Re: For package maintainers: dlls which use fdopen
Christopher Faylor wrote: cygXpm-X4.dll cygXpm-noX4.dll cygbz2-1.dll cygz.dll done. cygbz21.0.dll this is a compatibility library. It won't be re-compiled; no new apps should be using it. cygncurses7.dll stay tuned. -- Chuck
Re: [ITP] wtf
On Mon, Jul 28, 2003 at 08:17:11PM -0400, Igor Pechtchanski wrote: As per John Morrison's suggestion, I would like to contribute and maintain wtf (http://cronus.comp.utas.edu.au/~thsutton/computing/wtf.html). wtf(6) is a utility provided by some UNIX and UNIX-like systems including Slackware Linux and NetBSD. It translates acronyms and filename suffixes by looking up the definition of a term in various databases. I've added the OLOCA to the database directory (named %OLOCA to place it first in search order). There's also a perl script that can update that file from the web version of the OLOCA (also requires wget), but since this rule is totally optional (and won't be invoked in the default build), I didn't put perl and wget as prerequisites for building wtf. This is mentioned in wtf.README as well. http://cs.nyu.edu/~pechtcha/cygwin/wtf/wtf-0.0.3-1.tar.bz2 http://cs.nyu.edu/~pechtcha/cygwin/wtf/wtf-0.0.3-1-src.tar.bz2 http://cs.nyu.edu/~pechtcha/cygwin/wtf/setup.hint (also inline below) sdesc: translates acronyms and filename suffixes for you ldesc: wtf translates acronyms and filename suffixes for you. The wtf program looks-up the definition of a term. It supports a number of definition sources. In this version they are an acronyms database and qa filename suffixes database. requires: cygwin category: Text Doc This has my vote. -- +---+ | Marcel Telka e-mail: [EMAIL PROTECTED] | |homepage: http://telka.sk/ | |jabber: [EMAIL PROTECTED] | +---+
Re: Proposal: psutils (again)
Hi Daniel,
I got my source archive from a server in France, but I just noticed,
that the author's original site is up again:
http://knackered.knackered.org/angus/psutils/
Here you can download the source, and it does *not* contain the
Makefile. You probably have the Debian patched version. Maybe the
Debian-patch cannot be reversed :)
Defintely, sorry for the fuzz.
You package looks ready, but I found two last nit. When using the
$(prefix) variable, you also have to change the following lines:
original:
-MANDIR = /usr/local/share/man/man$(MANEXT)
current:
+MANDIR = /usr/man/man$(MANEXT)
better:
+MANDIR = ${prefix}/man/man$(MANEXT)
and:
current:
+ cp CYGWIN-PATCHES/psutils-1.17.README /usr/doc/Cygwin
better:
+ cp CYGWIN-PATCHES/psutils-1.17.README ${prefix}/doc/Cygwin/
Anyway, with both changes the package is ready to be uploaded.
Volker
--
PGP/GPG key (ID: 0x9F8A785D) available from wwwkeys.de.pgp.net
key-fingerprint 550D F17E B082 A3E9 F913 9E53 3D35 C9BA 9F8A 785D
pgp0.pgp
Description: PGP signature
RE: [ITP] wtf
Igor Pechtchanski wrote: As per John Morrison's suggestion, I would like to contribute and maintain wtf (http://cronus.comp.utas.edu.au/~thsutton/computing/wtf.html). wtf(6) is a utility provided by some UNIX and UNIX-like systems including Slackware Linux and NetBSD. It translates acronyms and filename suffixes by looking up the definition of a term in various databases. I've added the OLOCA to the database directory (named %OLOCA to place it first in search order). There's also a perl script that can update that file from the web version of the OLOCA (also requires wget), but since this rule is totally optional (and won't be invoked in the default build), I didn't put perl and wget as prerequisites for building wtf. This is mentioned in wtf.README as well. http://cs.nyu.edu/~pechtcha/cygwin/wtf/wtf-0.0.3-1.tar.bz2 http://cs.nyu.edu/~pechtcha/cygwin/wtf/wtf-0.0.3-1-src.tar.bz2 http://cs.nyu.edu/~pechtcha/cygwin/wtf/setup.hint (also inline below) sdesc: translates acronyms and filename suffixes for you ldesc: wtf translates acronyms and filename suffixes for you. The wtf program looks-up the definition of a term. It supports a number of definition sources. In this version they are an acronyms database and qa filename suffixes database. requires: cygwin category: Text Doc Igor Has my vote :) J. === Information in this email and any attachments are confidential, and may not be copied or used by anyone other than the addressee, nor disclosed to any third party without our permission. There is no intention to create any legally binding contract or other commitment through the use of this email. Experian Limited (registration number 653331). Registered office: Talbot House, Talbot Street, Nottingham NG1 5HF
RE: For package maintainers: dlls which use fdopen
On Mon, 28 Jul 2003, Robert McNulty Junior wrote: I thought Cygwin1.dll was the main dll. Should it not have already been rebuilt? I'll check out the sources tonight and look into this. Could have sworn you built the 1.5.1 cygwin1.dll. Robert McNulty Junior It has been re-built and Chris is the one doing the re-building (of Cygwin). Elfyn -- Elfyn McBratney, EMCB | http://www.nongnu.org/wwwauth/ http://www.emcb.co.uk | http://www.emcb.co.uk/webauth/ [EMAIL PROTECTED] | wwwauth-users AT nongnu DOT org
Re: [setup PATCH] next_dialog micropatch (2)
On Tue, 2003-07-29 at 06:50, Max Bowsher wrote: You realise that *all* I am doing with this patch is to change how data is passed between a function and it's caller? Yes, and its that aspect I objected to. I don't understand what was so unclear. Rob -- GPG key available at: http://members.aardvark.net.au/lifeless/keys.txt. --- signature.asc Description: This is a digitally signed message part
Re: Er I mean... (was: RE: [PATCH] next_dialog micropatch (3)(was: RE: [setup PATCH] next_dialog micropatch (2)))
On Tue, 2003-07-29 at 13:47, Gary R. Van Sickle wrote:
...this straightforward patch:
2003-07-28 Gary R. Van Sickle [EMAIL PROTECTED]
* dialog.h (do_fromcwd): Change function declaration.
* fromcwd.cc (do_fromcwd): Change return type to bool. Eliminate
use of next_dialog, return true or false instead.
* localdir.cc (LocalDirPage::OnNext): Use do_fromcwd()'s return
value instead of next_dialog.
Approved, subject to one caveat:
instead of
bool found_ini;
found_ini = do_fromcwd ()..
if (found_ini)
please use
if (do_fromcwd(...)) {
unneeded temporary variables are annoying.
Cheers,
Rob
--
GPG key available at: http://members.aardvark.net.au/lifeless/keys.txt.
---
signature.asc
Description: This is a digitally signed message part
RE: [setup PATCH] next_dialog micropatch (2)
On Tue, 2003-07-29 at 11:11, Gary R. Van Sickle wrote: Seems like a lot of work to go to just to eliminate a global variable. Getting the direction right gives huge long term benefits. This is precisely why I want small patches, as it lets me evaluate each chunk. Note the commits I got done over the weekend: each small and relatively isolated - even when more work simply cried out to be done immediately afterwards. The only complex patch was the one shifing PickView into the Window hierarchy, and that was as small as I could make it. (note too: The same discipline let me quickly identify and fix the problem Max reported - it was 5 minutes work once I returned home.) Rob -- GPG key available at: http://members.aardvark.net.au/lifeless/keys.txt. --- signature.asc Description: This is a digitally signed message part
Re: Proposal: psutils (again)
Hi Volker,
You are right, I forgot these lines.
What's next? Will someone upload the package?
Bye,
Daniel
Volker Quetschke wrote:
Hi Daniel,
I got my source archive from a server in France, but I just noticed,
that the author's original site is up again:
http://knackered.knackered.org/angus/psutils/
Here you can download the source, and it does *not* contain the
Makefile. You probably have the Debian patched version. Maybe the
Debian-patch cannot be reversed :)
Defintely, sorry for the fuzz.
You package looks ready, but I found two last nit. When using the
$(prefix) variable, you also have to change the following lines:
original:
-MANDIR = /usr/local/share/man/man$(MANEXT)
current:
+MANDIR = /usr/man/man$(MANEXT)
better:
+MANDIR = ${prefix}/man/man$(MANEXT)
and:
current:
+cp CYGWIN-PATCHES/psutils-1.17.README /usr/doc/Cygwin
better:
+cp CYGWIN-PATCHES/psutils-1.17.README ${prefix}/doc/Cygwin/
Anyway, with both changes the package is ready to be uploaded.
Volker
Re: [setup PATCH] next_dialog micropatch (2)
Gary R. Van Sickle wrote: On Mon, 28 Jul 2003, Gary R. Van Sickle wrote: Robert Collins wrote: On Tue, 2003-07-29 at 02:13, Max Bowsher wrote: If we do add a class, it should probably be a thread class from which all of our threaded tasks can derive. Regardless, I don't see that any further cleanup will reverse these changes. I can see how it may involve changing the same lines to a third form, but that is not a reversal. Are you thinking of the (IIR the name C) the 'completion object' pattern? That would work too. I have no idea what this means. It's Gang Of Four-speak, though I'm not sure this is actually original Gang Of Four, i.e. Design Patterns: Elements Of Reusable Object-Oriented Software by four guys and a foreward by Grady Booch. I don't recall running across it in there. I think Rob might have meant the Asynchronous Completion Token pattern: http://www.cs.wustl.edu/~schmidt/PDF/ACT.pdf (first Google match). This is definitely post-GOF. Seems like a lot of work to go to just to eliminate a global variable. Thankyou! It seems appropriate to point out that I have never read any design patterns books. Max.
Re: [setup PATCH] next_dialog micropatch (2)
Robert Collins wrote:
On Tue, 2003-07-29 at 06:50, Max Bowsher wrote:
You realise that *all* I am doing with this patch is to change how data
is
passed between a function and it's caller?
Yes, and its that aspect I objected to. I don't understand what was so
unclear.
I'm unclear on why you objected and what you want me to do instead.
Please bear in mind that I have never read any Design Patterns books, and
have no idea what an Asynchronous Completion Token is.
I'm also extremely puzzled, since you seem to be objecting to the use of a
return value to communicate from a callee to a caller function. Any other
method *must* be more complicated, and I do believe in avoiding unnecessary
complication.
Just in case I'm not explaining myself well enough in words:
int foo()
{
return 8;
}
int main()
{
printf(%u, foo());
return 0;
}
In the above, you wouldn't want me to make foo void, and find another way to
pass the value 8 back to main would you?
Max.
Re: Er I mean... (was: RE: [PATCH] next_dialog micropatch (3)(was: RE: [setup PATCH] next_dialog micropatch (2)))
Robert Collins wrote:
On Tue, 2003-07-29 at 13:47, Gary R. Van Sickle wrote:
...this straightforward patch:
2003-07-28 Gary R. Van Sickle [EMAIL PROTECTED]
* dialog.h (do_fromcwd): Change function declaration.
* fromcwd.cc (do_fromcwd): Change return type to bool. Eliminate
use of next_dialog, return true or false instead.
* localdir.cc (LocalDirPage::OnNext): Use do_fromcwd()'s return
value instead of next_dialog.
Approved, subject to one caveat:
instead of
bool found_ini;
found_ini = do_fromcwd ()..
if (found_ini)
please use
if (do_fromcwd(...)) {
unneeded temporary variables are annoying.
Committed, with the above change.
Max.
Re: [setup PATCH] next_dialog micropatch (2)
On Tue, 2003-07-29 at 19:32, Max Bowsher wrote:
Robert Collins wrote:
On Tue, 2003-07-29 at 06:50, Max Bowsher wrote:
You realise that *all* I am doing with this patch is to change how data
is
passed between a function and it's caller?
Yes, and its that aspect I objected to. I don't understand what was so
unclear.
I'm unclear on why you objected and what you want me to do instead.
Please bear in mind that I have never read any Design Patterns books, and
have no idea what an Asynchronous Completion Token is.
Thats ok - I thought you where referring to the ACT pattern is all.
I'm also extremely puzzled, since you seem to be objecting to the use of a
return value to communicate from a callee to a caller function. Any other
method *must* be more complicated, and I do believe in avoiding unnecessary
complication.
*must* is not strictly true here.
The thing I was objecting to was the use of the return value to return
program flow decisions, as opposed to the information needed to make the
decision.
Just in case I'm not explaining myself well enough in words:
int foo()
{
return 8;
}
int main()
{
printf(%u, foo());
return 0;
}
In the above, you wouldn't want me to make foo void, and find another way to
pass the value 8 back to main would you?
No, but it's not analogous:
* foo here is constant, it doesn't perform work, there is no decisions
to be made
* there is no current global to be 'fixed'.
A more useful analogy:
bool showUser = false;
int profit = 0;
void
getProfit()
{
/* do some outside 'stuff' */
profit = /* more outside stuff*/;
if (profit 45)
showUser = true;
}
int main()
{
getProfit()
if (showUser)
printf(profit: %d\n, profit);
exit 0;
}
here, I would not put the boolean into the return value of getProfit.
(Functions that perform an action (mutators) should not return a value.)
I would:
bool showUser = false;
class Report {
public:
Report() : profit(0) {}
void calculateProfit() {
/* do the stuff */
profit = /* ... */;
if (profit 45)
showUser = true;
}
int getProfit() { return profit;}
private:
int profit;
};
int main()
{
Report aReport;
aReport.calculateProfit();
if (showUser)
printf(profit: %d\n, aReport.getProfit());
exit 0;
}
as a first step. I'd probably then move the logic out of calculateProfit
into a 'bool showReport()' that does nothing but
bool showReport() {
return getProfit() 45;
}
The analogy here is that moving the direction for the windows to go into
the return code, increases the coupling to the UI from the underlying
logic, and would be reversed as soon as a class is introduced to remove
a mutating method with a return code.
Cheers,
Rob
--
GPG key available at: http://members.aardvark.net.au/lifeless/keys.txt.
---
signature.asc
Description: This is a digitally signed message part
gcj file input problem, available incorrect
Hi,
while using the gcj component of gcc (java compiler frontemd for gcc)
we have found a peculiar problem with FileInputStreams on Windows
(XP).
Given Freader.java:
import java.io.*;
public class Freader {
static public void main(String arg[]) {
try {
File f=new File(inp.txt);
FileInputStream fis=new FileInputStream(f);
int inp;
int count=0;
while ((inp=fis.read())!=-1) {
count++;
}
fis.close();
System.out.println(#1 + count);
fis=new FileInputStream(f);
count=0;
fis.available();
while ((inp=fis.read())!=-1) {
count++;
}
fis.close();
System.out.println(#2 + count);
} catch (Exception e) {
e.printStackTrace();
}
}
}
File inp.txt:
weapons of mass destruction
that should report the same number of bytes read from the file
inp.txt in both cases. And does so using javac/java on both Linux
and Windows.
With gcj (v 3.2) on Linux the program works fine.
On Windows, however, the call to 'available' seem to consume input,
see output below.
Is this a known issue with gcc/gcj on Windows?
Output on windows:
a.exe
#1 31
#2 30
$ gcj --version
gcj (GCC) 3.2 20020927 (prerelease)
Copyright (C) 2002 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
The installation is from cygwin/redhat appears to be up-to-date.
Regards,
--
/jørgen nørgaard
SmartTV A/S
[EMAIL PROTECTED]
Phone: +45 3332 5770
http://smarttv.dk/
Re: [setup PATCH] next_dialog micropatch (2)
Robert Collins wrote: On Tue, 2003-07-29 at 19:32, Max Bowsher wrote: I'm also extremely puzzled, since you seem to be objecting to the use of a return value to communicate from a callee to a caller function. Any other method *must* be more complicated, and I do believe in avoiding unnecessary complication. *must* is not strictly true here. The thing I was objecting to was the use of the return value to return program flow decisions, as opposed to the information needed to make the decision. Ah. I can very easily understand the above objection, but do not believe it applies to my patch. My point is, that currently, program flow decisions are returned in a global variable. I'm trying to change it so that program flow decisions are returned in a return value. I.e., I'm working on the global variable ugliness, not the returns program flow decisions ugliness. The fact that I am solving the former should not compel me to solve the latter simultaneously. cheekyAfter all, patches should be modular, and as small as possible! :-)/cheeky Hence my confusion about the rejection of this patch. I'm happy to try and tackle the returns program flow decisions issue (in fact, I already have planned the changes to ini.cc), but I consider this to be a separate issue, meriting a separate patch. A more useful analogy: Studied. (Hopefully) memorized. Max.
