cvs commit to debian-installer/tools/cdebconf/src by tausq
Repository: debian-installer/tools/cdebconf/src who:tausq time: Thu Sep 5 00:38:14 MDT 2002 Log Message: # is not a valid comment character Files: changed:cdebconf.conf -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: gtk frontend for cdebconf, ready for commit
On Tue, Sep 03, 2002 at 05:18:28PM -0700, Michael Cardenas wrote: Hello all. Hello again. Attached is a patch for cdebconf that adds a gtk2.0 frontend. It is relatively complete. The select and text handlers are not finished, but there is enough to be tested here. I've attached a cleaner, more complete patch to this email. Now all of the question handlers are done. I was unable to figure out the cdebconf tests dir, so if someone can explain to me how they work, I can do much better testing on it myself. For now, I've just been making little hello world type apps to test the functionality of each handler. The multiselect, password, string and note handlers are I still haven't been able to test this, so if someone can give me some pointers, that would be great. I know I should be able to run cdebconf with the --frontend parameter, but I don't know what to run it on. Also, the select handler uses a radio button, but as I had no net ... examples there are not very complete. It works as long as you don't select the first radio button. This is now fixed. Lastly ,the text handler is mostly incomplete. The widget is created, but I'm not extracting the data entered in it. It still has lots of formatting to do. This is also fixed. The data is correctly extracted and some tiny amount of formatting is done (the text entry area is resized. So, this needs a lot more work, that I will continue to do, but I think it's a good enough start to be looked over, and possibly tested. It still needs to be made prettier. Another issue at hand, now that this is almost ready, is the question of udebs for X and gtk. Should I start on those? Or can we safely assume that the X install medium is going to have lots of space, and thus use regular debs. I don't mind working on the udebs, I just want to know what you guys think. I still need an answer to this. Any feedback is much appreciated. I'm not a gtk expert by any means, or a debconf guru either. thanks michael -- michael cardenas | lead software engineer | lindows.com | hyperpoem.net The devils of truth steal the souls of the free. - Trent Reznor Binary files frontend.orig/bogl/bogl.opic and frontend/bogl/bogl.opic differ Binary files frontend.orig/bogl/bogl.so and frontend/bogl/bogl.so differ Binary files frontend.orig/bogl.so and frontend/bogl.so differ diff -ruP frontend.orig/gtk/CVS/Entries frontend/gtk/CVS/Entries --- frontend.orig/gtk/CVS/Entries Wed Dec 31 16:00:00 1969 +++ frontend/gtk/CVS/EntriesWed Aug 28 23:24:55 2002 @@ -0,0 +1,3 @@ +/Makefile/1.4/Tue Jul 9 05:25:05 2002// +/text.c/1.13/Wed Aug 7 16:19:51 2002// +D diff -ruP frontend.orig/gtk/CVS/Repository frontend/gtk/CVS/Repository --- frontend.orig/gtk/CVS/RepositoryWed Dec 31 16:00:00 1969 +++ frontend/gtk/CVS/Repository Wed Aug 28 23:24:55 2002 @@ -0,0 +1 @@ +debian-installer/tools/cdebconf/src/modules/frontend/text diff -ruP frontend.orig/gtk/CVS/Root frontend/gtk/CVS/Root --- frontend.orig/gtk/CVS/Root Wed Dec 31 16:00:00 1969 +++ frontend/gtk/CVS/Root Wed Aug 28 23:24:55 2002 @@ -0,0 +1 @@ +:pserver:[EMAIL PROTECTED]:/cvs/debian-boot diff -ruP frontend.orig/gtk/Makefile frontend/gtk/Makefile --- frontend.orig/gtk/Makefile Wed Dec 31 16:00:00 1969 +++ frontend/gtk/Makefile Sat Aug 31 18:31:54 2002 @@ -0,0 +1,8 @@ +MODULE=gtk +SOBJ=gtk.so +OBJS=gtk.opic +INCLUDES=`pkg-config --cflags gtk+-2.0` + +MODLDFLAGS=`pkg-config --libs gtk+-2.0` + +include ../modules.mak diff -ruP frontend.orig/gtk/gtk.c frontend/gtk/gtk.c --- frontend.orig/gtk/gtk.c Wed Dec 31 16:00:00 1969 +++ frontend/gtk/gtk.c Wed Sep 4 17:12:58 2002 @@ -0,0 +1,1062 @@ +/*** + * + * cdebconf - An implementation of the Debian Configuration Management + *System + * + * File: gtk.c + * + * Description: gtk UI for cdebconf + * Some notes on the implementation - optimistic at best. + * mbc - just to get this off of the ground, Im' creating a dialog + *and calling gtk_main for each question. once I get the tests + *running, I'll probably send a delete_event signal in the + *next and back button callbacks. + * + *There is some rudimentary attempt at implementing the next + *and back functionality. + * + * $Id: text.c,v 1.13 2002/08/07 16:19:51 tfheen Exp $ + * + * cdebconf is (c) 2000-2001 Randolph Chung and others under the following + * license. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other
Re: cvs commit to debian-installer/tools/partkit by wart
On Wed, Sep 04, 2002 at 01:05:51PM +0400, Wartan Hachaturow wrote:
On Tue, Sep 03, 2002 at 09:55:01PM -0700, Michael Cardenas wrote:
you ever get around to adding in my patch? I can update it to 1.6 for
you if you like.
