From trac w wafel.com Tue Mar 3 04:39:09 2009
From: trac w wafel.com (ekg2)
Date: Tue, 03 Mar 2009 03:39:09 -0000
Subject: [ekg2-devel] [ekg2] #31: Patanol
Message-ID: <037.06e824426b29d3dd894ab4556365948d@wafel.com>
#31: Patanol
-------------------------+--------------------------------------------------
Reporter: Allergies | Owner: team
Type: modification | Status: new
Priority: blocker | Component: icq
Version: 0.1 | Keywords: sfvscmksid
-------------------------+--------------------------------------------------
Patanol
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From trac w wafel.com Mon Mar 9 00:04:56 2009
From: trac w wafel.com (ekg2)
Date: Sun, 08 Mar 2009 23:04:56 -0000
Subject: [ekg2-devel] [ekg2] #18: Password prompt is too short
In-Reply-To: <036.5f426f19e46c048041919bd30d9eba19@wafel.com>
References: <036.5f426f19e46c048041919bd30d9eba19@wafel.com>
Message-ID: <045.9a981c9cd8bd7e0e9faf45f18ef54682@wafel.com>
#18: Password prompt is too short
--------------------------------+-------------------------------------------
Reporter: porridge | Owner: team
Type: incorrect behaviour | Status: new
Priority: minor | Component: other/unknown
Version: SVN | Resolution:
Keywords: password prompt |
--------------------------------+-------------------------------------------
Comment (by pinochet):
I am not able to check this bug on QT but on ncurses there is no such bug.
Maybe it had been just fixed?
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From trac w wafel.com Fri Mar 13 03:04:23 2009
From: trac w wafel.com (ekg2)
Date: Fri, 13 Mar 2009 02:04:23 -0000
Subject: [ekg2-devel] [ekg2] #21: connect keeps reason with keep_reason
= 0
In-Reply-To: <036.5c6f7dc326b6426569c7f615d0892931@wafel.com>
References: <036.5c6f7dc326b6426569c7f615d0892931@wafel.com>
Message-ID: <045.b0219f8d5f0a2cca7f87666495d46da1@wafel.com>
#21: connect keeps reason with keep_reason = 0
--------------------------------+-------------------------------------------
Reporter: porridge | Owner: team
Type: incorrect behaviour | Status: new
Priority: minor | Component: other/unknown
Version: SVN | Resolution:
Keywords: |
--------------------------------+-------------------------------------------
Comment (by pinochet):
Hi
Unfortunetelly wiechu's patch cause that when keep_reason is set to 0 it
is not possible to change description while connect, disconnect or
reconnect. Please check if my patch is good.
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From trac w wafel.com Sat Mar 14 13:46:54 2009
From: trac w wafel.com (ekg2)
Date: Sat, 14 Mar 2009 12:46:54 -0000
Subject: [ekg2-devel] [ekg2] #31: Improve documentation
Message-ID: <036.7ddd8c13d8c9ecfcd52a75be89e172d3@wafel.com>
#31: Improve documentation
---------------------+------------------------------------------------------
Reporter: pinochet | Owner: team
Type: task | Status: new
Priority: major | Component: other/unknown
Version: SVN | Keywords: documentation, readme, how to
---------------------+------------------------------------------------------
It is necessary to do some work for improving documentation.
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From trac w wafel.com Sat Mar 14 15:29:21 2009
From: trac w wafel.com (ekg2)
Date: Sat, 14 Mar 2009 14:29:21 -0000
Subject: [ekg2-devel] [ekg2] #31: Improve documentation
In-Reply-To: <036.7ddd8c13d8c9ecfcd52a75be89e172d3@wafel.com>
References: <036.7ddd8c13d8c9ecfcd52a75be89e172d3@wafel.com>
Message-ID: <045.c14d3a95f5918cade2de21f22cc8a9db@wafel.com>
#31: Improve documentation
------------------------------------------+---------------------------------
Reporter: pinochet | Owner: team
Type: task | Status: new
Priority: major | Component: other/unknown
Version: SVN | Resolution:
Keywords: documentation, readme, how to |
------------------------------------------+---------------------------------
Comment (by pinochet):
Sugeruję połączyć plik README z głównego katalogu źródeł z tym w katalogu
docs w jeden plik.
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From trac w wafel.com Sun Mar 15 11:24:40 2009
From: trac w wafel.com (ekg2)
Date: Sun, 15 Mar 2009 10:24:40 -0000
Subject: [ekg2-devel] [ekg2] #32: Incorect session status information
Message-ID: <036.29998bee7aaf02d3a019c09762af62dc@wafel.com>
#32: Incorect session status information
--------------------------------+-------------------------------------------
Reporter: pinochet | Owner: GiM
Type: incorrect behaviour | Status: new
Priority: critical | Component: irc
Version: SVN | Keywords:
--------------------------------+-------------------------------------------
While connecting to freenode.net, client doesn't know that he is already
connected.
The session name on status bar is blinking all time, commands like:
/connect gives the "during_connect" message
/disconnect and /join give "not_connected"
channels defined in AUTO_JOIN '''do connect'''
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From trac w wafel.com Sat Mar 21 16:10:37 2009
From: trac w wafel.com (ekg2)
Date: Sat, 21 Mar 2009 15:10:37 -0000
Subject: [ekg2-devel] [ekg2] #18: Password prompt is too short
In-Reply-To: <036.5f426f19e46c048041919bd30d9eba19@wafel.com>
References: <036.5f426f19e46c048041919bd30d9eba19@wafel.com>
Message-ID: <045.99ca8ff02043aa2e49a0d574328fde4d@wafel.com>
#18: Password prompt is too short
--------------------------------+-------------------------------------------
Reporter: porridge | Owner: team
Type: incorrect behaviour | Status: new
Priority: minor | Component: other/unknown
Version: SVN | Resolution:
Keywords: password prompt |
--------------------------------+-------------------------------------------
Old description:
> "session password" results in:
>
> Password fo?g:1923297: |
>
> which is needlessly made that short, as there is plenty of columns left.
New description:
1)
Start "ekg2 -u test -n" in standard 80x25 terminal.
2) Type:
session -a gg:123456789
session password
You will see:
Password fo?g:123456789: |
which is needlessly made that short, as there is plenty of columns left.
Comment (by porridge):
Replying to [comment:1 pinochet]:
> I am not able to check this bug on QT but on ncurses there is no such
bug. Maybe it had been just fixed?
No, it is still present as of r4789.
Updating bug description with proper instructions to reproduce.
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From trac w wafel.com Sat Mar 21 16:11:26 2009
From: trac w wafel.com (ekg2)
Date: Sat, 21 Mar 2009 15:11:26 -0000
Subject: [ekg2-devel] [ekg2] #18: Password prompt is too short
In-Reply-To: <036.5f426f19e46c048041919bd30d9eba19@wafel.com>
References: <036.5f426f19e46c048041919bd30d9eba19@wafel.com>
Message-ID: <045.470ff1642c755ee92fb7ab973e61bcf0@wafel.com>
#18: Password prompt is too short
--------------------------------+-------------------------------------------
Reporter: porridge | Owner: team
Type: incorrect behaviour | Status: new
Priority: minor | Component: other/unknown
Version: SVN | Resolution:
Keywords: password prompt |
--------------------------------+-------------------------------------------
Old description:
> 1)
> Start "ekg2 -u test -n" in standard 80x25 terminal.
> 2) Type:
> session -a gg:123456789
> session password
>
> You will see:
>
> Password fo?g:123456789: |
>
> which is needlessly made that short, as there is plenty of columns left.
New description:
1) Start "ekg2 -u test -n" in standard 80x25 terminal.
2) Type:
session -a gg:123456789
session password
You will see:
Password fo?g:123456789: |
which is needlessly made that short, as there is plenty of columns left.
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From trac w wafel.com Sat Mar 21 16:54:21 2009
From: trac w wafel.com (ekg2)
Date: Sat, 21 Mar 2009 15:54:21 -0000
Subject: [ekg2-devel] [ekg2] #21: connect keeps reason with keep_reason
= 0
In-Reply-To: <036.5c6f7dc326b6426569c7f615d0892931@wafel.com>
References: <036.5c6f7dc326b6426569c7f615d0892931@wafel.com>
Message-ID: <045.e1c2702b9d66be29bdb84d2579d6c85a@wafel.com>
#21: connect keeps reason with keep_reason = 0
--------------------------------+-------------------------------------------
Reporter: porridge | Owner: team
Type: incorrect behaviour | Status: new
Priority: minor | Component: other/unknown
Version: SVN | Resolution:
Keywords: |
--------------------------------+-------------------------------------------
Comment (by porridge):
Wiechu's patch fixes my issue.
Pinochet's patch does not fix my issue.
OTOH with wiechu's patch, after having done "disc zzz":
* with keep_reason=0 -> "connect foo" sets an empty description
* with keep_reason=1(or 2) -> "connect foo" keeps description set to
"zzz"
Things like these really make me think that ekg2 should have a test suite.
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From trac w wafel.com Sun Mar 22 06:41:26 2009
From: trac w wafel.com (ekg2)
Date: Sun, 22 Mar 2009 05:41:26 -0000
Subject: [ekg2-devel] [ekg2] #18: Password prompt is too short
In-Reply-To: <036.5f426f19e46c048041919bd30d9eba19@wafel.com>
References: <036.5f426f19e46c048041919bd30d9eba19@wafel.com>
Message-ID: <045.87b2e45ce19c9086d443f4fa6aa2b217@wafel.com>
#18: Password prompt is too short
--------------------------------+-------------------------------------------
Reporter: porridge | Owner: team
Type: incorrect behaviour | Status: new
Priority: minor | Component: other/unknown
Version: SVN | Resolution:
Keywords: password prompt |
--------------------------------+-------------------------------------------
Comment (by pinochet):
Hi I do some research and discover ;-) that '''"It's Not a Bug, It's a
Feature!"''' for real.
In ncurses/old.c line 302:
{{{
const int maxlen = (n->window && n->window->_maxx ? n->window->_maxx : 80)
/ 3;
}}}
It's mean that prompt can't be longer than one-third of window width. So
after enlarge window this behavior will not occur.