Re: [setup PATCH] next_dialog micropatch (2)
On Tue, 2003-07-29 at 20:13, Max Bowsher wrote: I.e., I'm working on the global variable ugliness, not the returns program flow decisions ugliness. The fact that I am solving the former should not compel me to solve the latter simultaneously. cheekyAfter all, patches should be modular, and as small as possible! :-)/cheeky Yes they should be... but they should also all move in the same direction, not simply shuffle the work around. As you can see from the example, it is possible with only a touch of extra work to cleanup the globals without making more work to be tackled later. Anyway, this one is done to death.. next! Rob -- GPG key available at: http://members.aardvark.net.au/lifeless/keys.txt. --- signature.asc Description: This is a digitally signed message part
Re: [setup PATCH] OnAcceptActivation
On Tue, 2003-07-29 at 08:36, Max Bowsher wrote: Here at last... OnAcceptActivation! Extracted from the bulk of Gary's changes, bugfixed (the DWL_MSGRESULT logic was inverted), and also removing the bumper code in SplashPage and SourcePage which was substituting for this functionality. Actually, I'd like to propose the alternate name WantActivation - slightly more intuitive? if you use... bool wantsActivation() const; it's approved. Cheers, Rob -- GPG key available at: http://members.aardvark.net.au/lifeless/keys.txt. signature.asc Description: This is a digitally signed message part
Please upload: docbook-xsl-1.61.3-2
Hi. Please upload updated docbook-xsl packages: http://telka.sk/cygwin/docbook-xsl/docbook-xsl-1.61.3-2-src.tar.bz2 http://telka.sk/cygwin/docbook-xsl/docbook-xsl-1.61.3-2.tar.bz2 and remove obsolete 1.61.2-1 files. Thanks. -- +---+ | Marcel Telka e-mail: [EMAIL PROTECTED] | |homepage: http://telka.sk/ | |jabber: [EMAIL PROTECTED] | +---+
[setup PATCH] Rework do_ini [next_dialog removal (2a)]
Let's see if this approach is more satisfactory...
+2003-07-29 Max Bowsher [EMAIL PROTECTED]
+
+ * ini.cc (do_ini_thread): Return bool.
+ (do_ini_thread_reflector): Put returned bool in lParam.
+ * threebar.cc (WM_APP_SETUP_INI_DOWNLOAD_COMPLETE): Rework for new
+ meaning of lParam.
+
Index: ini.cc
===
RCS file: /home/max/cvsmirror/cygwin-apps-cvs/setup/ini.cc,v
retrieving revision 2.35
diff -u -p -r2.35 ini.cc
--- ini.cc 5 Jul 2002 05:17:21 - 2.35
+++ ini.cc 29 Jul 2003 13:17:34 -
@@ -203,7 +203,7 @@ do_remote_ini (HWND owner)
return ini_count;
}
-static void
+static bool
do_ini_thread (HINSTANCE h, HWND owner)
{
size_t ini_count = 0;
@@ -213,10 +213,7 @@ do_ini_thread (HINSTANCE h, HWND owner)
ini_count = do_remote_ini (owner);
if (ini_count == 0)
-{
- next_dialog = source == IDC_SOURCE_CWD ? IDD_S_FROM_CWD : IDD_SITE;
- return;
-}
+return false;
if (get_root_dir ().cstr_oneuse())
{
@@ -266,7 +263,7 @@ do_ini_thread (HINSTANCE h, HWND owner)
note (owner, IDS_OLD_SETUP_VERSION, version, setup_version.cstr_oneuse());
}
- next_dialog = IDD_CHOOSE;
+ return true;
}
static DWORD WINAPI
@@ -275,10 +272,10 @@ do_ini_thread_reflector(void* p)
HANDLE *context;
context = (HANDLE*)p;
- do_ini_thread((HINSTANCE)context[0], (HWND)context[1]);
+ bool succeeded = do_ini_thread((HINSTANCE)context[0], (HWND)context[1]);
// Tell the progress page that we're done downloading
- Progress.PostMessage(WM_APP_SETUP_INI_DOWNLOAD_COMPLETE, 0, next_dialog);
+ Progress.PostMessage(WM_APP_SETUP_INI_DOWNLOAD_COMPLETE, 0, succeeded);
ExitThread(0);
}
Index: threebar.cc
===
RCS file: /home/max/cvsmirror/cygwin-apps-cvs/setup/threebar.cc,v
retrieving revision 2.5
diff -u -p -r2.5 threebar.cc
--- threebar.cc 25 Mar 2003 20:57:13 - 2.5
+++ threebar.cc 29 Jul 2003 13:33:23 -
@@ -206,14 +206,23 @@ bool
}
case WM_APP_SETUP_INI_DOWNLOAD_COMPLETE:
{
- if (lParam == IDD_S_FROM_CWD)
+ if (lParam)
{
- // There isn't actually a dialog template named this
- do_fromcwd (GetInstance (), GetHWND ());
+ GetOwner ()-SetActivePageByID (lParam);
}
else
{
- GetOwner ()-SetActivePageByID (lParam);
+ if (source == IDC_SOURCE_CWD)
+ {
+ // There was a setup.ini file (as found by do_fromcwd),
+ // but it had vanished by the time we came to read it.
+ exit(100);
+ }
+ else
+ {
+ // Download failed, try another site.
+ GetOwner ()-SetActivePageByID (IDD_SITE);
+ }
}
break;
}
Re: Proposal: psutils (again)
On Tue, 29 Jul 2003, Daniel Bößwetter wrote: Hi Volker, You are right, I forgot these lines. What's next? Will someone upload the package? Not yet, I'm afraid. The binaries in usr/bin (in the package) are not stripped. Please strip them. Also, you have 'TCM Port' at the end of your README. Please either remove or change that. Elfyn -- Elfyn McBratney, EMCB | http://www.nongnu.org/wwwauth/ http://www.emcb.co.uk | http://www.emcb.co.uk/webauth/ [EMAIL PROTECTED] | wwwauth-users AT nongnu DOT org
Re: [setup PATCH] OnAcceptActivation
Gary R. Van Sickle wrote: and also removing the bumper code in SplashPage and SourcePage which was substituting for this functionality. DOH! How the heck did I miss those?!?! Easily, apparently, since I only just noticed the similar arrangements for skipping RootPage in a download-only situation. I've added those changes to my commit. Max.
Re: Please upload: docbook-xsl-1.61.3-2
On Tue, 29 Jul 2003, Marcel Telka wrote: Hi. Please upload updated docbook-xsl packages: http://telka.sk/cygwin/docbook-xsl/docbook-xsl-1.61.3-2-src.tar.bz2 http://telka.sk/cygwin/docbook-xsl/docbook-xsl-1.61.3-2.tar.bz2 Uploaded. Please send an announcement in a couple of hours. Thanks, Elfyn -- Elfyn McBratney, EMCB | http://www.nongnu.org/wwwauth/ http://www.emcb.co.uk | http://www.emcb.co.uk/webauth/ [EMAIL PROTECTED] | wwwauth-users AT nongnu DOT org
Re: TEST: patch-2.5.8-5
Christopher Faylor writes: I've rebuilt Corinna's version of patch using 1.5.1 import libraries so it should now work correctly with a 1.5.1 or greater DLL. It still doesn't work for me. I get $ date | patch --dry-run patch: write error : Illegal seek It works if I recompile from sources, though. I compiled with $ ./configure --prefix=/usr make strip patch.exe \ make install I tried this on a recently installed Cygwin with only the Cygwin DLL at the Test level. Dave
Re: Proposal: psutils (again)
Hi Elfyn, Elfyn McBratney wrote: On Tue, 29 Jul 2003, Daniel Bößwetter wrote: Hi Volker, You are right, I forgot these lines. What's next? Will someone upload the package? Not yet, I'm afraid. The binaries in usr/bin (in the package) are not stripped. Please strip them. Also, you have 'TCM Port' at the end of your README. Please either remove or change that. Cough, just force of habit. I changed the README and added a strip-command to the Makefile. Source- and binary packages and patch are updated at http://home.in.tum.de/~boesswet/psutils_cygwin/. Regards, Daniel Elfyn
Re: gcj file input problem, available incorrect
Wrong list, redirecting. Please use the cygwin at cygwin dot com list
for all further discussion.
Igor
On Tue, 29 Jul 2003, Jørgen Nørgaard wrote:
Hi,
while using the gcj component of gcc (java compiler frontemd for gcc)
we have found a peculiar problem with FileInputStreams on Windows
(XP).
Given Freader.java:
import java.io.*;
public class Freader {
static public void main(String arg[]) {
try {
File f=new File(inp.txt);
FileInputStream fis=new FileInputStream(f);
int inp;
int count=0;
while ((inp=fis.read())!=-1) {
count++;
}
fis.close();
System.out.println(#1 + count);
fis=new FileInputStream(f);
count=0;
fis.available();
while ((inp=fis.read())!=-1) {
count++;
}
fis.close();
System.out.println(#2 + count);
} catch (Exception e) {
e.printStackTrace();
}
}
}
File inp.txt:
weapons of mass destruction
that should report the same number of bytes read from the file
inp.txt in both cases. And does so using javac/java on both Linux
and Windows.
With gcj (v 3.2) on Linux the program works fine.
On Windows, however, the call to 'available' seem to consume input,
see output below.
Is this a known issue with gcc/gcj on Windows?
Output on windows:
a.exe
#1 31
#2 30
$ gcj --version
gcj (GCC) 3.2 20020927 (prerelease)
Copyright (C) 2002 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
The installation is from cygwin/redhat appears to be up-to-date.
Regards,
--
http://cs.nyu.edu/~pechtcha/
|\ _,,,---,,_[EMAIL PROTECTED]
ZZZzz /,`.-'`'-. ;-;;,_[EMAIL PROTECTED]
|,4- ) )-,_. ,\ ( `'-' Igor Pechtchanski, Ph.D.
'---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow!
I have since come to realize that being between your mentor and his route
to the bathroom is a major career booster. -- Patrick Naughton
Re: Proposal: psutils (again)
On Tue, 29 Jul 2003, Daniel Bößwetter wrote: Hi Elfyn, Elfyn McBratney wrote: On Tue, 29 Jul 2003, Daniel Bößwetter wrote: Hi Volker, You are right, I forgot these lines. What's next? Will someone upload the package? Not yet, I'm afraid. The binaries in usr/bin (in the package) are not stripped. Please strip them. Also, you have 'TCM Port' at the end of your README. Please either remove or change that. Cough, just force of habit. I changed the README and added a strip-command to the Makefile. Source- and binary packages and patch are updated at http://home.in.tum.de/~boesswet/psutils_cygwin/. Uploaded. Please send an announcement in a couple of hours. Thanks, Elfyn -- Elfyn McBratney, EMCB | http://www.nongnu.org/wwwauth/ http://www.emcb.co.uk | http://www.emcb.co.uk/webauth/ [EMAIL PROTECTED] | wwwauth-users AT nongnu DOT org
XFree86 Build for 1.5.0/1
I have never subscribed to this mailing list before, but it looks like this is where the real business of Cygwin takes place. I started reading the archives of the 1.5.0/1 discussion today and realized that some other people might be waiting for me to release an updated build of XFree86 for 1.5.0/1. In any case, if I don't start working out the kinks of such a rebuild I will probably never get it right. Just wanted to drop a heads-up here that I will be starting work on this shortly. I am now subscribed to the list, so you can reach me here or on cygwin-xfree with questions/suggestions. Harold
Re: XFree86 Build for 1.5.0/1
On Tue, 29 Jul 2003, Harold L Hunt II wrote: I have never subscribed to this mailing list before, but it looks like this is where the real business of Cygwin takes place. I started reading the archives of the 1.5.0/1 discussion today and realized that some other people might be waiting for me to release an updated build of XFree86 for 1.5.0/1. In any case, if I don't start working out the kinks of such a rebuild I will probably never get it right. In a way it is, only you miss the mean'ness on this side of Cygwin land :-) This list (cygwin-apps) is where we're coordinating the move to Cygwin 1.5, release wise. If you could drop us a note when you update the XFree86 packages, no pressure intended, that would be great. Just wanted to drop a heads-up here that I will be starting work on this shortly. I am now subscribed to the list, so you can reach me here or on cygwin-xfree with questions/suggestions. Thanks for the heads up, Harold. Elfyn -- Elfyn McBratney, EMCB | http://www.nongnu.org/wwwauth/ http://www.emcb.co.uk | http://www.emcb.co.uk/webauth/ [EMAIL PROTECTED] | wwwauth-users AT nongnu DOT org
tetex-bin-2.0.2 for 1.3 and 1.5.1
I've built tetex-bin for 1.3 and 1.5.1. Previous tetex-bin releases must be removed. Applications that link to libkpathsea3 must be rebuilt. Jan. http://lilypond.org/cygwin/uploads/tetex/tetex-bin/setup.hint sdesc: The TeX text formatting system (binaries). test: 2.0.2-11 curr: 2.0.2-2 category: Text Publishing # There's no need to require a texmf tree, although not doing so # might be confusing. requires: bash clear cygutils cygwin ed diffutils fileutils findutils gawk grep libjpeg6b libjpeg62 libkpathsea3abi13 libkpathsea3 libncurses6 libncurses7 libpng12 libtiff3 libtiff4 perl sed sh-utils textutils termcap zlib #external-source: tetex-bin #suggests: tetex-tiny | tetex-base, tetex-extra, tetex-x11 ldesc: teTeX binary files This is teTeX, a TeX distribution for UNIX compatible systems. These are all the binaries for the teTeX system. You need at least tetex-tiny or tetex-base together with it. http://lilypond.org/cygwin/uploads/tetex/tetex-bin/tetex-bin-2.0.2-11-src.tar.bz2 http://lilypond.org/cygwin/uploads/tetex/tetex-bin/tetex-bin-2.0.2-11.tar.bz2 http://lilypond.org/cygwin/uploads/tetex/tetex-bin/tetex-bin-2.0.2-2-src.tar.bz2 http://lilypond.org/cygwin/uploads/tetex/tetex-bin/tetex-bin-2.0.2-2.tar.bz2 === http://lilypond.org/cygwin/uploads/tetex/tetex-devel/setup.hint sdesc: The TeX text formatting system (headers and static libraries). test: 2.0.2-11 curr: 2.0.2-2 category: Devel Libs requires: cygwin libkpathsea3abi13 libkpathsea3 external-source: tetex-bin ldesc: The TeX text formatting system (headers and static libraries) The teTeX development headers and kpathsea static library. http://lilypond.org/cygwin/uploads/tetex/tetex-devel/tetex-devel-2.0.2-11.tar.bz2 http://lilypond.org/cygwin/uploads/tetex/tetex-devel/tetex-devel-2.0.2-2.tar.bz2 === http://lilypond.org/cygwin/uploads/tetex/tetex-x11/setup.hint sdesc: The TeX text formatting system (X11 binaries). test: 2.0.2-11 curr: 2.0.2-2 category: Text Publishing # There's no need to require a texmf tree, although not doing so # might be confusing. # Also, xdvi and mfw require XFree86, obviously, but we don't want # to force a dependency like Debian does. requires: cygwin tetex-bin tetex-tiny XFree86-base ghostscript-x11 external-source: tetex-bin ldesc: teTeX X11 binary files This is teTeX, a TeX distribution for UNIX compatible systems. These are all the X11 binaries for the teTeX system. You need to install XFree86 and at least tetex-tiny or tetex-base together with it. http://lilypond.org/cygwin/uploads/tetex/tetex-x11/tetex-x11-2.0.2-11.tar.bz2 http://lilypond.org/cygwin/uploads/tetex/tetex-x11/tetex-x11-2.0.2-2.tar.bz2 === http://lilypond.org/cygwin/uploads/tetex/libkpathsea3abi13/setup.hint sdesc: The TeX text formatting system (runtime libraries). curr: 2.0.2-2 category: Text Publishing Libs requires: cygwin external-source: tetex-bin ldesc: The TeX text formatting system (runtime libraries). The teTeX shared object library libkpathsea. http://lilypond.org/cygwin/uploads/tetex/libkpathsea3abi13/libkpathsea3abi13-2.0.2-2.tar.bz2 === http://lilypond.org/cygwin/uploads/tetex/libkpathsea3/setup.hint sdesc: The TeX text formatting system (runtime libraries). test: 2.0.2-11 category: Text Publishing Libs requires: cygwin external-source: tetex-bin ldesc: The TeX text formatting system (runtime libraries). The teTeX shared object library libkpathsea. http://lilypond.org/cygwin/uploads/tetex/libkpathsea3/libkpathsea3-2.0.2-11.tar.bz2 === -- Jan Nieuwenhuizen [EMAIL PROTECTED] | GNU LilyPond - The music typesetter http://www.xs4all.nl/~jantien | http://www.lilypond.org
lilypond-1.6.11 for 1.3 and 1.7.29 for 1.5.1