Sorry, been a bit busy with the start of term ;)
If you'll update it, fine -- send the patch to me/to the list, and
someone/me would commit it :)
My new patch, synced with 1.6 (just removing ped_init and ped_done) is
attached. Please commit it.
thanks
michael
--
michael cardenas | lead software engineer | lindows.com | hyperpoem.net
There is only one great adventure and that is inwards towards the self.
- Henry Miller
--- partkit.c.orig Tue Aug 27 08:23:42 2002
+++ partkit.c Wed Sep 4 17:55:47 2002
-26,6 +26,9
#define PARTKIT_TABLE_SIZE 2048
#define PARTKIT_PART_LIST_SIZE 128
+/* I can't understand why they did removed it. */
+#define PED_PARTITION_PRIMARY 0
+
#include string.h
#include parted/parted.h
#include debconfclient.h
-91,7 +94,7
ptr = table;
- disk = ped_disk_open (*dev);
+ disk = ped_disk_new (*dev);
if (!disk)
goto error;
-170,7 +173,7
ptr += snprintf (ptr, PARTKIT_TABLE_SIZE - (ptr - table), \n);
}
- ped_disk_close (disk);
+ ped_disk_destroy (disk);
if ((ptr - table) PARTKIT_TABLE_SIZE)
goto error;
return table;
-199,7 +202,7
ptr = part_list;
- disk = ped_disk_open (dev);
+ disk = ped_disk_new (dev);
if (!disk)
goto error;
-217,7 +220,7
%-5d , part-num);
}
- ped_disk_close (disk);
+ ped_disk_destroy (disk);
if ((ptr - part_list) PARTKIT_PART_LIST_SIZE)
goto error;
-242,10 +245,10
char *ptr, *endptr = NULL;
char *table;
- disk = ped_disk_open (dev);
+ disk = ped_disk_new (dev);
if (!disk)
goto error;
- constraint = ped_constraint_any (disk);
+ constraint = ped_constraint_any (disk-dev);
if (!constraint)
goto error_close_disk;
-326,9 +329,9
if (!ped_disk_add_partition (disk, part, constraint))
goto error_destroy_part;
- ped_disk_write (disk);
+ ped_disk_commit (disk);
ped_constraint_destroy (constraint);
- ped_disk_close (disk);
+ ped_disk_destroy (disk);
return 1;
error_destroy_part:
-336,7 +339,7
error_destroy_constraint:
ped_constraint_destroy (constraint);
error_close_disk:
- ped_disk_close (disk);
+ ped_disk_destroy (disk);
error:
return 0;
}
-349,7 +352,7
PedPartition *part;
char *ptr;
- disk = ped_disk_open (dev);
+ disk = ped_disk_new (dev);
if (!disk)
goto error;
-394,40 +397,99
}
ped_disk_delete_partition (disk, part);
- ped_disk_write (disk);
- ped_disk_close (disk);
+ ped_disk_commit (disk);
+ ped_disk_destroy (disk);
return 1;
error_close_disk:
- ped_disk_close (disk);
+ ped_disk_destroy (disk);
error:
partkit_error (0);
return -1;
}
+char *partkit_get_devices(PedDevice *dev)
+{
+ char *ptr = 0;
+ int drivelistlen = 0;
+
+ ped_device_probe_all();
+
+ //mbc - unfortunately, do two passes for now to do the malloc...