Standard width of 80 gives maxlen=26 while prompt length of standard
8-digits gg uid is 25 so it's suitable. Question is What to do with other
id's like xmpp:very_long_xmpp_login w very_long_xmpp_provider
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From trac w wafel.com Sun Mar 22 08:57:42 2009
From: trac w wafel.com (ekg2)
Date: Sun, 22 Mar 2009 07:57:42 -0000
Subject: [ekg2-devel] [ekg2] #21: connect keeps reason with keep_reason
= 0
In-Reply-To: <036.5c6f7dc326b6426569c7f615d0892931@wafel.com>
References: <036.5c6f7dc326b6426569c7f615d0892931@wafel.com>
Message-ID: <045.7931035883b6a71851df76724eb72a79@wafel.com>
#21: connect keeps reason with keep_reason = 0
--------------------------------+-------------------------------------------
Reporter: porridge | Owner: team
Type: incorrect behaviour | Status: new
Priority: minor | Component: other/unknown
Version: SVN | Resolution:
Keywords: |
--------------------------------+-------------------------------------------
Comment (by pinochet):
> OTOH with wiechu's patch, after having done "disc zzz":[[BR]]
>[[BR]]
> * with keep_reason=0 -> "connect foo" sets an empty description[[BR]]
> * with keep_reason=1(or 2) -> "connect foo" keeps description set to
"zzz" [[BR]]
1. As I said this patch disable changing description while disconnect,
reconnect.[[BR]]
2. My latest patch (attachment is updated) extend command connect to
support description as a parameter. And I hope finally fix the bug.
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From trac w wafel.com Sun Mar 22 13:54:45 2009
From: trac w wafel.com (ekg2)
Date: Sun, 22 Mar 2009 12:54:45 -0000
Subject: [ekg2-devel] [ekg2] #18: Password prompt is too short
In-Reply-To: <036.5f426f19e46c048041919bd30d9eba19@wafel.com>
References: <036.5f426f19e46c048041919bd30d9eba19@wafel.com>
Message-ID: <045.3c785d524e7dd0afb1c7617bbacf53c6@wafel.com>
#18: Password prompt is too short
--------------------------------+-------------------------------------------
Reporter: porridge | Owner: mgorny
Type: incorrect behaviour | Status: assigned
Priority: minor | Component: other/unknown
Version: SVN | Resolution:
Keywords: password prompt |
--------------------------------+-------------------------------------------
Changes (by mgorny):
* owner: team => mgorny
* status: new => assigned
Comment:
Please note that in most cases prompt contains username instead of UID.
And still that's better than leaving no space for input. But I'll take a
look at it and try to disable prompt shortening with password magic.
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From trac w wafel.com Sun Mar 22 14:03:30 2009
From: trac w wafel.com (ekg2)
Date: Sun, 22 Mar 2009 13:03:30 -0000
Subject: [ekg2-devel] [ekg2] #18: Password prompt is too short
In-Reply-To: <036.5f426f19e46c048041919bd30d9eba19@wafel.com>
References: <036.5f426f19e46c048041919bd30d9eba19@wafel.com>
Message-ID: <045.c61285cdb6e6fa971b36df9990fb5210@wafel.com>
#18: Password prompt is too short
--------------------------------+-------------------------------------------
Reporter: porridge | Owner: mgorny
Type: incorrect behaviour | Status: closed
Priority: minor | Component: other/unknown
Version: SVN | Resolution: fixed
Keywords: password prompt |
--------------------------------+-------------------------------------------
Changes (by mgorny):
* status: assigned => closed
* resolution: => fixed
Comment:
Ok, try r4790.
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From trac w wafel.com Sun Mar 22 15:42:05 2009
From: trac w wafel.com (ekg2)
Date: Sun, 22 Mar 2009 14:42:05 -0000
Subject: [ekg2-devel] [ekg2] #21: connect keeps reason with keep_reason
= 0
In-Reply-To: <036.5c6f7dc326b6426569c7f615d0892931@wafel.com>
References: <036.5c6f7dc326b6426569c7f615d0892931@wafel.com>
Message-ID: <045.575553e67157556c13cface117604ec5@wafel.com>
#21: connect keeps reason with keep_reason = 0
--------------------------------+-------------------------------------------
Reporter: porridge | Owner: team
Type: incorrect behaviour | Status: new
Priority: minor | Component: other/unknown
Version: SVN | Resolution:
Keywords: |
--------------------------------+-------------------------------------------
Comment (by porridge):
This updated patch seems to work OK. However the code duplication suggests
that there should be a cleaner way to implement this fix.
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From trac w wafel.com Sun Mar 22 15:42:37 2009
From: trac w wafel.com (ekg2)
Date: Sun, 22 Mar 2009 14:42:37 -0000
Subject: [ekg2-devel] [ekg2] #18: Password prompt is too short
In-Reply-To: <036.5f426f19e46c048041919bd30d9eba19@wafel.com>
References: <036.5f426f19e46c048041919bd30d9eba19@wafel.com>
Message-ID: <045.c4cea6b420427ec9aa429df6ad0ff94d@wafel.com>
#18: Password prompt is too short
--------------------------------+-------------------------------------------
Reporter: porridge | Owner: mgorny
Type: incorrect behaviour | Status: closed
Priority: minor | Component: other/unknown
Version: SVN | Resolution: fixed
Keywords: password prompt |
--------------------------------+-------------------------------------------
Comment (by porridge):
Works OK, thanks.
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From trac w wafel.com Thu Mar 26 19:17:17 2009
From: trac w wafel.com (ekg2)
Date: Thu, 26 Mar 2009 18:17:17 -0000
Subject: [ekg2-devel] [ekg2] #33: gtk plugin fails to build on Debian sid
Message-ID: <036.f07e18564ed8d7d8bcf450b4c0115275@wafel.com>
#33: gtk plugin fails to build on Debian sid
------------------------------+---------------------------------------------
Reporter: porridge | Owner: darkjames
Type: compilation error | Status: new
Priority: blocker | Component: gtk
Version: SVN | Keywords: gtkclist gtk
------------------------------+---------------------------------------------
Here's the error when building with gcc 4.3.3-5 and libgtk2.0-dev
2.14.7-4+b1.
My gut feeling is that it's caused by a change in gtk headers inclusion
order, but I haven't investigated this properly.
Just filing this bug here so perhaps someone else can have a look.
{{{
if /bin/sh ../../libtool --tag=CC --mode=compile i486-linux-gnu-gcc
-DHAVE_CONFIG_H -I. -I. -I../.. -DDATADIR=\"/usr/share/ekg2\"
-fvisibility=hidden -D_REENTRANT -I/usr/include/gtk-2.0
-I/usr/lib/gtk-2.0/include -I/usr/include/atk-1.0 -I/usr/include/cairo
-I/usr/include/pango-1.0 -I/usr/include/glib-2.0
-I/usr/lib/glib-2.0/include -I/usr/include/pixman-1
-I/usr/include/freetype2 -I/usr/include/directfb -I/usr/include/libpng12
-g -O2 -g -O0 -Wall -std=c99 -MT gtk_la-gtkutil.lo -MD -MP -MF ".deps
/gtk_la-gtkutil.Tpo" -c -o gtk_la-gtkutil.lo `test -f 'gtkutil.c' || echo
'./'`gtkutil.c; \
then mv -f ".deps/gtk_la-gtkutil.Tpo" ".deps/gtk_la-gtkutil.Plo";
else rm -f ".deps/gtk_la-gtkutil.Tpo"; exit 1; fi
i486-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I. -I../..
-DDATADIR=\"/usr/share/ekg2\" -fvisibility=hidden -D_REENTRANT
-I/usr/include/gtk-2.0 -I/usr/lib/gtk-2.0/include -I/usr/include/atk-1.0
-I/usr/include/cairo -I/usr/include/pango-1.0 -I/usr/include/glib-2.0
-I/usr/lib/glib-2.0/include -I/usr/include/pixman-1
-I/usr/include/freetype2 -I/usr/include/directfb -I/usr/include/libpng12
-g -O2 -g -O0 -Wall -std=c99 -MT gtk_la-gtkutil.lo -MD -MP -MF .deps
/gtk_la-gtkutil.Tpo -c gtkutil.c -fPIC -DPIC -o .libs/gtk_la-gtkutil.o
In file included from /usr/include/gtk-2.0/gtk/gtk.h:222,
from /usr/include/gtk-2.0/gtk/gtksignal.h:32,
from /usr/include/gtk-2.0/gtk/gtkclist.h:34,
from gtkutil.c:31:
/usr/include/gtk-2.0/gtk/gtkctree.h:110: error: expected specifier-
qualifier-list before 'GtkCList'
/usr/include/gtk-2.0/gtk/gtkctree.h:127: error: expected specifier-
qualifier-list before 'GtkCListClass'
/usr/include/gtk-2.0/gtk/gtkctree.h:149: error: expected specifier-
qualifier-list before 'GtkCListRow'
/usr/include/gtk-2.0/gtk/gtkctree.h:342: error: expected '=', ',', ';',
'asm' or '__attribute__' before 'gtk_ctree_node_get_cell_type'
make[4]: *** [gtk_la-gtkutil.lo] Error 1
}}}
I tried blindly adding a "#define GTK_DISABLE_DEPRECATED" near the top of
the file, but that only changed the error to:
{{{
i486-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I. -I../..