I've built lilypond for 1.3 and 1.5.1. Previous releases must be removed. Jan. http://lilypond.org/cygwin/uploads/lilypond/setup.hint sdesc: A program for printing sheet music test: 1.7.29-11 curr: 1.6.11-1 category: Publishing requires: bash cygwin fileutils findutils ghostscript libguile12abi13 libguile12 libiconv2 libintl2 libkpathsea3 libkpathsea3abi13 python tetex-bin tetex-tiny #requires: tetex-bin, tetex-tiny | tetex-base #suggests: emacs gsview lilypond-doc rxvt tetex-x11 XFree86-serv ldesc: A program for printing sheet music. LilyPond prints beautiful sheet music. It produces music notation from a description file. It excels at typesetting classical music, but you can also print pop-songs. LilyPond is part of the GNU Project. http://lilypond.org/cygwin/uploads/lilypond/lilypond-1.7.29-11-src.tar.bz2 http://lilypond.org/cygwin/uploads/lilypond/lilypond-1.7.29-11.tar.bz2 http://lilypond.org/cygwin/uploads/lilypond/lilypond-1.6.11-1-src.tar.bz2 http://lilypond.org/cygwin/uploads/lilypond/lilypond-1.6.11-1.tar.bz2 === http://lilypond.org/cygwin/uploads/lilypond/lilypond-doc/setup.hint sdesc: LilyPond documentation. test: 1.7.29-11 curr: 1.6.11-1 category: Doc ldesc: LilyPond Documentation in HTML, PS, PDF and DVI formats. This package contains the HTML, PostScript PDF and DVI documentation for the LilyPond music typesetting software. Info and man pages are in the lilypond package. http://lilypond.org/cygwin/uploads/lilypond/lilypond-doc/lilypond-doc-1.7.29-11.tar.bz2 http://lilypond.org/cygwin/uploads/lilypond/lilypond-doc/lilypond-doc-1.6.11-1.tar.bz2 === -- Jan Nieuwenhuizen [EMAIL PROTECTED] | GNU LilyPond - The music typesetter http://www.xs4all.nl/~jantien | http://www.lilypond.org
Re: XFree86 Build for 1.5.0/1
On Tue, 29 Jul 2003, David Rothenberger wrote: Harold L Hunt II wrote: I have never subscribed to this mailing list before, but it looks like this is where the real business of Cygwin takes place. I started reading the archives of the 1.5.0/1 discussion today and realized that some other people might be waiting for me to release an updated build of XFree86 for 1.5.0/1. In any case, if I don't start working out the kinks of such a rebuild I will probably never get it right. Just wanted to drop a heads-up here that I will be starting work on this shortly. I am now subscribed to the list, so you can reach me here or on cygwin-xfree with questions/suggestions. I would suggest starting with XFree86-bin, since this contains all the libraries, right? Once the libraries are released, other people can start compiling XFree86 packages for 1.5.x (or their own X apps). The actual apps can come along piecemeal after that, I think. I don't think X apps compiled with 1.5.1 against the old 1.3.x X libraries will work. But, I am not a Cygwin expert (IANACE), so hopefully others on this list will correct me if needed. I Am Not A Cygwin Expert? That's a new one of the OLOCA... Aaah yes. If the XFree86 packages use other packages' libraries then you will have to wait until they are updated before updating the XFree86 packages. Elfyn -- Elfyn McBratney, EMCB | http://www.nongnu.org/wwwauth/ http://www.emcb.co.uk | http://www.emcb.co.uk/webauth/ [EMAIL PROTECTED] | wwwauth-users AT nongnu DOT org
Re: XFree86 Build for 1.5.0/1
Elfyn McBratney wrote: On Tue, 29 Jul 2003, David Rothenberger wrote: Harold L Hunt II wrote: I have never subscribed to this mailing list before, but it looks like this is where the real business of Cygwin takes place. I started reading the archives of the 1.5.0/1 discussion today and realized that some other people might be waiting for me to release an updated build of XFree86 for 1.5.0/1. In any case, if I don't start working out the kinks of such a rebuild I will probably never get it right. Just wanted to drop a heads-up here that I will be starting work on this shortly. I am now subscribed to the list, so you can reach me here or on cygwin-xfree with questions/suggestions. I would suggest starting with XFree86-bin, since this contains all the libraries, right? Once the libraries are released, other people can start compiling XFree86 packages for 1.5.x (or their own X apps). The actual apps can come along piecemeal after that, I think. I don't think X apps compiled with 1.5.1 against the old 1.3.x X libraries will work. But, I am not a Cygwin expert (IANACE), so hopefully others on this list will correct me if needed. I Am Not A Cygwin Expert? That's a new one of the OLOCA... Aaah yes. If the XFree86 packages use other packages' libraries then you will have to wait until they are updated before updating the XFree86 packages. Right. And so by the same logic, the XFree86 libraries should be the first XFree86 package updated (as soon as the libraries they depend on are updated). Once the XFree86 libraries are updated, there is no real rush to get the apps themselves updated, unless they will really benefit from 64-bit I/O. (FYI, I'm personally interested because I want to recompile XEmacs, but can't without the updated XFree86 libraries.) Dave
Re: XFree86 Build for 1.5.0/1
Elfyn, I know we have a dependency on ncurses... and possibly jpeg, png, and/or tiff. However, I think all of those have been updated, no? Harold Elfyn McBratney wrote: On Tue, 29 Jul 2003, David Rothenberger wrote: Harold L Hunt II wrote: I have never subscribed to this mailing list before, but it looks like this is where the real business of Cygwin takes place. I started reading the archives of the 1.5.0/1 discussion today and realized that some other people might be waiting for me to release an updated build of XFree86 for 1.5.0/1. In any case, if I don't start working out the kinks of such a rebuild I will probably never get it right. Just wanted to drop a heads-up here that I will be starting work on this shortly. I am now subscribed to the list, so you can reach me here or on cygwin-xfree with questions/suggestions. I would suggest starting with XFree86-bin, since this contains all the libraries, right? Once the libraries are released, other people can start compiling XFree86 packages for 1.5.x (or their own X apps). The actual apps can come along piecemeal after that, I think. I don't think X apps compiled with 1.5.1 against the old 1.3.x X libraries will work. But, I am not a Cygwin expert (IANACE), so hopefully others on this list will correct me if needed. I Am Not A Cygwin Expert? That's a new one of the OLOCA... Aaah yes. If the XFree86 packages use other packages' libraries then you will have to wait until they are updated before updating the XFree86 packages. Elfyn
Re: guile-1.6.4 for 1.3 and 1.5.1
On 29 Jul 2003, Jan Nieuwenhuizen wrote: http://lilypond.org/cygwin/uploads/guile/guile-doc/setup.hint sdesc: The GNU extension language and Scheme interpreter (documentation) test: 1.6.4-11 curr: 1.6.4-2 category: doc requires: texinfo external-source: guile ldesc: The GNU extension language and Scheme interpreter (documentation) This package contains the documentation for guile, including both a reference manual (via `info guile'), and a tutorial (via `info guile-tut'). Should guile-doc not require guile? Elfyn -- Elfyn McBratney, EMCB | http://www.nongnu.org/wwwauth/ http://www.emcb.co.uk | http://www.emcb.co.uk/webauth/ [EMAIL PROTECTED] | wwwauth-users AT nongnu DOT org
Re: tetex-bin-2.0.2 for 1.3 and 1.5.1
On 29 Jul 2003, Jan Nieuwenhuizen wrote: sdesc: The TeX text formatting system (binaries). test: 2.0.2-11 curr: 2.0.2-2 category: Text Publishing # There's no need to require a texmf tree, although not doing so # might be confusing. requires: bash clear cygutils cygwin ed diffutils fileutils findutils gawk grep libjpeg6b libjpeg62 libkpathsea3abi13 libkpathsea3 libncurses6 libncurses7 libpng12 libtiff3 libtiff4 perl sed sh-utils textutils termcap zlib #external-source: tetex-bin #suggests: tetex-tiny | tetex-base, tetex-extra, tetex-x11 ldesc: teTeX binary files This is teTeX, a TeX distribution for UNIX compatible systems. These are all the binaries for the teTeX system. You need at least tetex-tiny or tetex-base together with it. zlib is on the (unofficial) list of DLL's that need to be re-built under Cygwin 1.5.1 . Unless I've mis-understood things, you'll have to wait until zlib is re-built. Elfyn -- Elfyn McBratney, EMCB | http://www.nongnu.org/wwwauth/ http://www.emcb.co.uk | http://www.emcb.co.uk/webauth/ [EMAIL PROTECTED] | wwwauth-users AT nongnu DOT org
Re: guile-1.6.4 for 1.3 and 1.5.1
Elfyn McBratney [EMAIL PROTECTED] writes: category: doc requires: texinfo external-source: guile ldesc: The GNU extension language and Scheme interpreter (documentation) This package contains the documentation for guile, including both a reference manual (via `info guile'), and a tutorial (via `info guile-tut'). Should guile-doc not require guile? Technically not, it's just info files. Maybe suggests:? If people think a requirement should be added, that's fine with me. Jan, -- Jan Nieuwenhuizen [EMAIL PROTECTED] | GNU LilyPond - The music typesetter http://www.xs4all.nl/~jantien | http://www.lilypond.org
Re: [setup PATCH] OnAcceptActivation
On Tue, 2003-07-29 at 23:59, Max Bowsher wrote: What is the rationale for the lowercase wants? http://sources.redhat.com/ml/cygwin-apps/2003-03/msg00750.html It should be fairly clear from the context, but I can enlarge: With C++ it can be hard to tell if you are calling a static or non static member function. Likewise, if you are accessing a global object or variable. The guidelines are constructed, so that at a glance you can tell that 'wantsActivation()' is a non-static method. Cheers, Rob -- GPG key available at: http://members.aardvark.net.au/lifeless/keys.txt. signature.asc Description: This is a digitally signed message part
Re: tetex-bin-2.0.2 for 1.3 and 1.5.1
On Tue, 29 Jul 2003, Elfyn McBratney wrote: On 29 Jul 2003, Jan Nieuwenhuizen wrote: sdesc: The TeX text formatting system (binaries). test: 2.0.2-11 curr: 2.0.2-2 category: Text Publishing # There's no need to require a texmf tree, although not doing so # might be confusing. requires: bash clear cygutils cygwin ed diffutils fileutils findutils gawk grep libjpeg6b libjpeg62 libkpathsea3abi13 libkpathsea3 libncurses6 libncurses7 libpng12 libtiff3 libtiff4 perl sed sh-utils textutils termcap zlib #external-source: tetex-bin #suggests: tetex-tiny | tetex-base, tetex-extra, tetex-x11 ldesc: teTeX binary files This is teTeX, a TeX distribution for UNIX compatible systems. These are all the binaries for the teTeX system. You need at least tetex-tiny or tetex-base together with it. zlib is on the (unofficial) list of DLL's that need to be re-built under Cygwin 1.5.1 . Unless I've mis-understood things, you'll have to wait until zlib is re-built. Sorry for the noise, I see zlib was updated this morning. Elfyn -- Elfyn McBratney, EMCB | http://www.nongnu.org/wwwauth/ http://www.emcb.co.uk | http://www.emcb.co.uk/webauth/ [EMAIL PROTECTED] | wwwauth-users AT nongnu DOT org
Re: tetex-bin-2.0.2 for 1.3 and 1.5.1
Elfyn McBratney [EMAIL PROTECTED] writes: requires: bash clear cygutils cygwin ed diffutils fileutils findutils gawk grep libjpeg6b libjpeg62 libkpathsea3abi13 libkpathsea3 libncurses6 libncurses7 libpng12 libtiff3 libtiff4 perl sed sh-utils textutils termcap zlib zlib is on the (unofficial) list of DLL's that need to be re-built under Cygwin 1.5.1 . Unless I've mis-understood things, you'll have to wait until zlib is re-built. As I understood it, 1.5.0-applications or -libraries using fdopen should be relinked for 1.5.1. Because tetex and guile both use fdopen, I rebuilt them; but I assumed that the zlib exporting interface would not change? Jan. -- Jan Nieuwenhuizen [EMAIL PROTECTED] | GNU LilyPond - The music typesetter http://www.xs4all.nl/~jantien | http://www.lilypond.org
Error in termcap setup.hint
There is an error in the termcap setup.hint. The listed curr version doesn't exist, and the 20020930-1 version isn't published to setup.ini at all. The result is that a [test] termcap is installed, since it is in Base. Max.
Re: Error in termcap setup.hint
On Tue, 29 Jul 2003, Max Bowsher wrote: There is an error in the termcap setup.hint. The listed curr version doesn't exist, and the 20020930-1 version isn't published to setup.ini at all. The result is that a [test] termcap is installed, since it is in Base. I hope I'm not tredding on anyone's toes by fixing it. curr: 20020930-1 test: 20021106-2 Elfyn -- Elfyn McBratney, EMCB | http://www.nongnu.org/wwwauth/ http://www.emcb.co.uk | http://www.emcb.co.uk/webauth/ [EMAIL PROTECTED] | wwwauth-users AT nongnu DOT org
Re: XFree86 Build for 1.5.0/1
Harold L Hunt II wrote: Elfyn, I know we have a dependency on ncurses... and possibly jpeg, png, and/or tiff. However, I think all of those have been updated, no? jpeg, png, tiff, and zlib have been updated to 1.5.0 zlib has been updated to 1.5.1 ncurses has been updated to 1.5.0 -- and NEEDS to be updated to 1.5.1, so hold off. I ran into some issues last night rolling up to T.E.Dickey's lastest interim patches. Stay tuned. -- Chuck
Re: XFree86 Build for 1.5.0/1
Charles Wilson wrote: Harold L Hunt II wrote: Elfyn, I know we have a dependency on ncurses... and possibly jpeg, png, and/or tiff. However, I think all of those have been updated, no? jpeg, png, tiff, and zlib have been updated to 1.5.0 zlib has been updated to 1.5.1 ncurses has been updated to 1.5.0 -- and NEEDS to be updated to 1.5.1, so hold off. I ran into some issues last night rolling up to T.E.Dickey's lastest interim patches. Stay tuned. Okay, thanks for the update. Harold
Re: Who's who of packagers?
I think cygwin-apps is more appropriate for what I wrote.
How does one check who the maintainer of a package is; e.g.
to make sure it is not oneself?
And how does one check who the maintainer of a package is ?
I was looking for the Links one. I had to check the January
2002 cygwin-announce archives to find
http://sources.redhat.com/ml/cygwin-apps/2002-01/msg00357.html
BTW, what's the status of this particular package ? Will it be
removed, updated, or replaced by ELinks ?
Anyway, I compiled 0.99pre8 and there were only 2 noticeable
problems.
1- Accentuation doesn't work (at least the ouput) without the
following change in os_dep.h (suggested a year ago by the
ELinks maintainer):
-#elif defined(_WIN32) || defined(__CYGWIN__)
+#elif defined(_WIN32)
Then I had to comment
void terminate_osdep ()
{
kill (w32_input_pid, SIGINT) ;
}
in win32.c because it failed to link. I'm not a programmer, but
that worked and also fixed another problem. Links froze / hung
when started from rxvt.
CC'ing the package maintainer if he's still alive. I already
sent a report to the Links maintainer.
irc is another package that needs to be updated (April 2001)
or replaced, IMO by EPIC or irssi. I use the latter on Cygwin.
Too bad I'm not a programmer, or I could volunteer to maintain
them.
RE: [XFree86-4.2.0] Now that we have an improved ld, please makelibXt a shared library.
On Mon, 28 Jul 2003, Ralf Habacker wrote:
I changed the type of _XtInherit to a variable instead of a function.
It compiles but I've not tested it completely.
#ifdef SUNSHLIB
/*
* _XtInherit needs to be statically linked since it is compared against as
* well as called.
*/
void _XtInherit()
{
extern void __XtInherit();
__XtInherit();
}
#define _XtInherit __XtInherit
+ #elif defined(CYGWIN)
+ void (_XtInherit)(void) = __XtInherit;
+ #define _XtInherit __XtInherit
#endif
With this way we have a symbol which contains the address of the XtInherit
function in libXt and is directly exported. This requires the pseudo-reloc
feature of ld.
Any comments on this? Will this work or are there still problems?
