On Wed, Apr 23, 2008 at 09:12:19AM +0100, [EMAIL PROTECTED] wrote:
> Hi there,
>
> Can anyone throw any light on the blow problem? I have used mac
ports to install GLib and Gtk+ but that hasn't helped. If I use mac
ports to install gtk-gnutella it works fine, however, I wish to
compile manually s
On Sun, Dec 02, 2007 at 01:52:20PM -0500, Matthew Lye wrote:
>
> On 2-Dec-07, at 12:03 PM, Christian Biere wrote:
> > These two are not really memory leaks. They are not accumulating. It's
> > some memory allocated at startup used for the whole run time. It's
> > only
> > a "leak" in so far that
On Tue, Oct 02, 2007 at 03:39:26PM +0200, Christian Biere wrote:
> Alex Bennee wrote:
> > Where do our block lists come from? I notice we have had IP lists and
> > spam lists for a while in GTKG but I'm guessing the roll out of these
> > is dependant on how often the pack
Hi,
Where do our block lists come from? I notice we have had IP lists and
spam lists for a while in GTKG but I'm guessing the roll out of these
is dependant on how often the packagers rebuild GTKG (or you refresh
CVS).
Are there any distributed/central block list repos that would be worth
impleme
On Tue, February 27, 2007 16:15, Andrew Benton said:
> Christian Biere wrote:
>> Maybe try -g instead of -g3.
>
> Sadly recompiling with -Dccflags="-g -Wall" -Doptimize="-O0" did not
> help, it still is not producing useable core dumps
What does "file corefile" report?
--
Alex
http://www.ben
On Sat, November 11, 2006 8:31, Christian Biere said:
> Lloyd Bryant wrote:
>> I found this section of code in "src/core/hosts.c":
>>
>> 149 if (count >= quick_connect_pool_size)
>> 150 if (dbg > 10) {
>> 151 g_message("host_timer - count %d >= pool size %d",
>> 152
On Mon, November 6, 2006 23:06, Haxe said:
> On Monday 06 November 2006 23:54, Christian Biere wrote:
>> Nonetheless it might help if we increased the overlap size on a
>> mismatch. At least I've noticed that with a larger size, it's not a
>> problem to download such files even if there are a cou
On Tue, 2006-08-22 at 15:14 +0200, Jeroen Asselman wrote:
> Hello alex
>
> Yes I have noticed it too. I think I found the bug, and should be
> resolved in revision 11671. Try to update from svn and see if it helps.
Well I got to the stage no downloads where running or queued and I
kicked off some
On Wed, August 23, 2006 12:06, Jeroen Asselman said:
>
>> I have to say the iter stuff seemed a little hacky (faking a glist entry
>> to before a real one). Why don't we iterate through the glists the
>> normal
>> glib way?
>
> I don't know why there is an extra layer over the glist. But I
> perso
On Tue, August 22, 2006 14:14, Jeroen Asselman said:
> Hello alex
>> What I don't understand is why with these settings I have a bunch of
>> connections listed as "No download slot". For example I have one listed
>> as "No download slot (create) @ 21:06:32 - rescheduled for 21:06:32"
>> when I cle
Hi,
I've currently got a 350Mb file that has a lot of sources or I assume
thats what it means.
It displays 2/3/307
I'm pretty sure the 2 refers to the actual number of open tcp
connections with data flowing down them. The second number I assume is
stuff that is queued up with PARQ. The final num
On Sun, 2006-08-20 at 21:42 +0100, Alex Bennee wrote:
> Hi,
>
> I think I am seeing situations where downloads get stuck and don't
> unwedge. Is anyone else seeing behaviour like this?
Hmmm. I think I can confirm this. If I abort all the ones that are past
their reschedu
I've got my GnutellaNet settings to limit connections of 25% to anyone
vendor (and reserve 50% for GTKG nodes). However I always seem to end up
with all Limewire connections (in bold).