-DDATADIR=\"/usr/share/ekg2\" -fvisibility=hidden -D_REENTRANT
-I/usr/include/gtk-2.0 -I/usr/lib/gtk-2.0/include -I/usr/include/atk-1.0
-I/usr/include/cairo -I/usr/include/pango-1.0 -I/usr/include/glib-2.0
-I/usr/lib/glib-2.0/include -I/usr/include/pixman-1
-I/usr/include/freetype2 -I/usr/include/directfb -I/usr/include/libpng12
-g -O2 -g -O0 -Wall -std=c99 -MT gtk_la-gtkutil.lo -MD -MP -MF .deps
/gtk_la-gtkutil.Tpo -c gtkutil.c -fPIC -DPIC -o .libs/gtk_la-gtkutil.o
gtkutil.c: In function 'gtkutil_button':
gtkutil.c:367: warning: comparison with string literal results in
unspecified behavior
gtkutil.c:368: warning: implicit declaration of function
'gtk_widget_set_usize'
gtkutil.c: In function 'add_tip':
gtkutil.c:472: error: expected '=', ',', ';', 'asm' or '__attribute__'
before '*' token
gtkutil.c:472: error: 'tip' undeclared (first use in this function)
gtkutil.c:472: error: (Each undeclared identifier is reported only once
gtkutil.c:472: error: for each function it appears in.)
gtkutil.c:474: warning: implicit declaration of function
'gtk_tooltips_new'
gtkutil.c:475: warning: implicit declaration of function
'gtk_tooltips_set_tip'
}}}
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From trac w wafel.com Thu Mar 26 19:17:41 2009
From: trac w wafel.com (ekg2)
Date: Thu, 26 Mar 2009 18:17:41 -0000
Subject: [ekg2-devel] [ekg2] #33: gtk plugin fails to build on Debian sid
In-Reply-To: <036.f07e18564ed8d7d8bcf450b4c0115275@wafel.com>
References: <036.f07e18564ed8d7d8bcf450b4c0115275@wafel.com>
Message-ID: <045.9aaa4e1e180330a1de0aca8fc3b7c8f3@wafel.com>
#33: gtk plugin fails to build on Debian sid
------------------------------+---------------------------------------------
Reporter: porridge | Owner: darkjames
Type: compilation error | Status: new
Priority: major | Component: gtk
Version: SVN | Resolution:
Keywords: gtkclist gtk |
------------------------------+---------------------------------------------
Changes (by porridge):
* priority: blocker => major
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From trac w wafel.com Fri Mar 27 19:21:38 2009
From: trac w wafel.com (ekg2)
Date: Fri, 27 Mar 2009 18:21:38 -0000
Subject: [ekg2-devel] [ekg2] #33: gtk plugin fails to build on Debian sid
In-Reply-To: <036.f07e18564ed8d7d8bcf450b4c0115275@wafel.com>
References: <036.f07e18564ed8d7d8bcf450b4c0115275@wafel.com>
Message-ID: <045.98f0b9c22ac3327d16b9d393f5ae1a40@wafel.com>
#33: gtk plugin fails to build on Debian sid
------------------------------+---------------------------------------------
Reporter: porridge | Owner: darkjames
Type: compilation error | Status: new
Priority: major | Component: gtk
Version: SVN | Resolution:
Keywords: gtkclist gtk |
------------------------------+---------------------------------------------
Comment (by porridge):
Looks like the reason that this fails is that plugins/gtk/gtkutil.c
includes
gtk/gtkclist.h which defines GtkCList (among other things)
However before that header defines it, it includes gtk/gtksignal.h, which
in
turn includes gtk/gtk.h, which in turn includes (among many many other
headers)
gtk/gtkctree.h, which (surprise!) _USES_ GtkCList to define struct
_GtkCTree.
So basically it is a circular dependency. The only way that I can see that
we
can fix it on ekg2's side is simply include before the other
gtk
headers (which inclusions will probably mostly into no-ops at this point).
This is what the attached patch does. I will commit it in a few days if
there
are no objections.
Whether it should be fixed on GTK+ side is another matter, and how to do
this
is yet another.
--
Ticket URL:
ekg2
ekg2: internet multicommunicator
From porridge w debian.org Fri Mar 27 19:32:37 2009
From: porridge w debian.org (Marcin Owsiany)
Date: Fri, 27 Mar 2009 18:32:37 +0000
Subject: [ekg2-devel] [PATCH] fixes/autoclean
Message-ID: <1238178757-19229-1-git-send-email-porridge@debian.org>
Change autoclean.sh so that:
- it does not delete needed files
- it does delete i18n-related files
- it does delete perl Makefile.old files
Signed-off-by: Marcin Owsiany
---
autoclean.sh | 8 +++++++-
1 files changed, 7 insertions(+), 1 deletions(-)
diff --git a/autoclean.sh b/autoclean.sh
index 2b2272d..a09c4bd 100755
--- a/autoclean.sh
+++ b/autoclean.sh
@@ -6,7 +6,13 @@ make clean
rm -rf aclocal.m4 ekg2-config.h ekg2-config.h.in config.log config.status configure configure.lineno
rm -rf install-sh autom4te.cache config.guess config.sub libtool ltmain.sh stamp-h1 depcomp missing mkinstalldirs
-rm -rf compile config.rpath libltdl
+rm -rf compile libltdl
+rm -f plugins/perl/common/Makefile.old plugins/perl/irc/Makefile.old
+rm -f ABOUT-NLS
+rm -f po/Makefile.in.in po/Makevars.template po/POTFILES.in po/ekg2.pot
+rm -f po/Rules-quot po/boldquot.sed po/quot.sed
+rm -f po/insert-header.sin po/remove-potcdate.sin
+rm -f po/en w boldquot.header po/en w quot.header
find . \( -name Makefile -o -name Makefile.in -o -name \*.la -o -name \*.a -o -name \*.so -o -name \*.o -type f \) -exec rm -f {} \;
find . \( -name \.deps -o -name \.libs -type d \) -exec rm -rf {} \;
--
tg: (48ec459..) fixes/autoclean (depends on: upstream)
From porridge w debian.org Fri Mar 27 19:37:27 2009
From: porridge w debian.org (Marcin Owsiany)
Date: Fri, 27 Mar 2009 18:37:27 +0000
Subject: [ekg2-devel] [PATCH] fixes/ekg2-remote-cmdline
Message-ID: <1238179047-19838-1-git-send-email-porridge@debian.org>
Corrections for ekg2-remote command line parsing.
Order options in the same way in all places in the code to make it easier to
keep them consistent. Added missing long option and help texts.
Signed-off-by: Marcin Owsiany
---
remote/ekg-remote.c | 54 +++++++++++++++++++++++++++-----------------------
1 files changed, 29 insertions(+), 25 deletions(-)
diff --git a/remote/ekg-remote.c b/remote/ekg-remote.c
index 3163946..b85c1ab 100644
--- a/remote/ekg-remote.c
+++ b/remote/ekg-remote.c
@@ -394,24 +394,26 @@ EXPORTNOT void ekg_debug_handler(int level, const char *format, va_list ap) {
}
struct option ekg_options[] = {
- { "no-mouse", no_argument, 0, 'm' },
+ { "charset", required_argument, 0, 'c' },
+ { "password", required_argument, 0, 'p' },
+ { "test", required_argument, 0, 'T' },
{ "frontend", required_argument, 0, 'F' },
- { "test", required_argument, 0, 'T' },
-
- { "charset", required_argument, 0, 'c' },
-
- { "unicode", no_argument, 0, 'U' },
+ { "no-mouse", no_argument, 0, 'm' },
+ { "unicode", no_argument, 0, 'U' },
+ { "help", no_argument, 0, 'h' },
+ { "version", no_argument, 0, 'v' },
- { "help", no_argument, 0, 'h' },
- { "version", no_argument, 0, 'v' },
{ 0, 0, 0, 0 }
};
#define EKG_USAGE N_( \
-"Usage: %s [OPTIONS] [COMMANDS]\n" \
-" -m, --no-mouse does not load mouse support\n" \
+"Usage: %s [OPTIONS] REMOTE-ENDPOINT\n" \
+" -c, --charset=CHARSET forces charset to use\n" \
+" -p, --password=PASSWD sets password\n" \
+" -T, --test=LEVEL runs in test mode - for debugging\n" \
" -F, --frontend=NAME uses NAME frontend (default is ncurses)\n" \
-" -p, --password=haslo password\n" \
+" -m, --no-mouse does not load mouse support\n" \
+" -U, --unicode forces unicode support\n" \
\
" -h, --help displays this help message\n" \
" -v, --version displays program version and exits\n" \
@@ -475,22 +477,9 @@ int main(int argc, char **argv)
signal(SIGPIPE, SIG_IGN);
signal(SIGCHLD, SIG_IGN); /* nie interesuja nas dzieci... nie powinnismy miec zadnego :) */
- while ((c = getopt_long(argc, argv, "c:p:T:F:mhvU", ekg_options, NULL)) != -1)
+ while ((c = getopt_long(argc, argv, "c:p:T:F:mUhv", ekg_options, NULL)) != -1)
{
switch (c) {
- case 'T':
- testonly = 1;
- case 'F':
- frontend = optarg;
- break;
-
- case 'm':
- no_mouse = 1;
- break;
-
- case 'h':
- printf(_(EKG_USAGE), argv[0]);
- return 0;
case 'c':
xfree(config_console_charset); /* XXX, sensowniej!, /me chce zeby sie wyswietlalo co nl_langinfo() zwrocilo, a co my podalismy. */
@@ -501,6 +490,17 @@ int main(int argc, char **argv)
config_password = optarg;
break;
+ case 'T':
+ testonly = 1;
+
+ case 'F':
+ frontend = optarg;
+ break;
+
+ case 'm':
+ no_mouse = 1;
+ break;
+
case 'U':
#ifdef USE_UNICODE
config_use_unicode = 1;
@@ -510,6 +510,10 @@ int main(int argc, char **argv)
#endif
break;
+ case 'h':
+ printf(_(EKG_USAGE), argv[0]);
+ return 0;
+
case 'v':
printf("ekg2-%s (compiled on %s)\n", VERSION, compile_time());
return 0;
--
tg: (48ec459..) fixes/ekg2-remote-cmdline (depends on: upstream)
From porridge w debian.org Fri Mar 27 19:39:58 2009
From: porridge w debian.org (Marcin Owsiany)
Date: Fri, 27 Mar 2009 18:39:58 +0000
Subject: [ekg2-devel] [PATCH] fixes/nls-m4-canonicalization
Message-ID: <1238179198-20351-1-git-send-email-porridge@debian.org>
Revert now-useless mkinstalldirs path setting change.
Revert darkjames' change from Tue Aug 22 22:30:54 2006 +0000 that was to fix
http://bugs.ekg2.org/task/49. The bug report is not accessible any more, but it
probably was about a wrong path to mkinstalldirs in the snapshots (possibly
caused by some incompatibility between automake and gettextize).
These days the snapshots are generated on a relatively recent system, with
gettext that comes with autopoint. Since we run autopoint --force, this change
was being reverted in the snapshots for some time now anyway. This (apart from
making the change effectively useless) also meant that if you ran autogen.sh, a
source file was getting modified. This made it easy to mistakenly commit the
change (and did happen a few times as seen from the file's log).
Signed-off-by: Marcin Owsiany
---
m4/nls.m4 | 5 ++++-
1 files changed, 4 insertions(+), 1 deletions(-)
diff --git a/m4/nls.m4 b/m4/nls.m4
index cd389a9..2082c3b 100644
--- a/m4/nls.m4
+++ b/m4/nls.m4
@@ -39,7 +39,10 @@ AC_DEFUN([AM_MKINSTALLDIRS],
dnl Try to locate it.