I will have to change the declaration of _XtInherit in the header too, but
this will be in the real patch which I will send at the end of week.
bye
ago
--
[EMAIL PROTECTED]
http://www.gotti.org ICQ: 126018723
RE: [XFree86-4.2.0] Now that we have an improved ld, please makelibXt a shared library.
On Tue, 29 Jul 2003, Alexander Gottwald wrote: + void (_XtInherit)(void) = __XtInherit; oops, missed one character. It must be + void (*_XtInherit)(void) = __XtInherit; hope it'n now correct *g* bye ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
RE: [XFree86-4.2.0] Now that we have an improved ld, please make libXt a shared library.
#ifdef SUNSHLIB
/*
* _XtInherit needs to be statically linked since it is compared against as
* well as called.
*/
void _XtInherit()
{
extern void __XtInherit();
__XtInherit();
}
#define _XtInherit __XtInherit
+ #elif defined(CYGWIN)
+ void (_XtInherit)(void) = __XtInherit;
+ #define _XtInherit __XtInherit
#endif
With this way we have a symbol which contains the address of the XtInherit
function in libXt and is directly exported. This requires the pseudo-reloc
feature of ld.
Why ? Does client code access _XtInherit+offset at any place ? Only in that
case the pseudo-reloc stuff is needed.
Ralf
unable to change us keyboard in french keyboard
I have done the installation of cygwin-xfree under W2000. It works perfectly but I get an us keyboard in a graphic window. I tried to put the line run setxkbmap.exe -layout fr in startxwin.bat but with that line, it is no more possible to open a graphic window. I have also tried the command setxkbmap.exe -layout fr in a graphic window and the window was shut. All the previous work in the case of Windows98. I would like some help. Thank you.
Cygwin.rules change
Add clean target to ResourceObjectRule. Affects Cygwin only. Harold cygwin.rules-20030729-1117.diff Description: Binary data
Test 92 observations
Hi Harold, et. al. - I just installed Test 92 and it seems to work fine. Just a few notes, for what they're worth. I'm on Win2K, SP2 - I use the Dvorak keyboard layout and I have to do a run setxkbmap dvorak in my batch file that start XWin.exe. I don't know if it should automatically detect my Windows keyboard layout or not. - I start XWin.exe with start Xwin -multiwindow -multiplemonitors -clipboard. I have 2 monitors on my system, both with the same resolution, the secondary with a lower bit depth. When I drag windows to the secondary monitor, they just draw the whole window in white. I'm not sure how far along the -multiplemonitors support is, so I'm not sure what to expect, but I thought I'd mention it. Many thanks to all for XFree on Windows. What a treat to have all these great apps and tools. -- Jason Dufair - [EMAIL PROTECTED] http://www.dufair.org/ Time flies like an arrow. Fruit flies like a banana. -- Groucho Marx
RE: [XFree86-4.2.0] Now that we have an improved ld, please makelibXt a shared library.
On Tue, 29 Jul 2003, Ralf Habacker wrote: Why ? Does client code access _XtInherit+offset at any place ? Only in that case the pseudo-reloc stuff is needed. No. It never uses this indirect access (it would not make sense either since _XtInherit is a function and (f + 4)() does not make much sense) Maybe I got confused by the tests I've done yesterday *g* But the pseudo-reloc is needed anyhow by other parts of the Xt library. bye ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
Re: Test 92 observations
On Tue, 29 Jul 2003, Jason Dufair wrote: - I use the Dvorak keyboard layout and I have to do a run setxkbmap dvorak in my batch file that start XWin.exe. I don't know if it should automatically detect my Windows keyboard layout or not. dvorak is not one of the preconfigured layouts. Run the program from http://www.tu-chemnitz.de/~goal/xfree/keyboard.tar.gz and send me the output. I'll add dvorak to the list of preconfigured layouts. bye ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
Re: [ANNOUNCEMENT] Server Test 92
HLH == Harold L Hunt [EMAIL PROTECTED] writes: HLH 3) winmultiwindowclass.c - Add checks for NULL pointers. HLH Should help fix some crashes in MultiWindow mode. (Ralf HLH Habacker) Several minutes into an X session with Test 92, while I was using a remote emacs session, the server bombed with the following message: --8---cut here---start-8--- $ xinit -- -multiwindow -clipboard winMutliWindowWMIOErrorHandler! xinit: connection to X server lost. xterm: fatal IO error 104 (Connection reset by peer) or KillClient on X server :0.0 --8---cut here---end---8--- Is the winMutliWindowWMIOErrorHandler error handler triggered when one of those null-pointer checks you mentioned finds a null pointer? Thanks for releasing this build! Regards, --Joe -- Joseph Barillari -- http://barillari.org
Re: [ANNOUNCEMENT] Server Test 92
Joe, Don't know for sure what happened in your case. It appears that all X Clients (including the inernal window manager) lost communication with the X Server at about the same time. The *IOErrorHandler's are called when communication with the X Server is lost. Sounds like you had a hiccup in your TCP/IP network stack on your Windows machine. Lets just forget this unless it happens to you on a regular basis. Harold Joseph Barillari wrote: HLH == Harold L Hunt [EMAIL PROTECTED] writes: HLH 3) winmultiwindowclass.c - Add checks for NULL pointers. HLH Should help fix some crashes in MultiWindow mode. (Ralf HLH Habacker) Several minutes into an X session with Test 92, while I was using a remote emacs session, the server bombed with the following message: --8---cut here---start-8--- $ xinit -- -multiwindow -clipboard winMutliWindowWMIOErrorHandler! xinit: connection to X server lost. xterm: fatal IO error 104 (Connection reset by peer) or KillClient on X server :0.0 --8---cut here---end---8--- Is the winMutliWindowWMIOErrorHandler error handler triggered when one of those null-pointer checks you mentioned finds a null pointer? Thanks for releasing this build! Regards, --Joe
RE: XLT
Harold, XLT is a set of widgets used in Motif/Lesstif development environment. I'm running cygwin on XP and trying to install the XLT widget classes. It uses a configure script which is giving me errors. Gary -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Harold L Hunt II Sent: Monday, July 28, 2003 6:27 PM To: [EMAIL PROTECTED] Subject: Re: XLT XLT? Gary Nicholson wrote: Is this the right place to ask for help installing the XLT package? If not, where should I go? Thank you.
Re: [ANNOUNCEMENT] Server Test 92
after installing the new binary, netscape stop crashings. thanks for the good work. my env: MS Win2k, SP3. chiew Harold L Hunt II wrote: I would like some feedback as to whether or not this release fixes any crashing with -multiwindow and -clipboard. Harold Harold L Hunt II wrote: Links: I just posted Test 92 to the server development page: http://xfree86.cygwin.com/devel/shadow/ You can install the Test 92 package via setup.exe by selecting the following version of the XFree86-xserv package:4.2.0-43 Server Test Series binary and source code releases are now available via the sources.redhat.com ftp mirror network (http://cygwin.com/mirrors.html) in the pub/cygwin/xfree/devel/shadow/ directory. You may wish to note the desired filename in the links below, then download from your closest mirror (http://cygwin.com/mirrors.html). Server binary, direct link: http://www.msu.edu/~huntharo/xwin/shadow/XWin-Test92.exe.bz2 (1210 KiB) Server source, direct link: http://www.msu.edu/~huntharo/xwin/shadow/xwin-20030728-2010.tar.bz2 (118 KiB) xc/programs/Xserver/hw/xwin (all files) diff against Test91 source code: http://www.msu.edu/~huntharo/xwin/shadow/xwin-Test91-to-Test92.diff (9 KiB) Changes: 1) winconfig.c - Add new defaults for keyboardlayout: German (Switzerland) and English (USA, International). (Alexander Gottwald) 2) XWin.rc - Move Exit button to the bottom of the tray icon's menu. (List suggestion) 3) winmultiwindowclass.c - Add checks for NULL pointers. Should help fix some crashes in MultiWindow mode. (Ralf Habacker) 4) General - Add global variable that tracks whether XInitThreads () has been called so that it is only called once per process. Calling XInitThreads multiple times may have been contributing to crashes. (Harold L Hunt II) Enjoy, Harold
Re: XLT
Gary, Yes, this is the right place. Please post some details of your problem. Thanks, Harold Gary Nicholson wrote: Harold, XLT is a set of widgets used in Motif/Lesstif development environment. I'm running cygwin on XP and trying to install the XLT widget classes. It uses a configure script which is giving me errors. Gary -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Harold L Hunt II Sent: Monday, July 28, 2003 6:27 PM To: [EMAIL PROTECTED] Subject: Re: XLT XLT? Gary Nicholson wrote: Is this the right place to ask for help installing the XLT package? If not, where should I go? Thank you.
4.3.0.1 Possible Release and Keeping XFree86 CVS Up to Date
I am working on pulling the 4.3.0.1 tree: export CVS_RSH=ssh export [EMAIL PROTECTED]:/cvs cvs -z3 checkout -r xf-4_3_0_1 xc cvs-checkout.log 21 I will be copying over the latest hw/xwin files and a few misc. patches and building the 4.3.0.1 tree for possible release. I am going to try my hand at packaging this up again to see if I run into any problems. If all goes well, I might post it as a test release... but it will probably be a test release dependent upon Cygwin1.dll v 1.5.1 (or whatever the most recent release is). As for keeping the XFree86 CVS tree up to date, am I right that I should be sending patches against XFree86 CVS xf-4_3-branch, instead of the xf-4_3_0_1 snapshot? Could I do this with: cd xc/programs/Xserver/hw cvs diff -U3 -r xf-4_3-branch xwin xwin.diff The main question here is can I make diffs against the specified branch rather than the branch that I have locally? I assume the answer is yes, but some confirmation from others before I waste time looking into it would be helpful. Harold
XFree86 fonts from 4.2.0 to 4.3.0
Can anyone think of any changes to the font packages between 4.2.0 and 4.3.0 that would require me to repackage and distribute 4.3.0 font sets? If not, I would really like to just leave the current 4.2.0 fonts in place since it will save most people from downloading between 15 and 50 MB of the same stuff. Of course, the version numbers will look a little funny, but I think most users would be pleased that they didn't have to download the fonts again. Of course, if anyone knows otherwise, please speak up now so I don't make a terrible mistake. Harold
Re: [ANNOUNCEMENT] Server Test 92
HLH == Harold L Hunt [EMAIL PROTECTED] writes: HLH Joe, Don't know for sure what happened in your case. It HLH appears that all X Clients (including the inernal window HLH manager) lost communication with the X Server at about the HLH same time. The *IOErrorHandler's are called when HLH communication with the X Server is lost. Sounds like you had HLH a hiccup in your TCP/IP network stack on your Windows HLH machine. HLH Lets just forget this unless it happens to you on a regular HLH basis. Hm. It just happened again in emacs: $ xinit -- -multiwindow -clipboard hours pass winMutliWindowWMIOErrorHandler! xterm: fatal IO error 32 (Broken pipe) or KillClient on X server :0.0 xinit: connection to X server lost. I was running emacs on a machine whose network connection blocks for several seconds every few hours. Usually, the X server (either exceed or XFree86) just tolerates this, but sometimes it blocks for too long and the SSH session times out, killing the X processes tunneled over it. I doubt that this is related, since you said that winMutliWindowWMIOErrorHandler is a consequence of /all/ clients losing their connections, but I thought it might be of some significance. Best, --Joe -- Joseph Barillari -- http://barillari.org
Re: [ANNOUNCEMENT] Server Test 92
I'm running Win2k SP4. I have a problem with winclip exceptions. If I close down the X session when I have multiple remote terminals open, I almost always get an exception in xwinclip. It is still happening with Test92. I run XWin -multiwindow -clipboard. I get the same xwinclip error at home on win98, but I haven'y upgraded to this server test version yet. -Pete Harold L Hunt II wrote: I would like some feedback as to whether or not this release fixes any crashing with -multiwindow and -clipboard. Harold Harold L Hunt II wrote: Links: I just posted Test 92 to the server development page: http://xfree86.cygwin.com/devel/shadow/ You can install the Test 92 package via setup.exe by selecting the following version of the XFree86-xserv package:4.2.0-43 Server Test Series binary and source code releases are now available via the sources.redhat.com ftp mirror network (http://cygwin.com/mirrors.html) in the pub/cygwin/xfree/devel/shadow/ directory. You may wish to note the desired filename in the links below, then download from your closest mirror (http://cygwin.com/mirrors.html). Server binary, direct link: http://www.msu.edu/~huntharo/xwin/shadow/XWin-Test92.exe.bz2 (1210 KiB) Server source, direct link: http://www.msu.edu/~huntharo/xwin/shadow/xwin-20030728-2010.tar.bz2 (118 KiB) xc/programs/Xserver/hw/xwin (all files) diff against Test91 source code: http://www.msu.edu/~huntharo/xwin/shadow/xwin-Test91-to-Test92.diff (9 KiB) Changes: 1) winconfig.c - Add new defaults for keyboardlayout: German (Switzerland) and English (USA, International). (Alexander Gottwald) 2) XWin.rc - Move Exit button to the bottom of the tray icon's menu. (List suggestion) 3) winmultiwindowclass.c - Add checks for NULL pointers. Should help fix some crashes in MultiWindow mode. (Ralf Habacker) 4) General - Add global variable that tracks whether XInitThreads () has been called so that it is only called once per process. Calling XInitThreads multiple times may have been contributing to crashes. (Harold L Hunt II) Enjoy, Harold -- Peter W. Colovas RNC Traffic Processing Department cPSB Development Lucent Technologies Naperville, IL +1 630 224-2181
Re: [ANNOUNCEMENT] Server Test 92
What do you mean you get an xwinclip execption when using -clipboard. You aren't running xwinclip AND -clipboard, are you? If so, I would certainly expect crashes, since both xwinclip and -clipboard do exactly the same thing. Please tell me that you just made a mistake in your description and that you are not actually doing this. Harold Peter Colovas wrote: I'm running Win2k SP4. I have a problem with winclip exceptions. If I close down the X session when I have multiple remote terminals open, I almost always get an exception in xwinclip. It is still happening with Test92. I run XWin -multiwindow -clipboard. I get the same xwinclip error at home on win98, but I haven'y upgraded to this server test version yet. -Pete Harold L Hunt II wrote: I would like some feedback as to whether or not this release fixes any crashing with -multiwindow and -clipboard. Harold Harold L Hunt II wrote: Links: I just posted Test 92 to the server development page: http://xfree86.cygwin.com/devel/shadow/ You can install the Test 92 package via setup.exe by selecting the following version of the XFree86-xserv package:4.2.0-43 Server Test Series binary and source code releases are now available via the sources.redhat.com ftp mirror network (http://cygwin.com/mirrors.html) in the pub/cygwin/xfree/devel/shadow/ directory. You may wish to note the desired filename in the links below, then download from your closest mirror (http://cygwin.com/mirrors.html). Server binary, direct link: http://www.msu.edu/~huntharo/xwin/shadow/XWin-Test92.exe.bz2 (1210 KiB) Server source, direct link: http://www.msu.edu/~huntharo/xwin/shadow/xwin-20030728-2010.tar.bz2 (118 KiB) xc/programs/Xserver/hw/xwin (all files) diff against Test91 source code: http://www.msu.edu/~huntharo/xwin/shadow/xwin-Test91-to-Test92.diff (9 KiB) Changes: 1) winconfig.c - Add new defaults for keyboardlayout: German (Switzerland) and English (USA, International). (Alexander Gottwald) 2) XWin.rc - Move Exit button to the bottom of the tray icon's menu. (List suggestion) 3) winmultiwindowclass.c - Add checks for NULL pointers. Should help fix some crashes in MultiWindow mode. (Ralf Habacker) 4) General - Add global variable that tracks whether XInitThreads () has been called so that it is only called once per process. Calling XInitThreads multiple times may have been contributing to crashes. (Harold L Hunt II) Enjoy, Harold
Re: -multiwindow crashes XWin.exe, twm does not
Thomas L Roche wrote: WebSphere Studio products typically put up a configuration dialog, then display the full UI. When my startxwin.bat has - start XWin -multiwindow -clipboard -unixkill I get my xterm, in which I run # local xhost + 1.2.3.4 ssh -X -l userid 1.2.3.4 # remote cd /opt/IBM/WebSphere/Express/SiteDeveloper ./wasexpress51 This launches the dialog. When I hit OK (taking the harmless defaults), taskmgr shows some activity, until XWin.exe GPFs. But if I use + start XWin -clipboard -unixkill + REM Startup an xterm, using bash as the shell. + run xterm -sl 1000 -sb -rightbar -ms red -fg white -bg blue -e /usr/bin/bash + REM Startup the twm window manager. + REM WARNING: Do not use an external window manager in conjunction + REM with the ``-multiwindow'' command-line parameter for XWin. + REM Doing so would start two window managers, which is never + REM supposed to happen. + run twm in my startxwin.bat, and run the same xterm session above, both the dialog and the full IDE launch properly. Harold L Hunt II Mon, 28 Jul 2003 19:50:53 -0400 - Please try -multiwindow without -clipboard. That may work. Otherwise, please try the next test release that I make... it has some NULL pointer checks that may relate to your problem. Harold L Hunt II Mon, 28 Jul 2003 20:37:11 -0400 You can install the Test 92 package via setup.exe by selecting the following version of the XFree86-xserv package: 4.2.0-43 So I installed that package from setup.exe and tried: * start XWin -multiwindow -clipboard -unixkill No change: I still get the startup dialog, then a GPF. * start XWin -multiwindow -unixkill Same: dialog, then GPF. * start XWin -clipboard -unixkill snip run twm Still works.
