IRC log of #zope for Tuesday, 2010-05-04

*** giampaolo has joined #zope00:16
*** TKtiddle has joined #zope00:17
*** mgedmin has quit IRC00:19
*** benji has quit IRC00:23
*** TomBlockley has quit IRC00:24
*** fdrake has joined #zope00:26
*** dayne has joined #zope00:27
fdrakeTresEquis: I suspect I should be able to update the importance of bugs for zope.tal and zope.pagetemplate, but don't appear to have appropriate permissions on those.00:27
fdrakeIs that something you know how to change, or do you know who I should bug about it?00:28
TresEquisyou need to be a member of ztk-steering-group, likely00:28
TresEquiswhich would give you perms for all the ZTK packages00:28
TresEquisI can add you, if you like00:29
*** jim_SFU has quit IRC00:29
fdrakeEgads, that would be a lot of packages!00:29
fdrakeI think I'll be content to just ignore those two packages as much as I can for now.  :-)00:30
fdrakeI do expect a number of developers would be interested in being involved in only a subset of the packages in the ZTK; I'm not sure if there's a way to handle that on LP.00:31
*** dayne has quit IRC00:32
*** RaFromBRC is now known as RaFromBRC|away00:32
*** RaFromBRC|away has quit IRC00:33
*** evilbungle has quit IRC00:34
*** fdrake has left #zope00:38
TKtiddleHi could anyone be so kind as to look at this script I've registered as a view but I'm getting an error with it.Script:  error:
*** sunew has quit IRC00:42
*** Omega_ has joined #zope00:45
*** mr_jolly has joined #zope00:46
Omega_Can anyone point me to a good tutorial on creating my own indexes for zodb?  Probably something surrounding using OOBTrees?00:46
r0verOmega_: what version of Zope are you using ?00:48
Omega_No zope, just zodb.00:48
r0verOmega_: zodb version ? :)00:49
Omega_Ack!  Best way to check? :)00:49
Omega_I'm hoping the latest, let me see what I get when I try to reinstall it with easy_install...00:49
r0verOmega_: did you check at ?00:50
Omega_Yeah, I've scoured around.  OOBTree and all it's friends don't get much mention.00:51
Omega_Not that I could find.00:51
Omega_Especially not in the context of just simply creating the indexes and tossing them into the root of the DB.00:51
Omega_(which is to me is the most primitive use -- an effective "hello world" of btree indexing with zodb)00:51
Omega_Which is probably what I'm looking for... :(00:51
*** regebro has quit IRC00:53
Omega_If I want to create an object index by a string value found on that object, do I use an IOBTree or an OOBTree?00:53
davisagliOmega_: I don't know about a tutorial, but maybe look at the code of zope.catalog and zope.index ... or of cornerstone.soup00:53
davisagliOmega_: OOBTree; IOBTree uses integer keys00:54
Omega_That's what I assumed.00:54
Omega_Regarding existing projects, they all seem to build off of zope's catalogs.  Difficult to grok what's going on in the code.00:57
Omega_This is really holding me up, I wish the zodb project would have a bit more hand holding ;)00:57
Omega_In the documentation that is.00:58
*** daMaestro has quit IRC00:59
*** kosh has quit IRC01:00
r0verOmega_: i have a similar prolbem recently and i solved watching zope's code, in particular tests cases01:06
Omega_Well, there's this:
Omega_Which is probably spot on for what I want to do, but it doesn't seem to have coalesced into a project I can rely on.01:06
r0verOmega_: either you want to create a new kind of index or an instance of an already existant index, watching code from Products.PluginIndexes.FieldIndex.FieldIndex helped a lot01:07
Omega_Ideally I want all my data classes to know how to index themselves upon creation.01:07
