IRC log of #schooltool for Monday, 2006-11-27

*** th1a has quit IRC01:10
*** th1a has joined #schooltool01:11
*** th1a has quit IRC01:20
*** th1a has joined #schooltool01:20
*** ignas has joined #schooltool01:23
*** jstraw has joined #schooltool03:02
*** wrobel has quit IRC03:15
*** ignas has quit IRC05:12
*** alga has quit IRC05:19
*** ignas has joined #schooltool05:20
*** srichter has joined #schooltool07:40
*** vidasp has joined #schooltool08:44
*** vidasp has quit IRC08:44
*** vidasp has joined #schooltool10:01
*** SteveA has joined #schooltool10:19
*** thisfred has joined #schooltool10:27
SteveAth1a: hi10:31
*** Aiste has quit IRC11:29
*** jinty has joined #schooltool11:40
*** vidasp has quit IRC12:00
*** lisppaste5 has quit IRC12:16
*** Aiste has joined #schooltool12:19
*** lisppaste5 has joined #schooltool12:21
*** Aiste has quit IRC12:44
*** thisfred has left #schooltool12:46
*** mgedmin has joined #schooltool12:47
*** Aiste has joined #schooltool13:43
*** alga has joined #SchoolTool14:07
*** alga_ has joined #SchoolTool14:08
*** eukreign has joined #schooltool15:41
*** alga_ has quit IRC15:51
th1aSteveA: Hi.16:16
SteveAhi tom16:18
SteveAth1a: I was about to take a short nap -- still kind of timezone lagged from my trip to the US.  chat later?16:20
th1aSteveA: Sure.16:20
ignasth1a: hi16:25
th1aignas: Hi.16:25
th1aHm... where's jfroche?16:26
ignasgot Lyceum to sign the contract16:26
ignas(don't have the papers yet though, as i am pretty ill today)16:26
ignasi'll get them tomorow, and leave them to Aiste to either send or fax to you16:27
th1aignas: Great!16:27
th1aI don't have a fax machine.16:27
ignasthen it's snail mail i guess16:28
*** thisfred has joined #schooltool16:30
th1aCan you scan it?16:30
ignashmm, it would be possible16:30
th1aI could get the fax number at school.16:31
ignasscanning will be good enough i guess16:32
th1ahrm... can't do much today without jfroche.16:32
th1aignas: Are you ready to go to Belgium?16:34
ignasready as in?16:34
th1aBags packed?  ;-)16:35
ignasno, not yet. I'll have enough time for it tomorrow.16:36
ignasat the moment i am staying in bet to get as much not-sick as possible ;)16:36
ignasas for the progress - i got PersistentTimetableEvents editable and changing the scedule of a section adds/removes them properly16:40
ignaswhat's missing is the integration of the system with the current "date"16:41
th1aWhat do you mean?16:42
ignasif you change the schedule, only events past today are modified16:42
ignasnot sure we need it now16:42
ignasi guess the first deployment will not have this working ...16:43
th1aThat seems necessary to me.16:43
th1aBut I imagine we could live without it.16:44
th1aWhen can we merge what you've got now?16:44
ignasdifficult to say, depends on how much do we care about it being nice and clean16:46
ignasremoving timetables from ovelrays will break a lot of functional tests16:46
ignasdeleting functionality is easier than adding it though, even with tests16:47
ignasas for lyceum, i'll deploy it earlier, they do not need editable terms16:48
ignasso i can skim on some usecases16:48
th1aSo are you worrying less about tests on the lyceum branch?16:50
ignasi can just disable parts of the system16:50
ignaslike - remove edit-term view, remove tests for term editing16:51
*** jstraw has left #schooltool16:51
ignasi don't care about regressing functionality on lyceum branch16:52
ignasas for the Person work16:54
ignasi want to get Persons easier to customize16:54
ignasat the moment you can change the demographics stuff16:54
ignasbut it is too difficult to work with all the places person list is being displayed ...16:54
ignaswhich limits customisations to the Person container, Person internals16:55
th1aWill this replace some of the demographics code?16:57
ignasdon't know yet16:58
ignasi think i'll be touching the Sorting/paging/displaying in a table parts a lot more ...17:00
th1aOK.  That could definitely use some work.17:01
ignasat the moment you can display only titles of an object in most lists :/17:03
*** jstraw has joined #schooltool17:21
ignasth1a: as for the conference - i am afraid that it might not be toll free from lithuania ...17:22
ignasth1a: maybe we could use some VoIP service ?17:22
th1aOh, if you have to place the call it won't be free.17:23
th1aI can ask Dave if he can call you.17:23
ignasi see17:24