Re: [ANNOUNCEMENT] Server Test 92
Howdy... At 04:13 PM 7/29/2003 -0500, Peter Colovas wrote: I'm running Win2k SP4. I have a problem with winclip exceptions. If I close down the X session when I have multiple remote terminals open, I almost always get an exception in xwinclip. It is still happening with Test92. I run XWin -multiwindow -clipboard. I get the same xwinclip error at home on win98, but I haven'y upgraded to this server test version yet. Do you mean that you get this when you are trying to kill the server from the notification icon? I too see that occasionally, but haven't taken the time to run under GDB and see why the xwinclip thread is GPFing during server shutdown. In any case it's benign since you're killing the app anyway, right? -Earle F. Philhower, III [EMAIL PROTECTED] cdrlabel - ZipLabel - FlpLabel http://www.cdrlabel.com
1.5.1: lseek64 woes (was Re: 1.5.1: unzip problems with binary files in archive)
Charles Wilson writes:
Christopher Faylor wrote:
I'm getting a strange error from unzip when trying to extract an
archive that contains binary files. Everything works fine if the
archive contains just text files.
Can't duplicate it. Sorry.
I --can-- duplicate it. But I don't know why it happens; I'll have to
debug it...but not until after I update to 1.5.1. FWIW, zip does use
fdopen().
I've done some debugging and I have a STC. I have no idea why it
fails, and why Chris doesn't see it, but here's the info.
The STC is composed of the two attached files: works.c and fails.c.
Both files essentially do
int fd = open(works.c, 0);
lseek(fd, 1, 0);
but works.c does this directly in main(), while fails.c does this in
unzip() called from main(). As you can guess, lseek() returns -1 in
fails.c and 1 in works.c.
The strace output is interesting.
$ gcc -o works works.c
$ strace --mask=syscall ./works | grep lseek
[*] 100 47928 [main] works 1536 lseek64: lseek (3, 1, 0)
227 48155 [main] works 1536 lseek64: 1 = lseek (0, 3, 0)
$ gcc -o fails fails.c
$ strace --mask=syscall ./fails | grep lseek
[*] 96 3 [main] fails 1908 lseek64: lseek (3, 1, 0)
[*] 79 36745 [main] fails 1908 lseek64: dir is invalid (1627411387)
76 36821 [main] fails 1908 lseek64: -1 = lseek (-1, 3, 0)
[*] I added these:
Index: syscalls.cc
===
RCS file: /cvs/src/src/winsup/cygwin/syscalls.cc,v
retrieving revision 1.278
diff -u -p -r1.278 syscalls.cc
--- syscalls.cc 26 Jul 2003 04:53:59 - 1.278
+++ syscalls.cc 29 Jul 2003 05:18:01 -
@@ -541,8 +541,11 @@ lseek64 (int fd, _off64_t pos, int dir)
_off64_t res;
sigframe thisframe (mainthread);
+ syscall_printf(lseek (%d, %lld, %d), fd, pos, dir);
+
if (dir != SEEK_SET dir != SEEK_CUR dir != SEEK_END)
{
+ syscall_printf(dir is invalid (%d), dir);
set_errno (EINVAL);
res = -1;
}
The output shows that dir is fine right before the if statement, but
not after.
(The output above was generated using the dll from CVS HEAD compiled
with mknetrel. When compiled with -g only, both cases fail,
although the incorrect dir value for the works.c is 2048, much lower
than value for fails.c, which is of the same magnitude as the value
above.)
The tests work as originally described with the official 1.5.1 dll,
although without the added syscall_printf()'s. However, even the
stock strace output is weird. The output is
79 123229 [main] fails 2308 lseek64: -1 = lseek (-1, 3, 0)
but the printf is
syscall_printf (%d = lseek (%d, %D, %d), res, fd, pos, dir);
which means
res = -1 (okay)
fd = -1 (should be 3)
pos = 3 (should be 1)
dir = 0 (okay)
At least I think so. I'm not sure what %D means
I've attached another cygcheck output because the original report
was generated from a fresh 1.5.1 installation while this one is CVS
HEAD with all test packages installed.
HTH,
Dave
#include stdio.h
int main()
{
int fd;
long retval;
fd = open(works.c, 0);
retval = lseek(fd, 1, 0);
printf(retval in unzip = %ld\n, retval);
}
#include stdio.h
void unzip()
{
int fd;
long retval;
fd = open(fails.c, 0);
retval = lseek(fd, 1, 0);
printf(retval in unzip = %ld\n, retval);
}
int main()
{
unzip();
return(0);
}
Cygwin Win95/NT Configuration Diagnostics
Current System Time: Mon Jul 28 22:10:02 2003
Windows 2000 Professional Ver 5.0 Build 2195 Service Pack 4
Path: h:\users\drothe\bin
C:\cygwin\usr\local\bin
C:\cygwin\bin
C:\cygwin\bin
c:\WINNT\system32
c:\WINNT
c:\WINNT\System32\Wbem
c:\Program Files\PGPNT
c:\Program Files\Executive Software\DiskeeperWorkstation\
c:\Program Files\SSH Communications Security\SSH Secure Shell
C:\cygwin\usr\X11R6\bin
Output from C:\cygwin\bin\id.exe (nontsec)
UID: 1000(drothe) GID: 513(None)
513(None)
Output from C:\cygwin\bin\id.exe (ntsec)
UID: 1000(drothe) GID: 513(None)
513(None) 0(root)
545(Users)
SysDir: C:\WINNT\system32
WinDir: C:\WINNT
HOME = `h:\users\drothe'
MAKE_MODE = `unix'
PWD = `/c/temp/unzip'
USER = `drothe'
ALLUSERSPROFILE = `C:\Documents and Settings\All Users'
APPDATA = `C:\Documents and Settings\drothe\Application Data'
COLORFGBG = `0;default;15'
COLORTERM = `rxvt-xpm'
COMMONPROGRAMFILES = `C:\Program Files\Common Files'
COMPUTERNAME = `TELA'
COMSPEC = `C:\WINNT\system32\cmd.exe'
CVSROOT = `:ext:[EMAIL PROTECTED]:/home/drothe/cvsroot'
CVS_RSH = `/usr/local/bin/ssh-noescape'
DISKEEPERICON = `C:\Program Files\Executive Software\DiskeeperWorkstation\'
DISPLAY = `:0'
HOMEDRIVE = `C:'
HOMEPATH = `\'
LOGONSERVER = `\\TELA'
LS_COLORS =
Re: 1.5.1: lseek64 woes (was Re: 1.5.1: unzip problems with binary files in archive)
Darn it! I forgot to change the disposition on those attachments. Sorry about that, guys. Getting late for me. Dave -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: tcflush waits
Having now looked at the source I feel it is like this by design but I don't understand why as there are no comments.If this is indeed a best approximation to the intended behaviour should it be documented as such ? Jon On 28 Jul 2003 at 14:49, [EMAIL PROTECTED] wrote: I have an app that opens a serial port and sets VTIME to 1, VMIN to 0. When it does tcflush(,TCIOFLUSH) or tcflush(,TCIFLUSH) it takes exactly 100ms. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Launching cygwin with tcsh
Hello i wanted to know how could i launch tcsh-shell instead of bash-shell when i run cygwin ... If i run tcsh-shell, is there any possibility to create a .tcshrc for configuring path and other environment variables? ___ 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 Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: upgrade question
Thank you for this. However, I managed to find a disaster in there despite everything. :o( I renamed the whole e:\cygwin tree to e:\old_cygwin. I had to rerun the install several times, for reasons I need not go into. Eventually it got to the end and said I found an old cygwin, shall I delete it? I assumed it meant one from the immediately previous installation, which had also completed, but with errors, so innocently pressed OK. Mistake. It deleted the whole e:\old_cygwin tree, including a load of stuff that wasn't actually cygwin at all and was only there for historical reasons. And none of it is recoverable, nor backed up, of course. Boo hoo. Rob Larry Hall wrote: Igor Pechtchanski wrote: On Fri, 25 Jul 2003, Rob Clack wrote: snip Can I just rename my old cygwin directory tree (E:\cygwin) to something else and install the latest in its place? I did read somewhere that you shouldn't have two cygwin1.dlls, hence the question. Thanks in advance Rob Yes, you can. If you want to be completely safe, rename the old cygwin1.dll to, say, old-cygwin1.dll (I have a few on my system, no problems yet). Right. The issue isn't duplicate DLLs. The issue is using duplicate DLLs. Just make sure that you're new installation can't see the old and that you don't run apps that can find the other DLL. There can be only 1 Cygwin DLL loaded at a time. -- ~~ Trying to think. Nothing happens. ~~ Rob ClackAcedb Development, Informatics Group email: [EMAIL PROTECTED]Wellcome Trust Sanger Institute Tel: +44 1223 494780 Wellcome Trust Genome Campus Fax: +44 1223 494919 Hinxton CambridgeCB10 1SA -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: upgrade question
Rob Clack wrote: Thank you for this. However, I managed to find a disaster in there despite everything. :o( I renamed the whole e:\cygwin tree to e:\old_cygwin. I had to rerun the install several times, for reasons I need not go into. Eventually it got to the end and said I found an old cygwin, shall I delete it? I assumed it meant one from the immediately previous installation, which had also completed, but with errors, so innocently pressed OK. Mistake. It deleted the whole e:\old_cygwin tree, including a load of stuff that wasn't actually cygwin at all and was only there for historical reasons. And none of it is recoverable, nor backed up, of course. Boo hoo. Huh? Cygwin setup.exe doesn't do this. It only deletes old cygwin1.dll files. Not directory trees. Max. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: upgrade question
On Tue, 2003-07-29 at 20:22, Max Bowsher wrote: Rob Clack wrote: Thank you for this. However, I managed to find a disaster in there despite everything. :o( I renamed the whole e:\cygwin tree to e:\old_cygwin. I had to rerun the install several times, for reasons I need not go into. Eventually it got to the end and said I found an old cygwin, shall I delete it? I assumed it meant one from the immediately previous installation, which had also completed, but with errors, so innocently pressed OK. Mistake. It deleted the whole e:\old_cygwin tree, including a load of stuff that wasn't actually cygwin at all and was only there for historical reasons. And none of it is recoverable, nor backed up, of course. Boo hoo. Huh? Cygwin setup.exe doesn't do this. It only deletes old cygwin1.dll files. Not directory trees. In fact, thats not even a cygwin setup prompt. Rob, you ran something other than the current cygwin installer for sure... Cheers, Rob -- GPG key available at: http://members.aardvark.net.au/lifeless/keys.txt. --- signature.asc Description: This is a digitally signed message part
profile folder missing in docbook-xsl package
Hi Marcel, the profile folder is missing in the latest docbook-xsl-1.61.3-1 package, breaking the various profile stylesheets :o( Did you drop the folder intentionally from the distribution or was it a mere oversight? Patrick -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Local printer access question..
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Igor Pechtchanski On Mon, 28 Jul 2003, Rick Rankin wrote: --- Igor Pechtchanski [EMAIL PROTECTED] wrote: On Mon, 28 Jul 2003, Hannu E K Nevalainen (garbage mail) wrote: From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Igor Pechtchanski On Sun, 27 Jul 2003, Ishwar Rattan wrote: Is there a way to access the printer (Deskjet 694C on parallel port)?? If your Windows printer name is Deskjet 694C, try 'lpr -P Deskjet 694C'. You can also 'export PRINTER=Deskjet 694C'. Igor I wish it was that easy... I have a Deskjet 720C which should be about the same thing as the one above. I've tried all of the possible ways described in man lpr. All I get is a document showing up Start Settings Printers HP720C indicating that it is printing. The printer itself does nothing. This seems not be cygwin failing but windows or the driver - as the behaviour is the same from cmd.exe using PRINT /D:all possibilities tried Well, it *is* that easy for me on Win2k. I haven't tried it on other systems, there might be some peculiarities on 9x/ME/XP. I still have my W98SE runnable, but I am almost constantly using the dual-boot W2K currently. All this has been tried there. $ uname -a CYGWIN_NT-5.0 P450 1.3.22(0.78/3/2) 2003-03-18 09:20 i686 unknown unknown Cygwin setup.exe used very recently (Yesterday? no test/exp-packages). Some things to try that I can think of are: renaming your printer, or making it shared and using the UNC notation to access it (e.g., export PRINTER='\\MYCOMPUTER\HP720C'). This doesn't sound like a naming problem to me. Hannu indicates that the document shows up on the print queue, but the printer does nothing. That tells me that lpr and the DOS print command are finding the print queue properly. Exactly. Was my writing unclear? Please tell me where if so - private email please! - In hope this will improve my english. What kind of file are you trying to print and *exactly* what command are you using to print it? See below for an example. --Rick Oh. Yes, it seems that I misunderstood Hannu's message... Thanks, Rick. Hannu, Well, the obvious question first: is the printer paused? Nope. Can you print from Windows apps? Yes I can. :-} Also, try setting up a dummy printer that prints to file and see if that works. Igor Hrm... why didn't I think of that! :-P Well, here we go: Start Settings Printers Add Printer - Add printer wizard Next - Local Printer DONT Automatically Detect ... Next - Use following port FILE: Next - Manufacturer and model Generic Generic / Text only Next - Printer name: Generic / Text only Default: No Next - Share as: GT Next - Location and Comment: Next - Print a test page: No Next - Finish F: print /D:\\P450\GT C:\Autoexec.bat Filename: d:\zz.txt $ diff -up /cygdrive/d/zz.txt /cygdrive/c/AUTOEXEC.BAT --- /cygdrive/d/zz.txt 2003-07-29 13:05:50.0 +0200 +++ /cygdrive/c/AUTOEXEC.BAT2003-05-18 20:41:22.0 +0200 @@ -59,4 +59,3 @@ Rem TShoot: echo Autoexec done... - \ No newline at end of file $ lpr -P //P450/GT /cygdrive/c/AUTOEXEC.BAT [EMAIL PROTECTED] ~ bash PID=2028, Last xc=0 $ diff -up /cygdrive/d/zz.txt /cygdrive/c/AUTOEXEC.BAT --- /cygdrive/d/zz.txt 2003-07-29 13:09:20.0 +0200 +++ /cygdrive/c/AUTOEXEC.BAT2003-05-18 20:41:22.0 +0200 @@ -59,4 +59,3 @@ Rem TShoot: echo Autoexec done... - $ Works like a charm... so this should prove one thing: Cheap HP printers aren't up to all the tasks you could expect. :-/ I suspect there is a problem WRT printer sharing in this driver too. I've tried this on W98SE earlier - got very few pages out of it, then it died. No resurrection possible whatever I tried. And then, you ask whether there is a more recent driver: http://h20015.www2.hp.com/en/document.jhtml?docName=bpd09190lc=en There is no need to download a driver from the Web site if you selected a download link and were brought here. The most current driver is already integrated into Windows 2000 and it is already on every Windows 2000 PC. Follow the appropriate instructions below to install the correct printer driver using the Add Printer Wizard. NOTE: The above webpage gives indication of that the Network Printer stuff SHOULD work. I don't know what to think about it. The descriptions are very general and matches my actions exactly. (I have tried to vary from the basics, where possible and where I have deemed apropriate - no go. Details forgotten by now). /Hannu E K Nevalainen, B.Sc EE/Microcomputer systems, 59?14'N, 17?12'E. --END OF MESSAGE-- -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Launching cygwin with tcsh
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of philippe guillaume Hello i wanted to know how could i launch tcsh-shell instead of bash-shell when i run cygwin ... Change C:\cygwin\cygwin.bat (assuming default install path) to your fulfill your needs. If i run tcsh-shell, is there any possibility to create a .tcshrc for configuring path and other environment variables? You have to make sure that you start tcsh with the apropriate flags/options to make it a login shell. This really isn't a cygwin issue (i.e. not a subject for this mailing list), so please read up on it using e.g: $ info tcsh $ man tchs http://www.google.com/ +tchs +login +shell might give you good info. You might also find something special by adding site:cygwin.com here. /Hannu E K Nevalainen, B.Sc EE/Microcomputer systems, 59°14'N, 17°12'E. --END OF MESSAGE-- -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: profile folder missing in docbook-xsl package
Hi Patrick. On Tue, Jul 29, 2003 at 01:01:06PM +0200, Patrick Eisenacher wrote: the profile folder is missing in the latest docbook-xsl-1.61.3-1 package, breaking the various profile stylesheets :o( I've repackaged docbook-xsl with 'profiling' directory included. The new version (1.61.3-2) should be available on the mirrors shortly. Thank you for your feedback. -- +---+ | Marcel Telka e-mail: [EMAIL PROTECTED] | |homepage: http://telka.sk/ | |jabber: [EMAIL PROTECTED] | +---+ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Source code for binaries offered at http://thinstall.com/ ?