Omega_Just write one base data class and then when it's extended, they highlight which properties are to be indexed.01:08
Omega_Which as a result creates the btrees based on the Class and property name.01:08
Omega_(property or attribute, whatever ;)  )01:08
r0verthat page is kinda old, but general concepts may be applied: "This page was last updated Aug 30, 2005."01:08
Omega_Exactly! :)01:09
Omega_Just that I'm somewhat newish to doing this indexing manually - so it'll take a while for me to understand what's being done.01:10
Omega_Which is kind of what drew me here to ask about tutorials, which might do better at explaining the concepts as they are applied.01:10
*** TKtiddle has quit IRC01:10
MatthewWilkesOmega_: I don't think there are many people doing this kind of indexing manually.  Most people that use ZODB will be either using Zope to handle it and the few that aren't and need indexing will probably think they know all the answers anyway ;)01:11
Omega_all too true.  Trying to learn is definitely an unforgiving process ;)01:11
MatthewWilkesOmega_: There will probably be good stuff to crib from in ZCatalog, the simpler indexes especially shouldn't require any zope stuff01:12
Omega_Alright, I'll put that on my hit list next then.01:12
mcdoncthere's not much tutorial grist around but you might get use out of reading the code at
mcdoncthis is effectively a zope2 style "catalog" without the zope201:13
mcdonc(it uses a package named "zope.index" under the hood to implement actual indexes)01:13
mcdoncend user docs are at
*** TKtiddle has joined #zope01:16
mcdonc(reading this will almost certainly be more productive than reading zope2 catalog code, as that code has zope2-isms in it that you wil trip over)01:17
*** mcdonc has quit IRC01:19
Omega_That was my concern when it came to the concept of "catalogs".01:19
Omega_I'd rather just deal with fields or text being indexed.01:20
*** Arfrever has quit IRC01:23
*** TresEquis has quit IRC01:34
*** menesis has quit IRC01:35
*** RaFromBRC has joined #zope01:40
*** r0ver has left #zope01:41
*** igbun has joined #zope01:45
*** r0ver has joined #zope01:46
*** fredvd has joined #zope01:48
*** menesis has joined #zope01:48
Omega_Thanks for the help all!01:51
*** Omega_ has left #zope01:51
*** menesis has quit IRC01:58
*** menesis has joined #zope02:00
*** RaFromBRC has quit IRC02:06
*** kosh has joined #zope02:11
*** TKtiddle has quit IRC02:13
*** RaFromBRC has joined #zope02:13
*** redir has quit IRC02:18
*** sashav has quit IRC02:20
*** jan_s_ has left #zope02:20
*** mr_jolly has left #zope02:35
*** dayne has joined #zope02:40
*** RaFromBRC has quit IRC02:42
*** dayne has quit IRC02:44
*** mcdonc has joined #zope02:47
*** cwarner has quit IRC02:49
*** davisagli has left #zope03:05
*** davisagli has joined #zope03:07
*** RaFromBRC has joined #zope03:09
*** davisagli has quit IRC03:12
*** r0ver has left #zope03:15
*** giampaolo has quit IRC03:16
*** tiwula has quit IRC03:27
*** menesis1 has joined #zope03:51
*** menesis has quit IRC03:51
*** tiwula has joined #zope04:00
*** Guest88457 has quit IRC04:06
*** davisagli has joined #zope04:12
*** aaronv has quit IRC04:15
*** baijum has joined #zope04:17
*** astoon has joined #zope04:20
*** r0ver has joined #zope04:37
*** tiwula has quit IRC04:38
*** matthewwilkes_ has joined #zope04:51
*** MatthewWilkes has quit IRC04:51
*** matthewwilkes_ is now known as MatthewWilkes04:51
*** r0ver has left #zope05:29
*** huajie has joined #zope05:41
*** davisagli has quit IRC05:51
*** tiwula has joined #zope05:58
*** fredvd has quit IRC05:59
*** igbun has quit IRC06:05
*** davisagli has joined #zope06:35
*** davisagli has quit IRC06:37
*** menesis1 has quit