#foswiki 2016-02-23,Tue

↑back Search ←Prev date Next date→ Show only urls(Click on time to select a line by its url)

WhoWhatWhen
***Vampire0_ has quit IRC (Ping timeout: 244 seconds) [02:08]
..................... (idle for 1h44mn)
ChanServ sets mode: +o Lynnwood_ [03:52]
.................... (idle for 1h39mn)
GuilainC_away is now known as GuilainC [05:31]
gac410 has left [05:40]
................... (idle for 1h32mn)
WikiRingBot has quit IRC (Remote host closed the connection) [07:12]
ChanServ sets mode: +o MichaelDaum [07:19]
GuilainC is now known as GuilainC_away [07:27]
ChanServ sets mode: +o CDot [07:32]
.................. (idle for 1h28mn)
eddie3182 has left [09:00]
................. (idle for 1h24mn)
foswiki_irc5Hi, Totally novice to foswiki. Need to install foswiki on windows. Have downloaded xampp. however cannot load any foswiki default site. get an error for missing dependencies
CGI::Session.pm, Crypt::PassswdMD5 and error.pm
[10:24]
................ (idle for 1h15mn)
EisNerdmoin
is subnumbering available in Foswiki w/o plugin? (e.g. 1 ...\ 1.1 ...)? or having (1 ...\ a ...)
[11:39]
***eddie3182 has left [11:54]
ColasHomeEisNerd: you can do http://foswiki.org/Sandbox/NumberingTest but not 1a 1b 1c if that is your question
otherise you have http://foswiki.org/Extensions/ExplicitNumberingPlugin
[11:55]
............ (idle for 59mn)
EisNerdthx
no idea why this hasn't worked when I tried it the first time
[12:55]
......... (idle for 42mn)
***ChanServ sets mode: +o gac410 [13:38]
gac410MichaelDaum: Are you sure that my change breaks json and rest?
Unit test suite passes 100% and configure, which uses jsonrpc heavily is fully functional
It was actually an extremely minor change. *all* it did was not extract the web/topic from the path_info to use as the default. The rest of the parsing is still there.
I also tested from cli to make sure that comments could post, etc.
By not extracting an "initial topic" from the path_info, the rest of the parsing processes, but uses the home web / home topic as the default instead of the CommentPlugin/comment or the like recovered from path_info.
[13:40]
MichaelDaumgac410, it breaks jsonrpc as I tried to explain. [13:45]
gac410why is configure still fully functional? [13:46]
MichaelDaumbecause it does not deal with a web.topic location [13:46]
gac410gac410 goes back to read your comments. [13:46]
MichaelDaumconfigure's jsonrpc isn't setting any topic url param or topic property to the jsonrpc's param object
it simply does not care about it
however some jsonrpc's do
like jsonrpc/AngularPlugin/tmpl ... which renders an html fragment for a specific web.topic ... and this is of course different depending on the web.topic ... specified in the jsonrpc param object
[13:47]
gac410So the fix is wrong, but the old implementation is bogus. jsonrpc was previosly passed "AngularPlugin/tmpl as the web/topic. which isn't right either. [13:49]
MichaelDaumthats right [13:49]
gac410A web of AngularPlugin is right??? [13:49]
MichaelDaumthe _way_ the web.topic property is derived from the request is different
^your finding is right^
[13:50]
gac410okay, but I thin you are saying that there cannot be *any* default for jsonrpc (which was satisfiled by passing a totally bogus web/topic by accident) [13:50]
MichaelDaumevery jsonrpc ... that needs a web.topic properly set needs to put it into the param object of the jsonrpc request
... which is totally jsonrpc specific
as I tried to illustrate at http://foswiki.org/Tasks/Item13969
[13:51]
gac410okay, but why does correcting a completely incorrect default webname of "TheJSONNamespace" then break jsonrpc's ability to set the topic. [13:52]
MichaelDaumno jsonrpc will ever be able to put a topic url parameter in a position the core would successfully pick it up. so far so bad.
like { json-rpc: "2.0", topic:"SomeWeb.SomeTopic", params: { payload } } is an INVALID json-rpc request
[13:53]
gac410right. that's fine. so core falls back to Home/WebHome instead of Angular:plugin/unknown jsonrpc shoudl still override. I did not change the url param [13:54]
MichaelDaumso with your change it is executed at Main.WebHome [13:54]
gac410vs excuted as Bogus.bogus? [13:54]
MichaelDaumit will happily pick up any preferences in action at Main.WebHome
and that breaks things
[13:55]
gac410gac410 things jsonrpc is broken as core starting out with an initial valid topic seems compeltely correct.