My concern is that issues surrounding Mr. Clark's use of cygwin goes well beyond just supplying source on his website. It appears that the raison d'etre for his software is to simplfy software distribution, and can up processes and process combinations for very tailored purposes - a worthwhile endeavor. However, looking at it from a different angle, what he's actually done is create a kind of super compiler that links executables together as if they were methods that one would find in the cygwin1.dll. In fact, neither he nor anyone else can run bash without the cygwin1.dll. By slurping up the cygwin1.dll into a proprietary binary structure and combining it not only with the proprietary environment that enables it's operation, but potentially with other standalone binary utilities that may run the gamut from open source, to shareware to extremely proprietary - i.e - one license per machine for sequential, not simultaneous processing only - how does that differ conceptually from a compiler linking a proprietary method with a cygwin method found in the cygwin1.dll into a stand-alone executable??? Such a use - if one hopes to distribute such a binary for a fee - requires a contractual agreement with Red Hat, along with payment of a rather NOT insignificant (if I may say so myself) licensing fee. This utility looks like a great way to trojan horse licensed software into a binary structure that would mask it's very existence. A pirate's dream come true perhaps - at least when used and examined by unsophisticated users - which make up the vast majority. This is why I urged Mr. Clark to consult an experienced attorney. He needs to inform his user base that licensing restrictions apply to everything bundled with his utility - and even if the final executable is not sold for a fee, but distributed in such a way that the stand-alone executable may change hands many times, it may even be necessary that the source for any and all GPL'd software be included IN the executable itself in such a way that it can be exported to a text file with a command line switch. Certainly including the source on a distribution cd and the website goes a long way towards satisfying this requirement. However, one way such a binary would be useful would be in a process chain - where the bundle actually is used to support a distributed operation B2B, over the internet, etc. In such a chain the final package would arrive after traveling through many highways. Traversing such journeys with source in tow may not always be practical. Furthermore, new and uneducated recipient's of such an executable NEED to know about it's contents, the licensing restrictions of those contents, and prohibitions of duplicating the bundled executable without satisfying ALL licensing restrictions of ALL its contents - INCLUDING the distribution of source for any and all bundled GPL software. Brian Kelly Christopher Faylor [EMAIL PROTECTED]@cygwin.com on 07/28/2003 11:19:04 PM Please respond to [EMAIL PROTECTED] Sent by:[EMAIL PROTECTED] To:[EMAIL PROTECTED] cc: (bcc: Brian Kelly/WTC1/Empire) Subject:Re: Source code for binaries offered at http://thinstall.com/ ? On Mon, Jul 28, 2003 at 10:12:07PM -0400, Jon A. Lambert wrote: From: Carlo Florendo [EMAIL PROTECTED] From: Jonathan Clark [EMAIL PROTECTED] Subject: RE: Source code for binaries offered at http://thinstall.com/ ? Hello Max, This is a good point. I have the source downloaded and in backups somewhere around here - so it can be located if needed. While I'm in the process of putting together a new archive that contains more GPL disclaimers (and more utilities), I wish to be in compliance by offering a CDROM for those that desire it. http://www.gnu.org/licenses/gpl-faq.html#SourceAndBinaryOnDifferentSites GPL 3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following: ... b) Accompany it with a written offer, valid for at least three ^ This would require sending a written offer to every person who purchases your software. Written is not email. It's an actual letter. I'll say it again: We have consulted with experts. The FSF FAQ entry is really correct here. If you are offering binaries on a web site, you need to offer sources on the web site, too. cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ WellChoice, Inc. made the following annotations on 07/29/2003 08:55:49 AM -- Attention! This electronic message contains information that may be legally confidential and/or privileged. The information is intended
And now a problem with tcl
I've been searching for a long time about this
problem.
The command runs normally on Unix but not on Cygwin :(
The problem concerns a command button which runs a
sh-script ...
Here is the syntax :
button .frPrincipal.thug\
-activebackground #beccbe \
-background #d2e0d2 \
-command {exec scriptsh } \
-cursor {hand1} \
-font $font \
-height {2} \
-text {run scriptsh} \
-width {22}
The directory containing the scriptsh file is defined
in the PATH both on Cygwin and Unix.
But it runs on Unix and not on Cygwin.
Actually, i must type exec sh /home/.../scriptsh to
run it ...
So please, what can i do solve this problem ?
___
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
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
how to activate registry changes
Hi, I have the problem, that I change values in the registry to activate a screensaver. But after I wrot the new values into the registry nothing happend. I can see the new values in the registry. When I open the dialog in Windows the old values are still shown. How can I force windows to accept the new values? Thanks Franz __ ComputerBild 15-03 bestaetigt: Den besten Spam-Schutz gibt es bei WEB.DE FreeMail - Deutschlands beste E-Mail - http://s.web.de/?mc=021121 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: how to activate registry changes
On Tue, Jul 29, 2003 at 04:05:45PM +0200, [EMAIL PROTECTED] wrote: Hi, I have the problem, that I change values in the registry to activate a screensaver. But after I wrot the new values into the registry nothing happend. I can see the new values in the registry. When I open the dialog in Windows the old values are still shown. How can I force windows to accept the new values? Reboot ;-) (SCNR). Gruss Olaf Föllinger -- Olaf Föllinger Berater S.E.S.A. Software und Systeme AG Alt-Moabit 91a D-10559 Berlin Germany Tel: +49 30 390722 -291 Fax: +49 30 390722 -222 Mobil: +49 173 6227080 http://www.sesa.de mailto: [EMAIL PROTECTED] -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: how to activate registry changes
On Tue, 29 Jul 2003 [EMAIL PROTECTED] wrote: Hi, I have the problem, that I change values in the registry to activate a screensaver. But after I wrot the new values into the registry nothing happend. I can see the new values in the registry. When I open the dialog in Windows the old values are still shown. How can I force windows to accept the new values? This does not seem to be Cygwin related what so ever, so please take your question elsewhere, to say, a Windows forum or newsgroup. Elfyn -- Elfyn McBratney, EMCB | http://www.nongnu.org/wwwauth/ http://www.emcb.co.uk | http://www.emcb.co.uk/webauth/ [EMAIL PROTECTED] | wwwauth-users AT nongnu DOT org -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
And now a problem with tcl
I've been searching for a long time about this
problem.
The command runs normally on Unix but not on Cygwin :(
The problem concerns a command button which runs a
sh-script ...
Here is the syntax :
button .frPrincipal.thug\
-activebackground #beccbe \
-background #d2e0d2 \
-command {exec scriptsh } \
-cursor {hand1} \
-font $font \
-height {2} \
-text {run scriptsh} \
-width {22}
The directory containing the scriptsh file is defined
in the PATH both on Cygwin and Unix.
But it runs on Unix and not on Cygwin.
Actually, i must type exec sh /home/.../scriptsh to
run it ...
So please, what can i do solve this problem ?
___
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
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: DLL 1.3.22 - etags 5.5.4 fails on XP Pro:K@N@!:
Hannu E K Nevalainen (garbage mail) [EMAIL PROTECTED] writes: The solution which should have been much more obvious to me is to use ctags.exe with the '-e' option. I'm assuming etags is a link to ctags. Jeff From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Igor Pechtchanski --8-- What you could do is create an etags.bat that runs 'bash -c etags %1 %2 %3 %4 %5 %6 %7 %8 %9', put it in your PATH, and you should be able to call that from NTEmacs. Igor Hmm... I can't remember if there is a equiv. of $@ in DOS/BAT-files... $ man bash /\$\@ /Hannu E K Nevalainen, B.Sc EE/Microcomputer systems, 59°14'N, 17°12'E. --END OF MESSAGE-- -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ -- Thanks, Jeff , | Jeffery B. Rancier | | Softechnics | a METTLER TOLEDO company ` -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Limit to size of pipe
Hello!
Further to an earlier message I sent, I now attach an example. This
tries to send a message of a fixed sized down a pipe. Under windows I
can't seem to send much more than 25k down in one go, although I can
send more if I chop it up into sections. Under linux however I can send
as large amounts as I like. While it is possible to work around it, I
thought I would mention it in case it was easy to fix, just no-one had
requested it :)
Example program follows: setting MSGSIZE25000ish on my computer causes
fail (ie pipeval=-1)
#include unistd.h
#include stdio.h
#define MSGSIZE 23000
char *msg1 = message;
void main(void) {
char inbuf[MSGSIZE];
int p[2], j;
int pid;
if(pipe(p) == -1) exit(1);
switch(pid = fork()){
case -1:exit(2);
case 0: /* if child then write down pipe */
close(p[0]); /* first close the read end of the pipe */
write(p[1], msg1, MSGSIZE);
break;
default: /* parent reads pipe */
close(p[1]); /* first close the write end of the pipe */
int pipeval=read(p[0], inbuf, MSGSIZE);
printf(pipeval:%d\n,pipeval);
printf(Parent sent: %s\n, inbuf);
wait(NULL);
}
}
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
[ANNOUNCEMENT] Updated: docbook-xsl-1.61.3-2
I've updated the docbook-xsl package to version 1.61.3-2. docbook-xsl package contains XSL stylesheets for the DocBook XML DTD created by Norman Walsh and others. Changes since 1.61.3-1: - Added 'profiling' directory To update your installation, click on the Install Cygwin now link on the http://cygwin.com/ web page. This downloads setup.exe to your system. Then, run setup and answer all of the questions. If you have questions or comments, please send them to the Cygwin mailing list at: cygwin at cygwin dot com. I would appreciate it if you would use this mailing list rather than emailing me directly. -- +---+ | Marcel Telka e-mail: [EMAIL PROTECTED] | |homepage: http://telka.sk/ | |jabber: [EMAIL PROTECTED] | +---+ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: 1.5.1: lseek64 woes (was Re: 1.5.1: unzip problems with binary files in archive)
On Mon, Jul 28, 2003 at 11:41:43PM -0700, David Rothenberger wrote:
Content-Description: message body text
Charles Wilson writes:
Christopher Faylor wrote:
I'm getting a strange error from unzip when trying to extract an
archive that contains binary files. Everything works fine if the
archive contains just text files.
Can't duplicate it. Sorry.
I --can-- duplicate it. But I don't know why it happens; I'll have to
debug it...but not until after I update to 1.5.1. FWIW, zip does use
fdopen().
I've done some debugging and I have a STC. I have no idea why it
fails, and why Chris doesn't see it, but here's the info.
The STC is composed of the two attached files: works.c and fails.c.
Both files essentially do
int fd = open(works.c, 0);
lseek(fd, 1, 0);
but works.c does this directly in main(), while fails.c does this in
unzip() called from main(). As you can guess, lseek() returns -1 in
fails.c and 1 in works.c.
And here's why fails.c fails:
--- lseek-fails.c~ 2003-07-29 11:43:29.428769079 -0400
+++ lseek-fails.c 2003-07-29 11:46:01.911652780 -0400
@@ -1,4 +1,5 @@
#include stdio.h
+#include unistd.h
void unzip()
{
You need to include the prototype for lseek.
cgf
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: DLL 1.3.22 - etags 5.5.4 fails on XP Pro:K@N@!:
Exactly. That's why I mentioned using ctags directly in http://cygwin.com/ml/cygwin/2003-07/msg01331.html Doing so should remove the magic, mystery, and the problem. Larry Jeffery B. Rancier wrote: Hannu E K Nevalainen (garbage mail) [EMAIL PROTECTED] writes: The solution which should have been much more obvious to me is to use ctags.exe with the '-e' option. I'm assuming etags is a link to ctags. Jeff From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Igor Pechtchanski --8-- What you could do is create an etags.bat that runs 'bash -c etags %1 %2 %3 %4 %5 %6 %7 %8 %9', put it in your PATH, and you should be able to call that from NTEmacs. Igor Hmm... I can't remember if there is a equiv. of $@ in DOS/BAT-files... $ man bash /\$\@ /Hannu E K Nevalainen, B.Sc EE/Microcomputer systems, 59°14'N, 17°12'E. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Limit to size of pipe
chris writes:
Hello!
Further to an earlier message I sent, I now attach an example. This
tries to send a message of a fixed sized down a pipe. Under windows I
can't seem to send much more than 25k down in one go, although I can
send more if I chop it up into sections. Under linux however I can send
as large amounts as I like. While it is possible to work around it, I
thought I would mention it in case it was easy to fix, just no-one had
requested it :)
Example program follows: setting MSGSIZE25000ish on my computer causes
fail (ie pipeval=-1)
#include unistd.h
#include stdio.h
#define MSGSIZE 23000
char *msg1 = message;
I modified the program to allocate the message buffer to send on the
heap and to initialize the entire thing. The test seems to work for
any size at that point.
Dave
#include unistd.h
#include stdio.h
char *msg1;
void initmessage (int msgsize)
{
msg1 = (char*) malloc(sizeof(char) * msgsize);
int i;
for (i=0; imsgsize; ++i) {
msg1[i] = i%255+1;
}
}
int main (int argc, char** argv)
{
char *inbuf;
int p[2], j;
int pid;
int msgsize = atoi(argv[1]);
inbuf = (char*) malloc(sizeof(char) * msgsize);
initmessage(msgsize);
if(pipe(p) == -1) exit(1);
switch(pid = fork()){
case -1:exit(2);
case 0: /* if child then write down pipe */
close(p[0]); /* first close the read end of the pipe */
write(p[1], msg1, msgsize);
break;
default: /* parent reads pipe */
close(p[1]); /* first close the write end of the pipe */
int pipeval=read(p[0], inbuf, msgsize);
printf(pipeval:%d\n,pipeval);
if (pipeval 60) {
inbuf[60] = '\0';
}
printf(Parent sent: %s\n, inbuf);
wait(NULL);
}
}
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: And now a problem with tcl
On Tue, 29 Jul 2003, philippe guillaume wrote:
I've been searching for a long time about this
problem.
The command runs normally on Unix but not on Cygwin :(
The problem concerns a command button which runs a
sh-script ...
Here is the syntax :
button .frPrincipal.thug\
-activebackground #beccbe \
-background #d2e0d2 \
-command {exec scriptsh } \
-cursor {hand1} \
-font $font \
-height {2} \
-text {run scriptsh} \
-width {22}
The directory containing the scriptsh file is defined
in the PATH both on Cygwin and Unix.
But it runs on Unix and not on Cygwin.
Actually, i must type exec sh /home/.../scriptsh to
run it ...
So please, what can i do solve this problem ?
IIRC, tcl doesn't understand POSIX paths. So, your solution (exec sh ...)
is correct (you could also most likely omit the full path from scriptsh,
i.e, use exec sh scriptsh, provided scriptsh *is* in the PATH).
Igor
--
http://cs.nyu.edu/~pechtcha/
|\ _,,,---,,_[EMAIL PROTECTED]
ZZZzz /,`.-'`'-. ;-;;,_[EMAIL PROTECTED]
|,4- ) )-,_. ,\ ( `'-' Igor Pechtchanski, Ph.D.
'---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow!
I have since come to realize that being between your mentor and his route
to the bathroom is a major career booster. -- Patrick Naughton
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: 1.5.1: lseek64 woes (was Re: 1.5.1: unzip problems with binary files in archive)
Christopher Faylor wrote:
On Mon, Jul 28, 2003 at 11:41:43PM -0700, David Rothenberger wrote:
Content-Description: message body text
Charles Wilson writes:
Christopher Faylor wrote:
I'm getting a strange error from unzip when trying to extract an
archive that contains binary files. Everything works fine if the
archive contains just text files.