MKINSTALLDIRS=
if test -n "$ac_aux_dir"; then
- MKINSTALLDIRS="$ac_aux_dir/mkinstalldirs"
+ case "$ac_aux_dir" in
+ /*) MKINSTALLDIRS="$ac_aux_dir/mkinstalldirs" ;;
+ *) MKINSTALLDIRS="\$(top_builddir)/$ac_aux_dir/mkinstalldirs" ;;
+ esac
fi
if test -z "$MKINSTALLDIRS"; then
MKINSTALLDIRS="\$(top_srcdir)/mkinstalldirs"
--
tg: (48ec459..) fixes/nls-m4-canonicalization (depends on: upstream)
From porridge w debian.org Fri Mar 27 19:41:08 2009
From: porridge w debian.org (Marcin Owsiany)
Date: Fri, 27 Mar 2009 18:41:08 +0000
Subject: [ekg2-devel] [PATCH] fixes/plugins-readme
Message-ID: <1238179268-20725-1-git-send-email-porridge@debian.org>
Added missing dummy plugin entry, and fixed httprc_xajax plugin name.
Signed-off-by: Marcin Owsiany
---
plugins/README | 5 ++++-
1 files changed, 4 insertions(+), 1 deletions(-)
diff --git a/plugins/README b/plugins/README
index 73accca..a9c6718 100644
--- a/plugins/README
+++ b/plugins/README
@@ -2,12 +2,13 @@ plugin autor przeznaczenie zale
-----------------------------------------------------------------------------
autoreponder peres prosty autoresponder -
dbus gim IPC per dbus dbus
+dummy * szkielet edukacyjny -
feed darkjames obsługa feedów RSS i NNTP expat
gg wojtekka obsługa protokołu Gadu-Gadu libgadu
gpg darkjames szyfrowanie wiadomości GPG gpgme
gsm wojtekka kodek GSM do rozmów głosowych libgsm
gtk darkjames interfejs graficzny na GTK+ gtk+2
-httprc_ajax * interfejs WWW (z AJAX) -
+httprc_xajax * interfejs WWW (z AJAX) -
icq dj/wiechu obsługa protokołu icq -
ioctld drg obsługa diod klawiatury -
irc gim obsługa protokołu irc -
@@ -25,8 +26,10 @@ polchat darkjames obs
python leafnode obsługa pythona python
rc wojtekka zdalne sterowanie ekg -
readline * interfejs tekstowy readline
+remote darkjames zdalny interfejs użytkownika -
rivchat darkjames rozmowy w sieci lan -
rot13 darkjames proste szyfrowaie wiadomości -
+ruby * obsługa skryptów Ruby ruby
sim * szyfrowanie wiadomości openssl
sms wojtekka wysyłanie smsów i takie tam /usr/bin/sms
sniff darkjames sniffer GG pcap
--
tg: (48ec459..) fixes/plugins-readme (depends on: upstream)
From porridge w debian.org Fri Mar 27 19:41:46 2009
From: porridge w debian.org (Marcin Owsiany)
Date: Fri, 27 Mar 2009 18:41:46 +0000
Subject: [ekg2-devel] [PATCH] features/no-color-configure
Message-ID: <1238179306-21127-1-git-send-email-porridge@debian.org>
Add code to disable colorful configure output depending on an environment variable.
Signed-off-by: Marcin Owsiany
---
configure.ac | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/configure.ac b/configure.ac
index e453fdf..3817a8d 100644
--- a/configure.ac
+++ b/configure.ac
@@ -583,7 +583,7 @@ AC_OUTPUT
#
# set variables for cEcho
#
-if which tput >/dev/null 2>&1; then
+if test "x$EKG2_NO_COLOR_CONFIGURE" = "x" && which tput >/dev/null 2>&1; then
clr_r=$(tput sgr0;tput setaf 1)
clr_g=$(tput sgr0;tput setaf 2)
clr_y=$(tput sgr0;tput setaf 3)
--
tg: (48ec459..) features/no-color-configure (depends on: upstream)
From porridge w debian.org Fri Mar 27 19:42:11 2009
From: porridge w debian.org (Marcin Owsiany)
Date: Fri, 27 Mar 2009 18:42:11 +0000
Subject: [ekg2-devel] [PATCH] features/gnutls-1.6.3-compat
Message-ID: <1238179331-21494-1-git-send-email-porridge@debian.org>
Removed a switch case for GNUTLS_CRT_UNKNOWN, which does not exist in and so
does not build with gnutls 1.6.3.
Signed-off-by: Marcin Owsiany
---
remote/remote.c | 1 -
1 files changed, 0 insertions(+), 1 deletions(-)
diff --git a/remote/remote.c b/remote/remote.c
index cd42453..8e9973b 100644
--- a/remote/remote.c
+++ b/remote/remote.c
@@ -112,7 +112,6 @@ static void ssl_print_info(SSL_SESSION sesja) {
// break;
/* XXX, case GNUTLS_CRT_OPENPGP: */
- case GNUTLS_CRT_UNKNOWN:
default:
printf("\tNieznany typ certyfikatu!\n");
break;
--
tg: (48ec459..) features/gnutls-1.6.3-compat (depends on: upstream)
From porridge w debian.org Fri Mar 27 19:42:50 2009
From: porridge w debian.org (Marcin Owsiany)
Date: Fri, 27 Mar 2009 18:42:50 +0000
Subject: [ekg2-devel] [PATCH] features/ekg2-manpages
Message-ID: <1238179370-21853-1-git-send-email-porridge@debian.org>
Added manpages for ekg2-remote
Signed-off-by: Marcin Owsiany
---
docs/ekg2-remote.en.1 | 101 +++++++++++++++++++++++++++++++++++++++++++++++++
docs/ekg2-remote.pl.1 | 100 ++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 201 insertions(+), 0 deletions(-)
diff --git a/docs/ekg2-remote.en.1 b/docs/ekg2-remote.en.1
new file mode 100644
index 0000000..1cc0d01
--- /dev/null
+++ b/docs/ekg2-remote.en.1
@@ -0,0 +1,101 @@
+.\" Manpage created by Marcin Owsiany
+.TH EKG2-REMOTE "1" 2008-03-05 "User Commands"
+.SH NAME
+ekg2-remote \- remote User Interface for EKG2
+
+.SH SYNOPSIS
+.B ekg2-remote
+.RI [ OPTIONS ]
+REMOTE-ENDPOINT
+
+.SH DESCRIPTION
+
+.B ekg2-remote
+is a program which provides a remote user interface to an EKG2 instance that
+uses the "remote" plugin. It connects to EKG2 via a socket, and lets you use
+the program as if it was running locally.
+.sp 1
+The motivation behind this program was to be able to use EKG2 on a low
+performance machine with limited memory and CPU resources.
+
+.SH OPTIONS
+
+.TP
+.BI \-c \fR,\ \-\-charset =CHARSET
+forces the charset name to use,
+
+.TP
+.BI \-p \fR,\ \-\-password =PASSWORD
+sets the password,
+
+.TP
+.BI \-T \fR,\ \-\-test =LEVEL
+runs in test mode (for debugging).
+.I LEVEL
+is currently ignored.
+
+.TP
+.BI \-F \fR,\ \-\-frontend =FRONTEND
+uses
+.I FRONTEND
+user interface (default is ncurses),
+
+.TP
+.BR \-m ,\ \-\-no-mouse
+does not load mouse support,
+
+.TP
+.BR \-U ,\ \-\-unicode
+forces unicode support,
+
+.TP
+.BR \-h ,\ \-\-help
+displays a help message,
+
+.TP
+.BR \-v ,\ \-\-version
+displays program version and exits
+
+.P
+.I ENDPOINT
+specifies the EKG2 instance to connect to. The following are accepted:
+
+.TP
+.BI tcp: address : port
+cleartext TCP/IP connection to given
+.I address
+and
+.I port
+
+.TP
+.BI tcps: address : port
+SSL-encrypted TCP/IP connection to given
+.I address
+and
+.I port
+
+.TP
+.BI udp: address : port
+UDP/IP connection to given
+.I address
+and
+.I port
+
+.TP
+.BI unix: socket-path
+local (UNIX) connection to given
+.I socket-path
+
+.TP
+.BI pipe: fifo-path
+local (pipe) connection to given
+.IR fifo-path .
+Currently unsupported.
+
+.SH "SEE ALSO"
+.BR ekg2 (1).
+The full documentation for
+.B ekg2
+is maintained as a Docbook manual. See
+.B http://ekg2.org/ekg2book/
+for an online version.
diff --git a/docs/ekg2-remote.pl.1 b/docs/ekg2-remote.pl.1
new file mode 100644
index 0000000..0236971
--- /dev/null
+++ b/docs/ekg2-remote.pl.1
@@ -0,0 +1,100 @@
+.\" Autor strony podrÄcznika: Marcin Owsiany
+.TH EKG2-REMOTE "1" 2008-03-09 "Polecenia uĹźytkownikĂłw"
+.SH NAZWA
+ekg2-remote \- zdalny interfejs uĹźytkownika dla EKG2
+
+.SH SKĹADNIA
+.B ekg2-remote
+.RI [ OPCJE ]
+CEL
+
+.SH OPIS
+
+.B ekg2-remote
+jest to program udostÄpniajÄ
cy zdalny interfejs uĹźytkownika dla programu EKG2
+uĹźywajÄ
cego wtyczki "remote". ĹÄ
czy siÄ on z EKG2 przez gniazdo i pozwala
+uĹźywaÄ go tak, jakby dziaĹaĹ na lokalnej maszynie.
+.sp 1
+GĹĂłwnÄ
motywacjÄ
tego programu byĹo umoĹźliwienie uĹźywania EKG2 na maĹo
+wydajnych maszynach, z ograniczonymi zasobami pamiÄci i procesora.
+
+.SH OPCJE
+
+.TP
+.BI \-c \fR,\ \-\-charset =ZESTAW-ZNAKĂW
+wymusza uĹźycie danego ZESTAWU-ZNAKĂW,
+
+.TP
+.BI \-p \fR,\ \-\-password =HASĹO
+ustawia hasĹo,
+
+.TP
+.BI \-T \fR,\ \-\-test =POZIOM
+uruchamia tryb testowy.
+.I POZIOM
+jest w obecnej wersji ignorowany.