IRC06:50
*** astoon has quit IRC06:53
*** davisagli has joined #zope07:06
*** rTheJeste has joined #zope07:16
*** smita has quit IRC07:19
*** kosh has quit IRC07:19
*** kosh has joined #zope07:20
*** smita has joined #zope07:22
*** JaRoel|4D has quit IRC07:25
*** chaoflow_ has quit IRC07:25
*** TheJester has quit IRC07:25
*** JaRoel|4D has joined #zope07:27
*** chaoflow_ has joined #zope07:27
*** skt_ has joined #zope07:29
*** baijum has quit IRC07:35
*** shinygiftson has joined #zope07:37
*** lcpfnvc has quit IRC07:53
*** shinygiftson has quit IRC08:10
*** Vq has quit IRC08:10
*** vigith has joined #zope08:12
*** shinygiftson has joined #zope08:15
*** Vq has joined #zope08:15
*** vigith has quit IRC08:18
*** pavanbabu has joined #zope08:23
*** timte has joined #zope08:24
*** RaFromBRC has quit IRC08:51
*** MatthewWilkes has quit IRC08:52
*** tiwula has quit IRC08:56
*** JaRoel|4D has quit IRC09:12
*** baijum has joined #zope09:12
*** phimic has joined #zope09:24
*** menesis has joined #zope09:25
*** astoon has joined #zope09:50
*** jan_s has joined #zope09:58
*** JaRoel|4D has joined #zope09:59
*** Kabz|4D has joined #zope10:00
*** akm1 has joined #zope10:06
*** menesis has quit IRC10:08
*** menesis1 has joined #zope10:08
*** mr_jolly has joined #zope10:13
*** wosc has joined #zope10:18
*** factor has quit IRC10:19
*** menesis has joined #zope10:32
*** planetzopebot has quit IRC10:33
*** menesis1 has quit IRC10:33
*** planetzopebot has joined #zope10:34
*** neo|4D has joined #zope10:35
*** matleh has joined #zope10:37
*** agroszer has joined #zope10:38
*** menesis1 has joined #zope10:43
*** menesis has quit IRC10:43
*** agroszer has quit IRC11:07
*** digitalmortician has joined #zope11:07
*** tisto has joined #zope11:09
*** jakke has joined #zope11:11
*** jakke has quit IRC11:12
*** menesis1 has quit IRC11:24
*** regebro has joined #zope11:37
*** davisagli has left #zope11:42
*** igbun has joined #zope11:44
*** menesis has joined #zope11:49
*** menesis has quit IRC11:51
*** smita has quit IRC11:53
*** smita has joined #zope11:56
*** menesis has joined #zope11:57
*** matleh has quit IRC11:58
*** emrojo has joined #zope12:22
*** ramonski has joined #zope12:29
*** fredvd has joined #zope12:31
*** igbun has quit IRC12:34
*** igbun has joined #zope12:34
*** igbun_ has joined #zope12:35
*** igbun has quit IRC12:35
*** igbun_ is now known as igbun12:35
*** eperez has joined #zope12:38
*** sim_ has joined #zope12:41
*** menesis has quit IRC12:46
*** evilbungle has joined #zope12:46
*** MJ has joined #zope12:55
*** tisto is now known as tisto|lunch12:59
*** menesis has joined #zope13:01
*** TomBlockley has joined #zope13:02
*** thetet has joined #zope13:07
*** pavanbabu has left #zope13:16
*** pavanbabu has joined #zope13:17
pavanbabuHi all13:18
*** giampaolo has joined #zope13:19
*** hever has joined #zope13:39
*** agroszer has joined #zope13:42
*** hever has quit IRC13:44
*** JaRoel|4_ has joined #zope13:49
*** JaRoel|4D has quit IRC13:49
*** sim_ has quit IRC13:52
*** sim_sim has joined #zope13:53
*** mgedmin has joined #zope14:03
*** sunew has joined #zope14:05
*** teix has joined #zope14:06
*** igbun has quit IRC14:08
*** igbun has joined #zope14:09
*** astoon has quit IRC14:19
*** aaronv has joined #zope14:20
*** purserj has quit IRC14:22
*** purserj has joined #zope14:22
*** goschtl has joined #zope14:35
*** igbun has quit IRC14:58
*** emrojo has quit IRC14:59
*** shinygiftson has quit IRC15:00
*** benji has joined #zope15:06
*** Charlie_X has joined #zope15:11
*** emrojo has joined #zope15:13
pavanbabuhi all can any body tell zeopack how it works15:17