Can't duplicate it. Sorry.
I --can-- duplicate it. But I don't know why it happens; I'll have to
debug it...but not until after I update to 1.5.1. FWIW, zip does use
fdopen().
I've done some debugging and I have a STC. I have no idea why it
fails, and why Chris doesn't see it, but here's the info.
The STC is composed of the two attached files: works.c and fails.c.
Both files essentially do
int fd = open(works.c, 0);
lseek(fd, 1, 0);
but works.c does this directly in main(), while fails.c does this in
unzip() called from main(). As you can guess, lseek() returns -1 in
fails.c and 1 in works.c.
And here's why fails.c fails:
--- lseek-fails.c~ 2003-07-29 11:43:29.428769079 -0400
+++ lseek-fails.c 2003-07-29 11:46:01.911652780 -0400
@@ -1,4 +1,5 @@
#include stdio.h
+#include unistd.h
void unzip()
{
You need to include the prototype for lseek.
cgf
This is also why unzip fails.
--- unzip-5.50/unzpriv.h2003-07-29 09:10:41.325502400 -0700
+++ unzip-5.50-fixed/unzpriv.h 2003-07-29 09:08:48.783675200 -0700
@@ -200,6 +200,10 @@
typedef unsigned int extent;
#endif /* ?MODERN */
+#ifdef __CYGWIN__
+# include unistd.h
+#endif
+
#ifndef MINIX/* Minix needs it after all the other includes
(?) */
# include stdio.h
I spent WAY too much time on this for such a simple problem. Darn!
Next time I'll know to look for prototypes if the strace output is
strange.
Dave
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: Limit to size of pipe
On Tue, 29 Jul 2003, chris wrote: David Rothenberger wrote: chris writes: Hello! Further to an earlier message I sent, I now attach an example. This tries to send a message of a fixed sized down a pipe. Under windows I can't seem to send much more than 25k down in one go, although I can send more if I chop it up into sections. Under linux however I can send as large amounts as I like. While it is possible to work around it, I thought I would mention it in case it was easy to fix, just no-one had requested it :) Example program follows: setting MSGSIZE25000ish on my computer causes fail (ie pipeval=-1) #include unistd.h #include stdio.h #define MSGSIZE 23000 char *msg1 = message; I modified the program to allocate the message buffer to send on the heap and to initialize the entire thing. The test seems to work for any size at that point. snip program Thanks! I notice (by some fiddling) that it seems I have to instansiate (at least most of) the buffer before I send it. Is there some rule that you should instansiate memory before reading it? of course doing so is sensible, but I didn't know nessasary? Out of interest (because I couldn't actually find a 'mission plan'), what is the plan of Cygwin? to create a system wherebye any linux / unix program will compile without changes under windows? Or to simply make it much easier to convert but not try to support stupid activities? Having said that, I've been convinced I should change the program I'm converting from *BSD to not send stupidly large mostly-empty buffers anyway :) Chris Cygwin's purpose is to provide a POSIX layer that helps in compiling *portable* programs. Relying on the C runtime to initialize automatic arrays is non-portable, and probably won't work on many systems (including Cygwin). When in doubt, look at the SUS (which Cygwin tries to follow) -- if some implicit behavior is unspecified there, it's very likely to be different on various Unix systems anyway, and you should explicitly provide the behavior your program expects (e.g., initialize memory). Igor -- http://cs.nyu.edu/~pechtcha/ |\ _,,,---,,_[EMAIL PROTECTED] ZZZzz /,`.-'`'-. ;-;;,_[EMAIL PROTECTED] |,4- ) )-,_. ,\ ( `'-' Igor Pechtchanski, Ph.D. '---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow! I have since come to realize that being between your mentor and his route to the bathroom is a major career booster. -- Patrick Naughton -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Source code for binaries offered at http://thinstall.com/unix_tools ?
I will write up some documentation soon about the GPL issue, but a quick clarification: 1. Some emails have mentioned unix_tools as being commercial and that I have customers who purchased it. unix_tools is free (as in beer), always has been, always will be - it is something I made to be a handy tool and to show off the power of Thinstall and make freely available at no charge. I've found a lot of people find it is a handy utility to have on hand, especially since it is runnable from a floppy. I realize the fact that unix_tools is free does not change any GPL issues, but just wanted to point this out. 2. Thinstall does not modify the binary structure (i.e. bytes) of files it links together other than to provide compression. In this manner it's no different from zip or tar. All files can be copied from the compressed file system to hard drive with their exact original contents. This is easily demonstrated by running the bash example and typing the command cp -R * /cygdrive/c. The term link is used in documentation to illustrate the point that the files may be used without extracting to disk - however in fact, there is no link between any files except that they are all located in the same archive. compile, link, etc are used as marketing terms to illustrate this is new technology - not a self-extracting ZIP where files must first be copied outside the archive - but Thinstall in no way resembles a compiler or linker at the implementation level. The Thinstall OS is able to run EXEs and load DLLs both inside and outside of the archive. Simply running an GPLed EXE located on in a compressed filesystem can not invoke GPL on the loading OS or most operating systems would fall under this umbrella. Likewise, having an OS located in the same archive as GPL software would not somehow require it to be GPLed or we could open source Windows tomorrow because I do this all the time with VMWare. :) Thinstall works with all EXEs, DLLs, and files, and has no direct tie or dependency on any GPL software or components. More user-info related to this discussion: http://thinstall.com/help/index.html?virtualoperatingsystem.htm (The OS) http://thinstall.com/help/createprocessshellexecuteo.htm (Loading external EXEs) http://thinstall.com/help/index.html?externaldllloading.htm (Loading external DLLs) Best Regards, Jonathan Clark -Original Message- From: Williams, Gerald S (Jerry) [mailto:[EMAIL PROTECTED] Sent: Tuesday, July 29, 2003 4:58 AM To: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Subject: RE: Source code for binaries offered at http://thinstall.com/ ? Brian Kelly wrote: Please, seek out and consult with an experienced software and intellectual property rights attorney at your earliest possible opportunity. You've got a very nifty little utility, but then again, so did Napster. [ ... ] I know enough to see that there are lot of legal issues you have not investigated in depth - and I STRONGLY advise that you do so with the assistance of an experienced counselor. Jonathan, I don't want to tie up our mailing lists with this, and I'm sure the GPL licensing lists address it better, but please listen to him if you haven't already done so. Your online help indicates that Thinstall *links* target libraries, executables, and such. That certainly sounds like it would trigger the GPL (i.e., everything else you link into that executable would have to fall within the GPL guidelines). Personally, I'd get something in writing from the FSF or RMS or somebody saying that this doesn't trigger the GPL, refer to that fact liberally throughout the help file, *and* change the wording to make it read more like creating a self-extracting archive (even though it may be linking the files underneath). gsw P.S. It really does look like a neat program you've got. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Local printer access question..
--- Hannu E K Nevalainen (garbage mail) [EMAIL PROTECTED] wrote: From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Igor Pechtchanski On Mon, 28 Jul 2003, Rick Rankin wrote: --- Igor Pechtchanski [EMAIL PROTECTED] wrote: On Mon, 28 Jul 2003, Hannu E K Nevalainen (garbage mail) wrote: From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Igor Pechtchanski On Sun, 27 Jul 2003, Ishwar Rattan wrote: Is there a way to access the printer (Deskjet 694C on parallel port)?? If your Windows printer name is Deskjet 694C, try 'lpr -P Deskjet 694C'. You can also 'export PRINTER=Deskjet 694C'. Igor I wish it was that easy... I have a Deskjet 720C which should be about the same thing as the one above. I've tried all of the possible ways described in man lpr. All I get is a document showing up Start Settings Printers HP720C indicating that it is printing. The printer itself does nothing. This seems not be cygwin failing but windows or the driver - as the behaviour is the same from cmd.exe using PRINT /D:all possibilities tried Well, it *is* that easy for me on Win2k. I haven't tried it on other systems, there might be some peculiarities on 9x/ME/XP. I still have my W98SE runnable, but I am almost constantly using the dual-boot W2K currently. All this has been tried there. $ uname -a CYGWIN_NT-5.0 P450 1.3.22(0.78/3/2) 2003-03-18 09:20 i686 unknown unknown Cygwin setup.exe used very recently (Yesterday? no test/exp-packages). Some things to try that I can think of are: renaming your printer, or making it shared and using the UNC notation to access it (e.g., export PRINTER='\\MYCOMPUTER\HP720C'). This doesn't sound like a naming problem to me. Hannu indicates that the document shows up on the print queue, but the printer does nothing. That tells me that lpr and the DOS print command are finding the print queue properly. Exactly. Was my writing unclear? Please tell me where if so - private email please! - In hope this will improve my english. What kind of file are you trying to print and *exactly* what command are you using to print it? See below for an example. --Rick Oh. Yes, it seems that I misunderstood Hannu's message... Thanks, Rick. Hannu, Well, the obvious question first: is the printer paused? Nope. Can you print from Windows apps? Yes I can. :-} Also, try setting up a dummy printer that prints to file and see if that works. Igor Hrm... why didn't I think of that! :-P Well, here we go: Start Settings Printers Add Printer - Add printer wizard Next - Local Printer DONT Automatically Detect ... Next - Use following port FILE: Next - Manufacturer and model Generic Generic / Text only Next - Printer name: Generic / Text only Default: No Next - Share as: GT Next - Location and Comment: Next - Print a test page: No Next - Finish F: print /D:\\P450\GT C:\Autoexec.bat Filename: d:\zz.txt $ diff -up /cygdrive/d/zz.txt /cygdrive/c/AUTOEXEC.BAT --- /cygdrive/d/zz.txt 2003-07-29 13:05:50.0 +0200 +++ /cygdrive/c/AUTOEXEC.BAT2003-05-18 20:41:22.0 +0200 @@ -59,4 +59,3 @@ Rem TShoot: echo Autoexec done... - \ No newline at end of file $ lpr -P //P450/GT /cygdrive/c/AUTOEXEC.BAT [EMAIL PROTECTED] ~ bash PID=2028, Last xc=0 $ diff -up /cygdrive/d/zz.txt /cygdrive/c/AUTOEXEC.BAT --- /cygdrive/d/zz.txt 2003-07-29 13:09:20.0 +0200 +++ /cygdrive/c/AUTOEXEC.BAT2003-05-18 20:41:22.0 +0200 @@ -59,4 +59,3 @@ Rem TShoot: echo Autoexec done... - $ Works like a charm... so this should prove one thing: Cheap HP printers aren't up to all the tasks you could expect. :-/ I suspect there is a problem WRT printer sharing in this driver too. I've tried this on W98SE earlier - got very few pages out of it, then it died. No resurrection possible whatever I tried. And then, you ask whether there is a more recent driver: http://h20015.www2.hp.com/en/document.jhtml?docName=bpd09190lc=en There is no need to download a driver from the Web site if you selected a download link and were brought here. The most current driver is already integrated into Windows 2000 and it is already on every Windows 2000 PC. Follow the appropriate instructions below to install the correct printer driver using the Add Printer Wizard. NOTE: The above webpage gives indication of that the Network Printer stuff SHOULD work. I don't know what to think about it. The descriptions are very general and matches my actions exactly. (I have tried to vary from the basics, where possible and where I have deemed apropriate - no go. Details forgotten by now). I don't know about the DOS print command, but lpr is just a raw spooler. It is intended to be used as a filter
RE: Local printer access question..
On Tue, 29 Jul 2003, Rick Rankin wrote: --- Hannu E K Nevalainen (garbage mail) [EMAIL PROTECTED] wrote: From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Igor Pechtchanski On Mon, 28 Jul 2003, Rick Rankin wrote: --- Igor Pechtchanski [EMAIL PROTECTED] wrote: On Mon, 28 Jul 2003, Hannu E K Nevalainen (garbage mail) wrote: From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Igor Pechtchanski On Sun, 27 Jul 2003, Ishwar Rattan wrote: Is there a way to access the printer (Deskjet 694C on parallel port)?? If your Windows printer name is Deskjet 694C, try 'lpr -P Deskjet 694C'. You can also 'export PRINTER=Deskjet 694C'. Igor I wish it was that easy... I have a Deskjet 720C which should be about the same thing as the one above. I've tried all of the possible ways described in man lpr. All I get is a document showing up Start Settings Printers HP720C indicating that it is printing. The printer itself does nothing. This seems not be cygwin failing but windows or the driver - as the behaviour is the same from cmd.exe using PRINT /D:all possibilities tried Well, it *is* that easy for me on Win2k. I haven't tried it on other systems, there might be some peculiarities on 9x/ME/XP. I still have my W98SE runnable, but I am almost constantly using the dual-boot W2K currently. All this has been tried there. $ uname -a CYGWIN_NT-5.0 P450 1.3.22(0.78/3/2) 2003-03-18 09:20 i686 unknown unknown Cygwin setup.exe used very recently (Yesterday? no test/exp-packages). Some things to try that I can think of are: renaming your printer, or making it shared and using the UNC notation to access it (e.g., export PRINTER='\\MYCOMPUTER\HP720C'). This doesn't sound like a naming problem to me. Hannu indicates that the document shows up on the print queue, but the printer does nothing. That tells me that lpr and the DOS print command are finding the print queue properly. Exactly. Was my writing unclear? No, no, it was my late night reading... Please tell me where if so - private email please! - In hope this will improve my english. What kind of file are you trying to print and *exactly* what command are you using to print it? See below for an example. --Rick Oh. Yes, it seems that I misunderstood Hannu's message... Thanks, Rick. Hannu, Well, the obvious question first: is the printer paused? Nope. Can you print from Windows apps? Yes I can. :-} Also, try setting up a dummy printer that prints to file and see if that works. Igor Hrm... why didn't I think of that! :-P Well, here we go: Start Settings Printers Add Printer - Add printer wizard Next - Local Printer DONT Automatically Detect ... Next - Use following port FILE: Next - Manufacturer and model Generic Generic / Text only Next - Printer name: Generic / Text only Default: No Next - Share as: GT Next - Location and Comment: Next - Print a test page: No Next - Finish F: print /D:\\P450\GT C:\Autoexec.bat Filename: d:\zz.txt $ diff -up /cygdrive/d/zz.txt /cygdrive/c/AUTOEXEC.BAT --- /cygdrive/d/zz.txt 2003-07-29 13:05:50.0 +0200 +++ /cygdrive/c/AUTOEXEC.BAT2003-05-18 20:41:22.0 +0200 @@ -59,4 +59,3 @@ Rem TShoot: echo Autoexec done... - \ No newline at end of file $ lpr -P //P450/GT /cygdrive/c/AUTOEXEC.BAT [EMAIL PROTECTED] ~ bash PID=2028, Last xc=0 $ diff -up /cygdrive/d/zz.txt /cygdrive/c/AUTOEXEC.BAT --- /cygdrive/d/zz.txt 2003-07-29 13:09:20.0 +0200 +++ /cygdrive/c/AUTOEXEC.BAT2003-05-18 20:41:22.0 +0200 @@ -59,4 +59,3 @@ Rem TShoot: echo Autoexec done... - $ Works like a charm... so this should prove one thing: Cheap HP printers aren't up to all the tasks you could expect. :-/ I suspect there is a problem WRT printer sharing in this driver too. I've tried this on W98SE earlier - got very few pages out of it, then it died. No resurrection possible whatever I tried. And then, you ask whether there is a more recent driver: http://h20015.www2.hp.com/en/document.jhtml?docName=bpd09190lc=en There is no need to download a driver from the Web site if you selected a download link and were brought here. The most current driver is already integrated into Windows 2000 and it is already on every Windows 2000 PC. Follow the appropriate instructions below to install the correct printer driver using the Add Printer Wizard. NOTE: The above webpage gives indication of that the Network Printer stuff SHOULD work. I don't know what to think about it. The descriptions are very general and matches my actions exactly. (I have tried to vary from the basics, where possible
RE: Source code for binaries offered at http://thinstall.com/unix_tools ?