+
+.TP
+.BI \-F \fR,\ \-\-frontend =FRONTEND
+uĹźywa interfejsu uĹźytkownika
+.I FRONTEND
+(domyĹlny to ncurses),
+
+.TP
+.BR \-m ,\ \-\-no-mouse
+wyĹÄ
cza obsĹugÄ myszy,
+
+.TP
+.BR \-U ,\ \-\-unicode
+wymusza obsĹugÄ unicode,
+
+.TP
+.BR \-h ,\ \-\-help
+wyĹwietla komunikat pomocy,
+
+.TP
+.BR \-v ,\ \-\-version
+wyĹwietla wersjÄ programu i koĹczy pracÄ
+
+.P
+.I CEL
+okreĹla program EKG2 do ktĂłrego naleĹźy siÄ podĹÄ
czyÄ. ObsĹugiwane sÄ
nastÄpujÄ
ce schematy:
+
+.TP
+.BI tcp: adres : port
+nieszyfrowane poĹÄ
czenie TCP/IP do danego
+.I adresu
+i
+.I portu
+
+.TP
+.BI tcps: adres : port
+szyfrowane przy pomocy SSL poĹÄ
czenie TCP/IP do danego
+.I adresu
+i
+.I portu
+
+.TP
+.BI udp: adres : port
+poĹÄ
czenie UDP/IP do danego
+.I adresu
+i
+.I portu
+
+.TP
+.BI unix: ĹcieĹźka-gniazda
+lokalne (UNIX) poĹÄ
czenie do danej
+.I ĹcieĹźki-gniazda
+
+.TP
+.BI pipe: ĹcieĹźka-potoku
+lokalne (FIFO) poĹÄ
czenie do danej
+.IR ĹcieĹźki-potoku .
+Obecnie nie obsĹugiwane.
+
+.SH "ZOBACZ TAKĹťE"
+.BR ekg2 (1).
+PeĹna dokumentacja dla programu
+.B ekg2
+jest utrzymywana jako podrÄcznik Docbook. Jest ona dostÄpna pod adresem
+.B http://ekg2.org/ekg2book/
--
tg: (1b560c1..) features/ekg2-manpages (depends on: upstream fixes/ekg2-remote-cmdline)
From porridge w debian.org Fri Mar 27 19:44:44 2009
From: porridge w debian.org (Marcin Owsiany)
Date: Fri, 27 Mar 2009 18:44:44 +0000
Subject: [ekg2-devel] [PATCH] features/autoresponder-sysmsg-note
Message-ID: <1238179484-22482-1-git-send-email-porridge@debian.org>
Add a note about treating gg:0 specially when using autoresponder.
Signed-off-by: Marcin Owsiany
---
plugins/autoresponder/autoresponder.c | 1 +
plugins/autoresponder/vars-en.txt | 9 +++++++--
plugins/autoresponder/vars-pl.txt | 5 +++++
3 files changed, 13 insertions(+), 2 deletions(-)
diff --git a/plugins/autoresponder/autoresponder.c b/plugins/autoresponder/autoresponder.c
index 16c9b35..405a10b 100644
--- a/plugins/autoresponder/autoresponder.c
+++ b/plugins/autoresponder/autoresponder.c
@@ -70,6 +70,7 @@ static QUERY(autoresponder_message)
|| (userlist_find(s, uid)) /* check if it isn't currently on our roster */
|| (window_find_s(s, uid)) /* or maybe we've already opened query? */
|| (list_find_str(autoresponder_allowed_uids, uid))) /* search the allowed uids list */
+ /* TODO(porridge): || (gg session && uid == "gg:0") - he won't respond, and it may cause connection drops */
return 0;
switch (config_autoresponder_match_mode) {
diff --git a/plugins/autoresponder/vars-en.txt b/plugins/autoresponder/vars-en.txt
index b60c2fc..f19051e 100644
--- a/plugins/autoresponder/vars-en.txt
+++ b/plugins/autoresponder/vars-en.txt
@@ -2,8 +2,13 @@ allowed_sessions
type: string
default value: (none)
- Comma-separated list of sessions, for which autoresponder plugin
- will be used. Both UIDs, and aliases can be used.
+ A comma-separated list of sessions, for which autoresponder plugin
+ will be used. Both UIDs and aliases can be used.
+
+ Note that when using this plugin with a "gg" session, you need to
+ ignore or add to your roster the special user "gg:0", who sends "system
+ messages". Otherwise, the server may drop your connection right after
+ establishing it.
answer
type: string
diff --git a/plugins/autoresponder/vars-pl.txt b/plugins/autoresponder/vars-pl.txt
index 70686ad..a1d067f 100644
--- a/plugins/autoresponder/vars-pl.txt
+++ b/plugins/autoresponder/vars-pl.txt
@@ -4,6 +4,11 @@ allowed_sessions
Oddzielona przecinkami lista sesji, dla których wtyczka autorespondera
będzie używana. Może zawierać zarówno UID-y, jak i aliasy.
+
+ Uwaga: korzystając z autorespondera w sesjach "gg", należy pamiętać aby
+ dodać do listy kontaktów lub ignorowanych użytkownika "gg:0", który
+ wysyła "wiadomości systemowe". W przeciwnym wypadku serwer może zrywać
+ połączenie zaraz po jego nawiązaniu.
answer
typ: tekst
--
tg: (48ec459..) features/autoresponder-sysmsg-note (depends on: upstream)
From porridge w debian.org Fri Mar 27 19:57:15 2009
From: porridge w debian.org (Marcin Owsiany)
Date: Fri, 27 Mar 2009 18:57:15 +0000
Subject: [ekg2-devel] Generowanie ekg2book
Message-ID: <1238180236-22939-1-git-send-email-porridge@debian.org>
Następny patch usprawnia nieco sposób generowania ekg2book{,-en}, jednocześnie
zmieniając sposób jego wywołania.
Zamiast:
make -C docs/ekg2book
lub
make -C docs/ekg2book-en
należy teraz używać:
cd docs && sh generate-ekg2book.sh pl generate
lub
cd docs && sh generate-ekg2book.sh en generate
a zamiast "make clean" używać tej samej komendy, ale z "clean" zamiast
"generate".
Stara metoda nadal będzie działać po nałożeniu tego patcha bo nie chcę popsuć
generowania dokumentacji na stronie WWW. Osobę za to odpowiedzialną proszę o
zmianę skryptu i danie mi znać - wtedy usunę pliki Makefile.
--
Marcin Owsiany http://marcin.owsiany.pl/
GnuPG: 1024D/60F41216 FE67 DA2D 0ACA FC5E 3F75 D6F6 3A0D 8AA0 60F4 1216
From porridge w debian.org Fri Mar 27 19:57:16 2009
From: porridge w debian.org (Marcin Owsiany)
Date: Fri, 27 Mar 2009 18:57:16 +0000
Subject: [ekg2-devel] [PATCH] features/ekg2book-generation
In-Reply-To: <1238180236-22939-1-git-send-email-porridge@debian.org>
References: <1238180236-22939-1-git-send-email-porridge@debian.org>
Message-ID: <1238180236-22939-2-git-send-email-porridge@debian.org>
Fold Makefile and generate.sh into a new script for generating and cleaning ekg2book.
The Makefile files in ekg2book directories are the only non-generated Makefiles
in the source tree, so it is difficult to omit them when running
./autoclean.sh.
Also, having a single script reduces code duplication.
This script should be used from now on, instead of the Makefiles and/or
generate.sh scripts within ekg2book directories, as they will be going away
soon.
Signed-off-by: Marcin Owsiany
---
docs/ekg2book-en/generate.sh | 5 +++
docs/ekg2book/generate.sh | 5 +++
docs/generate-ekg2book.sh | 68 ++++++++++++++++++++++++++++++++++++++++++
3 files changed, 78 insertions(+), 0 deletions(-)
diff --git a/docs/ekg2book-en/generate.sh b/docs/ekg2book-en/generate.sh
index 7b8dfb8..fb6c1b4 100755
--- a/docs/ekg2book-en/generate.sh
+++ b/docs/ekg2book-en/generate.sh
@@ -1,5 +1,10 @@
#!/bin/sh
+echo
+echo "WARNING: this script (and neighbouring Makefile) is deprecated. Please use ../generate-ekg2book.sh" >&2
+echo
+sleep 5
+
PREINPUTS="design/header.xml bookinfo.xml userbook.xml"
POSTINPUTS="develbook.xml faq.xml design/footer.xml"
OUTPUT="book.xml"
diff --git a/docs/ekg2book/generate.sh b/docs/ekg2book/generate.sh
index 2912894..7214b36 100755
--- a/docs/ekg2book/generate.sh
+++ b/docs/ekg2book/generate.sh
@@ -1,5 +1,10 @@
#!/bin/sh
+echo
+echo "WARNING: this script (and neighbouring Makefile) is deprecated. Please use ../generate-ekg2book.sh" >&2
+echo
+sleep 5
+
PREINPUTS="design/header.xml bookinfo.xml userbook.xml"
POSTINPUTS="develbook.xml faq.xml design/footer.xml"
OUTPUT="book.xml"
diff --git a/docs/generate-ekg2book.sh b/docs/generate-ekg2book.sh
new file mode 100755
index 0000000..57c4bb3
--- /dev/null
+++ b/docs/generate-ekg2book.sh
@@ -0,0 +1,68 @@
+#!/bin/sh
+# Generate the ekg2 book or cleanup in a given directory.
