Timestamps are in GMT/BST.
[0:23] * toad_ (n=toad@) Quit (Remote closed the connection)
[1:09] * kgblol (i=kgbol@) has joined #freenet
[1:09] <kgblol> hello
[1:11] <Luke771> hi
[1:11] * NEOatNHNG1 (n=neo@) has joined #freenet
[1:15] * Luke771 (n=luke@) Quit ("[add smart/funny quote here]")
[1:15] <kgblol> freenet is *always* eating 100m ram and 100m of swap. [Vista, Java 1.6.0_07-b06] No matter what options (memory realted) i will change, the memory usage is always the same :| It is any way to minime it ? (My current freenet databse is about 1gb)
[1:16] <kgblol> i just wish to leave freenet running 24h/7 but its eating alot too much memory.And migrating to linux is nt
[1:17] <kgblol> not an option :(
[1:30] * NEOatNHNG (n=neo@) Quit (Read error: 113 (No route to host))
[1:33] * infinity0 (n=infinity@) Quit ("Leaving.")
[1:38] * kgblol (i=kgbol@) has left #freenet
[1:47] * FrinkC (n=FrinkC@) Quit ("Bye")
[1:53] <TheSeeker> toad: note... when I used my merged up db4o jar the existing persistent requests didn't seem to convert to new requests in db4o properly. nuking the node.db4o abd re-starting the downloads after creation seems fine though.
[1:56] <TheSeeker> Don't seem to be able to change priorities of fproxy downloads using Thaw anymore. permission issue maybe? Will try with a few downloads started with thaw.
[1:59] <TheSeeker> hmm, nevermind, just lots of lag. forgot about that. database gets clogged with requests, and fcp stuff get shoved to the back of the queue along with everything else it seems.
[2:11] * archypetro (n=hellequi@) has joined #freenet
[2:19] * gasi_ (n=gasi@) has joined #freenet
[2:24] * sanity_ (n=ian@) has joined #freenet
[2:24] * ChanServ sets mode +o sanity_
[2:25] * archangelpetro (n=hellequi@) Quit (Read error: 110 (Connection timed out))
[2:32] * gasi (n=gasi@) Quit (Read error: 110 (Connection timed out))
[2:47] * n0ob (n=travis@) has joined #freenet
[2:50] * TheBishop_ (n=TheBisho@) Quit ("Verlassend")
[2:58] * sanity_ (n=ian@) Quit ()
[3:19] * kalkin-- (n=kalkin-@) has joined #freenet
[3:35] * kalkin- (n=kalkin-@) Quit (Read error: 110 (Connection timed out))
[3:44] * Ratchet__ (n=xxxx@) has joined #freenet
[3:59] * Ratchet (n=xxxx@) Quit (Read error: 110 (Connection timed out))
[4:39] * FrinkC (n=FrinkC@) has joined #freenet
[5:33] * n0ob (n=travis@) Quit ("Now onto something more interesting")
[6:13] * vi390 (n=fc@) has joined #freenet
[6:23] * Dieppe (n=clement@) Quit (Read error: 113 (No route to host))
[6:44] * Lykos (n=wolfgang@) has joined #freenet
[6:45] <Lykos> Hi how do I tell my freenet on linux to regulate the bandwith on the TERMINAL? Without the "web"-interface?
[6:46] <Lykos> to add this to the chrontab to regulate the bandwith depending to the daytime?
[6:58] <TheSeeker> it's probably possible via FCP ... but I don't know how it's done. maybe sniff some node<->Thaw traffic and mess with the node options in Thaw?
[7:00] <TheSeeker> http://wiki.freenetproject.org/FCP2p0ClientHello
[7:01] <TheSeeker> http://wiki.freenetproject.org/FCP2p0ModifyConfig
[7:04] * Ratchet__ (n=xxxx@) Quit ("Reconnecting")
[7:04] * Ratchet (n=xxxx@) has joined #freenet
[7:06] <Lykos> I think i have just found it... not nice::a "hack", but it will work
[7:07] <Lykos> thx bye
[7:07] * Lykos (n=wolfgang@) has left #freenet
[7:45] * saces (n=saces@) Quit (Read error: 110 (Connection timed out))
[7:46] * saces (n=saces@) has joined #freenet
[7:46] * ChanServ sets mode +o saces
[7:51] * Johan^mlg (n=bllarf@) has joined #freenet
[8:05] * JazZz99 (n=coltrane@) has joined #freenet
[8:09] * zerwas (n=zerwas@) has joined #freenet
[8:13] * findeton (n=findeton@) has joined #Freenet
[8:14] <findeton> hi ya all
[8:18] <zerwas> hi findeton
[8:18] <zerwas> wow, downloading with 280 kb/s
[8:19] <zerwas> can't be true ;)
[8:20] <findeton> XMLLibrarian doesn't work to me
[8:20] <findeton> I'm new to freenet
[8:20] <findeton> I enter on XMLLibrarian and then search something
[8:21] <findeton> but the web doesn't load
[8:21] <findeton> I'm even on insecure mode to be faster
[8:27] * findeton (n=findeton@) Quit ("Konversation terminated!")