ignasi guess we're done with the meeting17:42
th1a'fraid so...17:42
ignasi'll get on irc 2pm EST17:44
th1aOK.  See you then.17:44
jintyth1a: did you see the mail about the server migration and Time to live dns stuff?17:46
th1ajinty:  Yes.  I'll put that in motion.17:46
*** Aiste has quit IRC17:47
* jinty actually reads his mail....17:47
*** jfroche has joined #schooltool17:52
jfrocheignas: th1a: sorry for being late17:52
th1aAh, yes, kind of hard to have a meeting with my two main developers when one is missing.17:53
*** aelkner has joined #schooltool17:54
jfrochesorry about that got a late train17:54
th1aI see.17:55
jfrocheignas is ill ?17:57
*** wdickers has joined #schooltool17:58
jfrocheready to fetch him tom but would need a photo or something cause i ll miss him maybe at the airport :)17:58
wdickersare you there aelkner?17:59
aelknerwdickers: hey there.17:59
jstrawhi all17:59
*** jstraw is now known as Lumiere17:59
aelknercheck out FILE_STRUCTURE on maddog.17:59
aelknerI'm working on not only documenting the code, but building stategy for reorg.18:00
aelknerwhat does kk stand for?18:00
wdickerssorry, used to speaking on MSN. 'kk' means 'okay'18:01
wdickersso it looks like you took a closer look at the config file and how it is used. What is stored there?18:04
aelkneractually, i didn't get into that level of detail yet.18:05
aelkneryou can look at the config file now and see how it matches the code.18:05
th1aThe file should be very straightforward.18:06
aelknerThat's right.18:06
aelknerTom makes use of many python concepts in his code, so you can learn a lot just by reading it.18:06
aelknerAmongst the things used are:18:07
aelkner1) config file parsing18:07
aelkner2) logging: google 'python' to learn more about that18:07
aelkner3) of course, client, server, and zodb, as we already discussed18:08
wdickerswhat exactly is the config file though? The actual config I mean18:11
wdickersElkner says high by the way18:11
aelknerHey Elk.18:12
aelknertinyzis.conf is what you want to look at.18:12
wdickersAh, I see it now18:12
aelknerWe need to study this to see what an agent's config would have in common.18:13
wdickerswell port, mode, hostname for starters18:14
aelknerthere's a problem with hostname, for instance.18:14
wdickershow so?18:14
aelkneron the server, it would be localhost.18:14
aelkneron the agent, a remote host18:14
wdickersright, so there would be a conf for the agents and one for the server18:14
wdickersI mean, they'll all be on different machines, right?18:15
aelknerone of the key problems with the code right now is its use of the conf object throughout.18:15
aelknerIf the config is going to be different between agent and server, then we need a common set of values that we could pass around.18:16
aelknerI'm not sure how this would be done as yet.18:16
aelknerI'm still researching this18:16
aelknerTOm and I will be meeting at 10:00 wed to work on this.18:17
aelknerYou could join us when you get in that day.18:17
wdickersWhat values would have to be 'passed around'?18:17
wdickersI definitely will18:17
aelknerhave you noticed how conf is passed aroung everywhere?18:17
aelknerthla: do you follow?18:20
wdickersYeah. But you say it like the server will have to pass a conf variable to the agent. If they are two conf files, then they can be set separately18:20
aelknerNo, I'm saying that the code as it is uses the conf object which is built by parsing tinyzis.conf.18:21
aelknerAny code that we want to have in common for the agents would not work with that assumption in place.18:21
aelknerFor starters, message building and parsing.18:22
wdickersOh, I see now18:22
jfrocheth1a: ignas is ready ?18:23 is what we need to get working first for both server and agents.18:24
aelknerOne thing to consider is that the conf object for the server, although different than the agent, could have common member variables.18:25
wdickersRight now it seems pretty universal. The conf just needs to have SIFns18:26
aelknerif message,py were to use only those vars that are common, then it wouldn't matter that one conf is not identical to the other18:26
aelkneralso supported versions18:26
wdickersSo both confs have to have those two. How is that a problem?18:27
aelknerSo far, that's all.  But as I said, I haven't finished researching the places that there might need to be code sharing18:28