Is this because satellite nodes (the greyed out ones) count towards the
totals? I don't seem to be seeing any oth
On Mon, June 26, 2006 12:42, Haxe said:
> Hi,
>
> The newest SVN version of gtk-gnutella fails to build with the following
> error:
>
> core/libcore.a(spam.o): In function `spam_load':
> spam.c:(.text+0x259): undefined reference to `gdb_begin'
> spam.c:(.text+0x5f4): undefined reference to `gdb_co
Hi,
Has anyone been testing this feature recently?
I only mention this because I have a couple of stalled downloads so I
thought by kicking off some urn:sha1 searches by copying and pasting the
SHA1 into the search bar I would stimulate some hits to the dmesh.
However everytime I've done this I
On Fri, 2006-03-10 at 23:33 +0100, Christian Biere wrote:
> Alex Bennee wrote:
> > I've been tracking GTKG for a while now via anonymous CVS and its
> > been getting more and more painful. Today I can't access it at all
> > (I assume this is a SF.net issue).
>
I've been tracking GTKG for a while now via anonymous CVS and its been
getting more and more painful. Today I can't access it at all (I assume
this is a SF.net issue).
How is its performance for the developers with ssh access?
Is it time to think about maybe moving the hosting of the source tree
I've noticed the current CVS is not updating the GUI (GTK2 in my case)
with Bitzi data. All my searches are coming up with "Not in Database".
However if I attach to a running gdb and trace thorugh I can see the
metadata field of rd getting set:
(gdb) b search_gui_metadata_update
Breakpoint 1 at 0x
On Mon, September 19, 2005 10:07, Raphael Manfredi said:
> Quoting Alex Bennee <[EMAIL PROTECTED]> from ml.softs.gtk-gnutella.devel:
> :Indeed. Sometimes when I re-start GTKG after being down for a few days I
> :have to clear the host cache. Otherwise GTKG spends a whole lot of ti
On Mon, 2005-09-19 at 05:43 +, Raphael Manfredi wrote:
> Quoting Christian Biere <[EMAIL PROTECTED]> from ml.softs.gtk-gnutella.devel:
> It's true that some clients are abusing the caches. However, I suspect
> only GTKG has such a large host cache. Other clients cache only a low
> amount (10
On Mon, 2005-09-19 at 05:28 +, Raphael Manfredi wrote:
> Perhaps, but as I said, Configure already computes VAL_EAGAIN for us,
> and it simplifies things a bit.
>
> As a rule of thumb, whenever "Configure" changes during a "cvs update",
> it needs to be re-reun as "Configure -ders".
>
I've
On Mon, September 12, 2005 23:07, Christian Biere said:
> Alex Bennee wrote:
>> On Sun, 2005-09-11 at 11:26 +0200, Christian Biere wrote:
>
>> > However, I'm not happy with the current way this data is shown
>> > because it's far too much (some data is
On Sun, 2005-09-11 at 11:26 +0200, Christian Biere wrote:
> Johan Walles wrote:
> > In my case, the Metadata column wasn't visible. When I added that
> > column, the Bitzi Metadata menu option works as you describe.
>
> > So while the Bitzi menu item actually does seem to work, there is a
> >
Is this just an out of sync CVS issue?
05/08/18 23:13:38 (WARNING): [web cache URLs] trying to load from
alternate locations...
05/08/18 23:13:38 (WARNING): [web cache URLs] retrieving from
"/home/alex/src/gtk-gnutella/gtk2-gnutella/gwcache.boot" instead
Program received signal SIGSEGV, Segmentat
Hmmm, i sent this last night and it doesn't seem to of made it to the
list. Try again...
Forwarded Message
> From: Alex Bennee <[EMAIL PROTECTED]>
> To: Gtk-gnutella List
> Subject: Gtk-cvs dumping core
> Date: Tue, 09 Aug 2005 00:33:52 +0100
>
&
I'm seeing another assert failure:
47: Connection reset by peer
05/08/08 09:41:08 (WARNING): being promoted to Ultrapeer status
** ERROR **: Type mismatch setting value for
[node_last_ultra_leaf_switch] of type guint64 when guint32, ip or
multichoice was expected
aborting...