[9:27] * sbc (n=ca@) has joined #freenet
[9:53] * archypetro (n=hellequi@) Quit (Remote closed the connection)
[10:07] * Mathiasdm (n=Mathias@) has joined #freenet
[10:55] * localhost1 (n=Chris@) has joined #freenet
[10:58] * infinity0 (n=infinity@) has joined #freenet
[11:00] * localhost (n=Chris@) Quit (Read error: 60 (Operation timed out))
[11:09] * apophis (n=apophis@) has joined #freenet
[11:37] * amphibian (n=toad@) has joined #freenet
[11:44] * NEOatNHNG1 (n=neo@) Quit ("Leaving.")
[11:46] <amphibian> anyone testing trunk? any complaints?
[11:47] * Nakeido (n=gmtyjk65@) has joined #freenet
[11:47] * amphibian is now known as toad_
[11:47] * ChanServ sets mode +o toad_
[11:48] <kork> toad_: the 141 gig node that was being converted crashed. to be precise, the server itself went down due to power failure... the conversion should have left the original databases intect, shouldn't it?
[11:49] <kork> intact*
[11:49] <toad_> probably
[11:49] <toad_> there are 6 stores
[11:49] <toad_> it should migrate one at a time and then maybe delete the old files - but i'm not sure if it does delete the old files
[11:50] * TimothyA (n=gmtyjk65@) Quit (Connection reset by peer)
[11:50] * Nakeido (n=gmtyjk65@) Quit (Read error: 104 (Connection reset by peer))
[11:51] <toad_> nope, it doesn't afaics
[11:53] <kork> well /storage isn't mounted yet because I only have the password at home
[11:53] <kork> we'll see what happens tonight
[11:53] <toad_> :)
[11:54] <toad_> anyone here have any objection to me releasing 1165?
[11:54] <toad_> it seems to work
[11:54] <toad_> i've asked for wider testing but i don't really want to wait for the weekend...
[11:58] * Nakeido (n=gmtyjk65@) has joined #freenet
[11:58] <toad_> released 1165
[11:59] * toad_ changes topic to 'http://freenetproject.org/download.html (1165 mandatory soon), please read before asking for help here. #freenet-refs is no more, turn on insecure mode if you don't have any friends using Freenet | http://wiki.freenetproject.org/FrequentlyAskedQuestions | logs: http://emu.freenetproject.org/irc/ | Tor blocked due to spam (get an op to voice you) | #freenet-fr #freenet-se #freenet-es #freenet-it #freenet-de | paste at http://code.bulix.org'
[12:01] <FreenetLogBot> r22843 (1165) was built successfully on emu, mirrors are updating
[12:02] <kork> let's update this sucker
[12:03] * Nakeido is now known as TimothyA
[12:07] <kork> I have just increased the http://downloads.freenetproject.org/alpha/freenet-r22843-snapshot.jar
[12:08] <kork> whoops lol
[12:08] <zerwas> updated :)
[12:08] <kork> I have just increased the node.storeSize from 30 to 45GiB
[12:08] <kork> in order to increase the capacity of the cache
[12:08] <kork> from 460334 to 690499 keys
[12:08] * Zarggg (n=z@) Quit ()
[12:08] * Hadaka (i=naked@) Quit (wolfe.freenode.net irc.freenode.net)
[12:08] <kork> /dev/mapper/freenet 40G 20G 19G 52% /home/kork/freenet
[12:09] <kork> it's now actually overbooked, but I'll keep an eye on it
[12:09] * p0s (n=1mn57@) has joined #freenet
[12:09] * ChanServ sets mode +o p0s
[12:09] <p0s> toad_: 2582 seems gone for me.
[12:13] <toad_> can you mention that on the bug?
[12:14] <toad_> kork: well you'll need to turn preallocation off
[12:14] <p0s> toad_: do i have to re-open it for that?
[12:14] <toad_> hmmm?
[12:14] <toad_> you mean restart? i dunno
[12:15] <p0s> toad_: ah commenting is possible without re-opening. okay.
[12:15] <toad_> it should be
[12:16] <toad_> was it closed?
[12:16] <p0s> toad_: you closed it.
[12:16] <p0s> toad_: oh no the SVN tool did.
[12:16] <toad_> not intentionally... nextgens' svn integration being too clever for its own good...