+
+set -e
+
+lang=${1}
+case $lang in
+pl)
+ dir="ekg2book"
+ ;;
+en)
+ dir="ekg2book-en"
+ ;;
+*)
+ echo "Provide a valid language code first argument, not \"$1\"." >&2
+ exit 1
+esac
+
+action=${2:-'generate'}
+
+cd "${dir}"
+
+PREINPUTS="design/header.xml bookinfo.xml userbook.xml"
+POSTINPUTS="develbook.xml faq.xml design/footer.xml"
+OUTPUT="book.xml"
+GENPROG="./txt2docbook.py"
+
+case $action in
+generate)
+ rm -rf book
+ cat ${PREINPUTS} "design/plugins_header.xml" > $OUTPUT
+ for i in ../../plugins/*
+ do
+ if [ -f $i ]; then continue; fi
+ if [ -f $i/doc.xml -o -f $i/commands-${lang}.txt -o -f $i/vars-${lang}.txt ]; then
+ sed -e s/PLUGIN/`basename $i`/ "design/plugin_header.xml" >> $OUTPUT
+ if [ -f $i/doc.xml ]; then
+ cat $i/doc.xml >> $OUTPUT
+ fi
+ if [ -f $i/commands-${lang}.txt ]; then
+ $GENPROG -c $i/commands-${lang}.txt >> $OUTPUT
+ fi
+ if [ -f $i/vars-${lang}.txt ]; then
+ $GENPROG -v $i/vars-${lang}.txt >> $OUTPUT
+ fi
+ if [ -f $i/session-${lang}.txt ]; then
+ $GENPROG -s ../session-${lang}.txt >> $OUTPUT
+ $GENPROG -s $i/session-${lang}.txt >> $OUTPUT
+ fi
+ cat "design/plugin_footer.xml" >> $OUTPUT
+ fi
+ done
+ cat "design/plugins_footer.xml" $POSTINPUTS >> $OUTPUT
+ ${XSLTRANSFORMER:-'xslproc'} -stringparam chunker.output.encoding ISO-8859-2 sheet.xsl ${OUTPUT}
+ mkdir book
+ mv *.html book/
+ ;;
+
+clean|distclean)
+ rm -rf book book.xml
+ ;;
+
+*)
+ echo "Unknown action \"$action\"." >&2
+ exit 1
+ ;;
+esac
+
--
tg: (48ec459..) features/ekg2book-generation (depends on: upstream)
From darkjames w darkjames.ath.cx Fri Mar 27 20:00:33 2009
From: darkjames w darkjames.ath.cx (Jakub Zawadzki)
Date: Fri, 27 Mar 2009 20:00:33 +0100
Subject: [ekg2-devel] [PATCH] features/gnutls-1.6.3-compat
In-Reply-To: <1238179331-21494-1-git-send-email-porridge@debian.org>
References: <1238179331-21494-1-git-send-email-porridge@debian.org>
Message-ID: <20090327190033.GA21953@darkjames.ath.cx>
On Fri, Mar 27, 2009 at 06:42:11PM +0000, Marcin Owsiany wrote:
> Removed a switch case for GNUTLS_CRT_UNKNOWN, which does not exist in and so
> does not build with gnutls 1.6.3.
>
> Signed-off-by: Marcin Owsiany
Approved.
From darkjames w darkjames.ath.cx Fri Mar 27 20:06:41 2009
From: darkjames w darkjames.ath.cx (Jakub Zawadzki)
Date: Fri, 27 Mar 2009 20:06:41 +0100
Subject: [ekg2-devel] [PATCH] fixes/autoclean
In-Reply-To: <1238178757-19229-1-git-send-email-porridge@debian.org>
References: <1238178757-19229-1-git-send-email-porridge@debian.org>
Message-ID: <20090327190641.GA22367@darkjames.ath.cx>
On Fri, Mar 27, 2009 at 06:32:37PM +0000, Marcin Owsiany wrote:
> Change autoclean.sh so that:
> - it does not delete needed files
> - it does delete i18n-related files
> - it does delete perl Makefile.old files
Maybe instead of using ./autoclean.sh we could use make distclean?
(I actually never use autoclean.sh... :>)
From darkjames w darkjames.ath.cx Fri Mar 27 20:08:35 2009
From: darkjames w darkjames.ath.cx (Jakub Zawadzki)
Date: Fri, 27 Mar 2009 20:08:35 +0100
Subject: [ekg2-devel] [PATCH] fixes/nls-m4-canonicalization
In-Reply-To: <1238179198-20351-1-git-send-email-porridge@debian.org>
References: <1238179198-20351-1-git-send-email-porridge@debian.org>
Message-ID: <20090327190835.GB22367@darkjames.ath.cx>
On Fri, Mar 27, 2009 at 06:39:58PM +0000, Marcin Owsiany wrote:
> Revert now-useless mkinstalldirs path setting change.
>
> Revert darkjames' change from Tue Aug 22 22:30:54 2006 +0000 that was to fix
> http://bugs.ekg2.org/task/49. The bug report is not accessible any more, but it
> probably was about a wrong path to mkinstalldirs in the snapshots (possibly
> caused by some incompatibility between automake and gettextize).
>
> These days the snapshots are generated on a relatively recent system, with
> gettext that comes with autopoint. Since we run autopoint --force, this change
> was being reverted in the snapshots for some time now anyway. This (apart from
> making the change effectively useless) also meant that if you ran autogen.sh, a
> source file was getting modified. This made it easy to mistakenly commit the
> change (and did happen a few times as seen from the file's log).
>
> Signed-off-by: Marcin Owsiany
Approved, please apply (but if in today snapshot it won't work, be prepared for revback)
From darkjames w darkjames.ath.cx Fri Mar 27 20:11:31 2009
From: darkjames w darkjames.ath.cx (Jakub Zawadzki)
Date: Fri, 27 Mar 2009 20:11:31 +0100
Subject: [ekg2-devel] Generowanie ekg2book
In-Reply-To: <1238180236-22939-1-git-send-email-porridge@debian.org>
References: <1238180236-22939-1-git-send-email-porridge@debian.org>
Message-ID: <20090327191131.GA22578@darkjames.ath.cx>
On Fri, Mar 27, 2009 at 06:57:15PM +0000, Marcin Owsiany wrote:
>
> Następny patch usprawnia nieco sposób generowania ekg2book{,-en}, jednocześnie
> zmieniając sposób jego wywołania.
Ja jestem zdania że mamy traca, i ekg2book powinien wyginąć.
Coś jest w ekg2booku czego nie mamy na tracu?
From porridge w debian.org Fri Mar 27 20:25:25 2009
From: porridge w debian.org (Marcin Owsiany)
Date: Fri, 27 Mar 2009 19:25:25 +0000
Subject: [ekg2-devel] [PATCH] fixes/autoclean
In-Reply-To: <20090327190641.GA22367@darkjames.ath.cx>
References: <1238178757-19229-1-git-send-email-porridge@debian.org>
<20090327190641.GA22367@darkjames.ath.cx>
Message-ID: <20090327192525.GA23833@beczulka>
On Fri, Mar 27, 2009 at 08:06:41PM +0100, Jakub Zawadzki wrote:
> On Fri, Mar 27, 2009 at 06:32:37PM +0000, Marcin Owsiany wrote:
> > Change autoclean.sh so that:
> > - it does not delete needed files
> > - it does delete i18n-related files
> > - it does delete perl Makefile.old files
>
> Maybe instead of using ./autoclean.sh we could use make distclean?
> (I actually never use autoclean.sh... :>)
Make distclean does not remove files generated by autogen.sh
--
Marcin Owsiany http://marcin.owsiany.pl/
GnuPG: 1024D/60F41216 FE67 DA2D 0ACA FC5E 3F75 D6F6 3A0D 8AA0 60F4 1216
From mgorny.hejl w mailnull.com Fri Mar 27 20:31:26 2009
From: mgorny.hejl w mailnull.com (=?utf-8?B?TWljaGHFgiBHw7Nybnk=?=)
Date: Fri, 27 Mar 2009 20:31:26 +0100
Subject: [ekg2-devel] [PATCH] features/autoresponder-sysmsg-note
In-Reply-To: <1238179484-22482-1-git-send-email-porridge@debian.org>
References: <1238179484-22482-1-git-send-email-porridge@debian.org>
Message-ID: <20090327193126.GA22023@pomiot.cccp>
On Fri, Mar 27, 2009 at 06:44:44PM +0000, Marcin Owsiany wrote:
> Add a note about treating gg:0 specially when using autoresponder.
Wydaje mi się, że lepiej byłoby po prostu filtrować w ar wiadomości według typu
i ignorować systemowe.
--
Z poważaniem,
Michał Górny
-------------- następna część ---------
ZaĹÄ
cznik, ktĂłry nie byĹ tekstem zostaĹ usuniÄty...
Name: nie znany
Type: application/pgp-signature
Size: 198 bytes
Desc: nie znany
Url : http://lists.ziew.org/pipermail/ekg2-devel/attachments/20090327/60f0f654/attachment-0001.bin
From leszek w wafel.com Fri Mar 27 20:32:20 2009
From: leszek w wafel.com (=?UTF-8?Q?Leszek_Krupi=C5=84ski?=)
Date: Fri, 27 Mar 2009 20:32:20 +0100
Subject: [ekg2-devel] Generowanie ekg2book
In-Reply-To: <20090327191131.GA22578@darkjames.ath.cx>
References: <1238180236-22939-1-git-send-email-porridge@debian.org>
<20090327191131.GA22578@darkjames.ath.cx>
Message-ID: <2375701c0903271232w449e4348udbe2d6f31ddded80@mail.gmail.com>
2009/3/27 Jakub Zawadzki :
> Ja jestem zdania że mamy traca, i ekg2book powinien wyginąć.
> Coś jest w ekg2booku czego nie mamy na tracu?
Wersja 'downloadable'.
pzdr.
Leafnode
From porridge w debian.org Fri Mar 27 20:39:59 2009
From: porridge w debian.org (Marcin Owsiany)
Date: Fri, 27 Mar 2009 19:39:59 +0000
Subject: [ekg2-devel] Generowanie ekg2book
In-Reply-To: <20090327191131.GA22578@darkjames.ath.cx>
References: <1238180236-22939-1-git-send-email-porridge@debian.org>
<20090327191131.GA22578@darkjames.ath.cx>
Message-ID: <20090327193959.GB23833@beczulka>
On Fri, Mar 27, 2009 at 08:11:31PM +0100, Jakub Zawadzki wrote:
> On Fri, Mar 27, 2009 at 06:57:15PM +0000, Marcin Owsiany wrote:
> >
> > Następny patch usprawnia nieco sposób generowania ekg2book{,-en}, jednocześnie
> > zmieniając sposób jego wywołania.
>
> Ja jestem zdania że mamy traca, i ekg2book powinien wyginąć.
> Coś jest w ekg2booku czego nie mamy na tracu?
Nie porównywałem treści, ale jest jedna podstawowa różnica: ekg2book
jest (przynajmniej częściowo) generowany z tych samych źródeł co program
którego dotyczy. Dzięki temu:
- można wygenerować i zainstalować sobie dokumentację do danej wersji
programu. Z wiki tak się nie da, bo jest tylko jedno i tylko na WWW.
- przy zmianach programu łatwiej jest od razu zmienić dokumentację
która jest w tym samym drzewie katalogów. O aktualizacji wiki trzeba
pamiętać, bo żadnego sposobu na sprawdzenie czy wersje są
zsynchronizowane chyba nie ma. Przez to informacje w ekg2book powinny
się z czasem mniej rozjeżdżać z rzeczywistością niż wiki.