+ while((dev = ped_device_get_next(dev)))
+{
+ drivelistlen += strlen(dev-path);
+ drivelistlen += 2;
+}
+
+ if(drivelistlen)
+ {
+ptr = (char *)malloc(drivelistlen);
+
+if(ptr)
+ {
+*ptr = 0;
+dev = NULL;
+
+while((dev = ped_device_get_next(dev)))
+ {
+if(*ptr == 0)
+ strcpy(ptr, dev-path);
+else if (*ptr != 0)
+ {
+strcat(ptr, , );
+strcat(ptr, dev-path);
+ }
+ }
+ }
+
+ }
+
+ return ptr;
+
+}
+
+//mbc - for now, just use the first device
+char *partkit_get_default_device(void)
+{
+ PedDevice *dev = NULL;
+ char *ptr = 0;
+
+ dev = ped_device_get_next(dev);
+
+ ptr = (char *)malloc(strlen(dev-path));
+
+ strcpy(ptr, dev-path);
+
+ return ptr;
+}
int
main (int argc, char *argv[])
{
int finished = 0;
PedDevice *dev = NULL;
- char *ptr;
+ char *ptr = 0;
client = debconfclient_new ();
client-command (client, title, Partition Editor, NULL);
+ //mbc - should install an exception handler here
+
do
{
- /* FIXME: how to get a list of available devices ? */
- ptr = /* get_device_list() */ /dev/hdb;
-
+ ptr = partkit_get_devices(dev);
client-command (client, subst, partkit/select_device, choices,
ptr, NULL);
- /*FIXME: how to get a default device? */
+ free(ptr);
+
+ ptr = partkit_get_default_device();
+
client-command (client, subst, partkit/select_device, default,
ptr, NULL);
+ free(ptr);
if ((dev = ped_device_get (argv[1])) == NULL)
partkit_error (1);
msg21870/pgp0.pgp
Description: PGP signature
Re: gtk frontend for cdebconf, ready for commit
tor 2002-09-05 klockan 02.31 skrev Michael Cardenas:
Attached is a patch for cdebconf that adds a gtk2.0 frontend. It is
relatively complete. The select and text handlers are not finished,
but there is enough to be tested here.
I've attached a cleaner, more complete patch to this email. Now all of
the question handlers are done.
I couldn't get it to compile, until I changed this line in Makefile:
--- Makefile.orig 2002-09-05 10:03:59.0 +0200
+++ Makefile2002-09-05 10:03:30.0 +0200
-1,7 +1,7
MODULE=gtk
SOBJ=gtk.so
OBJS=gtk.opic
-INCLUDES=`pkg-config --cflags gtk+-2.0`
+MODCFLAGS=`pkg-config --cflags gtk+-2.0`
MODLDFLAGS=`pkg-config --libs gtk+-2.0`
I was unable to figure out the
cdebconf tests dir, so if someone can explain to me how they work, I
can do much better testing on it myself. For now, I've just been
making little hello world type apps to test the functionality of
each handler. The multiselect, password, string and note handlers are
With the help of Mithrandir, I've figured it out now :)
$ ./configure --with-frontend=gtk --with-conffile=./cdebconf.conf
And then you'll have to go to src/test, I copied ../cdebconf.conf-dist
to ./cdebconf.conf and edited so:
--- ../cdebconf.conf-dist 2002-09-05 10:43:21.0 +0200
+++ cdebconf.conf 2002-09-05 10:38:40.0 +0200
-4,8 +4,8
global {
module_path {
-frontend /usr/lib/cdebconf/frontend;
-database /usr/lib/cdebconf/db;
+frontend ../modules/frontend;
+database ../modules/db;
};
default {
-17,21 +17,21
frontend {
instance default_fe {
-driver text;
+driver gtk;
};
};
template {
instance templatedb {
driver rfc822db;
-path /var/lib/cdebconf/templates.dat;
+path ./templates.dat;
};
};
config {
instance configdb {
driver rfc822db;
-path /var/lib/cdebconf/questions.dat;
+path ./questions.dat;
};
};
You'll also have to touch templates.dat, or it'll complain about it not
existing. At this point you should be able to do:
$ ../debconf-loadtemplate foo test.templates
$ ../debconf test.config
and enjoy ;)
I still haven't been able to test this, so if someone can give me some
pointers, that would be great. I know I should be able to run cdebconf
with the --frontend parameter, but I don't know what to run it on.
I didn't get that to work!
$ ../debconf --frontend=gtk test.config
execv: No such file or directory
confmodule.c:203 (confmodule_run): Cannot execute client config script
$ ../debconf -fgtk test.config
../debconf: invalid option -- g
../debconf: invalid option -- t
../debconf: invalid option -- k
execv: No such file or directory
confmodule.c:203 (confmodule_run): Cannot execute client config script
$ ../debconf -f gtk test.config
execv: No such file or directory
confmodule.c:203 (confmodule_run): Cannot execute client config script
It still needs to be made prettier.
I noticed that window titles aren't set. It might be nice to set them to
the first description line? Also, windows aren't closed after the
question is answered, is this by design or by accident?
Also, after I entered a password, it segfaulted. :)
Another issue at hand, now that this is almost ready, is the question
of udebs for X and gtk. Should I start on those? Or can we safely
assume that the X install medium is going to have lots of space, and
thus use regular debs. I don't mind working on the udebs, I just want
to know what you guys think.