[12:16] <p0s> :)
[12:16] <toad_> Status assigned
[12:16] <kork> toad_: sure? I had set it to 30GiB, but only 20G of the partition are used even though the cache was actually at its key capacity
[12:16] <toad_> so it's open atm
[12:17] <toad_> kork: as of 1165, it will preallocate unless you tell it not to
[12:17] <p0s> toad_: you might have noticed that i closed some very old bugs. is that okay? i think it'd be a good idea to keep the bug tracker clean so it can actually be used for finding work.
[12:17] <toad_> p0s: if we have reasonable grounds to close them then sure
[12:17] <toad_> for example?
[12:17] <p0s> toad_: i tried only to close bugs where i'm sure that they're gone.
[12:17] <kork> toad_: oh, well, there's still only 20G used...
[12:18] <toad_> I don't believe in closing bugs purely because they're old
[12:18] <p0s> toad_: no i closed the ones which were certainly fixed a long time ago.
[12:18] <toad_> however if they're old, and relate to code that's been rewritten since, and the reporter has disappeared, and we can't reproduce...
[12:18] <toad_> then resolve->unable to reproduce or something similar is quite acceptable
[12:19] * toad_ doesn't actually use closed atm
[12:19] <p0s> i use "resolved"
[12:19] <toad_> ok
[12:19] <p0s> example: https://bugs.freenetproject.org/view.php?id=83
[12:19] <toad_> yep, that's fixed
[12:20] <toad_> granted there are a few where we don't do this yet
[12:20] <toad_> but we do at least have the infrastructure
[12:21] <p0s> well you know i like the idea of having a cleaned-up todo list to see how the progress of a project is...
[12:21] <p0s> and the bugtracker is nice.
[12:21] <toad_> you left the duplicate open ... you might want to open one and use it for those settings that need to be made NodeRestartNeeded aware??
[12:21] <p0s> didnt see the duplicate.
[12:21] <p0s> i will post your quote there.
[12:21] * Mathiasdm (n=Mathias@) Quit (Read error: 110 (Connection timed out))
[12:22] * Naked (i=naked@) has joined #freenet
[12:23] <p0s> toad_: about the file transfer and instant messengers...
[12:23] <p0s> toad_: if the two involved chat partners both use the official instant messenger of the network their on, for example ICQ 6, it will work i guess.
[12:23] <p0s> toad_: because ICQ for example is able to route the transfer over their server if the ports are blocked
[12:24] <toad_> right
[12:24] <p0s> toad_: but as sone as one of the involved persons uses an ALTERNATE client, having the ports free is a requirement
[12:24] <toad_> and *that is the common case*
[12:24] <p0s> toad_: because the alternate clients usually do not implement the transfer over the server
[12:24] <toad_> i mean the first is - usually both clients are using MSN, because MSN is microsoft and therefore popular :|
[12:25] <toad_> or ICQ because it was one of the first, or AIM because it's AOL, etc etc
[12:25] <p0s> toad_: and now we get to the point where my opinion is different
[12:25] <p0s> toad_: usually, an "expert" user, or whatever we should call it, is trying to convince a "normal user" to use freenet
[12:25] <p0s> toad_: and the "experts" use alternate IM clients and the "normal" guys use icq6
[12:26] <p0s> toad_: and then the damn filetransfers do nnot work.
[12:26] <toad_> ummm so they email the file?
[12:26] <p0s> toad_: which is annoying :)
[12:27] <toad_> big chunks of base64 text are intimidating and clumsy
[12:27] <p0s> toad_: actually the point is: i see no reason why the noderef should require linebreaks.
[12:27] <toad_> imho we need a very good reason
[12:27] <toad_> uh because it's a SimpleFieldSet
[12:27] <p0s> toad_: you could leave it in cleartext and just change the linebreaks and line beginnings to html-like tags.
[12:27] <toad_> granted it doesn't have to be
[12:27] <toad_> ugh
[12:28] <p0s> toad_: then having the line breaks still intact will be nice but broken ones will not fuck it up
[12:28] <toad_> <I AM A LINEBREAK>physical.udp=123.45.67.8:12345<I AM A LINEBREAK>...
[12:28] <p0s> toad_: yea like that :)
[12:29] <p0s> just with something else than "<" and ">"
[12:29] <toad_> SFS is designed to be a flexible format, that's why it's text based...
[12:29] <toad_> why not <> ?
[12:30] <p0s> toad_: as soon as people try to exchange refs over some bulletin board or anything else thats close to html then it will not work
[12:30] <toad_> don't they all escape <>'s anyway?
[12:31] <toad_> imho a user ever coming into contact with a noderef as a block of text is a failure
[12:31] <p0s> toad_: well does it matter for us what characters we use instead of <>? if we just use something that will not fuck up for sure then we will be on the save side
[12:31] <toad_> they'll have an automatic reaction: aiiiieeeeeeeeeeeeeeeeeeeeeeeeEEEEEEEEEEEEE!!!11!!1!!!