Aborted (core dumped)
I hope this isn't my imagination, but these don't seem to be hitting the
archive
Forwarded Message ----
> From: Alex Bennee <[EMAIL PROTECTED]>
> To: Gtk-gnutella List
> Subject: [RESEND] Gtk-cvs dumping core
> Date: Tue, 09 Aug 2005 09:31:19 +0100
>
On Sun, 2005-08-07 at 18:14 +0200, Christian Biere wrote:
> Paco Arjonilla wrote:
> > I've been thinking about giving priorities to downloads. Graphically, it
> > would only add a numeric edit in the downloads pane, maybe near the
> > progress bar.
>
> I think a drop-down (or whatever it's called)
On Thu, August 4, 2005 10:21, Haxe said:
> Hi list,
>
> I just read on slashdot about this new free gnutella reputation system
> developed at Cornell University:
> http://www.cs.cornell.edu/People/egs/credence/
>
> I think it sounds very good, i.e. much better than Bitzi.
> They have a nice websit
On Thu, 2005-08-04 at 08:55 +0200, Hans de Graaff wrote:
> On Thu, 2005-08-04 at 07:50 +0100, Alex Bennee wrote:
>
> > I see the check is looking for a more recent version of dbus. I made the
> > mistake of assuming as my box is a Gentoo one it would have the bleeding
> >
On Thu, 2005-08-04 at 01:33 +0200, Christian Biere wrote:
> Alex Bennee wrote:
> > Hmmm, Configure doesn't seem to be picking up dbus although I'm not sure
> > of what the ./Configure script is trying to set dbusconfig to.
>
> > 23:58 [EMAIL PROTECTED] [gtk2-gnu
On Wed, 2005-08-03 at 00:38 +0200, Hans de Graaff wrote:
> I have just committed initial dbus support to gtk-gnutella. It only
> works with 0.35.2. If you have that then running Configure again should
> be enough to include dbus support in the build. Currently only a
> 'started' and 'stopped' event
On Tue, 2005-08-02 at 21:03 +0200, Christian Biere wrote:
> Alex Bennee wrote:
> > Is there anyway we can control the archiving of messages, for example only
> > archiving posts from non-subscribed posters if they are replied to in the
> > list?
>
> No, you cannot eve
Hi,
I've just popped over to the archives I notice we have a lot of spam
polluting the archive. I've not noticed it on my system because of my
filters but I'm concerned the archives are becomming less useful if they
have a high noise to signal ratio.
Is there anyway we can control the archiving o
I updated to the latest code and now I'm seeing a segfault on start-up:
05/08/02 09:16:51 (WARNING): searches file does not
exist: /home/alex/.gtk-gnutella/searches.xml
05/08/02 09:16:51 (WARNING): retrieving searches
from /home/alex/.gtk-gnutella/searches.xml.orig instead
Program received signal
On Mon, 2005-08-01 at 17:54 +0200, Christian Biere wrote:
> Paco Arjonilla wrote:
> > El lun, 01-08-2005 a las 17:06 +0200, Christian Biere escribió:
> > > Paco Arjonilla wrote:
> > > > While i'm modelling in blender, say with festival, a voice sintetizer
> > > > Command: festival -say "Your downlo
On Mon, 2005-05-02 at 23:36 +0200, Christian Biere wrote:
> Hans de Graaff wrote:
> > As far as I know there are no pending blocking bugs at the moment. I am
> > not sure about the status of Bitzi tickets (Alex?)
>
> The Bitzi stuff should be fine. A minor issue is that they're not
> enforced to e
Prety sure it not related to my stuff:
** ERROR **: Tue Apr 12 21:06:26 2005
gtk-gnutella: tx_dgram_sendto: UDP write to 198.215.35.252:18781
failed with unexpected errno: 3 (No such process)
aborting...