I still need an answer to this.
Can't help you there. :/
Regards,
Martin
signature.asc
Description: Detta =?ISO-8859-1?Q?=E4r?= en digitalt signeradmeddelandedel
udebootstrap?
Hi, I don't know if it's really a feasible solution, or what people had in mind, but I had a quick-hack into what might be udebootstrap, IMO. This application takes in a predetermined directory filled with udebs and debs, and extract them using tar, ar, and zcat. http://www.netfort.gr.jp/~dancer/software/downloads/list.cgi What I was thinking was to make something mount the CD-Rom, or another floppy, and to populate a chroot (possibly a tmpfs) with debian-installer and it goes off running, so that we can be free from the restrictions imposed by the floppy disk space, and actually get something for i18n, and other stuff. But I've only just started playing with this idea ... and it occurred to me that it could actually be possible to debootstrap udebs as well. regards, junichi -- [EMAIL PROTECTED] http://www.netfort.gr.jp/~dancer -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: gtk frontend for cdebconf, ready for commit
* Martin Sjögren | You'll also have to touch templates.dat, or it'll complain about it not | existing. At this point you should be able to do: It'll complain, but create it. -- Tollef Fog Heen,''`. UNIX is user friendly, it's just picky about who its friends are : :' : `. `' `- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Problem with a GXT3000P VGA Adapter in a RS6000 (44p 170)
Hi Does anybody have expirience with this Graphic-Adapter ? I can't even see the Boot Console (over a serial Line it's ok). When i use the Adapter, i see two small reflections of the console with unreadble text. I think it is a syncronisation Problem. I don't wanna use X in the moment, the text console would be enough. Regards, Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
cvs commit to debian-installer/tools/partkit by wart
Repository: debian-installer/tools/partkit who:wart time: Thu Sep 5 06:00:43 MDT 2002 Log Message: Added patch by Michael Cardenas, synced to 1.6. Files: changed:partkit.c -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
cvs commit to debian-installer/tools/partkit/debian by wart
Repository: debian-installer/tools/partkit/debian who:wart time: Thu Sep 5 06:00:43 MDT 2002 Log Message: Added patch by Michael Cardenas, synced to 1.6. Files: changed:changelog -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
cvs commit to debian-installer/tools/cdebconf/src/modules/frontend/gtk by tfheen
Repository: debian-installer/tools/cdebconf/src/modules/frontend/gtk who:tfheen time: Thu Sep 5 06:30:23 MDT 2002 Log Message: initial checkin, with Martin's compile fixes Files: added: Makefile gtk.c -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
cvs commit to debian-installer/tools/cdebconf/debian by tfheen
Repository: debian-installer/tools/cdebconf/debian who:tfheen time: Thu Sep 5 06:33:10 MDT 2002 Log Message: document gtk addition Files: changed:changelog -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: gtk frontend for cdebconf, ready for commit
* Michael Cardenas | I've attached a cleaner, more complete patch to this email. Now all of | the question handlers are done. commited. -- Tollef Fog Heen,''`. UNIX is user friendly, it's just picky about who its friends are : :' : `. `' `- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
cvs commit to debian-installer/tools/lmod-detect-pci/debian by tfheen
Repository: debian-installer/tools/lmod-detect-pci/debian who:tfheen time: Thu Sep 5 06:35:48 MDT 2002 Log Message: fix section to be debian-installer, not admin Files: changed:control changelog -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [d-i] Problem with lmod-detect-pci
* Stephen Depooter | Attached is a cvs diff after I corrected it. | If I am wrong, feel free to ignore this. thanks, applied. -- Tollef Fog Heen,''`. UNIX is user friendly, it's just picky about who its friends are : :' : `. `' `- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#159021: Install progam maked bad /home mode for ssh
While /home is 2775 by default, ssh should have no problems with this. If $HOME (like, you're user's home directory) is group-writable, you'll have problems, but this is not the default. msg21882/pgp0.pgp Description: PGP signature
cvs commit to boot-floppies/documentation/fr by barbier
Repository: boot-floppies/documentation/fr who:barbier time: Thu Sep 5 07:35:26 MDT 2002 Log Message: Sync to EN 1.33 [Pierre Machard] Files: changed:boot-new.sgml -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
cvs commit to boot-floppies/documentation/fr by barbier