For jsonrpc, we could force it to be strings "undefined.undefined" ???
if "rest" then webtopic = '' elsif 'jsonrpc" webtopic="unspecified"
[13:55]
MichaelDaumthe real fix is a better way to parse requests: (1) normal as per path_info (2) viewfile / xsendfile stype ... stripping off PubUrl (3) from a topic url param (4) from a jsonrpc param object
(5) ignore any
[13:58]
gac410If the "AngularPlugin" web happened to exist as a valid web, then jsonrpc would run with a valid web/topic which would confuse it as well.
using the old way
[13:59]
MichaelDaumsure thing. broken as well.
point is jsonrpc/AngularPlugin/tmpl _does_ specify a topic param in its param object....alas it is never read by the core at its early stage.
At a late point in the tmpl service there is a pushContext as a work around this deficiencies in the Foswiki core
[13:59]
gac410But it's down in the POST data, not in the request, which needs much deeper processing. [14:01]
MichaelDaum^thats what I try to say all the time
even viewfile does its own processing
we should rething this all
and make it more flexible
so that jsonrpc could kick in doing its own thing
[14:01]
gac410well I can easily revert it back to it's broken state, or for now set a "undefined/undefined" for jsonrpc to block the default. [14:02]
MichaelDaumundef undef is fine [14:02]
gac410Without any unit tests, it's somewhat impossible for me to fix. [14:02]
MichaelDaumi.c. [14:02]
gac410I use configure as my last ditch test case.
tests, pass configure runs. SHIP IT :D
At least I didn't say Compiles ... Ship it
[14:03]
MichaelDaumnote there currenly is no abstraction in Foswiki::Request that would jsonrpc allow to override a getWebTopic thingy [14:03]
FoswikiBothttp://trunk.foswiki.org/System/PerlDoc?module=Foswiki::Request [14:04]
gac410I don't think we even have any sanity checks in the TestCases web that would have helped find the issue. [14:04]
MichaelDaumthere are no test cases ... and there is no code to test either, that implements a better parsing of the request properties in different cases [14:05]
gac410Anyway, I have to go out today very shortly. revert if you want, or change it to undef/undef for jsonrpc. Or I'll get to it tonight.
well, jsonrpc was added as a default extension, it *should* have come with unit tests. I suppose I could have blocked the release :P
[14:05]
MichaelDaumI hope you understand what I was going to explain
settling on a jsonrpc param containing a topic param ... which is then parsed ... this is not there yet.
[14:06]
gac410I understand that jsonrpc and xsendfile use a 3rd / 4th? method of setting the BASEWEB/BASETOPIC which core does not handle if that's what you are explaining. [14:08]
MichaelDaumno surprise that there are no unit tests for that either
xsendfile and viewfile do exactly the same parsing. they are in the same class.
not so jsonrpc
infact xsendfile re-implements the way viewfile parses path_info ... this is redundant
instead of reimplementing it or working around the standard hard-coded way of parsing path_info ... we should improve the core to do it right ootb
by delegating it to a path_info parser specific to a rest, jsonrpc, whatsoever script
[14:08]
gac410gac410 has a plane to catch in one hour. and I'm the pilot. :) So I can't fix this right now. [14:11]
MichaelDaumnice! [14:12]
........ (idle for 37mn)
***ChanServ sets mode: +o CDot [14:49]
foswiki_irc7hi all i'm trying to migrate a foswiki installation made from the default virtual machine file to a new machine i've created do you have any idea what to copy in order to make it work? thanks [14:50]
jastdid you customize the configuration or set of installed plugins or stuff like that?
