Re: [LinPrj] experiences with tzafrir's r2l-0.1.0

2001-09-11 Thread mulix

On Wed, 12 Sep 2001, Kohn Emil Dan wrote:

> > i saw some other errors when compiling with gcc3.0. i'll make a patch
> > later.
>
> Are those errors or just warnings?

warnings, and none of them were in refreshd. i fixed the warnings in
r2llib and biditext (missing #include ).

in the plugins dir i get:

ui_plugin_terminal.c: In function `ui_init':
ui_plugin_terminal.c:56: warning: implicit declaration of function
`strlen'
ui_plugin_terminal.c:59: warning: implicit declaration of function
`strcpy'

a missing #include , most likely.
-- 
mulix

http://www.advogato.com/person/mulix
http://www.sf.net/projects/syscalltrack



-
Haifa Linux Club Projects Mailing List (http://linuxclub.il.eu.org)
To unsub send an empty message to [EMAIL PROTECTED]





Re: [LinPrj] experiences with tzafrir's r2l-0.1.0

2001-09-11 Thread Kohn Emil Dan


Hi muli,

On Tue, 11 Sep 2001, mulix wrote:

> On Tue, 11 Sep 2001, Tzafrir Cohen wrote:
> 
 
> 
> > BTW: emil: the only warning I saw was:
> >
> > g++ -g -D_REENTRANT  -Wall -pedantic
> > -I/home/tzafrir/projects/LinClub/BiditextGui/r2l-tarball/r2l-0.1.1/include
> > -I/home/tzafrir/projects/LinClub/BiditextGui/r2l-tarball/r2l-0.1.1/include
> > -O2 -c communication.cc
> > In file included from /usr/include/sys/socket.h:34,
> >  from communication.cc:5:
> > /usr/include/bits/socket.h:226: warning: ANSI C++ forbids zero-size array 
>`__cmsg_data'
> 


I know this warning. However I don't think I am to blame for it. The guys
that wrote the above header file declared an array of zero size. That file
gets included whenever you #include 

> i saw some other errors when compiling with gcc3.0. i'll make a patch
> later.

Are those errors or just warnings?

> 
 
> 
> 

Emil


-
Haifa Linux Club Projects Mailing List (http://linuxclub.il.eu.org)
To unsub send an empty message to [EMAIL PROTECTED]





Re: [LinPrj] experiences with tzafrir's r2l-0.1.0

2001-09-11 Thread Tzafrir Cohen

On Tue, 11 Sep 2001, mulix wrote:

> On Tue, 11 Sep 2001, Tzafrir Cohen wrote:
>
> > I've updated only the r2l-package and r2l tarballs (the rpm managed to
> > build just fine anyway)
> >
> > http://www.technion.ac.il/~tzafrir/R2L/r2l-tarball/r2l-0.1.1.tar.gz
> > http://www.technion.ac.il/~tzafrir/R2L/r2l-tarball/r2l-package.0.1.1.tar.gz
>
> r2l-package gives me a 404. what is it?

Sorry:

http://www.technion.ac.il/~tzafrir/R2L/r2l-tarball/r2l-package-0.1.1.tar.gz

-- 
Tzafrir Cohen
mailto:[EMAIL PROTECTED]
http://www.technion.ac.il/~tzafrir



-
Haifa Linux Club Projects Mailing List (http://linuxclub.il.eu.org)
To unsub send an empty message to [EMAIL PROTECTED]





Re: [LinPrj] experiences with tzafrir's r2l-0.1.0

2001-09-11 Thread mulix

On Tue, 11 Sep 2001, Tzafrir Cohen wrote:

> I've updated only the r2l-package and r2l tarballs (the rpm managed to
> build just fine anyway)
>
> http://www.technion.ac.il/~tzafrir/R2L/r2l-tarball/r2l-0.1.1.tar.gz
> http://www.technion.ac.il/~tzafrir/R2L/r2l-tarball/r2l-package.0.1.1.tar.gz

r2l-package gives me a 404. what is it?