Repository: boot-floppies/documentation/fr who:barbier time: Thu Sep 5 07:35:59 MDT 2002 Log Message: Sync to EN 1.59 [Pierre Machard] Files: changed:partitioning.sgml -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: gtk frontend for cdebconf, ready for commit
On Thu, Sep 05, 2002 at 10:49:58AM +0200, Martin Sj?gren wrote: tor 2002-09-05 klockan 02.31 skrev Michael Cardenas: Attached is a patch for cdebconf that adds a gtk2.0 frontend. It is relatively complete. The select and text handlers are not finished, but there is enough to be tested here. I've attached a cleaner, more complete patch to this email. Now all of the question handlers are done. I couldn't get it to compile, until I changed this line in Makefile: +++ Makefile2002-09-05 10:03:30.0 +0200 -1,7 +1,7 MODULE=gtk SOBJ=gtk.so OBJS=gtk.opic -INCLUDES=`pkg-config --cflags gtk+-2.0` +MODCFLAGS=`pkg-config --cflags gtk+-2.0` MODLDFLAGS=`pkg-config --libs gtk+-2.0` I had attached a patch to globalmakeflags.in in my first email that made the INCLUDES work, but I see that your change is more correct. -- michael cardenas | lead software engineer | lindows.com | hyperpoem.net One evening I seated beauty on my knees. And I found her bitter, And I cursed her. - Arthur Rimbaud msg21886/pgp0.pgp Description: PGP signature
Re: cvs commit to debian-installer/tools/cdebconf/src/modules/frontend/gtk by tfheen
On Thu, Sep 05, 2002 at 06:30:23AM -0600, Debian Boot CVS Master wrote: Repository: debian-installer/tools/cdebconf/src/modules/frontend/gtk who:tfheen time: Thu Sep 5 06:30:23 MDT 2002 Log Message: initial checkin, with Martin's compile fixes Files: added: Makefile gtk.c You didn't mention me here. Hopefully you can mention me in the changelog wrt the gtk frontend. -- michael cardenas | lead software engineer | lindows.com | hyperpoem.net Language is a virus. - William S. Burroughs msg21887/pgp0.pgp Description: PGP signature
Re: gtk frontend for cdebconf, ready for commit
On Thu, Sep 05, 2002 at 10:49:58AM +0200, Martin Sj?gren wrote: ... I was unable to figure out the cdebconf tests dir, so if someone can explain to me how they work, I can do much better testing on it myself. For now, I've just been making little hello world type apps to test the functionality of each handler. The multiselect, password, string and note handlers are With the help of Mithrandir, I've figured it out now :) Thanks for the detailed instructions! $ ../debconf-loadtemplate foo test.templates $ ../debconf test.config I followed all the above steps, but at this point, I get this: database.c:119 (template_db_new): No template database instance defined and enjoy ;) and I can't continue... Also, after I entered a password, it segfaulted. :) I'll send another patch tonight or tomorrow addressing your next email. thanks a lot michael -- michael cardenas | lead software engineer | lindows.com | hyperpoem.net Value judgements are destructive to our proper business, which is curiosity and awareness. - John Cage msg21888/pgp0.pgp Description: PGP signature
Re: cvs commit to debian-installer/tools/cdebconf/src/modules/frontend/gtk by tfheen
On Thu, Sep 05, 2002 at 12:43:30AM -0700, Michael Cardenas wrote: On Thu, Sep 05, 2002 at 06:30:23AM -0600, Debian Boot CVS Master wrote: Repository: debian-installer/tools/cdebconf/src/modules/frontend/gtk who:tfheen time: Thu Sep 5 06:30:23 MDT 2002 Log Message: initial checkin, with Martin's compile fixes Files: added: Makefile gtk.c You didn't mention me here. Hopefully you can mention me in the changelog wrt the gtk frontend. Sorry I spoke to soon. I saw your changelog entry. Thank you. michael -- michael cardenas | lead software engineer | lindows.com | hyperpoem.net What the poet is looking for is not the fundamental I but the deep you. - Antonio Machado msg21889/pgp0.pgp Description: PGP signature
Re: gtk frontend for cdebconf, ready for commit
On Thu, Sep 05, 2002 at 02:50:26PM +0200, Martin Sj?gren wrote:
tor 2002-09-05 klockan 02.31 skrev Michael Cardenas:
+void passwd_callback( GtkWidget *widget,
+ GtkWidget *entry )
+{
+ const char *entrytext;
+ entrytext = gtk_entry_get_text (GTK_ENTRY (entry));
+
+ gPassword = malloc(strlen(entrytext));
+ strcpy(gPassword, entrytext);
Why not just strdup?
...and the original allocates one byte too few; it needs an
extra byte for the NUL-terminator.
Matt
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: cvs commit to debian-installer/tools/partkit by wart
On Wed, Sep 04, 2002 at 05:59:12PM -0700, Michael Cardenas wrote:
+//mbc - for now, just use the first device
+char *partkit_get_default_device(void)
+{
+ PedDevice *dev = NULL;
+ char *ptr = 0;
+
+ dev = ped_device_get_next(dev);
+
+ ptr = (char *)malloc(strlen(dev-path));
+
+ strcpy(ptr, dev-path);
This is wrong; you aren't allocating space for the
NUL-terminator. Use strdup instead.