*** digitalmortician has quit IRC15:17
Charlie_XLooks like a pack command for Zeo-based setups.15:20
*** hever has joined #zope15:25
*** Theuni1 has joined #zope15:26
*** TomBlockley has quit IRC15:28
*** j-w_ has joined #zope15:31
*** menesis has quit IRC15:33
*** menesis has joined #zope15:34
fredvdpavanbabu: If you delete an object in your zope site it's marked as deleted in the ZODB but not removed.15:46
fredvdif you run a zeopack the ZODB is compacted by removing all the objects that have been marked as deleted.15:46
pavanbabu fredvd:thanks15:47
pavanbabufredvd:but i am using buildout which recipe does it support?15:48
*** r0ver has joined #zope15:56
*** mcdonc has quit IRC16:00
*** igbun has joined #zope16:01
*** mcdonc has joined #zope16:05
*** j-w_ has quit IRC16:13
*** j-w_ has joined #zope16:13
*** hever has quit IRC16:15
*** hever has joined #zope16:15
*** seppo14 has joined #zope16:16
seppo14while installing 2.12.5 i get error: Installed distribution pytz 2010h conflicts with requirement pytz<=2010b16:16
betabugseppo14: you'll probably have to tell people in which way you try to install16:17
*** baijum has quit IRC16:17
seppo14damn my mistake16:18
betabugyour system python has a different version of pytz installed16:18
betabugyou want a cleaner python :-)16:18
seppo14just noticed i did not switch to my other env16:18
seppo14thanks for the help, i guess that's it :)16:18
*** grahal_ has joined #zope16:20
*** grahal has quit IRC16:21
betabugno problem :-)16:25
*** dabaR has joined #zope16:29
*** tisto|lunch is now known as tisto16:30
*** skt_ has quit IRC16:30
*** dabaR has left #zope16:34
*** alvaro has joined #zope16:36
*** j-w_ has quit IRC16:36
*** j-w_ has joined #zope16:37
*** hannosch has joined #zope16:41
*** TomBlockley has joined #zope16:43
*** hever has quit IRC16:45
*** r0ver has left #zope16:46
*** baijum has joined #zope16:59
*** fredvd is now known as fredvd|away17:00
*** Theuni1 has quit IRC17:02
hannoschccomb: ping? are you here?17:04
*** fdrake has joined #zope17:05
*** tarek has quit IRC17:05
*** fdrake has left #zope17:08
*** fdrake has joined #zope17:08
*** allisterb has quit IRC17:08
*** pepeu has joined #zope17:10
*** giampaolo has quit IRC17:13
*** allisterb has joined #zope17:25
*** j-w__ has joined #zope17:26
*** Theuni1 has joined #zope17:27
hannoschj-w__: looks like Christophe isn't here today. shall we aim for Thursday?17:28
j-w__hannosch: that's ok with me17:28
*** tiwula has joined #zope17:28
*** timte has quit IRC17:28
hannoschj-w__: ok, hope your move did go well :)17:29
j-w__it will actually give me a bit more time to attend17:29
j-w__hannosch: it went quite ok17:29
*** j-w_ has quit IRC17:29
hannoschok, that doesn't sounds all that happy ;)17:30
j-w__moving is always messy :-)17:30
j-w__so much cruft that you gather in an office you know17:30
j-w__loads and loads of paper, broken hardware that you didn;t throw away17:30
hannoschright, the trick is to throw stuff away well in advance :)17:31
j-w__old computers you cannot say goodbye to17:31
j-w__right :-)17:31
hannoschyeah. they'll remind you how heavy things used to be :)17:31
j-w__and then making yourself at home again in the new office17:31
j-w__exactly! :-)17:31
hannoschyep. but it's an improvement, right?17:32
j-w__and actually we will be working from two different locations to limit the amount of commuting17:32
j-w__so today we setup the second locations17:32
j-w__in the end it is an improvement :)17:32
hannoschah, ok. sounds like you are going to have fun for a while :)17:35
j-w__and you fully recovered?17:36
j-w__it sounded like being ill caught you in exactly the wrong moment17:37
j-w__(it almost always does of course...)17:37