[12:31] <p0s> toad_: indeed. then go program the one-time-shortrefs :)
[12:32] <toad_> sure, eventually we will
[12:32] <p0s> toad_: but still the normal refs are useful. i keep mine hosted on my webserver so i can usually access it at other peoples places if i need it
[12:32] <toad_> well you could easily do that as a file
[12:32] <toad_> it doesn't need to be text for that
[12:32] <p0s> toad_: yea of course.
[12:33] <p0s> a textfile :)
[12:33] <toad_> a file with mime type application/x-freenet-reference
[12:33] <toad_> so opaque
[12:33] <p0s> mmh okay.
[12:33] <toad_> that's what we give if you click on "My Node Reference"
[12:34] <toad_> i'm seriously considering getting rid of the text version in the name of simplifying the UI
[12:34] <p0s> well i'd just do everything we are talking about. provide single-time references, prove opaque x-freenet-reference files and provide text noderefs which are copy&paste resistant, i.e. which do not rely on linebreaks. maybe base-64 encode them so that the lines are not as damn long as they are right now.
[12:35] <toad_> it's all base64 anyway
[12:35] <toad_> it's just that the field names and the line breaks aren't base64
[12:35] <p0s> toad_: i mean a large text block of base-64 encoded stuff looks kinda "handy", more handy than the current noderefs.
[12:35] <p0s> toad_: it just looks like a single beautiful piece of data :D
[12:35] <toad_> hmmm maybe
[12:36] <toad_> maybe we could just prefix each line with an ^
[12:36] <p0s> toad_: there is the code to remove the IRC nicknames in front of each line anyway
[12:36] <toad_> a consideration for UI design: we'll probably have a separate page for this
[12:37] <p0s> toad_: you could recycle that code for that purpose
[12:37] <toad_> we'll have a button somewhere for Add a New Friend
[12:37] <p0s> yea good idea.
[12:37] <toad_> and then that will go to a page which explains about noderefs, and allows you to download yours and add theirs
[12:37] <p0s> and then a subpage where people can choose how to add them.
[12:38] <toad_> well if it's a single page then we at least have room for the different kinds
[12:38] <p0s> yes.
[12:38] <toad_> and of course lots of explanatory text
[12:39] <p0s> :D
[12:40] <toad_> "To connect to your friend's node, you each need to send each other a noderef. This is a small file containing information about your node. Ideally you should exchange this out of band, for example on a USB key, so that eavesdroppers can't see it. \n\nClick here to download your noderef.\n\nClick here to add a friend's noderef.\n\nYour nodes will only connect when both have added the other's noderef."
[12:40] <p0s> nice.
[12:41] <toad_> something like that
[12:41] <p0s> you could also add an email link
[12:41] <toad_> well it wouldn't be click here to add a friend's noderef, it would be upload a friend's noderef here
[12:41] <toad_> email is hard - everyone uses webmail, and all webmails are slightly different
[12:41] <p0s> i think for email links its also possible to set the subject and the body of the mail
[12:41] <p0s> mmh yes thats right :(
[12:41] <toad_> yeah but then it'll get added to spam filters
[12:43] <p0s> it would also be nice to write a binary format for the noderef which only contains absolutely necessary information, in binary, and make the text version a base64 encode of the binary. then it should be much shorter than the current text one.
[12:44] <toad_> yeah there are various proposals and bugs regarding that
[12:44] <toad_> the main advantage is the ability to exchange noderefs over the phone
[12:44] <p0s> yep.
[12:44] <toad_> hence out of band (for most plausible attackers) and in real time
[12:44] <toad_> in real time is important for instant gratification
[12:45] <toad_> but also because IPs may change
[12:45] <toad_> hmmm which bug were you talking about?
[12:45] <toad_> w.r.t. short noderefs etc?
[12:45] <p0s> i was just being creative :)
[12:45] <toad_> yeah but where did you file your comments?
[12:45] <toad_> on which bug?
[12:45] <p0s> here:
[12:46] * toad_ needs to link it to 2587
[12:46] <p0s> https://bugs.freenetproject.org/view.php?id=1343
[12:47] <toad_> ok
[12:47] <toad_> have you upgraded yet?
[12:47] <p0s> i am planning to do that now.
[12:47] <toad_> to 1165 i mean?
[12:47] <toad_> ok
[12:47] <toad_> find out whether it blows up spectacularly
[12:47] <p0s> haha why should it :)
[12:48] <toad_> finnagle's law?
[12:48] <p0s> :)
[12:48] <p0s> is it not uploaded as stable yet?
[12:48] <toad_> yes
[12:48] * HyperDimensions (n=me@) Quit ("Leaving")
[12:48] <toad_> it is 1165, it has been inserted
[12:48] <p0s> okay.