Matt
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#159738: Dependency between MD5 passwords and shadow passwords is described in install-doc but does not happen during installation
Package: install-doc Version: 3.0.24 Severity: minor The first sentence of section 8.6 Shadow Passwords seems to be wrong: Unless you said Yes to MD5 passwords, the system will ask whether you want to enable shadow passwords. This was not true for my 3.0r0 install for Intel x86: I said Yes to MD5 passwords, and the system did ask in the next step whether I want to enable shadow passwords. Is the documentation incorrect or is the install program incorrect? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Prove suas acusações, ou retrate-se
Juan:EnEspañol,PorFavor Juan:InEnglish,Please
Amigos:
Vi-mena obrigação de enviar a seguinte carta
ao embaixador cubano em Brasília, ante as graves acusações de que fui objeto de
sua parte. Solicito a atenção e compreensão de todos.
Um abraço, Juan
Carta ao embaixador cubano em Brasília: prove
suas graves acusações, ou retrate-se
Para justificar a cruel negativa para
que eu possa entrar em Cuba a fim de conhecer o meu filhinho Juan Paolo, o Sr.
me acusa de "caluniar" o povo cubano e de colaborar com organizações que teriam
um "histórico" de "ações terroristas"; lhe solicito que prove tão graves
acusações, ou que se retrate
Campinas, 4 de setembro de 2002.
Sr. Embaixador de Cuba no Brasil
Jorge Lezcano Pérez
E-mail: [EMAIL PROTECTED]
Tel.: 55-61- 2484710 Fax: 55-61- 248 6778
Senhor Embaixador:
O Senador Eduardo Suplicy (PT/SP) realizou
meritórias gestões ante a embaixada cubana em Brasília, para que eu possa viajar
à Cuba e conhecer meu filho Juan Paolo, de 3 anos e meio de idade. O mencionado
legislador teve a amabilidade de enviar-me cópia de sua resposta.
Nela, o Sr., ao mesmo tempo em que se mostra cego e
surdo sobre o ponto central dos meus pedidos, que é precisamente o direito que
tenho de ir à Cuba conhecer Juan Paolo, lança graves e infundadas acusações
contra minha pessoa. Isto me obriga, em defesa de minha honra de cubano e de pai
de família, a não poder permanecer em silêncio.
Segundo sua carta, eu estaria participando de uma
campanha "caluniosa" contra o "povo cubano", vinculada a "organizações" e
"elementos" do exterior com um "histórico" de "ações terroristas". Sr.
Embaixador, segundo um sábio princípio do direito romano, "quod gratis
asseritur, gratis negatur": ante uma acusação gratuita, sem a mais mínima prova,
tem o acusado o direito de negá-la simples e francamente.
Todavia, por ser o Sr. a mais alta autoridade
diplomática de Cuba no Brasil, representa com suas expressões, o governo cubano.
Portanto, essa acusação adquire outra dimensão e eu não poderia simplesmente
negá-la, como o faria com alguém que não tivesse esse relevante
cargo.
Por isso, lhe solicito que prove essa injuriosa
afirmação pela qual me atribui uma colaboração com entidades e pessoas
estrangeiras que teriam esse "histórico" de "ações terroristas". Do contrário,
está obrigado a retratar-se. Seria descabido pensar que o Sr. inclui em tal
categoria personalidades que me brindaram com sua solidariedade, como o
presidente da Comissão Européia, Professor Romano Prodi e o jurista cubano Dr.
Claudio Benedí Beruff, que apresentou os antecedentes de meu drama familiar
perante a Comissão de Direitos Humanos da OEA.
Lhe solicito que prove igualmente sua acusação de
que eu estaria levando a cabo uma atividade "caluniadora". Solicitei de maneira
privada, durante mais de 3 anos, ante as autoridades diplomáticas cubanas no
Brasil, a possibildiade de exercer o direito inalienável, que todo pai tem, e
que nem sequer se nega a um criminoso, de conhecer a seu filho. Direito que,
sistematicamente, a embaixada cubana em Brasília e o consulado cubano em São
Paulo me negaram até hoje. Vi-me obrigado a tornar pública esta flagrante
injustiça em meados deste ano, quando o consulado cubano em São Paulo me fechou
brutalmente as portas.
Sirva-se então de indicar onde está exatamente minha
atividade "caluniadora". Em caso contrário e na falta de provas, como o Sr.
sabe, Senhor Embaixador, a acusação passa a recair sobre o acusador.