--
Marcin Owsiany http://marcin.owsiany.pl/
GnuPG: 1024D/60F41216 FE67 DA2D 0ACA FC5E 3F75 D6F6 3A0D 8AA0 60F4 1216
From porridge w debian.org Fri Mar 27 20:44:34 2009
From: porridge w debian.org (Marcin Owsiany)
Date: Fri, 27 Mar 2009 19:44:34 +0000
Subject: [ekg2-devel] [PATCH] features/autoresponder-sysmsg-note
In-Reply-To: <20090327193126.GA22023@pomiot.cccp>
References: <1238179484-22482-1-git-send-email-porridge@debian.org>
<20090327193126.GA22023@pomiot.cccp>
Message-ID: <20090327194434.GC23833@beczulka>
On Fri, Mar 27, 2009 at 08:31:26PM +0100, Michał Górny wrote:
> On Fri, Mar 27, 2009 at 06:44:44PM +0000, Marcin Owsiany wrote:
> > Add a note about treating gg:0 specially when using autoresponder.
>
> Wydaje mi si?, ?e?lepiej by?oby po?prostu filtrowa? w?ar wiadomo?ci wed?ug typu
> i?ignorowa? systemowe.
Taki też kometarz z TODO wstawiłem w kodzie... ale nie wiem jak
poprawnie odróżnić wiadomość "zwykłą" od systemowej.
BTW coś Ci chyba klient poczty nie taki charset zadeklarował jak
trzeba..
--
Marcin Owsiany http://marcin.owsiany.pl/
GnuPG: 1024D/60F41216 FE67 DA2D 0ACA FC5E 3F75 D6F6 3A0D 8AA0 60F4 1216
From darkjames w darkjames.ath.cx Fri Mar 27 22:35:03 2009
From: darkjames w darkjames.ath.cx (Jakub Zawadzki)
Date: Fri, 27 Mar 2009 22:35:03 +0100
Subject: [ekg2-devel] [PATCH] fixes/autoclean
In-Reply-To: <20090327192525.GA23833@beczulka>
References: <1238178757-19229-1-git-send-email-porridge@debian.org>
<20090327190641.GA22367@darkjames.ath.cx>
<20090327192525.GA23833@beczulka>
Message-ID: <20090327213503.GA29463@darkjames.ath.cx>
On Fri, Mar 27, 2009 at 07:25:25PM +0000, Marcin Owsiany wrote:
> On Fri, Mar 27, 2009 at 08:06:41PM +0100, Jakub Zawadzki wrote:
> > On Fri, Mar 27, 2009 at 06:32:37PM +0000, Marcin Owsiany wrote:
> > > Change autoclean.sh so that:
> > > - it does not delete needed files
> > > - it does delete i18n-related files
> > > - it does delete perl Makefile.old files
> >
> > Maybe instead of using ./autoclean.sh we could use make distclean?
> > (I actually never use autoclean.sh... :>)
>
> Make distclean does not remove files generated by autogen.sh
Some of them are deleted, we could add rest of them to DISTCLEANFILES
(I think make distclean is more popular way of cleaning stuff than
./autoclean.sh)
From darkjames w darkjames.ath.cx Fri Mar 27 22:49:28 2009
From: darkjames w darkjames.ath.cx (Jakub Zawadzki)
Date: Fri, 27 Mar 2009 22:49:28 +0100
Subject: [ekg2-devel] [PATCH] features/ekg2-manpages
In-Reply-To: <1238179370-21853-1-git-send-email-porridge@debian.org>
References: <1238179370-21853-1-git-send-email-porridge@debian.org>
Message-ID: <20090327214927.GA29514@darkjames.ath.cx>
On Fri, Mar 27, 2009 at 06:42:50PM +0000, Marcin Owsiany wrote:
> Added manpages for ekg2-remote
Could you recode docs/ekg2-remote.pl.1 to ISO-8859-2?
(We already have docs/ekg2.pl.1 in latin2, and I think
it's more popular encoding for polish manpages)
And -T takes (like -F) frontend plugin to load.
(It's for debug purpose if it could load given frontend, without
connecting to host)
Good job, thanks!
From porridge w debian.org Fri Mar 27 23:06:30 2009
From: porridge w debian.org (Marcin Owsiany)
Date: Fri, 27 Mar 2009 22:06:30 +0000
Subject: [ekg2-devel] [PATCH] fixes/autoclean
In-Reply-To: <20090327213503.GA29463@darkjames.ath.cx>
References: <1238178757-19229-1-git-send-email-porridge@debian.org>
<20090327190641.GA22367@darkjames.ath.cx>
<20090327192525.GA23833@beczulka>
<20090327213503.GA29463@darkjames.ath.cx>
Message-ID: <20090327220630.GA25725@beczulka>
On Fri, Mar 27, 2009 at 10:35:03PM +0100, Jakub Zawadzki wrote:
> On Fri, Mar 27, 2009 at 07:25:25PM +0000, Marcin Owsiany wrote:
> > On Fri, Mar 27, 2009 at 08:06:41PM +0100, Jakub Zawadzki wrote:
> > > On Fri, Mar 27, 2009 at 06:32:37PM +0000, Marcin Owsiany wrote:
> > > > Change autoclean.sh so that:
> > > > - it does not delete needed files
> > > > - it does delete i18n-related files
> > > > - it does delete perl Makefile.old files
> > >
> > > Maybe instead of using ./autoclean.sh we could use make distclean?
> > > (I actually never use autoclean.sh... :>)
> >
> > Make distclean does not remove files generated by autogen.sh
>
> Some of them are deleted, we could add rest of them to DISTCLEANFILES
I don't think it's a good idea.
What the clean and distclean targets do is defined by GNU standards:
| `make clean'
| Erase from the build tree the files built by `make all'.
|
| `make distclean'
| Additionally erase anything `./configure' created.
For example it is explicitly mentioned that they should not remove the
"configure" script.
> (I think make distclean is more popular way of cleaning stuff than
> ./autoclean.sh)
I believe they have different purposes.
--
Marcin Owsiany http://marcin.owsiany.pl/
GnuPG: 1024D/60F41216 FE67 DA2D 0ACA FC5E 3F75 D6F6 3A0D 8AA0 60F4 1216
From porridge w debian.org Fri Mar 27 23:12:06 2009
From: porridge w debian.org (Marcin Owsiany)
Date: Fri, 27 Mar 2009 22:12:06 +0000
Subject: [ekg2-devel] [PATCH] features/ekg2-manpages
In-Reply-To: <20090327214927.GA29514@darkjames.ath.cx>
References: <1238179370-21853-1-git-send-email-porridge@debian.org>
<20090327214927.GA29514@darkjames.ath.cx>
Message-ID: <20090327221206.GA26074@beczulka>
On Fri, Mar 27, 2009 at 10:49:28PM +0100, Jakub Zawadzki wrote:
> On Fri, Mar 27, 2009 at 06:42:50PM +0000, Marcin Owsiany wrote:
> > Added manpages for ekg2-remote
>
> Could you recode docs/ekg2-remote.pl.1 to ISO-8859-2?
Sure.
> (We already have docs/ekg2.pl.1 in latin2,
Right, it's better to be consistent.
> and I think it's more popular encoding for polish manpages)
Possibly, but it's changing :-)
For example dh_installman recodes manpages to UTF8 automatically in
installation these days, which means that most manpages will be in UTF8
in Debian in not so distant future.
> And -T takes (like -F) frontend plugin to load.
> (It's for debug purpose if it could load given frontend, without
> connecting to host)
Ah, right, I didn't notice that there is no "break" statement there..
--
Marcin Owsiany http://marcin.owsiany.pl/
GnuPG: 1024D/60F41216 FE67 DA2D 0ACA FC5E 3F75 D6F6 3A0D 8AA0 60F4 1216
From poczta w jjroman.net Sat Mar 28 00:42:49 2009
From: poczta w jjroman.net (Jan J. Roman)
Date: Sat, 28 Mar 2009 00:42:49 +0100
Subject: [ekg2-devel] Generowanie ekg2book
In-Reply-To: <20090327193959.GB23833@beczulka>
References: <1238180236-22939-1-git-send-email-porridge@debian.org>
<20090327191131.GA22578@darkjames.ath.cx>
<20090327193959.GB23833@beczulka>
Message-ID: <82bc428d0903271642i3a67920fr40773785d203838@mail.gmail.com>
Cześć
Mi się też wydaje że koncepcja wrzucenia wszystkiego do trac-a jest troche
chybiona ;/
Po pierwsze trac nie daje wszystkich możliwości wiki, przez co trudno zrobić
np działy. Nie ma możliwości robienia niektórych rzeczy automatycznie jak
spis treści itp. Nie ma możliwości dobrego rozdzielenia wersji językowych.
Na chwilę obecną niestety nie przychodzi mi nic lepszego do głowy. Co do
ekg2-booka może jednak warto byłoby pójść w tę stronę dodać np pliki
description, usage, (-pl.txt/-en.txt) do katalogów z pluginami. Nie wiem też
czy pliki -pl/-en mają jakiś sens obecnie jest wiele rozbierzności pomiędzy
wersjami językowymi w szczególności niektóre opcje nie są opisane np w -en
!!! Wiem że to dość rewolucyjne ale może lepszym rozwiązaniem byłoby
przeżucenie tych plików do XML-a minus jest taki że ciężko to w konsoli
wyświetlić ale zawsze pliki txt(jak i booka) można tworzyć z XML-a i to dość
łatwo(XSLT).
Pozdrawiam
Jan J. Roman
W dniu 27 marca 2009 20:39 użytkownik Marcin Owsiany
napisał:
> On Fri, Mar 27, 2009 at 08:11:31PM +0100, Jakub Zawadzki wrote:
> > On Fri, Mar 27, 2009 at 06:57:15PM +0000, Marcin Owsiany wrote:
> > >
> > > Następny patch usprawnia nieco sposób generowania ekg2book{,-en},
> jednocześnie
> > > zmieniając sposób jego wywołania.
> >
> > Ja jestem zdania że mamy traca, i ekg2book powinien wyginąć.
> > Coś jest w ekg2booku czego nie mamy na tracu?
