\Program Options HKEY_LOCAL_MACHINE\Software\Cygnus Solutions HKEY_LOCAL_MACHINE\Software\Cygnus Solutions\Cygwin HKEY_LOCAL_MACHINE\Software\Cygnus Solutions\Cygwin\mounts v2 HKEY_LOCAL_MACHINE\Software\Cygnus Solutions\Cygwin\mounts v2\/ (default) = `C:/cygwin' flags = 0x000a HKEY_LOCAL_MACHINE\Software\Cygnus Solutions\Cygwin\mounts v2\/usr/bin (default) = `C:/cygwin/bin' flags = 0x000a HKEY_LOCAL_MACHINE\Software\Cygnus Solutions\Cygwin\mounts v2\/usr/lib (default) = `C:/cygwin/lib' flags = 0x000a HKEY_LOCAL_MACHINE\Software\Cygnus Solutions\Cygwin\Program Options a: fd N/AN/A c: hd FAT32 1017Mb 67% CPUN WIN_C d: cd UDF 4221Mb 100% UN UNDEFINED e: cd N/AN/A . /cygdrive userbinmode,noumount C:/cygwin / system binmode C:/cygwin/bin /usr/bin system binmode C:/cygwin/lib /usr/lib system binmode Found: C:\cygwin\bin\bash.exe Found: C:\cygwin\bin\cat.exe Found: C:\cygwin\bin\cpp.exe Found: C:\cygwin\bin\find.exe Found: c:\WIN\COMMAND\find.exe Warning: C:\cygwin\bin\find.exe hides c:\WIN\COMMAND\find.exe Found: C:\cygwin\bin\gcc.exe Not Found: gdb Found: C:\cygwin\bin\ld.exe Found: C:\cygwin\bin\ls.exe Found: C:\cygwin\bin\make.exe Found: C:\cygwin\bin\sh.exe 18k 2000/10/23 C:\cygwin\bin\cyggdbm.dll - os=4.0 img=1.0 sys=4.0 "cyggdbm.dll" v0.0 ts=2000/10/23 4:26 45k 2001/04/25 C:\cygwin\bin\cygform5.dll - os=4.0 img=1.0 sys=4.0 "cygform5.dll" v0.0 ts=2001/4/25 7:28 26k 2001/04/25 C:\cygwin\bin\cygmenu5.dll - os=4.0 img=1.0 sys=4.0 "cygmenu5.dll" v0.0 ts=2001/4/25 7:27 156k 2001/04/25 C:\cygwin\bin\cygncurses++5.dll - os=4.0 img=1.0 sys=4.0 "cygncurses++5.dll" v0.0 ts=2001/4/25 7:29 226k 2001/04/25 C:\cygwin\bin\cygncurses5.dll - os=4.0 img=1.0 sys=4.0 "cygncurses5.dll" v0.0 ts=2001/4/25 7:17 15k 2001/04/25 C:\cygwin\bin\cygpanel5.dll - os=4.0 img=1.0 sys=4.0 "cygpanel5.dll" v0.0 ts=2001/4/25 7:27 35k 2002/01/09 C:\cygwin\bin\cygform6.dll - os=4.0 img=1.0 sys=4.0 "cygform6.dll" v0.0 ts=2002/1/9 7:03 20k 2002/01/09 C:\cygwin\bin\cygmenu6.dll - os=4.0 img=1.0 sys=4.
Re: BUG with 1.3.9 : -mno-cygwin target still depends on cygwin1.dll
> I'm currently working on a cygwin32 port of DevIL (http://openil.sf.net). I get strange results when > trying to compile the libs with -mno-cygwin (we prefer to not depend on cygwin unix emulation as > we don't need it, we just use standard C functions) > >The problem is that i get dlls that still depend on cygwin1.dll ! but the compilation is done with > mno-cygwin flag. > > You have the entire libILU.dll compilation process in attachment (and the cygcheck output). It seems libtool is not passing the -mno-cygwin flag through when it starts linking the DLL. Regards Chris -- 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: How to create a windows link in script?