Lumierewouldn't 3 conf files solve this... tinyzif.conf tinyzif-server.conf tinyzif-agent.conf18:28
th1aOne thing is that we could just pass specific parameters instead of the whole conf.18:29
Lumierealso, wouldn't it make sense for the server to pass it's specific parameters to the agent on connection?18:29
wdickers@Lumiere: how would tinyzif be different from tinyzif-server?18:29
wdickersAnd what parameters would need to be passed? It seems like each are specific to what agent/sever they're attached to18:30
Lumierethe server would be server specific, the tinyzif would be shared common info18:30
aelknerlumiere: i like it.18:30
aelknerdeviding it that way makes it clear.  I love clarity18:31
Lumierebest choice i think would be to pass some of the shared stuff from server to agent18:32
Lumierethen all the agent would really need is connection info18:32
Lumiere(but that may not be in the sif standard)18:32
aelknerlike having a conf message?18:33
aelknerthere whould be a standard for that come to think of it.18:33
aelknerI mean, the agent needs to know what version the server is running?18:33
wdickersWhat other conf variables are necessary for both that aren't predefined?18:34
Lumierewell, the idea is that you shouldn't need to predefine them on BOTH sides18:34
Lumiereonly the server side18:34
aelknerTom might be in the best position to lay out the commonalities as well as the deifferences.18:35
wdickersAh, I see where you're going now18:35
th1aI think you guys are getting off base.18:35
aelknerhow so?18:35
th1aThe way the client and server exchange data is defined in the spec.18:35
th1aNo more and no less.18:35
LumiereI haven't read the spec :)18:36
LumiereI'm looking now18:36
aelknerOne can't a 500 page spec in ten minutes.18:36
Lumierebut I can skim to find messages available18:36
th1aI think aelkner wants to share more code than we'll be able to.18:36
th1aBasically we should be able to share code to parse and generate XML, and that's pretty much it.18:37
aelknerthat's where I wanted to get started.18:37
wdickersI'm afraid my time has run out. I'll see you same time tomorrow aelkner18:37
th1aRight, but it isn't going to go beyond that, I don't think.18:38
aelknersee you then18:38
wdickersBut for what it's worth I vote for having the agent's conf and server's seperate and mostly static. Perhaps when the message sends it's type the conf might be changed, I remember the messages have a 'version' attribute18:38
*** thisfred has quit IRC18:39
aelknerthla: it may be only the message parsing/building that needs to be common, true.18:40
*** wdickers has quit IRC18:40
aelknerthla: do you have a maddog account?18:42
aelknerwell, actually I suppose you don't need one if we use gobby, but it wouldn't hurt.18:43
th1aI don't believe I do.18:43
aelknerI'll get Jeff to set one up for you.  He wants to be tho only admin, the nazi wanker.18:45
aelknerwhat user id do you prefer, thla?18:46
aelknerI'm sorry th1a18:46
aelkner1s and ls look so similar.18:46
th1ahoffman is preferable for id.18:46
aelknerwill do.18:47
aelknerI figure my next step would be to make my own version of so that I don't break any existing tests.18:48
aelknerI think the request should not be part of the constructor, but rather passed to the parser.18:48
aelknerThe message object could created at startup and used repeatable as each message is built/parsed.18:49
aelknerWill and I could also have this object coevered with its own set of tests.18:50
aelknermessageObj.parse() could receive the xml and return a dcitionary of the extracted values.18:50
th1aaelkner: I have to be honest.  I have a very bad feeling about all this.18:52
aelknerhow so?18:52
th1aI think what you should do is write an agent from scratch and then we'll refactor from there.18:52
aelknerThat would work.  I just don't want to right hundreds of lines of code that we later throw away.18:53
th1aWell, as you get into the parts that you really want to copy, we can refactor then as well.18:54
aelknerWill and I could get the message building object working for our needs without much worrying about the server, true enough.18:54
th1aI think you're getting bogged down on pre-emptively refactoring the server.18:55
th1aThis could go on for a very long time.18:55
th1aI cannot stress enough that right now we've got a server that is basic & apparently working, but has NO CLIENTS TO TALK TO.18:56