> BTW: emil: the only warning I saw was:
>
> g++ -g -D_REENTRANT  -Wall -pedantic
> -I/home/tzafrir/projects/LinClub/BiditextGui/r2l-tarball/r2l-0.1.1/include
> -I/home/tzafrir/projects/LinClub/BiditextGui/r2l-tarball/r2l-0.1.1/include
> -O2 -c communication.cc
> In file included from /usr/include/sys/socket.h:34,
>  from communication.cc:5:
> /usr/include/bits/socket.h:226: warning: ANSI C++ forbids zero-size array 
>`__cmsg_data'

i saw some other errors when compiling with gcc3.0. i'll make a patch
later.

here's the patch for the 'r2l' crashes.

--- r2l-0.1.1/src/r2llib/r2l.c  Mon Aug 27 16:39:38 2001
+++ r2l-0.1.1-mulix/src/r2llib/r2l.cTue Sep 11 20:47:41 2001
@@ -90,7 +90,7 @@
  exit(LIBR2L_FAILED);
  }

- if (argc < 1 || argv[1] == NULL) {
+ if (argc < 2 || argv[1] == NULL) {
  print_usage_message("Command missing");
  r2l_destroy(rtl);
  exit(INCORRECT_ARGUMENT);
@@ -98,7 +98,7 @@

  switch (argv[1][0]) {
  case 'p': /* print */
- if ((argc < 2) || (argv[2] == NULL)) {
+ if ((argc < 3) || (argv[2] == NULL)) {
   print_usage_message("Missing Print argument");
   ret_val = INCORRECT_ARGUMENT;
   break;
@@ -138,9 +138,10 @@
  }
  break;
  case 's': /* change State */
- if (argc < 2 || argv[2] == NULL) {
+ if (argc < 3 || argv[2] == NULL) {
   print_usage_message("Missing State argument");
   ret_val = INCORRECT_ARGUMENT;
+  break;
  }
  if(!strcasecmp(argv[2],"disable")) {
   ret_val = r2l_disable(rtl);
@@ -160,9 +161,10 @@
  }
  break;
  case 'b': /* change Base direction */
- if (argc < 2 || argv[2] == NULL) {
+ if (argc < 3 || argv[2] == NULL) {
   print_usage_message("Missing Base direction argument");
   ret_val = INCORRECT_ARGUMENT;
+  break;
  }
  if(!strcasecmp(argv[2],"neutral")) {
   ret_val = r2l_set_biditext_base_state_neutral(rtl);

-- 
mulix

http://www.advogato.com/person/mulix
http://www.sf.net/projects/syscalltrack



-
Haifa Linux Club Projects Mailing List (http://linuxclub.il.eu.org)
To unsub send an empty message to [EMAIL PROTECTED]





Re: [LinPrj] experiences with tzafrir's r2l-0.1.0

2001-09-11 Thread Tzafrir Cohen

On Tue, 11 Sep 2001, mulix wrote:

> On Tue, 11 Sep 2001, Tzafrir Cohen wrote:

> thanks. checknig with it - refreshd works :)
> now i get a nice segmentation violation in 'r2l'. will you be acting as
> the clearing house for all r2l patches (r2llib, biditext, r2l-plugins,
> refreshd) or should we bite the bullet and open a sourceforge project?

I can do it. 

Note that the seperate packages are currently built as independent
tarballs (each can compile seperately). 'make unpack' unpacks all of those
makefiles to form the source tree. So you can either send a patch or post
the next version of the r2llib tarball.

-- 
Tzafrir Cohen
mailto:[EMAIL PROTECTED]
http://www.technion.ac.il/~tzafrir


-
Haifa Linux Club Projects Mailing List (http://linuxclub.il.eu.org)
To unsub send an empty message to [EMAIL PROTECTED]





Re: [LinPrj] experiences with tzafrir's r2l-0.1.0

2001-09-11 Thread mulix

On Tue, 11 Sep 2001, Tzafrir Cohen wrote:

> > * a rudimentary check of biditext shows that it works. a rudimentary
> > check of r2llib didnt take place because i had no idea how to test it -
> > emil? (my system has no hebrew specially set up, except maybe the
> > turkish patch for netscape).

[i meant refreshd above]

> You can get the hebrew text file from
> http://www.technion.ac.il/~tzafrir/R2L/hello_8bit.txt
>
> and view it in an X text editor with the font set to 'heb8x13' (e.g:
> 'xedit -fn heb8x13 hello_8bit.txt').
>
> You can also use netscape / xterm -e lynx

thanks. checknig with it - refreshd works :)
now i get a nice segmentation violation in 'r2l'. will you be acting as
the clearing house for all r2l patches (r2llib, biditext, r2l-plugins,
refreshd) or should we bite the bullet and open a sourceforge project?