[12:48] <toad_> brb
[12:48] <p0s> well i will CPU-upgrade and then freenet-upgrade
[12:51] <Ratchet> toad_: when did you release 1165?
[12:51] <Ratchet> an hour ago?
[12:52] <Ratchet> The update process seems to have been stuck for 1d17h; let's switch to UoM! SHOULD NOT HAPPEN! (1)
[12:52] <Ratchet> from wrapper log
[12:57] * sanity_ (n=ian@) has joined #freenet
[12:57] * ChanServ sets mode +o sanity_
[12:58] * cdent (n=cdent@) has joined #freenet
[13:00] * p0s (n=1mn57@) Quit ()
[13:02] <toad_> Ratchet: yeah that needs fixing at some point
[13:05] <Ratchet> ah
[13:12] <Ratchet> toad_: my dyndns name is now appearing in the noderef :-)
[13:13] <Ratchet> even with the local dns view of a lan ip
[13:13] <Ratchet> great, thx
[13:44] * ahuxley (n=ahuxley@) Quit ()
[13:57] * n0ob (n=travis@) has joined #freenet
[13:57] * n0ob (n=travis@) Quit (Read error: 104 (Connection reset by peer))
[13:57] * p0s (n=1mn57@) has joined #freenet
[13:58] <p0s> this should be fixed ASAP: https://bugs.freenetproject.org/view.php?id=2647
[14:00] <p0s> toad_: *poke*
[14:05] <Ratchet> toad_: what do you think about limiting the updater upload to half of gloabl bw limit? after an update, a node with low upload bw is "dead" for about 45-60minutes, because updater uses up all bw
[14:06] <p0s> the 2647 bug screws up the C:\WINDOWS folder by setting wrong permissions on it!
[14:08] <kork> jesus f* christ
[14:08] <kork> that one's bad
[14:08] <kork> I thought it just effed up the freenet dir?
[14:08] <p0s> obviously it does not.
[14:08] <kork> that's baaad
[14:08] <p0s> we need someone who is able to commit to update.cmd
[14:08] <p0s> fast.
[14:13] * phrosty (n=phrosty@) Quit ("He who asks a question is a fool for five minutes; he who does not ask a question remains a fool forever.")
[14:16] * NEOatNHNG (n=neo@) has joined #freenet
[14:38] * ddoc (n=chatzill@) Quit ("ChatZilla 0.9.83 [Firefox 2.0.0.17/2008082909]")
[14:46] * zerwas (n=zerwas@) Quit ()
[14:51] <toad_> p0s: do you have a fix for it?
[14:53] <toad_> line 47: cd /D %LOCATION%
[14:56] <toad_> p0s: how exactly did you invoke update.cmd ?
[14:56] <toad_> p0s: i can't see how it would be in the wrong directory
[14:56] <toad_> i mean, the update itself would break
[14:59] * toad_ tries it on a windows machine
[15:11] * toad__ (n=toad@) has joined #freenet
[15:12] <toad__> I suspect p0s's bug may be caused by calling update.cmd directly from cmd, and not from its own directory (if you double click it it should be run in its own dir iirc)...
[15:12] <toad__> the only problem i've encountered so far is that update.cmd waiting for the node to shut down hangs forever
[15:12] <toad__> which is bad as killing it during shutdown causes problems according to the script...
[15:14] * Nakeido (n=gmtyjk65@) has joined #freenet
[15:17] <toad__> "The service could not be controlled in its present state"
[15:17] <toad__> eh?
[15:18] <toad__> ok, it had to be killed by the wrapper :|
[15:25] <toad__> hmmm, when i run it from cmd.exe by typing the path in, it cd's to the correct directory first
[15:29] <toad__> p0s: i'm afraid i can't replicate your bug. i've fixed the first part - we need to add to the ACL not replace it - but I can't do anything about the second part - hitting the wrong target
[15:30] * vi390 (n=fc@) Quit ("Leaving")
[15:33] * TimothyA (n=gmtyjk65@) Quit (No route to host)
[15:33] <toad__> on the upside, a newly installed node seems to work pretty well
[15:34] <toad__> it's a little slow still, but at least it bootstraps fast, and it's not *hideously* slow
[15:37] <toad__> well, it is hideously slow... :|
[15:39] <TheSeeker> limit connections to K/5 instead of 20? :P
[15:39] <TheSeeker> K = output bandwidth (detected)
[15:44] <p0s> toad_: i double clicked
[15:45] <p0s> toad__: anyway, changing the "." to "%LOCATION%" is the proper way of doing things
[15:47] * caytchen (n=caytchen@) has joined #freenet
[15:50] <toad__> p0s: we cd to location
[15:50] <toad__> loads of code there relies on us being in the right place
[15:50] <toad__> and there is no other "cd"
[15:50] * Mathiasdm (n=Mathias@) has joined #freenet
[15:50] <toad__> we also have various checks that ensure that we are in the right place
[15:51] * caytchen (n=caytchen@) Quit (Read error: 104 (Connection reset by peer))
[15:51] <toad__> hmmm this is odd, the installer ...