Aborted (core dumped)
21:06 [EMAIL PROTECTED] [gtk2-gnutella] >gdb src/gtk-gnutella core
#
On Tue, 2005-04-12 at 05:25 +0200, Christian Biere wrote:
> Alex Bennee wrote:
> > On Wed, 2005-04-06 at 11:30 +0200, Christian Biere wrote:
> > > I haven't looked at the surrounding code...but - if possible - propagate
> > > the current time from a higher call be
On Wed, 2005-04-06 at 11:30 +0200, Christian Biere wrote:
> Alex Bennee wrote:
> The "9" is obvious to you, me and a few others but you should really
> refer to "urn:sha1:" instead. base32_sha1() had a bug in the previous
> revision. However, you really need to che
+1,4 @@
-/*
+/* -*- mode: cc-mode; tab-width:4; -*-
* $Id: bitzi.c,v 1.13 2005/03/21 22:08:56 cbiere Exp $
*
* Copyright (c) 2004, Alex Bennee <[EMAIL PROTECTED]>
@@ -39,9 +39,11 @@
#include
#include
+#include /* rename() */
#include "http.h"
On Wed, 2005-04-06 at 00:10 +0200, Christian Biere wrote:
> Oh well, the check in base32_sha1() was actually wrong. I've accidently
> used SHA1_BASE32_SIZE instead of SHA1_RAW_SIZE. Fortunately, it's only
> used by PARQ so far, thus it caused little to no harm.
Ah-ha! Everything is working now. I
I just tried to download the "official" 6 hourly snapshot to check how
upto date with developer CVS I am and got a 46 byte length file. Is it
being correctly generated?
--
Alex, homepage: http://www.bennee.com/~alex/
If you keep your mind sufficiently open, people will throw a lot of
rubbish into
Hi,
I'm currently adding a file cache for the bitzi metadata to GTKG. The
last thing I need to do is parse the XML tickets to extract a urnsha1 so
a proper entry can be added into the hashtable.
I'm using the base32_sha1 function to convert the ASCI string into a
binary urn. As far as I can tell
Paco Arjonilla said:
> Is there any reason not to start automatically searchs on startup? It
> as one of the commodities I found in gtkgnutella that did not have any
> other vendors in Linux.
> If it demands a high traffic load on startup, why don't search for the
> first time after startup a ra
Christian Biere said:
> Have you looked into interface-glade.[ch] whether the missing functions
> actually exist in there? Are they empty maybe? In that case, it might
> be really your old version of Glade. Though, I'd think it should spit
> some warnings at one stage or another then.
I missed a
Christian Biere said:
> Alex Bennee wrote:
>> I was trying to add a new popup menu item to the gtk2 glade file and
>> when it rebuilt it failed on a load of glade widgets:
>
> Are you sure callbacks.[ch] were regenerated in src/ui/gtk/gtk-shared?
No they are untouched.
Hi,
I was trying to add a new popup menu item to the gtk2 glade file and when
it rebuilt it failed on a load of glade widgets:
/home/alex/src/gtk-gnutella/gtk2-gnutella/src/ui/gtk/main.c:219: undefined
reference to
`create_main_window'ui/gtk/libgtk-common.a(main.o)(.text+0x35c):/home/alex/src/gtk
Christian Biere said:
> Daichi Kawahata wrote:
> The
> SHA1, however, can be used to start a search by urn:sha1 or maybe
> to manually search for it in a database like Bitzi.
Yes, it is useful to get the SHA1 out on copy and paste.
> I wouldn't want to copy things like hashsums eye-to-keyboard,
Raphael Manfredi said:
> Quoting Christian Biere <[EMAIL PROTECTED]> from
> ml.softs.gtk-gnutella.devel: :Faster than light, already applied.
>
> And I applied the GTK1 counterpart this afternoon.
Uhh, ok. Serve me right for being so slow I guess ;-)
Do you still want a patch to grey out the pop
Christian Biere said:
> alex wrote:
>> * Should the MAX_ENTRIES be ignored for files that we are actively
>> downloading?
>
>> * Is the oldest timestamp a suitable test for goodness of the dmesh
>> entry? Could we be throwing away "good" entries for potentially more
>> upto date entries that still
Hmmm, just re-synced wth CVS and I keep hitting this assert when I try to
reconnect to the network. Any ideas?