> I would like to be able to create a .lnk file to a > batch file (and ideally associate an icon with it) > from a bash script on a Windows 2000 machine. I've written a program that does just this. It's a fully scriptable and can set icons. I'm hoping to get it included in cygwin sometime, but I'm waiting until the new setup.exe comes out to bother developers with it. In the meantime, you can get it at: http://ns1.iocc.com/~joshua/mklink-0.1.tar.bz2 The archive is around 10K, source, (stripped) binary, and some examples included. I'd also love some feedback on possible improvements--or from anyone on Win9x since I've got a setup like Randall's (Win2k, NTFS, etc.) The MS docs say it the calls should work on any windows, but you know... I've not had any weird permission problems, though figuring out whether folders on the Desktop exist only for the current user or for all users can be tricky. __ Do You Yahoo!? Yahoo! Sports - Coverage of the 2002 Olympic Games http://sports.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: BUG with 1.3.9 : -mno-cygwin target still depends on cygwin1.dll
Chris January ([EMAIL PROTECTED]) wrote: > It seems libtool is not passing the -mno-cygwin flag through when it starts > linking the DLL. > > Regards > Chris > I forgot to say one important thing : Before we fixed some bugs to the code, the libtool was close to produce a cygwin independant library. The linker was failing because it didn't find some common symbols in C like : stricmp strcmp and ... But now it's not even able to produce this error :\ Btw, how can i pass the -mno-cygwin option to the libtool with the help of autoconf ? Is there a special macro (i didn't find one in the autotools book from red hat) ? Regards -- Edouard Gomez -- 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: BUG with 1.3.9 : -mno-cygwin target still depends on cygwin1.dll
On Wed, Feb 20, 2002 at 12:52:13AM +0100, Vincent GOMEZ wrote: >Hello, > >I'm currently working on a cygwin32 port of DevIL (http://openil.sf.net). Sorry, but I beg to differ. 1) It's "cygwin" not "cygwin32". 2) You're not working on a cygwin port if you are linking with -mno-cygwin. Assuming that you get this working with -mno-cygwin, you will have produced a executable which will be a straight windows program. It won't understand cygwin paths, signals, symbolic links, etc. >I get strange results when trying to compile the libs with -mno-cygwin >(we prefer to not depend on cygwin unix emulation as we don't need it, >we just use standard C functions) Sounds like you should move over to www.mingw.org, then. 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: How to create a windows link in script?
The Windows term for these things is 'shortcut'. Perhaps you want to call your tool mkshortcut? Yeah, it's longer, but it's less ambiguous than mklink. There's already enough confusion between linking executables from object modules and hard and symbolic links in the file system. Also, since the files created are .lnk files, mklink doesn't immediately make one see the connection. Perhaps mklnk would be more accurate. (And I haven't even had a chance to _look_ at the tool itself. stephan(); -Original Message- From: Joshua Franklin [mailto:[EMAIL PROTECTED]] Sent: Tuesday, February 19, 2002 4:05 PM To: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: Re: How to create a windows link in script? > I would like to be able to create a .lnk file to a > batch file (and ideally associate an icon with it) > from a bash script on a Windows 2000 machine. I've written a program that does just this. It's a fully scriptable and can set icons. I'm hoping to get it included in cygwin sometime, but I'm waiting until the new setup.exe comes out to bother developers with it. In the meantime, you can get it at: http://ns1.iocc.com/~joshua/mklink-0.1.tar.bz2 The archive is around 10K, source, (stripped) binary, and some examples included. I'd also love some feedback on possible improvements--or from anyone on Win9x since I've got a setup like Randall's (Win2k, NTFS, etc.) The MS docs say it the calls should work on any windows, but you know... I've not had any weird permission problems, though figuring out whether folders on the Desktop exist only for the current user or for all users can be tricky. __ Do You Yahoo!? Yahoo! Sports - Coverage of the 2002 Olympic Games http://sports.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/ -- 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/
clisp as a shell
I'd like to try clisp as a shell and am just fishing in the hope, that this has already been done. -- 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: BUG with 1.3.9 : -mno-cygwin target still depends on cygwin1.dll