-- 
mulix

http://www.advogato.com/person/mulix
http://www.sf.net/projects/syscalltrack



-
Haifa Linux Club Projects Mailing List (http://linuxclub.il.eu.org)
To unsub send an empty message to [EMAIL PROTECTED]





Re: [LinPrj] experiences with tzafrir's r2l-0.1.0

2001-09-11 Thread Tzafrir Cohen

On Tue, 11 Sep 2001, Tzafrir Cohen wrote:

> Hi
> 
> On Tue, 11 Sep 2001, mulix wrote:
> 
> > * first of all, the links in src/ are broken, which led to a spectacular
> > make fork bomb :)
> 
> Hmmm... I was sure that the ln command always tries to create relative
> symlinks. even if I give full pathes for both the source and te target. 
> But looking at the tarball it seems that the symlinks have full pathes.
> 
> Anyway, those symlinks have indeed a potential of a make loop (unless I
> check if 'cd' was successful), so I figure I'll replace that 'ln -s' with
> 'mv' .
> 
> > * make clean does some very funky things (log attached)
> 
> oops
> 
> clean_refreshd_hooks in the toplevel makefile accidentally depends on
> refershd_hook .
> 
> fixed
> 
> I'll try to have version 0.1.1 ready soon

I've updated only the r2l-package and r2l tarballs (the rpm managed to
build just fine anyway)

http://www.technion.ac.il/~tzafrir/R2L/r2l-tarball/r2l-0.1.1.tar.gz
http://www.technion.ac.il/~tzafrir/R2L/r2l-tarball/r2l-package.0.1.1.tar.gz


BTW: emil: the only warning I saw was:

g++ -g -D_REENTRANT  -Wall -pedantic
-I/home/tzafrir/projects/LinClub/BiditextGui/r2l-tarball/r2l-0.1.1/include
-I/home/tzafrir/projects/LinClub/BiditextGui/r2l-tarball/r2l-0.1.1/include
-O2 -c communication.cc
In file included from /usr/include/sys/socket.h:34,
 from communication.cc:5:
/usr/include/bits/socket.h:226: warning: ANSI C++ forbids zero-size array `__cmsg_data'

-- 
Tzafrir Cohen
mailto:[EMAIL PROTECTED]
http://www.technion.ac.il/~tzafrir



Re: [LinPrj] experiences with tzafrir's r2l-0.1.0

2001-09-11 Thread Tzafrir Cohen

Hi

On Tue, 11 Sep 2001, mulix wrote:

> * first of all, the links in src/ are broken, which led to a spectacular
> make fork bomb :)

Hmmm... I was sure that the ln command always tries to create relative
symlinks. even if I give full pathes for both the source and te target. 
But looking at the tarball it seems that the symlinks have full pathes.

Anyway, those symlinks have indeed a potential of a make loop (unless I
check if 'cd' was successful), so I figure I'll replace that 'ln -s' with
'mv' .

> * make clean does some very funky things (log attached)

oops

clean_refreshd_hooks in the toplevel makefile accidentally depends on
refershd_hook .

fixed

I'll try to have version 0.1.1 ready soon

> 
> * compiling all of the plugins is not a good idea - on my system
> compilation barfs because i dont have the wm headers. perhaphs (if not
> autoconf then) a simple script asking the user what he would like to
> install and doing rudimentary checks?

When I have some time...

> 
> * a rudimentary check of biditext shows that it works. a rudimentary
> check of r2llib didnt take place because i had no idea how to test it -
> emil? (my system has no hebrew specially set up, except maybe the
> turkish patch for netscape).

You can get the hebrew text file from
http://www.technion.ac.il/~tzafrir/R2L/hello_8bit.txt

and view it in an X text editor with the font set to 'heb8x13' (e.g:
'xedit -fn heb8x13 hello_8bit.txt').

You can also use netscape / xterm -e lynx

-- 
Tzafrir Cohen
mailto:[EMAIL PROTECTED]
http://www.technion.ac.il/~tzafrir