[15:51] <toad__> wrapper | The freenet-darknet service is not installed - The specified service
[15:51] <toad__> does not exist as an installed service. (0x424)
[15:52] <toad__> odd... it's actually freenet-darknet-8888
[15:52] <TheSeeker> won't that be whatever port is defined by the user in the wizard? or is the fproxy port not a user-defined option?
[15:52] <toad__> and now the service is still in Services ...
[15:53] <toad__> it autodetects it
[15:53] <toad__> it shut it down but it didn't remove it
[15:53] <toad__> and i can't remove it from Services
[15:54] * toad__ goes back to other computer for now...
[15:54] * toad__ (n=toad@) Quit ("My damn controlling terminal disappeared!")
[15:54] * caytchen (n=caytchen@) has joined #freenet
[16:10] * Mathiasdm (n=Mathias@) Quit ("bye")
[16:13] * Nakeido (n=gmtyjk65@) Quit (Read error: 110 (Connection timed out))
[16:19] * sich (n=sich@) has joined #freenet
[16:28] * batosai (n=julien@) has joined #freenet
[16:28] * ChanServ sets mode +v batosai
[16:48] * rogus000 (n=sc@) has joined #freenet
[16:50] * rogus000 (n=sc@) Quit ()
[16:51] * sc (n=sc@) has joined #freenet
[16:54] <- *sc* http://dark-code.bulix.org/xrhoiz-68486?raw
[16:54] * cdent (n=cdent@) Quit ()
[16:57] <- *sc* http://dark-code.bulix.org/xrhoiz-68486?raw
[16:58] * toad_ (n=toad@) Quit (Remote closed the connection)
[17:05] * skxpl (n=skx@) Quit ("changing servers")
[17:06] * cdent (n=cdent@) has joined #freenet
[17:06] * skxpl (i=skx@) has joined #freenet
[17:07] * mikusr (n=mikusrak@) has joined #freenet
[17:10] * toad_ (n=toad@) has joined #freenet
[17:10] * ChanServ sets mode +o toad_
[17:19] * NEOatNHNG (n=neo@) Quit (Read error: 110 (Connection timed out))
[17:20] * sbc (n=ca@) Quit ("Ex-Chat")
[17:20] * mikusr (n=mikusrak@) has left #freenet
[17:29] * NEOatNHNG (n=neo@) has joined #freenet
[17:37] * NEOatNHNG (n=neo@) Quit ("Leaving.")
[17:38] * sc (n=sc@) Quit (Remote closed the connection)
[17:52] * |sich| (n=sich@) has joined #freenet
[17:56] * p0s (n=1mn57@) Quit ()
[18:01] * a21 (n=chatzill@) Quit (Remote closed the connection)
[18:03] * sich (n=sich@) Quit (Nick collision from services.)
[18:03] * |sich| is now known as sich
[18:08] * a21 (n=chatzill@) has joined #freenet
[18:19] <toad_> reh
[18:19] <toad_> rehi
[18:19] <toad_> what's up?
[18:37] <kork> toad_: it has restarted the migration from scratch... yay :|
[18:37] <kork> damn power surges
[18:40] <toad_> not of the stuff it's completed already
[18:40] <toad_> CHK store or CHK cache?
[18:40] * greycat (i=rfc1413@) has joined #freenet
[18:40] <kork> migrating key 80896/426242
[18:40] <toad_> of?
[18:40] <kork> store
[18:40] <kork> it had already migrated that
[18:41] <toad_> you sure?
[18:41] <kork> 100%
[18:41] <toad_> hmmm
[18:41] <toad_> maybe because it doesn't delete the old data when it's migrated it
[18:41] <kork> maybe
[18:42] <kork> oh well, we gots time =)
[18:43] <toad_> get a UPS? :)
[18:43] <kork> it's not my server
[18:43] <kork> I just have an account there
[18:43] <toad_> :|
[18:45] * mulligan (n=user@) has joined #freenet
[18:47] <mulligan> i downloaded freenet07.tar.gz. when i `./run.sh start' i get "Exception in thread "main" java.lang.NoClassDefFoundError: freenet/node/NodeStarter"
[18:47] <mulligan> any ideas?
[18:48] <toad_> does freenet-cvs-snapshot.jar or freenet-stable-latest.jar or freenet.jar exist?
[18:49] <toad_> is it zero bytes?
[19:04] <toad_> how important is it not to break insert resuming?