*** hever has joined #zope17:38
hannoschj-w__: ah, it was just too many of us being at the same place. I blame it on continental viruses - and not any of the alcohol we drank ;)17:39
*** ramonski has left #zope17:39
j-w__hannosch: LOL - of course not!17:39
hannoschso it's just been a day at home sleeping a lot17:39
*** cwarner has joined #zope17:40
*** akm1 has quit IRC17:42
*** r0ver has joined #zope17:43
*** LeoRochael has joined #zope17:43
*** giampaolo has joined #zope17:44
*** phimic has quit IRC17:44
hannoschccomb: ah. you read your mails?17:49
ccombnot for a few hours17:49
ccombok reading17:49
*** Theuni1 has quit IRC17:49
LeoRochaelcan mr.developer checkout bzr branches?17:50
hannoschccomb: ok. j-w__ proposed to meet today. but he only has time for the next 40min. I guess we'll do the regular zope meeting instead17:50
j-w__hannosch: agreed17:50
hannoschccomb: does Thursday anytime work for you?17:52
ccombok I was not much available during the past week, I will have more time from tomorrow17:52
*** TresEquis has joined #zope17:52
ccombthursday I'm available only during the evening17:53
ccombI'm mean the french evening :P17:53
*** Theuni1 has joined #zope17:53
*** pavanbabu has left #zope17:54
ccombotherwise any other day is ok17:54
j-w__ccomb, hannosch thursday evening would work for me I think17:54
j-w__say between 20:00 and 21:00 (CEST)17:54
hannoschwould work for me as well17:55
j-w__i would not like to move to next week again, since I'm not sure about friday17:55
j-w__ok, thursday 20:00 - 21:00 (CEST) then!17:55
*** dunny has quit IRC17:56
* ccomb would like to have a TZ coprocessor in the brain17:56
*** regebro has quit IRC17:56
j-w__ccomb: German, French and Dutch time :)17:57
LeoRochaelccomb: is usually enough for me :-)17:57
ccombok thanks :P17:57
* Charlie_X says ccomb shouldn't worry about keeping time. We know you are French :-P17:57
hannoschccomb: what time are you currently using? not utc+2 ?17:57
ccombmmh I suppose so17:57
ccombit's 5pm17:58
hannoschccomb: indeed. then we are all on the same timezone17:58
Charlie_XCEST is UTC and valid from Spain to Poland (UK and Ireland excepted)17:58
bigkevmcdlet's just ignore Portugal ;-)17:59
hannoschCEST is UTC+217:59
*** J1m has joined #zope17:59
*** regebro has joined #zope17:59
TresEquiswe ready?18:00
*** Theuni1 has quit IRC18:00
* regebro has coffee18:00
*** deux_ has joined #zope18:00
* ccomb just finished a waffle18:00
regebroAnd Christian just lost connection. Nice timing. :)18:01
* TresEquis was thinking about chicken waffles this morning18:01
Charlie_XHi everyone. Theuni asked me to convene as he's in a train somewhere.18:01
Charlie_XFirst up KGS 3.4.1 release status18:01
TresEquisagroszer: are you the one with KGS 3.4.1 status update?18:01
agroszerI guess18:01
* agroszer was distracted18:02
agroszerI'm waiting on becoming a ZF member18:02
agroszerit seems like scp access to goes through that way18:02
hannoschhuh? is that what Jens said?18:02
agroszerotherwise tests pass, need to publish a 3.4.1b118:03
agroszernah, just seems like it's beter to be a ZF member18:03
regebroagroszer: It seems right, yeah. :)18:03
hannoschagroszer: ok. but it shouldn't block anything. just ask Jens18:03
baijumagroszer: ZF member is different from ZF committer ?18:03
agroszergood question18:04
hannoschyes. the two have nothing to do with each other18:04
regebroIt probably takes to long to become a member, last board meeting got postponed, I think.18:04
baijumah. ok18:04
agroszerregebro, I read 10 days18:04
* Charlie_X has commit rights but not a member of the foundation18:04
* baijum is not a ZF member. but has access to download.zope.org18:04