Christopher Faylor ([EMAIL PROTECTED]) wrote: > Sorry, but I beg to differ. 1) It's "cygwin" not "cygwin32". 2) You're not > working on a cygwin port if you are linking with -mno-cygwin. > 1 - Sorry for this mistake 2 - I'm working on a cygwin port because i use the cygwin environment (shell + autoconf + automake ...) but the resulting dlls don't need cygwin unix elulation stuff. And as the doc says, if you don't need unix emulation, use -mno-cygwin to get faster programs/libs. > Assuming that you get this working with -mno-cygwin, you will have produced > a executable which will be a straight windows program. It won't understand > cygwin paths, signals, symbolic links, etc. > I don't need this stuff > > Sounds like you should move over to www.mingw.org, then. > It's truely a cygwin problem, i use the cygwin gcc compiler, the cygwin tools the cygwin whateveryoulike but it doesn't do what it is expected to do. ie a cygwin1.dll independant lib. I was making a bug report so please don't flame me without good arguments. Regards -- Edouard Gomez -- 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: BUG with 1.3.9 : -mno-cygwin target still depends on cygwin1.dll
Edouard, You're using the cygwin tools to do what is basically a cross-compile. However, libtool in particular doesn't grok that "gcc -mno-cygwin" == "cygwin hosted mingw compiler". So, there are problems. Search the archives -- this was recently discussed on the main list. There was a big "argument" about whether "gcc -mno-cygwin" was a "mode" of a cygwin-hosted, cygwin-target compiler, or whether it was an actual cross-compiler "in disguise". It sounds like a trivial distinction, I know, but it makes a BIG difference in how the autotools behave. There were suggestions of using scripts similar to the following ---begin mingw-pc-gcc #!/bin/sh gcc -mno-cygwin $* ---end mingw-pc-gcc -- and similar(for all the binutils tools). Then you'd just configure as: ./configure --target=i686-pc-mingw and (hopefully) the autotools would figure it all out. There was also a suggestion that a "real" cygwin-hosted-mingw-target gcc/binutils be supplied as a cygwin package, that could install alongside the "normal" cygwin gcc. I dunno what the status of that is. Then, there was the simple suggestion to just 'export CC="gcc -mno-cygwin"' -- but libtool doesn't accept that. automake and autoconf kinda mostly work okay in this configuration -- but libtool, definitely not. But the point is, using 'gcc -mno-cygwin' really only works if you're doing the "work" by hand: hand-written makefiles, custom build commands, etc. Given the above problems, coupled with the *total* lack of autotoolability (?) in mingw, Earnie Boyd created a fork of cygwin called "msys". The goal of msys is to provide a cygwin-like environment (bash shells, scripting, autotools "work", etc) -- but everything in the environment is targetted toward building native, non-cygwin applications using the mingw compiler suite. Take a look at the mingw website. Or lobby for "someone" to provide a mingw-target cygwin-host cross-compiler suite as an official cygwin package. --Chuck Edouard Gomez wrote: > Christopher Faylor ([EMAIL PROTECTED]) wrote: > >>Sorry, but I beg to differ. 1) It's "cygwin" not "cygwin32". 2) You're not >>working on a cygwin port if you are linking with -mno-cygwin. >> >> > > 1 - Sorry for this mistake > 2 - I'm working on a cygwin port because i use the cygwin environment > (shell + autoconf + automake ...) but the resulting dlls don't need > cygwin unix elulation stuff. And as the doc says, if you don't need > unix emulation, use -mno-cygwin to get faster programs/libs. > > >>Assuming that you get this working with -mno-cygwin, you will have produced >>a executable which will be a straight windows program. It won't understand >>cygwin paths, signals, symbolic links, etc. >> >> > > I don't need this stuff > > >>Sounds like you should move over to www.mingw.org, then. >> >> > > It's truely a cygwin problem, i use the cygwin gcc compiler, the cygwin tools > the cygwin whateveryoulike but it doesn't do what it is expected to do. > ie a cygwin1.dll independant lib. > > I was making a bug report so please don't flame me without good arguments. > > Regards > > -- 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: clisp as a shell
I think you can just add it to /etc/shells and run chsh to set it as your default shell. Why do you want to do this? Bye, Neil On 20 Feb 2002, Guy Worthington wrote: > I'd like to try clisp as a shell and am just fishing in the hope, > that this has already been done. > > > -- > 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: cygwin!