th1aWe need a client.18:56
aelknerI understand.  We can create an agent without refactoring the server.18:57
aelknerHowever, it never hurts to discuss design.18:57
th1aI know.18:57
th1aBut the hardest thing here is not over-designing and getting bogged down.18:58
aelknerUnderstand that Will and I are new to the project, so there's going to be a learning curve anyway.18:58
th1aI know, but that's why I want you to focus on making a simple agent and learning SIF rather than redesigning TinyZIS.18:59
aelknerWe're learning as we look at the timyzis code.  We don't have to change it to learn from it.18:59
th1aSure.  I'm just worried you'll get bogged down on things like conf files.19:00
aelknerLet's agree that Will and I will leave the server alone and create our message builder that could/should be use later by the server.19:00
th1aI'd rather you take a fresh look at the problem.19:01
aelknerBTW, what's your opinion of the state of the server?  I don't know if the tests work as I see a lot of logging and can't reconcile the output.19:01
th1aThe state of the server is great.  In fact, Doug Daniels has implemented authentication and generally cleaned things up, and we're ready to release 0.2.19:03
aelknerSo the tests work, and my confusion over the output is because it logs to stderr?19:04
th1aYou can turn that off in test.conf, I think.19:06
th1aChange the logging level.19:06
aelknerChange to what?19:07
Lumiereno topic at all19:12
Lumiere(too used to irssi)19:12
aelknernever heard of it.19:14
th1aaelkner: Um... error?19:14
mgedmincan you please put the topic back?19:15
aelknerok, i'll try changing it to error and run the tests.19:15
Lumiereirssi is a unix console irc client19:17
LumiereSchoolTool development | IRC logs at | Dev meetings Mon, 14:30 UTC (16:30 EET)| CanDo dev meetings Tue, 4pm EST | Use for pasting19:18
th1aaelkner: Actually, you want to change [console] I think.19:18
Lumiereif someone ops me for a moment I can set the topic back19:18
aelknerunrelated to logging, python is yielding an error for me: no module named cryptlib_py19:19
th1aAlso, I suppose it may be throwing a bunch of errors because there are new requirements for 2.0, which are kind of tricky.19:19
th1aThat is, unfortunately, a bitch to install.19:19
LumiereSteveA: *ping*19:19
aelknerNo apt-get install available?19:19
SteveALumiere: bon soir19:20
th1aaelkner: No.  It is a problem.19:20
th1aBut not your problem.19:20
Lumierecan you update the topic?19:20
SteveAcan I update the topic?  I have no irc-fu19:20
Lumiereit seems to have been removed, and not reset by chanserv19:20
*** SteveA changes topic to "test"19:21
SteveAso /topic [your topic here] will do it19:21
Lumiere... /topic SchoolTool development | IRC logs at | Dev meetings Mon, 14:30 UTC (16:30 EET) | CanDo dev meetings Tue, 4pm EST | Use for pasting19:21
*** SteveA changes topic to "SchoolTool development | IRC logs at | Dev meetings Mon, 14:30 UTC (16:30 EET) | CanDo dev meetings Tue, 4pm EST | Use for pasting"19:21
th1aSteveA: Shall we discuss LaunchPad now?19:21
Lumierechanserv seems to have no irc-fu either19:21
SteveAth1a: okay19:22
aelknertaking a break for lunch.  be back later...19:22
Lumierecya at 2 aelkner19:22
SteveAth1a: you said that getting bugs into launchpad has been a roadblock19:23
SteveAI'm keen to see schooltool contributors give launchpad a fair trial.  I have resources to help get bugs into Launchpad and look into features that you find important.19:24
SteveAI think the cross-project collaboration features will be helpful19:24
SteveAwe'll be beta-testing a brand new UI for launchpad starting later this week, initially by invitation only, and I'd invite interested developers to evaluate and use the new UI19:25
th1aOK.  jfroche actually has exported all our old bugs as XML, but we'd gotten stuck having them imported.19:26
th1aI think he was emailing James Hentridge about it.19:26
jfrocheright and still no answer :(19:26
th1aYou were cc:ed.  Overall, it seemed to be lost in the shuffle.19:27
SteveAok. if you find things don't happen as fast as you'd like, you know you can call on me.19:27
th1aRight around the edgy release, as I recall.19:27
SteveAalso, note that email is lossy nowadays19:27
SteveAwith all the spam etc.19:27
th1aIt does seem like we're pretty bound to svn at this point.19:27