(by configuration I mean the stuff in the /bin/configure web interface or files in lib/ on the server)
[14:59]
foswiki_irc7i've done it in the first virtual machine but there the installation had only 3.7 gb of space and i couldn't find a way to resize it cause it's under lvm [15:01]
jastlvm is actually comparably easy to resize
it's the other stuff, the partitions and filesystems, that tends to be painful
where did you download the VM from, by the way? we don't have an official VM image for the newest release yet
[15:01]
foswiki_irc7the problem is that the root partition is always active and i've tried to use a live ubuntu to edit it but gparted doesn't allow me to expand the disk [15:03]
jastwell, it's safe to say this stuff is a little tricky [15:03]
foswiki_irc7i think it was the 2.0.3 [15:03]
jastI don't know the exact location the files live in the VM, probably /var/www/foswiki
the actual content is in the data and pub subdirs
[15:04]
foswiki_irc7ok and what about the apache configuration files and modules? i've tried to copy everithing but i couldn't make it work [15:06]
jastoh, you need to do all of the setup, right
there are a bunch of module dependencies and stuff like that
the VM has a number of packages pre-installed that foswiki depends on
[15:16]
foswiki_irc7so i follow the instruction for a fresh install exept that in the folder i put the one i have in the other installation right? [15:17]
jastone option is to do a fresh install and then copy in the data/pub directories
that will give you all of your existing content in what is otherwise a completely new wiki
but what you said is an option, too
[15:18]
foswiki_irc7ok i will try thanks a lot for your help [15:23]
jastgood luck :) [15:23]
......... (idle for 40mn)
***BenjaminMartin has left [16:03]
......... (idle for 41mn)
LavrHi guys. I finally upgraded my private installation to Foswiki 2.1 this weekend.
And it is running better than ever. Much faster. Especially the ability to turn off sessions for guests have meant a massive reduction on the server load.
in 1.1.9 the server was dragging 4000-8000 daily sessions - sometimes 20000 sessions from all sorts of bots and casual viewers. Now I have 2-3 sessions from the few that post things
I have not seen any negative impact on no guest sessions. here in Moto there is no guess access. All viewing is authenticated so all sessions are real.
[16:44]
gac410Cool You're the first I've heard of using that feature
We probably ought to think about turning it off on foswiki.org as well. I don't think we allow any guest activity that would require a session.
[16:48]
LavrGive it a try. [16:52]
gac410Right now foswiki.org has >7000 cgisess files in temp. So yeah, lots of guest sessions I'd expect [17:00]
MichaelDaum_: Unfortunately the parser needs to look at the file system to really know where to stop when pulling apart something like: web / topic / subdir / subdir / file.txt
and that's certainly not very forward looking toward db based stores. :(
Or, you end up calling the webExists routine for each part, trying to work out where to stop.
[17:12]
Anyway, viewfile certainly works fine, but core completely screws up the "baseweb" and "basetopic" assignments. [17:22]
..... (idle for 20mn)
LavrJust started my webnotify here at Moto. Ooof. That was a bug spam load because I had not copied over any history about previous notifications.
Well. This one time only.
They will forgive me one day ;-)
[17:42]
gac410;) yeah copying working/work_areas is critical for mailnotify [17:42]
LavrI always forget. I also had to fix some ActionTracker unique IDs because I forgot that little datafile in working.
In a normal upgrade where you copy on top of the existing it is not a problem
I found out that the problem with configure and using .. links everywhere is when you create a /var/www/Foswiki-2.1.0 directory and then create a synlink /var/www/foswiki that points to it.
When Apache config is setup for the symlink (that is the idea) hen Configure guesses all directories with paths with .. in them
[17:43]
gac410Well MichaelDaum_ ... I'm baffled by JsonRpcContrib. I can't see where it does anything different from core. If topic urlparam is not present, it sets it to the default homeweb [17:46]
LavrYesterday I renamed the Foswiki-2.1.0 dir to foswiki instead and then the guessing went better. it still double dotted in 1 place [17:46]
gac410Line 95 in JsonRpcContrib/Server.pm sets the topic overriding the namespace
did a grep of the whole JsonRpcContrib, and that's about the only place topic is mentioned of any significance
[17:46]
MichaelDaumgac410, I think you still don't get the situation for jsonrpc [17:48]
gac410not surprising :D [17:48]
MichaelDaumALL, params to a json-rpc call are encoded into a json object of a specific format with all payload inside the "params" section as outlined in http://foswiki.org/Tasks/Item13969
NO params are encoded neither in the path nor in any url params
[17:48]
gac410I see that. I see where it puts web/topic. I can't find where it reads that and sent it into the core. [17:49]
LavrBy the way. A long time ago I committed a feature of a configurable WikiWord and numbers so you could choose if numbers are upper or lower case. [17:49]
MichaelDaumfor NOW, what goes in inside the params json object is TOTALLY up to the jsonrpc endpoint
what we NEED is: agree on "topic" being required inside the json-rpc param object
so that the core can properly initialize
[17:49]
LavrWhat prevented me from making it was that we have several plugins that use name rule in Javascript. I need help with how we signal from perl to the JS which regex to use.