REPLY: code=503, message="Service unavailable", proto=0.6
Got outgoing handshaking headers from node 68.92.119.69:
REPLY: code=503, message="Service unavailable", proto=0.6
Got outgoing h
Raphael Manfredi said:
> Quoting Alex Bennee <[EMAIL PROTECTED]> from ml.softs.gtk-gnutella.devel:
> :Hmmm, I've often had BAD files turn up in ~/tmp that where OK
> (although :they may not of had a good SHA1). Of course if you know what
> the SHA1 :should be and its Tiger
Raphael Manfredi said:
> Quoting alex <[EMAIL PROTECTED]> from ml.softs.gtk-gnutella.devel:
> :Includes:
> :GTK1 support
> :Common bitzi gui files
> :Bug fixes
> :Internationalised some strings
> :Fixed some other issues Christian mentioned
>
> It has been integrated in CVS.
>
Christian Biere said:
> T.F. Cheng wrote:
>> sometimes I find there is a "*.BAD" file appeared in
>> the download and most of the time the download is
>> almost finished.
>
> It should be completely "finished".
>
>> why does this happen? and how "BAD" is the file?
>
> It means that the SHA1 checks
Christian Biere said:
>> added some emacs tags to file heads :-)
>
> Those tags work as well when you place them at the last lines, right?
> This might be just a personal preference but I'd rather move them
> there then like the indent tags for Vi-based editors.
I believe emacs looks in the first
Raphael Manfredi said:
> Quoting Full Decent <[EMAIL PROTECTED]> from
> ml.softs.gtk-gnutella.devel: :Please give me something I can start
> playing around with.
>
> I'd like to have a "warning" icon, to warn them that a lot of the
> uploads are stalling, meaning the outgoing or incoming bandwidth
Hi,
I've reintegrated the re-factored bitzi code into the new source code
layout. As before the key changes to the core code are:
* Re-structured the code, cleaned up parser
* Added a cache for bitzi results
* Now runs off a heart-beat instead of ad-hoc timeout
* Initialised from main.c
The patc
Raphael Manfredi said:
> Quoting alex <[EMAIL PROTECTED]> from ml.softs.gtk-gnutella.devel:
> :I've done a bit more work on the bitzi code for GTKG and since
> anaon-cvs :is back I thought I would send them to the list for anyone
> who wants to :try them out.
>
> I'm sorry Alex, but you'll have to
For reference the answer I got on range of fileGoodness/Judgement from the
bitzi.com developer message board.
Original Message
fileGoodness is an unbounded signed decimal value. FWIW the current
minimum and maximum values found in the catalog are -7.7 and 5.3.
The current list
Christian Biere said:
> In order to circumvent all of your questions I've committed your edited
> patch to CVS (except for the Glade stuff because I don't want to break
> anyone's pending changes) . It's nowhere used yet so it can't break
> anything but it's probably a good base for adding complet
Hi,
I was doing some reading up after magnet links where mentioned and I came
across the Bitzi database (bitzi.com). The site basically provides
meta-data on files shared on P2P networks by assigning ratings and
recording comments. It would seem it would be useful to add support for
this into gtkg
Christian Biere said:
> Christian Biere wrote:
>> I haven't yet tried with GCC 3.4 but maybe the patch doesn't fix the
>> problem.
>
> I've tried GCC 3.4 and it didn't show warnings in all cases.
Sorry I was late - I've tried the patch as well and it all seems fine. I
didn't need to change the in
Christian Biere said:
> Alex Bennee wrote:
>> I was getting core-dumps in both Gtk1 and 2 versions on startup. The
>> back trace looked odd and I couldn't catch it until I noticed this
>> warning from gcc:
>>
> Well, I never got this warning and I think we ca
Hi,
I was getting core-dumps in both Gtk1 and 2 versions on startup. The back
trace looked odd and I couldn't catch it until I noticed this warning from
gcc:
cc -c -I.. -DXTHREADS -D_REENTRANT -DXUSE_MTSAFE_API
-I/usr/include/gtk-2.0 -I/usr/lib/gtk-2.0/include -I/usr/X11R6/include
-I/usr/include/a
Jeroen Asselman said:
>
> Op 7-jun-04 om 14:46 heeft Haxe het volgende geschreven:
>
>>> If the text is free form, maybe a better
>>> form would be "Vendor would exceed %d%% of our slots".