[19:04] <toad_> there are optimisations that can be done to inserts ... but they could be quite a lot of work, and it might be easier to merge the db4o branch *before* doing the optimisations
[19:05] <toad_> but the catch is that if we do that, existing inserts will break
[19:06] <toad_> that could greatly annoy a lot of users, no?
[19:08] <saces> block adding new insert, convert if all insert done, continue normal
[19:08] <toad_> maybe
[19:08] <toad_> conversion might be possible yes
[19:10] <FreenetLogBot> Emu wasn't able to build r22851 (WoT) -- see error message on http://emu.freenetproject.org/~svn-build/build-r22851.txt
[19:10] <Tommy[D]> mulligan: what version of java installed? What OS?
[19:15] * cdent (n=cdent@) Quit ()
[19:18] * Ash-Fox (n=N@) has joined #FreeNET
[19:18] * archangelpetro (n=hellequi@) has joined #freenet
[19:21] <mulligan> Tommy[D]: it's working now. after rm -r freenet/ and a fresh untar. The first time i tried `java' wasn't in $PATH (after that ./run.sh didn't worked even with `java' in $PATH)
[19:21] <FreenetLogBot> Emu wasn't able to build r22852 (WoT) -- see error message on http://emu.freenetproject.org/~svn-build/build-r22852.txt
[19:24] <FreenetLogBot> r22853 (WoT) was built successfully on emu, mirrors are updating
[19:26] * Artefact2 (n=Romain@) has joined #freenet
[19:33] <toad_> Artefact2: hi
[19:33] <Artefact2> toad_: hi
[19:48] <toad_> hmmm
[19:52] * mulligan (n=user@) Quit (Read error: 104 (Connection reset by peer))
[19:55] <toad_> what is the bash syntax for comparing one string with another?
[19:55] <batosai> [ $1 -es $2 ]
[19:55] <batosai> [ $1 -eq $2 ]
[19:55] <batosai> sorry
[19:56] <batosai> damn
[19:56] <batosai> I'm half asleep
[19:56] <batosai> [ $var == "value" ]
[19:57] <greycat> ONE equal sign, not two.
[19:57] <greycat> [ "$var" = value ]
[19:57] <greycat> assuming this is for /bin/sh and not some weird java variant
[19:58] <toad_> -bash: [: too many arguments
[19:58] <toad_> -bash: [: ==: unary operator expected
[19:58] <toad_> lots of them
[19:59] <greycat> [ "$var" = value ]
[19:59] <greycat> http://wooledge.org/mywiki/BashPitfalls
[19:59] <toad_> ah you have to quote them right
[20:00] <greycat> bash will let the == work, but it'll blow up in /bin/sh
[20:00] <greycat> lack of quotes around the $var will blow up in any shell, though, if the variable contains whitespace
[20:00] * Nakeido (n=gmtyjk65@) has joined #freenet
[20:00] * Nakeido is now known as TimothyA
[20:01] * Artefact2 (n=Romain@) Quit (Read error: 104 (Connection reset by peer))
[20:01] * Artefact2 (n=Romain@) has joined #freenet
[20:01] <toad_> find -iname "*.java" | (while read filename; do cat "$filename" | (match=false; y="blah"; while read x; do if [ "$x" == "$y" ]; then match=true; fi; y=x; done; if [ "$match" == "true" ]; then echo Match: $filename; fi); done)
[20:01] <toad_> why does this not work?
[20:01] <toad_> i just want to find all files in which there is a duplicated line
[20:02] <greycat> you need a directory after "find" unless you're on GNU.
[20:02] <greycat> find . -iname ....
[20:02] <greycat> let me see....
[20:02] <toad_> yeah this is gnu
[20:02] * cdent (n=cdent@) has joined #freenet
[20:02] <greycat> ugh... it's too hard for me to read it like that :(
[20:03] <greycat> Is it just trying to find the names of all the *.java files that contain the string "blah"?
[20:04] <toad_> no
[20:04] <toad_> all the java files that contain the same line twice
[20:06] * caytchen_ (n=caytchen@) has joined #freenet
[20:07] <greycat> This is closer:
[20:07] <greycat> find . -name '*.java' -exec awk '{ if (s[$0]++) { print FILENAME } }' {} \;
[20:07] <toad_> well what's wrong with the one i pasted?
[20:07] <greycat> Not sure. It's kinda hard to read. :(
[20:08] <Artefact2> you should paste it on pastebin
[20:08] <greycat> The one I just posted will work, but it prints the filename N-1 times.
[20:08] <Artefact2> it would be easier to read
[20:08] <toad_> it prints it a LOT of times
[20:08] <greycat> ah, here's an improved version:
[20:08] <greycat> find . -name '*.java' -exec awk '{ if (s[$0]++) { print FILENAME; exit } }' {} \;
[20:08] <toad_> no, in fact, it's wrong
[20:09] <toad_> i want to find files where two consecutive lines have the same contents
[20:09] <toad_> not files where one line is repeated some time later
[20:09] <greycat> oh, you didn't say they had to be CONSECUTIVE!