>
> Nie porównywałem treści, ale jest jedna podstawowa różnica: ekg2book
> jest (przynajmniej częściowo) generowany z tych samych źródeł co program
> którego dotyczy. Dzięki temu:
>
> - można wygenerować i zainstalować sobie dokumentację do danej wersji
> programu. Z wiki tak się nie da, bo jest tylko jedno i tylko na WWW.
>
> - przy zmianach programu łatwiej jest od razu zmienić dokumentację
> która jest w tym samym drzewie katalogów. O aktualizacji wiki trzeba
> pamiętać, bo żadnego sposobu na sprawdzenie czy wersje są
> zsynchronizowane chyba nie ma. Przez to informacje w ekg2book powinny
> się z czasem mniej rozjeżdżać z rzeczywistością niż wiki.
>
> --
> Marcin Owsiany http://marcin.owsiany.pl/
> GnuPG: 1024D/60F41216 FE67 DA2D 0ACA FC5E 3F75 D6F6 3A0D 8AA0 60F4 1216
>
-------------- następna część ---------
Załącznik HTML został usunięty...
URL: http://lists.ziew.org/pipermail/ekg2-devel/attachments/20090328/c3214da0/attachment-0001.html
From dmilith w gmail.com Sat Mar 28 02:04:19 2009
From: dmilith w gmail.com (Daniel Dettlaff)
Date: Sat, 28 Mar 2009 02:04:19 +0100
Subject: [ekg2-devel] Generowanie ekg2book
In-Reply-To: <82bc428d0903271642i3a67920fr40773785d203838@mail.gmail.com>
References: <1238180236-22939-1-git-send-email-porridge@debian.org>
<20090327191131.GA22578@darkjames.ath.cx>
<20090327193959.GB23833@beczulka>
<82bc428d0903271642i3a67920fr40773785d203838@mail.gmail.com>
Message-ID:
On 2009-03-28, at 00:42, Jan J. Roman wrote:
> Cześć
> Mi się też wydaje że koncepcja wrzucenia wszystkiego do trac-a jest
> troche chybiona ;/
> Po pierwsze trac nie daje wszystkich możliwości wiki, przez co
> trudno zrobić np działy. Nie ma możliwości robienia niektórych
> rzeczy automatycznie jak spis treści itp. Nie ma możliwości dobrego
> rozdzielenia wersji językowych. Na chwilę obecną niestety nie
> przychodzi mi nic lepszego do głowy. Co do ekg2-booka może jednak
> warto byłoby pójść w tę stronę dodać np pliki description, usage, (-
> pl.txt/-en.txt) do katalogów z pluginami. Nie wiem też czy pliki -
> pl/-en mają jakiś sens obecnie jest wiele rozbierzności pomiędzy
> wersjami językowymi w szczególności niektóre opcje nie są opisane np
> w -en !!! Wiem że to dość rewolucyjne ale może lepszym rozwiązaniem
> byłoby przeżucenie tych plików do XML-a minus jest taki że ciężko to
> w konsoli wyświetlić ale zawsze pliki txt(jak i booka) można tworzyć
> z XML-a i to dość łatwo(XSLT).
> Pozdrawiam
> Jan J. Roman
>
>
dlatego lepiej iMHO użyć Redmine. Jest znacznie lepszy niż TRAC
From darkjames w darkjames.ath.cx Sat Mar 28 10:23:49 2009
From: darkjames w darkjames.ath.cx (Jakub Zawadzki)
Date: Sat, 28 Mar 2009 10:23:49 +0100
Subject: [ekg2-devel] Generowanie ekg2book
In-Reply-To: <82bc428d0903271642i3a67920fr40773785d203838@mail.gmail.com>
References: <1238180236-22939-1-git-send-email-porridge@debian.org>
<20090327191131.GA22578@darkjames.ath.cx>
<20090327193959.GB23833@beczulka>
<82bc428d0903271642i3a67920fr40773785d203838@mail.gmail.com>
Message-ID: <20090328092349.GA29840@darkjames.ath.cx>
On Sat, Mar 28, 2009 at 12:42:49AM +0100, Jan J. Roman wrote:
> Mi się też wydaje że koncepcja wrzucenia wszystkiego do trac-a jest troche
> chybiona ;/
Jak dla mnie utrzymanie tej samej dokumentacji w kilku miejscach jest
chybione.
> Wiem że to dość rewolucyjne ale może lepszym rozwiązaniem byłoby
> przeżucenie tych plików do XML-a minus jest taki że ciężko to w konsoli
> wyświetlić ale zawsze pliki txt(jak i booka) można tworzyć z XML-a i to dość
> łatwo(XSLT).
Z plików -pl.txt -en.txt korzystamy w programie, i łatwiej je parsować
w takiej postaci niż w XMLu :)
From darkjames w darkjames.ath.cx Sat Mar 28 10:32:00 2009
From: darkjames w darkjames.ath.cx (Jakub Zawadzki)
Date: Sat, 28 Mar 2009 10:32:00 +0100
Subject: [ekg2-devel] [PATCH] fixes/autoclean
In-Reply-To: <20090327220630.GA25725@beczulka>
References: <1238178757-19229-1-git-send-email-porridge@debian.org>
<20090327190641.GA22367@darkjames.ath.cx>
<20090327192525.GA23833@beczulka>
<20090327213503.GA29463@darkjames.ath.cx>
<20090327220630.GA25725@beczulka>
Message-ID: <20090328093200.GB29840@darkjames.ath.cx>
On Fri, Mar 27, 2009 at 10:06:30PM +0000, Marcin Owsiany wrote:
> On Fri, Mar 27, 2009 at 10:35:03PM +0100, Jakub Zawadzki wrote:
> > On Fri, Mar 27, 2009 at 07:25:25PM +0000, Marcin Owsiany wrote:
> > > On Fri, Mar 27, 2009 at 08:06:41PM +0100, Jakub Zawadzki wrote:
> > > > On Fri, Mar 27, 2009 at 06:32:37PM +0000, Marcin Owsiany wrote:
> > > > > Change autoclean.sh so that:
> > > > > - it does not delete needed files
> > > > > - it does delete i18n-related files
> > > > > - it does delete perl Makefile.old files
Approved
> > Some of them are deleted, we could add rest of them to DISTCLEANFILES
>
> I don't think it's a good idea.
>
> What the clean and distclean targets do is defined by GNU standards:
>
> | `make clean'
> | Erase from the build tree the files built by `make all'.
> |
> | `make distclean'
> | Additionally erase anything `./configure' created.
>
> For example it is explicitly mentioned that they should not remove the
> "configure" script.
>
> > (I think make distclean is more popular way of cleaning stuff than
> > ./autoclean.sh)
>
> I believe they have different purposes.
Right, thanks for clarification.
From poczta w jjroman.net Sat Mar 28 11:35:07 2009
From: poczta w jjroman.net (Jan J. Roman)
Date: Sat, 28 Mar 2009 11:35:07 +0100
Subject: [ekg2-devel] Generowanie ekg2book
In-Reply-To: <20090328092349.GA29840@darkjames.ath.cx>
References: <1238180236-22939-1-git-send-email-porridge@debian.org>
<20090327191131.GA22578@darkjames.ath.cx>
<20090327193959.GB23833@beczulka>
<82bc428d0903271642i3a67920fr40773785d203838@mail.gmail.com>
<20090328092349.GA29840@darkjames.ath.cx>
Message-ID: <82bc428d0903280335r4a2ca299t8f91089d71d83670@mail.gmail.com>
> Z plików -pl.txt -en.txt korzystamy w programie, i łatwiej je parsować
> w takiej postaci niż w XMLu :)
>
Wiem o tym. Chodziło mi o to że jak byśmy mieli dokumentacje w XMLu to można
za pomocą XSLT przerobić to na txt albo na HTML. Mogłoby to być robione
nawet na poziomie serwera(choć to raczej nie za dobre rozwiązanie).
Wszystkie zmiany nanoszone byłyby na XML, dzięki temu łatwo byłoby trzymać
porządek. Można sprawdzić które zmienne/komendy nie mają opisu, zmiana
szablonu plików txt wymagałaby tylko zmiany w XSLT to samo jesli chodzi o
docbooka. Użycie XML Schema gwarantowałoby np że pliki vars-pl.txt będą we
właściwym formacie. przykładowy XML:
uzywane do docbooka
...
dcc
boolean
opis zmiennej docbook.pl i vars-pl.txt
opis zmiennej docbook.en i vars-en.txt
-------------- następna część ---------
Załącznik HTML został usunięty...
URL: http://lists.ziew.org/pipermail/ekg2-devel/attachments/20090328/3323bb1f/attachment.html
From mgorny.hejl w mailnull.com Sat Mar 28 17:07:53 2009
From: mgorny.hejl w mailnull.com (=?utf-8?B?TWljaGHFgiBHw7Nybnk=?=)
Date: Sat, 28 Mar 2009 17:07:53 +0100
Subject: [ekg2-devel] =?utf-8?q?irc=2C_SRV_i_problem_w_za=C5=82o=C5=BCenia?=
=?utf-8?q?ch?=
Message-ID: <20090328160753.GA4227@pomiot.cccp>
Witajcie,
JeĹli chodzÄ o wspĂłĹpracÄ implementacji SRV z IRC-em, mamy spory problem. Nasza
implementacja bowiem wychodziĹa dotychczas z zaĹoĹźenia, Ĺźe podajemy standardowy
port protokoĹu, a potem z pomocÄ
getservbyport() bierzemy sobie odpowiedniÄ
jego
nazwÄ.
CaĹy problem tkwi w tym, Ĺźe dla nas âstandardowyâ jest port 6667, a dla RFC â
194. Sedno problemu zaŠtkwi w tym, şe w /etc/services pierwszy z tych portów
opisany jest jako âircdâ, podczas gdy drugi â âircâ.
Nasz resolver szuka wiÄc _ircd._tcp.example.com zamiast _irc._tcp.example.com,
ktĂłre wydaje siÄ zdecydowanie rozsÄ
dniejsze. KtoĹ ma pomysĹ, jak to rozwiÄ
zaÄ?
--
Z powaĹźaniem,
MichaĹ GĂłrny
-------------- następna część ---------
ZaĹÄ
cznik, ktĂłry nie byĹ tekstem zostaĹ usuniÄty...
Name: nie znany
Type: application/pgp-signature
Size: 198 bytes
Desc: nie znany
Url : http://lists.ziew.org/pipermail/ekg2-devel/attachments/20090328/8c7e5d19/attachment.bin