SteveAso sending a single email may not be enough sometimes.19:27
SteveAI think svn is a separate issue, but I'm interested to know why your bound to svn.19:28
th1aI think I'd need jinty to talk to you about that.19:28
SteveAso, if you're willing to give Launchpad a try, I'll ensure that James gets time and priority to work on the import of bugs.19:28
SteveAI need to know whether you want to use a demo server, to try it out among developers, or to have them imported directly to a live server.19:29
th1aI would be interested in trying the new UI.19:29
jintywe use zpkg to make releases, zpkg doesn't work with bzr AFAIK19:29
*** kitblake has joined #schooltool19:30
SteveAwhat does zpkg do that uses the VCS so closely?19:30
jintywe are bound to zpkg while zope3 is bound19:30
jintychecks out resources via svn19:30
jintyit is on my agenda to eggify schooltool once zope3 uses eggs, that will release that block19:31
SteveAjinty: I don't know a lot about this, but I work with people who do.  I would not like to consider it an open-and-shut case against evaluating bzr.  For example, we may find a way to make zpkg work as well with bzr as with svn.19:31
jintyI think my volunteer time would be better spent eggifying schooltool once I can rather than bashing my head against zpkg (which is a very dead project)19:33
th1aI just don't want to get caught up in infrastructural issues.  We've done enough of that already.19:33
SteveAI was thinking more that if zpkg is going to stand in the way of other projects trying out bzr, then that's something I should be looking to fix in zpkg19:33
SteveAbut, if you say it is more or less deprecated anyway...19:34
th1aYes, we're already in the unhappy space between zpkg and eggs.19:35
th1aI guess I still feel like if we've got to stick with svn we might as well use Trac in the interim as well.19:37
th1aAnd then move everything to Launchpad/bzr in about a year.19:37
SteveAI'd be disappointed if you chose trac over launchpad based on the infrastructural issue of zpkg vs eggs19:38
mgedminah, the trac idea19:38
mgedminwhat about spam?19:38
SteveAparticularly when I'm here, offering help to find how to make that infrastructural issue a non-issue19:39
mgedmintrac doesn't handle it19:39
th1amgedmin: You can password protect it.19:40
th1aSteveA:  Hm... so we're talking about what, making zpkg work with bzr?19:40
SteveAif that's the blocker, the motivator for this, then I guess we are19:41
mgedminanother blocker would be the svn2bzr conversion19:42
mgedminiirc ignas evaluated the various available tools and decreed that they were all too raw to be really usable19:42
mgedminlately I've been skimming the bzr mailing list archives on the web19:42
mgedmin(very inspirational code reviews there!)19:42
SteveAmgedmin: we have cutting-edge tools at canonical to do conversions19:42
mgedmindo they handle svn branches and various reorganisations?19:43
mgedmin(svn2bzr doesn't work with the latest bzr release; bzr-svn may have hope, but the version in edgy throws tracebacks right and left when I try to point it to any svn revision; tailor is a pain)19:44
th1aWe're a good test case for a hairy conversion.19:45
SteveAif someone will write for me the must-have stuff for a conversion to bzr, and the would-like stuff, then I'll get it checked out by the bzr experts.19:46
SteveAin this write-up, also explain why you need these things for schooltool, so that we can understand the context19:47
mgedminignas might now this stuff; he once played with tailor (iirc for svn -> darcs conversions)19:47
SteveAyou mean, your requirements?19:48
mgedminmy knowledge about the issue can be summed up like ths:19:48
mgedmin"hello, we are clueless users, we use svn and would like to switch to bzr because some smart people said it is better. uh, so what do we do?"19:49
* mgedmin started using bzr for maintaining a mini-fork of fbreader19:49
* mgedmin likes it:
SteveAok, so I will arrange a meeting with martin and/or david to figure out what will work best19:50
th1ajinty and ignas are the experts on our requirements at this point.19:50
mgedmindo we want a meeting?  are we ready to try to switch to bzr?19:50
SteveAI want to know your requirements in any case19:51
SteveAas, if there is stuff you need that we don't yet do, we can look at doing it.19:51
SteveAthen, you'll be well-informed when the time comes to seriously consider it19:51