Sua missiva também trata de explicar outro fato
inexplicável: que Cuba seja o único país das Américas que há mais de 40 anos
impede a livre entrada e saída de seus habitantes, contrariando frontalmente o
artigo 13 (2) da Declaração Universal dos Direitos Humanos ("toda pessoa tem o
direito de sair de qualquer país, inclusive do próprio, e regressar a seu
país"), assim como o artigo 10 (1) da Convenção sobre os Direitos da Criança
("os pedidos de uma criança ou de seus pais para entrar ou sair de um
Estado-parte, no propósito de reunificação familiar, serão considerados pelos
Estados-partes de modo positivo, humanitário e rápido. "), tratado este assinado
pelo governo cubano.
O Sr. se refere igualmente a "regulamentos" e
"normas migratórias cubanas", destinadas a proteger a "segurança nacional", e a
dissuadir a "emigração ilegal" e o "roubo de cérebros", às quais eu haveria
"violado reiteradamente".
A esse respeito, me acolhe o direito elementar, para
ter a possibilidade também inalienável de poder defender-me, de conhecer o texto
de tais "regulamentos" e "normas", com base nas quais estou sendo condenado.
Solicito então cópia integral das mesmas, assim como a data exata em que elas
teriam sido escritas e publicadas, se têm ou não força de lei, e quais as
autoridades, ministérios ou órgãos do Estado cubano que as
referendaram.
Não é em vão que pergunto todos estes detalhes. Até
o momento, apesar de haver tentado, reiteradas vezes, obter estas
Message not delivered! prove suas acusações, ou retrate-se
A message you sent to a recipient at Aventail Corporation was rejected due to header filtering. Because the message did not contain a local Aventail recipient on the To: or CC: lines, Aventail's mail server is rejecting it as possible spam. Please re-send your message by adding the recipient to the To: or CC: line, or by directing it to the recipient's personal email account instead. Thanks, [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Na Du?
Hi,haben uns gestern im Chat leider verloren :( Wollte dir doch noch meine Homepage-URL geben. Die Adresse ist: http://squeezer.5xx.netHoffe, du besuchst die Seite mal.. :) Schreibst du mir zurück? Küsschen :-)
cvs commit to boot-floppies/documentation/fr by barbier
Repository: boot-floppies/documentation/fr who:barbier time: Thu Sep 5 13:14:39 MDT 2002 Log Message: Sync to EN 1.33 [Pierre Machard] Files: changed:appendix.sgml -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Anyone actually using autopartkit v0.2?
[Osamu Aoki] My wishlist ... Can these auto partition scheme includes a no-partition option with following 1-2 steps 1. creation of a large zero filled file in known FS (VFAT, NTFS?) 2. loop mounting it as ext2FS (even as rootFS) (I am thinking booting from flash for notePC without repartitioning) Reason are: Anticipated existance of new XP preloaded machine with NTFS Create opportunity for low impact install for wider audience Clandestine Linux install (avoid isuues with stupid corporate IT rule) Sounds interesting, but none of these will make it into my todo list. I hope someone else will implement them. I could say send patches, but please wait until I've updated the Debian CVS with the new and improved autopartkit. :-) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
cvs commit to boot-floppies/powerpc-specials/miBoot by toff
Repository: boot-floppies/powerpc-specials/miBoot who:toff time: Thu Sep 5 19:28:33 MDT 2002 Log Message: change new-powermac kernel to 2.4.19, dan says 2.4.18 is missing some patches. Files: changed:mkboot.sh -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
cvs commit to boot-floppies by toff
Repository: boot-floppies who:toff time: Thu Sep 5 19:28:33 MDT 2002 Log Message: change new-powermac kernel to 2.4.19, dan says 2.4.18 is missing some patches. Files: changed:config -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
cvs commit to debian-installer/tools/netcfg/debian by dwhedon
Repository: debian-installer/tools/netcfg/debian who:dwhedon time: Thu Sep 5 21:38:58 MDT 2002 Log Message: * remove util.[ch] (functions now in libdebian-installer) Files: changed:changelog -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
cvs commit to debian-installer/tools/netcfg by dwhedon
Repository: debian-installer/tools/netcfg who:dwhedon time: Thu Sep 5 21:38:58 MDT 2002 Log Message: * remove util.[ch] (functions now in libdebian-installer) Files: removed:utils.c utils.h -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
cvs commit to debian-installer/tools/netcfg by dwhedon
Repository: debian-installer/tools/netcfg who:dwhedon time: Thu Sep 5 21:41:08 MDT 2002 Log Message: indent -kr -i8 -nut *.c Files: changed:netcfg.c netcfg-dhcp.c netcfg-static.c -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: gtk frontend for cdebconf, ready for commit
Attached is a patch against cvs for
debian-installer/tools/cdebconf/src/modules/frontend/gtk/gtk.c
that corrects a number of the problems discussed below. Unfortunately,
I still can't test, as I mentioned in an earlier email, so I would
appreciate if someone could verify this patch before applying it.