The perl part is dead easy.
[17:50]
gac410I had to set webtoipic as _Unknown/Unknown btw, to prevent core from setting it.
lavr, you have feature requests pending to implement that, I've been asking at release meetings if you plan to go ahead :)
If I make it unknown, it's stripped because it doesn't pass web validation rules, and then core asserts the default web.
[17:50]
LavrI would do if someone can help with the JS part. When I looked a year ago I got stuck with how to send the signal to the JS code running in the browser that the regex for WikiWord was this or that [17:52]
gac410Okay MichaelDaum I thnk what you are saying then is JsonRpc does *nothing* with the web/topic and it's currently up to each handler to figure out what it wants?
Sending the signals is fairly straight forward. Need to add it to the meta generated in the jquery "foswiki" plugin ... I think
Though the wikiword code has been changed iirc to the a jquery version.
[17:52]
***dldl-work has quit IRC (Quit: leaving) [17:57]
LavrI will also need to find out how many places now we have JS regex that check for WikiWord besides the wysiwyg editor [17:58]
gac410I think that the WebTopicCreator uses a common wikiword routine. I've lost track
Same thing used on registration page too
[17:59]
LavrIf we set number = uppercase should 4everYoung be a wikiword?
I sort of hate that idea
[18:00]
gac410I would think it passes if numbers are upper case [18:00]
LavrIt will unless I implement different [18:01]
gac410I think starting a topic name with anything other than an upper alpha will be a significant issue. Lots of places check for upperAlpha for web and topic names. [18:01]
LavrToday the regex is fixed [[:upper:]]+[[:lower:][:digit:]]+[[:upper:]]+[[:alnum:]]*
The configure setting would alter this to [[:upper:]]+[[:lower:]]+[[:upper:][:digit:]]+[[:alnum:]]* IMHO
ie. first char only alpha
And third option was - numbers are not legal in wikiwords.
That is simple.
Default will remain what we have and it will be an expert setting.
[18:02]
gac410Another option might be that numbers are both upper and lower - could serve as either. But that's probably even more difficult. [18:04]
LavrThe both option was discussed in the feature proposal and we were many that hated that idea once we thought it through. [18:05]
gac410okay. I don't intend to reopen any prior decisions. [18:06]
LavrJust explaining the rule in the docu made by head explode and the unwanted wikiword issue would explode. People may think they want that but they will hate it.
yeah and I am sure - numbers as uppercase is OK but not as first char. So doing that with a configure option is easy. That is perl.
I do not want to put an geek expert regex in the configure. We have too many of those. I will make it a value with 3 human values and then let the regex be defined in the Foswiki.pm perl code. Simple. Very simple.
The trick is getting that info to the JS.
[18:06]
gac410The magic happens in lib/Foswiki/Plugins/JQueryPlugin/FOSWIKI.pm [18:09]
LavrAh the JSON prefs thing [18:16]
So the idea would be to assign the preference to a real "MACRO" type preferences setting and send it along to the JS with all the rest
My problems start when I am in the JS receiving side. I have ZIP experience with JS other than 1 liners in-line
And that is where I need a JS expert like Michael it guide me.
to guide me
[18:21]
gac410The regexes are built here pub/System/JQueryPlugin/plugins/foswiki/jquery.foswiki.uncompressed.js [18:28]
LavrAlso in System/JavascriptFiles/foswiki.js [18:31]
gac410y, Thats what I was looking for ... thx
Hm JQueryPlugin doesn't use the WIKIWORD_REGEX built by foswiki.js.
[18:32]
LavrIs it my editor that is setup badly or are all many of the js files tabbed instead of spaced for indentation? [18:33]
***GuilainC_away is now known as GuilainC [18:33]
gac410They appear to be tabbed.
Not sure why. Michael would know I expect, but he left
[18:34]
LavrThat is why they impossible to read. Tabs in my editor is 4 spaces so there is no indentation because it is a mess of 4 spaces and tabs.