Joshua, I don't know what this is supposed to help with, because Cygwin will launch non-Cygwin programs just fine, and that includes the full suite of Java SDK tools, the JRE subset thereof and any program written in Java and requiring one of the previous two. There is exactly no advantage to using CMD.exe to initiate a Java SDK tool, or the Java interpreter of JRE. I have no idea what a Java ".exe" file might be, but if it is some sort of wrapper meant to create the illusion that a Java program is directly executable (by supplying the invocation of the Java interpreter), that too is irrelevant to whether the program is invoked from a Cygwin context (a shell or some other Cygwin-linked executable using one of the POSIX exec...() calls) or from a native Win32 program. If you want to pass Cygwin / POSIX names to a Java program or Java SDK tool, you'll have to translate them first by applying "cygpath -w" and / or "cygpath -w -p" to arguments or environment variables that will be interpreted by the Java program or the JVM (as file or directory names or as PATH-like variables). Randall Schulz Mountain View, CA USA At 12:00 2002-02-19, you wrote: >Well, a quick search of the cygwin mailing list about Java would have >unfortunately told you that Java is not "cygwin-aware". I run javac from >cmd.exe for this reason. You might try wrapping your Java program in a >script like: > >CMD.EXE /c "javaprog.exe" > >Hope it helps. > >--- "pramod m." <[EMAIL PROTECTED]> wrote: > > if i write a simple java program that uses a > > BufferedReader reader object wrapped from System.in > > and then try to read a line from user and then echo > > it back, cygwin for some reason already has stuff on > > the stream and so without waiting for the user to > > type something in the program fetches that and > > prints it out (it is usually) just a newline. why > > does this happen! > > > > i know it's not the program because it works in > > djgpp and other platforms. -- 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/
Avail for test: gdbm-1.8.0-4
TEST RELEASE. PLEASE test to see that this DLL doesn't break your existing gdbm-dependent applications -- like CVS. It seems okay to me, with my limited tests, but... Note that merely updating cyggdbm to this new version will NOT magically enable CVS to host repositories on text mounts; nor will it magically fix CVS's existing problems with CR/LF. This gdbm update may fix the gdbm database files within the CVSROOT repository, but CVS itself is still not text/binary clean. Workin' on it... Also, this new DLL is built using auto-import. It seems to work okay with existing binaries that I've tried -- like cvs.exe (e.g. I've seen no "symbol __foo can't be found in cyggdbm.dll" errors). If you find an existing binary that worked with the OLD cyggdbm.dll but doesn't work with this new one, let me know immediately. - The gdbm package has been updated to version 1.8.0-4. It is based on the official GNU Database Manager distribution from the FSF, gdbm-1.8.0.tar.gz. gdbm is a prerequisite for the cvs package. CHANGES: (vs. 1.8.0-3) 1. Internal changes to facilitate easier building/maintainance 2. gdbm is now built using the 'auto-import' features of recent binutils. 3. A few fixes so that gdbm database files may be hosted on text-mode mounts. (AJ Reins) 4. Fix gdbm.texinfo so that "install-info gdbm.info" doesn't issue warnings. (AJ Reins) Charles Wilson gdbm volunteer maintainer for cygwin INSTALLATION: To update your installation, click on the "Install Cygwin now" link on the http://sources.redhat.com/cygwin web page. This downloads setup.exe to your system. Run setup and answer all of the questions. You must choose a mirror, since direct downloads from 'cygwin.com' are no longer allowed. ftp://mirrors.rcn.net/pub/sourceware/cygwin/ (US) ftp://linux.sarang.net/mirror/development/compiler/cygwin/ (Korea) ftp://ftp.mirror.ac.uk/sites/sourceware.cygnus.com/pub/cygwin (UK) The setup.exe program will figure out what needs to be updated on your system and should install the gdbm package automatically. To install this TEST version of gdbm you'll have to select the experimental radio button in setup. If you have questions or comments, please send them to the Cygwin mailing list at: [EMAIL PROTECTED] . I would appreciate if you would use this mailing list rather than emailing me directly. This includes ideas and comments about the setup utility or Cygwin general. If you want to make a point or ask a question the Cygwin mailing list is the appropriate place. *** CYGWIN-ANNOUNCE UNSUBSCRIBE INFO *** If you want to unsubscribe to the cygwin-announce mailing list, look at the "List-Unsubscribe: " tag in the email header