agroszerunless someone does not want you18:04
agroszerokok, I'll talk to Jens18:05
Charlie_XThat's much faster than joining the PSF...18:05
regebroagroszer: 10 days is likely to not corresponding to actuality. :)18:05
Charlie_XOkay, so  doublecheck with Jens and press the button.18:05
* agroszer raising the finger...18:05
*** dayne has joined #zope18:05
Charlie_XAnymore on the KGS?18:06
ccomb9404 tests, 2 failures on python 2.418:06
ccombsame on 2.518:06
agroszerwhat's the BB address again?18:06
Charlie_XSounds like topic #2 Test runners/nightly builds18:07
Charlie_X   - Windows machines18:07
TresEquisccomb: do we need to backport Lennart's fixes?18:07
agroszerccomb, you're missing enscript18:07
*** tisto has quit IRC18:07
agroszerand using distribute instead of setuptools18:07
* Charlie_X wishes to give Lennart a round of applause for zope.testing on >= 3.118:07
regebroTresEquis: Doens't sound like it, because they didn't fail on 2.5.18:08
* ccomb aptitude install enscript18:08
regebroCharlie_X: Well, it doesn't exist yet. :)18:08
baijumAll our buildbots sends status to zope-tests list  ?
regebroExcept as a branch.18:08
ccombcan't we put an ellipsis on the setuptools/distribute line causing the failure?18:10
do3ccbaijum: I am working on getting better notifications, Its not done yet because I was in a long holiday18:10
baijumdo3cc: ok, fine18:11
agroszerccomb, actually it should be a (setuptools|distribute)18:11
*** hever has quit IRC18:11
agroszerccomb, maybe the normalizer can do it18:12
regebroThe normalizer can do it.18:12
regebroI added one on trunk already,18:12
ccombthat means releasing a new version of zc.recipe.testrunner?18:12
regebroccomb: Trunk *may* use zope.testrunner, so be careful. :)18:13
agroszerI'll look into that18:13
agroszerlet's keep the topic18:13
TresEquishannosch, agroszer, sidnei:  any update on the MSDN invites?  Did you get the license keys?18:13
agroszerTresEquis, nothing unless it went to gmail spam18:14
hannoschTresEquis: I filled in the info using the invitation key, but haven't heard anything back18:14
ccomb(ok so now I'm relaunching the KGS 3.4.1 tests with enscript and setuptools)18:14
agroszersame here18:14
hannoschTresEquis: could you poke your Microsoft contact about it?18:14
TresEquisOK, will do18:15
*** Theuni1 has joined #zope18:15
* TresEquis is still uncertain how those license keys will work with the EC2 image we are thinking to use18:15
TresEquisand just how that will mesh with the Health Agency's offer to host the image via vmware18:16
agroszerI think the requirement gathering thread went silent18:16
* Charlie_X thinks we should be looking at getting a whole MS Azure setup for the windows buids18:16
hannoschTresEquis: if we play by the Microsoft license rules, it is probably all impossible18:17
* TresEquis has enough blues already :)18:17
*** davisagli has joined #zope18:17
* Charlie_X we could be an MS poster child18:17
agroszersidnei is around?18:18
* mgedmin wonders why enscript is needed -- for z3c.coverage only?18:18
hannoschonce we get the actual concrete license, I can look up the license terms18:18
*** webmaven has joined #zope18:18
agroszermgedmin, yes18:18
* mgedmin mumbles about switching to pygments at some point18:18
agroszerTresEquis, I think we're waiting on sidnei to do the AMI18:19
*** wosc has quit IRC18:19
agroszerif I remember well18:19
Charlie_XOkay, so Tres to kick some MS butt re. licences18:20
Charlie_Xagroszer: what is AMI?18:20
regebroIt feels like the speed of this meeting halted.18:20
agroszermachine image18:20
agroszerCharlie_X, machine image18:20
Charlie_XThanks. Deliverable from Sidnei outstanding.18:20