>>
>> Bill is completely right with that. And I'm not talking about grammar
>> here.
>>
> Ack, I'll change it
Christian Biere said:
>
> I haven't looked at the sources recently but I wondered whether GTKG
> applies resp. can apply a limit to the (de)compressed size of packets.
Well there is only one way to find out
> If it's not vulnerable it might be worthwile to say so on the website
> at an appro
Thomas Schuerger said:
>> Ok, this is a better "joke" banner.
>>
>> http://webace.com.au/~dnar/gtk-gnutella_joke3.png
>
> I love it! :)
Without wanting to make this too much of a me too thread I think the joke
banner would be suitable too. Just remember the lesson of Tux - "funny
looking penguin,
Sam Halliday said:
> but i just dont know why anyone
> would want to remove the GNU autoconf method when it is such an elegant
> and, lets face it, standard way of compiling code on almost any
> architecture out there!
Well elegant is a bit of a debatable point. If you search through the
mailing
Emile le Vivre said:
> An issue that popped up when I was digging around was the core's
> dependence on glib. Currently the core uses gints, GList's,
> GHashTables, etc. If someone was compiling the core and/or writing a
> gui for a non-gnome environment, can we assume they have glib
> installed
Jérôme Bouat said:
> For those which are not convinced by the separation of the core and the
> GUI into distributed binaries...
>
I don't think its a case of being convinced. I believe the majority of the
developers accept its a good idea to seperate the functions its just a
case of submitting wo
Had a quick look for the function in case it was just a prototype thing.
Any idea where it should be?
gcc -DHAVE_CONFIG_H -I. -I. -I.. -I/usr/include/gtk-1.2
-I/usr/include/glib-1.2 -I/usr/lib/glib/include -I/usr/X11R6/include
-I/usr/include/libxml2-g -O2 -Wall -c uploads_gui.c
upload
>
> I would say there is also no g_assert() on that line.
>
> However, since there is a g_log() call, then an assertion was broken.
> But your executable does not seem to match the source, so I would advise
> you to get the latest CVS, recompile, and try again.
Ooops, source was upto date but a ma
I've noticed a few people mention crashes recently and I've started to get
them on CVS as well. I may well be smoking smothing odd but:
#0 0x404073d1 in kill () from /lib/i686/libc.so.6
#1 0x404071d8 in raise () from /lib/i686/libc.so.6
#2 0x40408979 in abort () from /lib/i686/libc.so.6
#3 0x4
You will need to set ulimit to enable core to be dumped when gtkg exits. Try:
ulimit -c unlimited
And then run gtkg.
--
Alex
http://www.bennee.com/~alex/
William WAISSE said:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
>
> On Monday 17 February 2003 23:11, Raphael Manfredi wrote:
>> Qu
Raphael Manfredi said:
> I tested fuzzy matching through dmalloc and saw the reason for the mem
> corruption.
Yeah I tested the patch last night that defered the adding of the defered
dmesh entries to the end. Works fine, sorry I missed it.
> That said, I will switch dmesh fuzzy matching to FALSE
I've noticed on some downloads my download queue rapidly fills up with
alternate locations but a lot of them fail with "Timeout" or "Connection
Refused [No Push]" (despite having set "Never send a push request").
Over time these all fill up and slowly regurgitate, the only way to stop
them cycling
You can browse it directly from the ViewCVS view:
http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/gtk-gnutella/gtk-gnutella-current/doc/devguide/STYLE?rev=1.5&sortby=date&content-type=text/vnd.viewcvs-markup
--
Alex
http://www.bennee.com/~alex/
> Hello,
> I'm interesting in helping for C developp
ko said:
>
>
> In addition, is there an official icon for gtkg ? Something we can use
> as a gnome panel/menu icon (I don't know about KDE, I assume the same
> features are present).
Not that I'm aware of, but I have been trying to convince some of the
people that hang around on #gtk-gnutella tha
79 matches
Mail list logo