of this message. Send email to the address specified there. It will be in the format: [EMAIL PROTECTED] NOTES: o Now uses the auto-import functionality of newer binutils, and doesn't use __declspec(dllimport). This means you no longer need "-DGDBM_STATIC" or "-DALL_STATIC" when compiling objects intended for static linking. Just compile as normal. HOWEVER, you need to use a special flags when linking statically: 'gcc -static'. For dynamic linking, you need no special link-time flags (assuming you're using binutils newer than 20011002, when --enable-auto-import was made the default). -- PRO: no compile time flags needed when building client programs; ONLY need a link-time flag linking to static libraries. NO special flags at compile-time nor link-time when linking to dynamic links. -- CON: (partial): if using binutils older than 20010930, you now need a special linktime flag for dynamic linking (-Wl,--enable-auto-import). However, with an up-to-date binutils, you don't need this. o The package includes several test and conversion utilities which are useful for evaluating gdbm, and converting older ndbm and dbm databases to the gdbm format. These utilities are dynamically linked, although the build process will make both dynamically and statically linked versions. o The package also includes compatibility headers, so that gdbm can be used as a drop-in replacement for ndbm and dbm. o Building the package from source requires gcc-2.95.3-5 and binutils-20011002-1 or later o If you are building a package that depends on libgdbm, and you wish to link *statically*, just use 'gcc -static' when linking your package. There no need to -Ddefine anything special when compiling your object files. o No special -Ddefine options or link options are needed when building a package that depends on libgdbm and you want to link dynamically (recommended). o The following packages have been converted from old style "-DFOO_STATIC" to the new autoimport style of DLL structure: ncurses readline gettext zlib tiff libpngxpm-nox jpegjbig gdbm The following have not (yet) been converted, and still need -DFOO_STAT
Re: cygwin!
DISCLAIMER: I am not a professional java developer >There is exactly no advantage to using CMD.exe to >initiate a Java SDK tool, or the Java interpreter of >JRE. I was under the impression that interacting with java's BufferedReader, etc. from within a cygwin bash shell was a Bad Idea because of different tty handling or something. I casually read the thread a while back, and I thought I'd run into the problem before I switched to java on linux. >I have no idea what a Java ".exe" file might be Me either. :) That was my fault, it was a quick suggestion to someone who emailed me off-list with a cygwin java question. (Obviously not a bright idea.) I should have said something like: cmd /c "java javaprog.java" but I don't even know if that would work since I've only got a JDK on a linux machine right now. > >--- "pramod m." <[EMAIL PROTECTED]> wrote: > > if i write a simple java program that uses a > > BufferedReader reader object wrapped from System.in > > and then try to read a line from user and then echo > > it back, cygwin for some reason already has stuff on > > the stream and so without waiting for the user to > > type something in the program fetches that and > > prints it out (it is usually) just a newline. why > > does this happen! > > > > i know it's not the program because it works in > > djgpp and other platforms. __ Do You Yahoo!? Yahoo! Sports - Coverage of the 2002 Olympic Games http://sports.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: How to create a windows link in script?
> The Windows term for these things is 'shortcut'. > Perhaps you want to call your tool mkshortcut? > Yeah, it's longer, but it's less ambiguous than > mklink. There's already enough confusion between > linking executables from object modules and hard and > symbolic links in the file system. Also, since the > files created are .lnk files, mklink doesn't > immediately make one see the connection. Perhaps > mklnk would be more accurate. > > (And I haven't even had a chance to _look_ at the > tool itself. Well, they're called shortcuts...or (in the MS documentation I read) "OLE Links". Oops. "mkshortcut" is a good idea, though, much more to the point of the app. The name came from a function call and .c file in cygwin's setup.exe source. In the archive the executable is actually prog.exe--easier to TAB-complete--because I wasn't sure of the name myself. I guess you noticed it was version 0.1 anyway... __ Do You Yahoo!? Yahoo! Sports - Coverage of the 2002 Olympic Games http://sports.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/