That is just bad coding style to mix tabs and spaces
Probably a mistake
[18:35]
gac410Seem to be readable in vim [18:35]
Lavrbecause your vim defaults a tab to 8 spaces [18:35]
gac410ah probably [18:36]
Lavrfoswiki_tiny also has some wikword stuff but I cannot see that it uses the real regex anyway
in TinyMCEPlugin javascript files
[18:40]
gac410well darn.... Just reviewed MichaelDaums translation work for JQuery strings. it *appears* to be totally outside of our weblate translation engine. That royally sucks ... or I don't understand what he did. [18:41]
LavrI need to learn the JS syntax to even read the code. var match = /^((?:\w+\.)*)(\w+)$/.exec(url); Eh??? [18:43]
Late. Need to go home for dinner [18:48]
gac410cu [18:48]
Item4992 needs review by our Translators. We have a well defined translation process. This needs to integrate. [18:59]
FoswikiBothttp://foswiki.org/Tasks/Item4992 [ Item4992: Work out how to translate Javascript strings ] [18:59]
***gac410 sets mode: +v WikiRingBot
gac410 sets mode: +v FoswikiBot
[18:59]
....... (idle for 34mn)
JulianLevensgac410: I've been looking at Foswiki::Time and commented out the line which has 'require Foswiki' and I'd had no impact so far [19:33]
FoswikiBothttp://trunk.foswiki.org/System/PerlDoc?module=Foswiki::Time [19:33]
JulianLevensThe only reason it needs Foswiki AFAICT is to access three cfg values. So we have a module that could so easily be completely stand-alone with clear separation of concerns
But alas isn't ...
indeed it brings in the whole of Foswiki :(
[19:35]
................. (idle for 1h22mn)
gac410Y. I think there is a way to load it without bringing in all of foswiki. Configure needed to do that ... back before all of CDot's work last year
if you load it with -nofoswiki, then it skips the big load.
The logger, and config checkers use use Foswiki::Time qw(-nofoswiki);
[20:57]
FoswikiBothttp://trunk.foswiki.org/System/PerlDoc?module=Foswiki::Time [20:58]
JulianLevensy, I spotted the -nofoswiki, it's still a shame it's required
In the last hour or so I've been reviewing Foswiki::Address only to conclude that it's unfinished and not really used anywhere
[21:06]
FoswikiBothttp://trunk.foswiki.org/System/PerlDoc?module=Foswiki::Address [21:07]
gac410yeah. Hm how does the test for UseLocale actually work if Foswiki is not always loaded. [21:07]
JulianLevensGood question, I wrote a small script to test it and that part did not barf [21:09]
gac410a couple of modules use Foswiki::Address - but I have no idea how extensively. Foswiki::Meta and Foswiki::Access::TopicACLAccess [21:09]
JulianLevensIn Foswiki::Meta it's just a comment [21:09]
FoswikiBothttp://trunk.foswiki.org/System/PerlDoc?module=Foswiki::Meta [21:09]
gac410figures. [21:10]
JulianLevensFoswiki::Access::TopicACLAccess in theory supports testing an address object for access [21:10]
FoswikiBothttp://trunk.foswiki.org/System/PerlDoc?module=Foswiki::Access::TopicACLAccess [21:10]
gac410This is something that was merged into trunk by PaulHarvey iirc. [21:10]
JulianLevensbut as nobody will pass a Foswiki::Address object that's redundant [21:10]
gac410One of the things that we could have reverted out when working on releasing 1.2/2.0 It was hard to figure what pieces in trunk were incompletee. [21:11]
JulianLevensFoswiki::Macros::META does use it but it's trivial [21:11]
FoswikiBothttp://trunk.foswiki.org/System/PerlDoc?module=Foswiki::Macros::META [21:11]
JulianLevensFortunately it is benign, but ideally it should be removed/reverted [21:12]
gac410so the only use of $Foswiki::cfg in Foswiki::Time is the static method formatTime. [21:12]
***GuilainC is now known as GuilainC_away [21:13]
gac410Well the other option is to try to finish off pharvey's work. Foswiki::Address does have a thorough set of unit tests.
gac410 likes code with unit tests
[21:13]
JulianLevensy, formatTime is the only place [21:14]
gac410Maybe Foswiki::Time::formatTime could be pulled back into Foswiki,
"# previous known as Foswiki::formatTime" probably should have been left there, or made part of Render
[21:14]
JulianLevensI was thinking of adding another param for the default time format, but that's not entirely elegant
$outputTimeZone is already a param
[21:17]
gac410It was probably a bit of excessive refactoring way-back-when [21:17]
JulianLevensIt this were Moo-ified then you'd create a new formatting object
The core could create such a beast with the defaults
[21:18]
gac410I know on older versions of Configure it was a really difficult situation. It would cause Foswiki to load, which loaded LocalSite.cfg on top of the in-process configuration. TimotheLitt fought with it a lot. [21:19]
JulianLevensAnother option would be to call an init function passed these two defaults during startup and store them locally [21:21]
gac410y that would work too. [21:21]
JulianLevensI'm very much aware we have spaghetti dependencies [21:22]
gac410I don't think the new configure has to care. It declares anything in process "local" so a separate space for the new config is maintained.