mgedmina stable and proven (in the real world) way (+ documentation) to convert Subversion repositories into bzr branches would do it19:51
mgedminI'm not even sure how the end result should look like19:52
SteveAmeanwhile, we can keep an up-to-date bzr branch up as a launchpad branch import19:52
mgedminprobably a shared repository with many branches, if I got the terminology right19:52
SteveAand sync that with work done in svn19:52
mgedminoh, cool19:52
mgedminhow do launchpad svn imports work?19:52
mgedminI mean, is there a FAQ that answers this question?19:52
SteveAmgedmin: have you read jamesh and andrew bennett's blog entries about using launchpad with bzr and teams for shared branches?19:52
mgedminjamesh blog entries about bzr were very very nice19:53
mgedminI don't remember andrew bennett's posts19:53
SteveAthey are linked from...19:53
mgedminis he syndicated on planet ubuntu/gnome/debian?19:53
mgedminalso, the bzr website has improved considerably over the last year19:54
SteveApoo, can't find it19:57
SteveAI mailed martin to ask him to make these more easily findable on the website19:58
SteveAmgedmin: it's in your bookmark list, actually19:59
mgedminin that case I have read them :)19:59
*** jinty has left #schooltool20:03
*** jinty has quit IRC20:03
SteveAnow I will go shopping for food.  I'll get jamesh working on the bugs import issues, and get martin pool in touch with y'all about looking at bazaar.20:05
SteveAthanks for conversing!20:05
th1aThanks SteveA.20:08
Lumierethat was fun...20:34
Lumierewelsh will be around in a second to prepare the 2pm20:34
Lumiereand to see if we can find a way to patch in ignas I guess20:35
aelkneraelkner is back and ready to listen in on the 2pm meeting.20:46
aelknerOh, it's a conference call not a #schooltool meeting, isn't it?20:47
th1aIt is on the phone.20:56
th1aignas:  The number is toll free in the US, not sure if that helps you.20:56
aelknerwhen should I make the call, now?20:56
Lumierewe're going to get on the call in 30s or so20:58
Lumierebut it's safe to predial20:58
Lumiereit just makes you listen to elevator music till we get on ;)20:58
Lumierewe're on...21:00
th1aignas: ayt?21:00
Lumiereeukreign: there?21:02
Lumiereour 2 pm is starting21:02
th1aignas:  What's your number?21:02
th1aignas:  Hm... why don't you just try the toll free number and invoice me if you're charged for it?21:04
th1aIs that ok?21:04
ignasdon't really know ... and Aiste is not there anymore ...21:05
*** dwelsh has joined #schooltool21:05
*** pcardune has joined #schooltool21:05
* mgedmin remembers an ages-old hotel bill that he still hasn't invoiced21:06
th1aignas: We'll just chat with you here...21:06
* mgedmin sucks at paperwork21:07
LumiereItems on Agenda: Infrastructure21:07
Lumiere-- LaunchPad (from the SteveA chat above)21:07
pcardunei can reach it on skype, but then I can't enter in the access code21:08
Lumiereupdates about repository updates21:08
dwelshhmmm.  don't you have a keypad21:09
dwelshdo you know the code?21:09
pcardunei got in with my cell phone finally...21:10
ignasth1a: now that i thought more, i'll try calling21:11
ignasth1a: it shouldn't be a problem to invoice you ;)21:11
dwelshit is toll free in us and canada21:11
ignasdwelsh: us and canada have pretty good connections when it comes to phone system ;)21:12
Lumiere<-- Jason as well (Lumiere is what I've used as my nick for 3 or 4 years)21:14
ignasth1a: as for the SIF - who is going to be working on it ?21:53
Lumieresounds like alga22:03
Lumiereerr aelkner22:03
Lumierexp on the l ;)22:03
ignashmm, what's his Zope3 background ?22:05
Lumieremaybe he's still around to talk?22:06
Lumierehe was on the call22:06
ignasaelkner: ayt?22:06
*** dwelsh has left #schooltool22:27
*** Lumiere has left #schooltool22:42
*** mgedmin has quit IRC23:01
th1aignas: I'm going to work on it with aelkner and probably some other CanDo folks.23:09
ignasth1a: i am worried as proper background thread integrations it is extremally difficult subject ...23:10
ignasdifficult as in - i would google for solutions to similar problems + consult someone experienced with Zope3 internals a lot ...23:13
th1aI'm on the phone...23:16
*** alga has quit IRC23:16
th1aignas: Luckily, it isn't too hard other than that issue.23:33
ignasyes, indeed23:34

Generated by 2.15.1 by Marius Gedminas - find it at!