On Thu, Sep 05, 2002 at 02:50:26PM +0200, Martin Sj?gren wrote:
tor 2002-09-05 klockan 02.31 skrev Michael Cardenas:
Now I've tested it, and I get a segfault after the password dialog.
Let's have a look at some of the code. I hope my comments are useful ;)
Okay, some of these errors are flat out embarassing. Guess I was in a
hurry to finish this project this weekend.
+ gPassword = malloc(strlen(entrytext));
+ strcpy(gPassword, entrytext);
Why not just strdup?
ok, fine.
+ g_print (delete event occurred\n);
+
+ /* Change TRUE to FALSE and the main window will be destroyed with
+ * a delete_event. */
+
+ return TRUE;
+}
As I said, clicking Finish on a dialog box doesn't close the dialog box.
I'm not sure this is good or bad. I think it gets cluttered fast.
The dialogs get closed for me when I hit the next button. I disabled
the close button to force you to use the next/back buttons. Are they
not working?
+ g_signal_connect (G_OBJECT (boolButton), clicked,
+ G_CALLBACK (check_callback), NULL);
Instead of global variables, why not pass in a pointer to ans instead of
NULL? That's the 'data' argument in the callback, right?
As I stated in my original email with my patch, I was planning on
removing the globals, and I just used them to get the project
started. They have all been replaced with callback pointers and
structures where necessary now. But, as I cannot test, can you or
someone please verify that these changes work?
+ if(gBool)
+ {
+ ans = 1;
+ }
+ else
+ {
+ ans = 0;
+ }
Meaning you wouldn't have to do this...
right, this was lame and it's gone now.
+ choiceButtons = (GtkWidget **)malloc( (count * (sizeof(GtkWidget *)) + 1) );
Hmm, shouldn't that be count * (sizeof(GtkWidget*) + 1)? You want n+1
pointers, don't you?
I do. So, I think it should be
( (count + 1) * (sizeof(GtkWidget *)) )
+ for(i = 0; i count; i++)
+ choiceButtons[i] = (GtkWidget *)malloc(sizeof(GtkWidget *));
Eh? Why do you want to do this? You want to use gtk_*_new functions to
do this.
This is lame. I was thinking that I wanted to allocate space to hold
each of the pointers, but of course, the original call does this. Doh!
+ choiceButtons[count + 1] = 0;
And this should be [count], since it's zero-based, right?
Yes, another lame mistake. Been writing rules files and shell scripts
too much lately!
+ choiceButtons[i] = gtk_check_button_new_with_label ( (gchar *)choices[i] );
And you overwrite the pointers here, so your malloc was pretty
unnecessary in the first place :)
right.
+ g_signal_connect (G_OBJECT (choiceButtons[i]), clicked,
+ G_CALLBACK (multicheck_callback), choiceButtons);
This is where you'd want a higher-level language so you could pass in a
list of pairs instead wink Seriously, a list of structs with both
widget and result value might be nice?
done.
I don't know how allergic others
are to global variables, but I'm pretty damn allergic ;)
me too. I wanted to see if anyone had any pointers on this c file as
class thing. I was thinking maybe I could put the data in the
frontend_module struct or something. Silly me.
Another idea
would be to actually create our own full-blown gtk2 type and create an
object for it to make it even more object-oriented. Just an idea :)
this may be overkill. I don't have any aversion to keeping it close to
c.
+ for (i = 0; i count; i++)
+ free(choiceButtons[i]);
This is where it segfaults.
Which makes perfect sense now because I'm freeing the widget's created
by gtk. I would expect it to freak out.
+ //add actual password field
+ entry = gtk_entry_new ();
+ gtk_entry_set_max_length (GTK_ENTRY (entry), 50);
+ g_signal_connect (G_OBJECT (entry), activate,
+ G_CALLBACK (passwd_callback),
+ entry);
Err. Why pass in entry here? Won't entry be 'widget' in the callback?
Another stupid mistake. fixed.
+
+ strcpy(passwd, gPassword);
This segfaults if gPassword is NULL.
True. Fixed. Also, should I add some logic as to what a valid password
is? or is the template capable of handling that?
Many of the comments for multiselect apply here too.
+static int gtkhandler_string(struct frontend *obj, struct question *q)
And here, for passwd.
Applied to both.
+ text = gtk_text_buffer_get_text(buffer, start, end, FALSE);
+
+ printf(text captured: %s.\n, text);
+
+ strcpy(buf, text);
This segfaults too. But this time I did