We probably could have revereted out the -nofoswiki as well once CDot refactored configure to json
[21:22]
JulianLevensIf Foswiki::Time completely removes it's use of Foswiki that becomes a moot point
And about locale, what does that do?
[21:23]
gac410Actually MichaelDaum and I were wrestling with the parsing of the "path" - which Foswiki::_parsePath() gets completely wrong for viewfile, xsendfile, rest and jsonrpc
Maybe all the work in Foswiki::Address could be applied there as well.
Not sure why the conditional useLocale was added to every module's begin block. jast might know.
[21:26]
***ChanServ sets mode: +o gac410 [21:28]
gac410damn. On my new system, pidgin occasionally loops [21:28]
JulianLevensReading around proposals related to Foswiki::Address it's clear that some concerns had not been address
addressed
[21:28]
gac410I'm not sure about useLocale in every BEGIN block. Maybe jast knows
jast made the change as part of Item11953:
[21:29]
FoswikiBothttp://foswiki.org/Tasks/Item11953 [ Item11953: {Site}{Locale} and tainted mode can't be used together. Ship with taint checking disabled. ] [21:30]
JulianLevensy, I'll ask jast tomorrow [21:30]
gac410I assume it was needed to get Foswiki to use locales uniformly if enabled in the config. but I've never really worked with them. [21:31]
JulianLevensy, in general terms I've not sure it's even useful anymore. Unicode covers a lot of i18n presumably there's some l10n to do as well
But locale is global, proper localisation would allow different users to localise differently
[21:35]
Anyway I off to dream about all this :P [21:44]
gac410jomo and I talked a bit about that when I changed all our sort compares to use NFKD .. That's certainly improved,
but true locales will still do a better job.
g'night
[21:44]
JulianLevensg'night [21:45]
***JulianLevens has left [21:45]
........... (idle for 52mn)
ChanServ sets mode: +o gac410 [22:37]
vrurggac410 or somebody familiar with tests – PluginHandlerTests, line 191. $this is used which is not available in this context. It should've been $tester, I guess. But the replacement breaks the tes.
Any idea on how it should be?
[22:44]
gac410hm Hi vrurg yeah that's strange. Maybe CDot would know.
Sorry... can't hang around - gotta run an errand for a couple of hours. back later.
[22:47]
vrurgOk, then it means waiting for tomorrow... No good. [22:48]
gac410The tests all pass on master, so not sure how $this gets resolved. [22:48]
vrurgI suspect a bug with prefs.
paths key must be refreshed when topic text is changed.
gac410: It's ignored.
Though not a bug on it's own but it masks a real bug with preferences reading.
[22:48]
***harlan has quit IRC (*.net *.split)
alexlist has quit IRC (*.net *.split)
jayen_ has quit IRC (*.net *.split)
uebera|| has quit IRC (*.net *.split)
Lavr has quit IRC (*.net *.split)
jast has quit IRC (*.net *.split)
rathier_afk has quit IRC (*.net *.split)
Colas has quit IRC (*.net *.split)
ColasHome has quit IRC (*.net *.split)
gmc has quit IRC (*.net *.split)
mephinet has quit IRC (*.net *.split)
rouilj has quit IRC (*.net *.split)
GuilainC_away has quit IRC (*.net *.split)
KryoStoffer has quit IRC (*.net *.split)
J-BBB has quit IRC (*.net *.split)
ChanServ has quit IRC (*.net *.split)
verne.freenode.net sets mode: +o SvenDowideit_
[22:58]
verne.freenode.net sets mode: +oooo ChanServ Lynnwood_ SvenDowideit_ Lavr
verne.freenode.net sets mode: +ov gac410 WikiRingBot
verne.freenode.net sets mode: +oovo ChanServ Lynnwood_ WikiRingBot gac410
verne.freenode.net sets mode: +oo Lavr SvenDowideit_
[23:11]

↑back Search ←Prev date Next date→ Show only urls(Click on time to select a line by its url)