[20:09] <toad_> e.g. a newline might be repeated
[20:09] <greycat> in that case, just use uniq(1)
[20:09] <toad_> uniq -d ?
[20:12] <greycat> find . -name '*.java' -exec bash -c 'while read -r; do if [[ $last && $REPLY = $last ]]; then echo "Match: $1"; fi; last=$REPLY; done < "$1"' _ {} \;
[20:12] <greycat> or use "prev" instead of "last" if you like that name better
[20:13] <greycat> (I'm using -name because my HP-UX box doesn't have -iname)
[20:16] <greycat> It's not often I get to put advanced #bash stuff in #freenet ;-)
[20:16] <greycat> That stuff I used is explained on http://wooledge.org/mywiki/UsingFind and in the Bash FAQ
[20:18] <greycat> uniq(1) looked like a decent choice at first glance (since it's normally crippled by the "duplicates must be adjacent", but that's what you wanted...) but I don't see any obvious way to make it print a filename and stop, instead of printing the matching line(s)
[20:18] <toad_> how do i read a line exactly as-is?
[20:18] <greycat> in bash, 'read -r' without a variable name. If you want to use a variable other than REPLY, you need something more complex, like IFS=$'\n' read -r varname which is a pain.
[20:19] <toad_> read -r still drops preceding whitespace
[20:19] <greycat> I really dislike the fact that omitting the variable name has different semantics than supplying one, but I didn't write the shell.
[20:19] <greycat> oh, does it? hmm....
[20:20] <greycat> It works for me.
[20:20] <greycat> while read -r; do echo "$REPLY"; done < foo
[20:21] <toad_> ah yeah
[20:21] <greycat> and now I can't remember exactly what the difference is between using a varname and the implicit REPLY
[20:24] <greycat> oh, there, I had IFS set for some reason. "read -r x" did indeed strip the leading space once I unset IFS. silly me.
[20:25] * caytchen (n=caytchen@) Quit (Read error: 110 (Connection timed out))
[20:35] <toad_> find -iname "*.java" | (while read filename; do echo "Processing $filename" > /dev/tty; cat "$filename" | (match=false; y="blah"; while read -r; do x="$REPLY"; if [ "$x" == "$y" ] && echo "$x" | grep "container.store" > /dev/null; then match="true"; else echo "$x"; fi; y=$x; done; if [ "$match" == "true" ]; then echo Match: $filename > /dev/tty; fi) > $filename.tmp; mv $filename.tmp $filename; done)
[20:35] <toad_> :)
[20:36] * cdent_ (n=cdent@) has joined #freenet
[20:38] <toad_> argh
[20:39] <toad_> if read is fed a line without a newline and with an eof on the end, it swallows the data, and doesn't return anything
[20:48] <nextgens> rehi
[20:50] <toad_> hi
[20:51] <Artefact2> hi
[20:51] <toad_> bbiab
[20:52] * cdent (n=cdent@) Quit (Read error: 113 (No route to host))
[20:55] * greycat (i=rfc1413@) Quit ("This time the bullet cold rocked ya / A yellow ribbon instead of a swastika")
[21:23] * dennisn (n=dennisn@) has joined #freenet
[21:25] <dennisn> would it be possible for downloads.freenetproject.org to host the fms tarball?
[21:25] <dennisn> err.. fms zipfile
[21:34] * cdent_ is now known as cdent
[22:15] * Zarggg (n=z@) has joined #freenet
[22:22] * kalkin-- (n=kalkin-@) Quit (Read error: 113 (No route to host))
[22:22] * cdent (n=cdent@) Quit ()
[22:26] <kork> migrating key 1331200/3983513
[22:26] <kork> one third done
[22:29] <toad_> dennisn: possibly
[22:29] <toad_> dennisn: but only if we reviewed it first
[22:29] <toad_> which would be significant work on its own
[22:30] * saces (n=saces@) Quit (Read error: 110 (Connection timed out))
[22:31] * saces (n=saces@) has joined #freenet
[22:31] * ChanServ sets mode +o saces
[22:31] * toad_ (n=toad@) Quit (Remote closed the connection)
[22:53] * Johan^mlg (n=bllarf@) Quit ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org")
[23:04] * Zarggg (n=z@) Quit ()
[23:15] * phrosty (n=phrosty@) has joined #freenet
[23:32] * dennisn (n=dennisn@) Quit ("Leaving")
[23:53] * Ash-Fox (n=N@) Quit (Read error: 104 (Connection reset by peer))
Irc logs of #freenet : 2008 2007 2006 2005
These logs were automatically created by FreenetLogBot on chat.freenode.net using the Java IRC LogBot.