As long as the *complete* binary contents can be extracted then I would essentially agree that Thininstall is fundamentally no different than an ordinary zip archive. No one has implied to my knowledge - especially me - that you were *actually* selling unix_tools - but that there certainly would be issues if one ever contemplated selling a binary of any kind that contained GPL'd software - even if the GPL'd software was used only to install a separate package and did not constitute the functionality actually being sold. And while being able to put unix_tools on a floppy is a cool thing, from a legal standpoint, anyone who permanently passes that floppy or makes a copy for others to use is in fact distributing GPL'd software and must also have the exact source *readily* available - i.e., on the floppy or archived within the Thininstall created binary (or archive if one so pleases) itself. Of course anyone who creates their own floppy for their own use is exempted. Brian Kelly Jonathan Clark [EMAIL PROTECTED]@cygwin.com on 07/29/2003 12:47:59 PM Sent by:[EMAIL PROTECTED] To:[EMAIL PROTECTED] cc: (bcc: Brian Kelly/WTC1/Empire) Subject:RE: Source code for binaries offered at http: //thinstall.com/unix_tools ? I will write up some documentation soon about the GPL issue, but a quick clarification: 1. Some emails have mentioned unix_tools as being commercial and that I have customers who purchased it. unix_tools is free (as in beer), always has been, always will be - it is something I made to be a handy tool and to show off the power of Thinstall and make freely available at no charge. I've found a lot of people find it is a handy utility to have on hand, especially since it is runnable from a floppy. I realize the fact that unix_tools is free does not change any GPL issues, but just wanted to point this out. 2. Thinstall does not modify the binary structure (i.e. bytes) of files it links together other than to provide compression. In this manner it's no different from zip or tar. All files can be copied from the compressed file system to hard drive with their exact original contents. This is easily demonstrated by running the bash example and typing the command cp -R * /cygdrive/c. The term link is used in documentation to illustrate the point that the files may be used without extracting to disk - however in fact, there is no link between any files except that they are all located in the same archive. compile, link, etc are used as marketing terms to illustrate this is new technology - not a self-extracting ZIP where files must first be copied outside the archive - but Thinstall in no way resembles a compiler or linker at the implementation level. The Thinstall OS is able to run EXEs and load DLLs both inside and outside of the archive. Simply running an GPLed EXE located on in a compressed filesystem can not invoke GPL on the loading OS or most operating systems would fall under this umbrella. Likewise, having an OS located in the same archive as GPL software would not somehow require it to be GPLed or we could open source Windows tomorrow because I do this all the time with VMWare. :) Thinstall works with all EXEs, DLLs, and files, and has no direct tie or dependency on any GPL software or components. More user-info related to this discussion: http://thinstall.com/help/index.html?virtualoperatingsystem.htm (The OS) http://thinstall.com/help/createprocessshellexecuteo.htm (Loading external EXEs) http://thinstall.com/help/index.html?externaldllloading.htm (Loading external DLLs) Best Regards, Jonathan Clark -Original Message- From: Williams, Gerald S (Jerry) [mailto:[EMAIL PROTECTED] Sent: Tuesday, July 29, 2003 4:58 AM To: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Subject: RE: Source code for binaries offered at http://thinstall.com/ ? Brian Kelly wrote: Please, seek out and consult with an experienced software and intellectual property rights attorney at your earliest possible opportunity. You've got a very nifty little utility, but then again, so did Napster. [ ... ] I know enough to see that there are lot of legal issues you have not investigated in depth - and I STRONGLY advise that you do so with the assistance of an experienced counselor. Jonathan, I don't want to tie up our mailing lists with this, and I'm sure the GPL licensing lists address it better, but please listen to him if you haven't already done so. Your online help indicates that Thinstall *links* target libraries, executables, and such. That certainly sounds like it would trigger the GPL (i.e., everything else you link into that executable would have to fall within the GPL guidelines). Personally, I'd get something in writing from the FSF or RMS or somebody saying that this doesn't trigger the GPL, refer to that fact liberally throughout the help file, *and* change the wording to make it read more like creating a self-extracting archive (even though it may be linking
Re: Source code for binaries offered at http://thinstall.com/unix_tools?
Jonathan Clark wrote: I will write up some documentation soon about the GPL issue, but a quick clarification: 1. unix_tools is free (as in beer) [...] I realize the fact that unix_tools is free does not change any GPL issues, [...] 2. Thinstall does not modify the binary structure (i.e. bytes) of files it links together other than to provide compression. In this manner it's no different from zip or tar. Thanks for clarifying this, and I'm sure the demo is a very useful tool. But it would perhaps be best if your demo was just a script or something to use Thinstall on a _customer-downloaded_ copy of cygwin to create the demo image. I.e. make available to your customers the script, and the instructions to: * Install cygwin base * Run the demo script to build the demo unix_tools binary * Enjoy. This way, everyone stays clear of the GPL issues. Alternatively, you can also make available the src tarballs for the version of cygwin and cygwin-ported tools that you created your unix_tools image with. -- Shankar. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Local printer access question..
--- Igor Pechtchanski [EMAIL PROTECTED] wrote: On Tue, 29 Jul 2003, Rick Rankin wrote: --- Hannu E K Nevalainen (garbage mail) [EMAIL PROTECTED] wrote: From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Igor Pechtchanski On Mon, 28 Jul 2003, Rick Rankin wrote: --- Igor Pechtchanski [EMAIL PROTECTED] wrote: On Mon, 28 Jul 2003, Hannu E K Nevalainen (garbage mail) wrote: From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Igor Pechtchanski On Sun, 27 Jul 2003, Ishwar Rattan wrote: Is there a way to access the printer (Deskjet 694C on parallel port)?? If your Windows printer name is Deskjet 694C, try 'lpr -P Deskjet 694C'. You can also 'export PRINTER=Deskjet 694C'. Igor I wish it was that easy... I have a Deskjet 720C which should be about the same thing as the one above. I've tried all of the possible ways described in man lpr. All I get is a document showing up Start Settings Printers HP720C indicating that it is printing. The printer itself does nothing. This seems not be cygwin failing but windows or the driver - as the behaviour is the same from cmd.exe using PRINT /D:all possibilities tried Well, it *is* that easy for me on Win2k. I haven't tried it on other systems, there might be some peculiarities on 9x/ME/XP. I still have my W98SE runnable, but I am almost constantly using the dual-boot W2K currently. All this has been tried there. $ uname -a CYGWIN_NT-5.0 P450 1.3.22(0.78/3/2) 2003-03-18 09:20 i686 unknown unknown Cygwin setup.exe used very recently (Yesterday? no test/exp-packages). Some things to try that I can think of are: renaming your printer, or making it shared and using the UNC notation to access it (e.g., export PRINTER='\\MYCOMPUTER\HP720C'). This doesn't sound like a naming problem to me. Hannu indicates that the document shows up on the print queue, but the printer does nothing. That tells me that lpr and the DOS print command are finding the print queue properly. Exactly. Was my writing unclear? No, no, it was my late night reading... Please tell me where if so - private email please! - In hope this will improve my english. What kind of file are you trying to print and *exactly* what command are you using to print it? See below for an example. --Rick Oh. Yes, it seems that I misunderstood Hannu's message... Thanks, Rick. Hannu, Well, the obvious question first: is the printer paused? Nope. Can you print from Windows apps? Yes I can. :-} Also, try setting up a dummy printer that prints to file and see if that works. Igor Hrm... why didn't I think of that! :-P Well, here we go: Start Settings Printers Add Printer - Add printer wizard Next - Local Printer DONT Automatically Detect ... Next - Use following port FILE: Next - Manufacturer and model Generic Generic / Text only Next - Printer name: Generic / Text only Default: No Next - Share as: GT Next - Location and Comment: Next - Print a test page: No Next - Finish F: print /D:\\P450\GT C:\Autoexec.bat Filename: d:\zz.txt $ diff -up /cygdrive/d/zz.txt /cygdrive/c/AUTOEXEC.BAT --- /cygdrive/d/zz.txt 2003-07-29 13:05:50.0 +0200 +++ /cygdrive/c/AUTOEXEC.BAT2003-05-18 20:41:22.0 +0200 @@ -59,4 +59,3 @@ Rem TShoot: echo Autoexec done... - \ No newline at end of file $ lpr -P //P450/GT /cygdrive/c/AUTOEXEC.BAT [EMAIL PROTECTED] ~ bash PID=2028, Last xc=0 $ diff -up /cygdrive/d/zz.txt /cygdrive/c/AUTOEXEC.BAT --- /cygdrive/d/zz.txt 2003-07-29 13:09:20.0 +0200 +++ /cygdrive/c/AUTOEXEC.BAT2003-05-18 20:41:22.0 +0200 @@ -59,4 +59,3 @@ Rem TShoot: echo Autoexec done... - $ Works like a charm... so this should prove one thing: Cheap HP printers aren't up to all the tasks you could expect. :-/ I suspect there is a problem WRT printer sharing in this driver too. I've tried this on W98SE earlier - got very few pages out of it, then it died. No resurrection possible whatever I tried. And then, you ask whether there is a more recent driver: http://h20015.www2.hp.com/en/document.jhtml?docName=bpd09190lc=en There is no need to download a driver from the Web site if you selected a download link and were brought here. The most current driver is already integrated into Windows 2000 and it is already on every Windows 2000 PC. Follow the appropriate instructions below to install the correct printer driver using the Add Printer Wizard. NOTE: The above webpage gives indication of that the
Problem call a cygwin command from a dos batch file
I am having a problem with trying to sort a pipe-delimited file using the cygwin sort command from a dos batch file. If I enter the command in cygwin environment, it works fine: sort -t \| -k 3 -k 11 -k 12 -o sortedfile.txt unsortedfile.txt But I have problems running it from a dos batch file. It appears to work in Windows XP but not in Windows 2000. The command in the batch file is: bash -c sort -t \| -k 3 -k 11 -k 12 -o sortedfile.txt unsortedfile.txt In Win XP it executes without a problem, but in Win 2000, I get a message the says, Input file specified two times, as if it does not recognize the -o option. If I put the sort command in an echo, I get the results I would expect, so the -o is not lost: bash -c echo sort -t \| -k 3 -k 11 -k 12 -o sortedfile.txt unsortedfile.txt Rearranging the arguements and adding extra escape characters in various combinations don't seem to help. Any ideas, please? --- Lisbeth Kellogg Office: 651-675-2610 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
cygwin.com unavailable from blueyonder/telewest uk
I'm fresh to this list so please excuse if this is a common kind of question. I can't get to cygwin.com from my home PC. I wondered if this is because of problems with my telewest/blueyonder ISP; looking at the site from another ISP I don't see any problems. Whatever: My question is: can I install cygwin without access to cygwin.com, and if so, how? Here is what I've done: found a mirror site and downloaded setup.exe ftp://ftp.mirror.ac.uk/sites/sources.redhat.com/pub/cygwin/ run setup.exe It can't get the list of mirrors, presumably because it can't contact cygwin.com. So setup fails. I tried downloading setup.bz2 as well in the hope that setup.exe would find it before trying to download it, but no joy. From the FAQ: If you are downloading from the internet, setup will fail if it cannot download the list of mirrors at http://cygwin.com/mirrors.html. It could be that the network is too busy. Similarly for an ftp download site that isn't working. Try another mirror, or try again later. I would try another mirror if I knew how. Is there any way of configuring setup.exe not to use cygwin.com to get its list of mirror sites, or to specify exactly which mirror site should be used? any help much appreciated, Matt. -- == . Matt Fairtlough . == -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: cygwin.com unavailable from blueyonder/telewest uk
On 29 Jul 2003, Matt Fairtlough wrote: I'm fresh to this list so please excuse if this is a common kind of question. I can't get to cygwin.com from my home PC. I wondered if this is because of problems with my telewest/blueyonder ISP; looking at the site from another ISP I don't see any problems. Whatever: I use both Blueyonder and Freeserve and have no problems here, though this isn't the first time someone has asked this. I would try pinging cygwin.com or even a traceroute (ping and tracert from cmd.exe/command.com) and see what you come up with. At a last shot, call the support line (of Blueyonder) and ask them what's up. My question is: can I install cygwin without access to cygwin.com, and if so, how? Here is what I've done: Unfortunately not as you have noticed below, setup.exe requires the file mirrors.lst, which is located on cygwin.com . found a mirror site and downloaded setup.exe ftp://ftp.mirror.ac.uk/sites/sources.redhat.com/pub/cygwin/ run setup.exe It can't get the list of mirrors, presumably because it can't contact cygwin.com. So setup fails. I tried downloading setup.bz2 as well in the hope that setup.exe would find it before trying to download it, but no joy. From the FAQ: If you are downloading from the internet, setup will fail if it cannot download the list of mirrors at http://cygwin.com/mirrors.html. It could be that the network is too busy. Similarly for an ftp download site that isn't working. Try another mirror, or try again later. I would try another mirror if I knew how. Is there any way of configuring setup.exe not to use cygwin.com to get its list of mirror sites, or to specify exactly which mirror site should be used? any help much appreciated, Sorry I can't help you more. Elfyn -- Elfyn McBratney, EMCB | http://www.nongnu.org/wwwauth/ http://www.emcb.co.uk | http://www.emcb.co.uk/webauth/ [EMAIL PROTECTED] | wwwauth-users AT nongnu DOT org -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Local printer access question..
-Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Igor Pechtchanski Sent: Tuesday, July 29, 2003 7:16 PM Was my writing unclear? No, no, it was my late night reading... Ahh... glad to hear that ;-) Also, try setting up a dummy printer that prints to file and see if that works. Igor Hrm... why didn't I think of that! :-P --8-- Works like a charm... so this should prove one thing: Cheap HP printers aren't up to all the tasks you could expect. :-/ I don't know about the DOS print command, but lpr is just a raw spooler. It is intended to be used as a filter to send data *already formatted for a specific printer* to a print queue. In other words, it is intended to send postscript data to a postscript printer, or PCL data to a PCL-capable printer. lpr knows nothing about any printer and makes no attempt to inject printer-specific formatting codes. *If and only if* your printer can directly print a text file will using a command like lpr -P some_device file.txt work. You might try the following. Using notepad or a similar *Windows* app, try printing your autoexec.bat file to your Deskjet printer, but in the print dialog, check the box that says Print to file so that the formatted data stream ends up in a file instead of being sent directly to the printer. Then, try spooling that file using lpr and/or the dos print command. Let me know what happens. --Rick Using a single page HTML print from IE6 in zz.prn These do work: D:\print /D:LPT1: zz.prn D:\zz.prn is currently being printed D:\print /D:\\P450\DJ720C zz.prn D:\zz.prn is currently being printed This one has problems: $ cat /cygdrive/d/zz.prn | lpr -l -PLPT1: The first doc prints. Any following documents causes the printer to hang - turning it off is enough to be able to print yet another doc. On one occassion I got something that I've never seen before: Power, no paper and low ink leds rapidly flashing in that sequence ;-7 This one works $ cat /cygdrive/d/zz.prn | lpr -l $ echo $PRINTER //P450/DJ720C $ with one exception on all of them; they do not eject the page at the end of printing. Unless you start another document print... Hannu, You could also try using 'a2ps' or 'enscript', both of which are available as Cygwin packages. Igor Hmm... a2ps - ascii to postscript; How do I make that readable for a deskjet; I believe it talks PCL. enscript - ps, html, HTML, RTF, ANSI... but no PCL. Now I _assume_ that this is PCL - or maybe some other native stuff: $ cat /cygdrive/d/zz.prn | od -t x1z | head 000 24 01 00 18 01 86 00 18 07 00 00 01 00 00 00 00 $... 020 00 02 00 00 00 00 01 f4 01 00 00 00 24 00 00 04 $... 040 de ad be ef 24 01 00 18 01 8f 00 18 07 00 00 02 $... 060 00 00 00 04 00 02 00 00 de ad be ef 02 00 00 00 100 24 01 00 20 01 83 00 20 05 00 00 01 00 00 00 00 $.. ... 120 00 02 00 00 28 2d 00 41 2d 32 00 46 2d 32 00 46 (-.A-2.F-2.F 140 2d 32 00 46 24 01 00 14 01 81 00 14 07 00 00 01 -2.F$... 160 00 00 00 00 00 02 00 00 01 01 46 50 24 00 10 8c ..FP$... 200 22 c1 80 0a 2d 02 c1 18 83 f8 9b 78 c2 fc ff 0a ...-..x 220 2d 04 c2 07 0c 1c c1 07 0a 2d 2d 2d c7 80 e0 f0 ---- I have a recollection of seeing something similar for PCL/Deskjets a few years back... If I just could remember the name... google! http://packages.debian.org/unstable/text/djtools.html Depends on: libc6 (= 2.3.1-1) GNU C Library: Shared libraries and Timezone data libgcc1 (= 1:3.2.1-1) GCC support library libpaper1 Library for handling paper characteristics libstdc++5 (= 1:3.2.1-1) The GNU Standard C++ Library v3 What do you say guys? ;-) does it fit within cygwin? I downloaded djtools_1.2.5.tar.gz - 158702 bytes... Will have a look at it ASAP. /Hannu E K Nevalainen, B.Sc EE/Microcomputer systems, 59?14'N, 17?12'E. --END OF MESSAGE-- -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