Charlie_XAnything else on this topic?18:20
TresEquiswe need18:21
TresEquisa grok smoketest being posted to zope-tests18:21
TresEquisnot all [the little grokcore.* ones18:21
TresEquisjust "is Grok 1.0" (or 1.1) "broken today?"18:22
TresEquisso we know if we broke something by changes to the ZTK18:22
Charlie_XAs per your e-mail earlier? hannosch will you take this up with Jan-Wijbrand on Thursday in case it isn't covered earlier?18:22
LeoRochaelIIUC grok versions for 1.0 are totally pinned, so unlikely to be broken. grok trunk OTOH, would make a lot of sense18:23
Charlie_XNext - Deprecating Python < 2.518:23
regebroI'm +/-0 on that.18:23
Charlie_XIt looks like we're starting to find areas where without deprecation we have considerable problems.18:23
Charlie_XI'm +1 on it. Python 2.4 is itself not maintained.18:24
Charlie_XAnd even Debian has moved off it18:24
ccombI few weeks ago I've made some changes to some doctests so that tests pass on python2.4 on bbkit18:25
regebroI don't think keeping it supported is going to be that tricky, but then again, we don't need it, so... +/-0. :)18:25
ccombit was just a matter of having output dicts ordered18:25
LeoRochaelAFAIU, this only came up because we were bumping on now-resolved compat problems in some tests18:25
LeoRochaelisn't the previous decision enough?18:25
ccombI wonder how different the ztk is from bbkit18:25
*** Arfrever has joined #zope18:25
ccombotherwise I don't care about python 2.418:26
ccombI'm ok for dropping it18:26
agroszeryah py2.4 is damn old18:26
LeoRochael(2.4 on the bycicle kit, 2.5 on the rest, not explicitly break 2.4 for a while)18:26
regebroWe'll need a bicycle toolkit buildbot.18:26
regebroWhich means we need to define what the bicycle tookit *is*. :)18:27
regebroRunning 2.4 to 3.118:28
ccombshould we have a bicycle.cfg in zopetoolkit ? :)18:28
regebroccomb: It's a possibility...18:29
Charlie_XI really would like to see someone make a case as to why support 2.4 isn't byte rot.18:29
Charlie_XAny more votes on this? Nearly time.18:29
regebroCharlie_X: as long as the tests pass it's not rotting...18:29
*** sunew has quit IRC18:30
hannoschit's creating more and more pain. without any real value18:30
baijum+1 for dropping 2.418:30
regebroWell, no -1 votes, and time is out.18:30
ccombok python 2.4 dropped18:30
Charlie_XLeoRochael: seemed -1 but outvoted18:30
* LeoRochael shrugs18:31
Charlie_XThanks everybody.18:31
regebroProcedure suggestion: We drop the Python 2.4 metatags, but doesn't change anything yet.18:31
baijumThanks Charlie_X !18:31
agroszerregebro, maybe a changelog entry?18:31
hannoschregebro: I'd shut down the Python 2.4 buildbots18:32
regebroagroszer: Yeah, sounds good.18:32
regebrohannosch: Right, we can do that.18:32
TresEquisregebro: BTK is zope.interface, zope.component, zope.configuration, and dependencies18:32
regebroAnd after ZTK 1.0 official release, we actually clean up any Python 2.4 specials.18:32
*** j-w__ has quit IRC18:33
baijumTresEquis: the "extra_requires" will not count for BTK, is it ?18:33
baijumI mean deps18:33
regebroTresEquis: OK, we need to list them up then. Maybe I'll do that next week, and set up a builbot, I have a server that isn't doing much.18:33
Charlie_XZTK status next week?18:33
TresEquisbaijum: right:  no security, no persistence, etc.18:33
hannoschCharlie_X: I cannot promise that we have anything by that time. And I'll be gone for the whole week at a customer18:34
*** daMaestro has joined #zope18:34
* ccomb will replace the python2.4 buildbot warning with a python2.7 one :D18:34
Charlie_XOkay. But that still leaves ccomb and j-w_18:34
TresEquisregebro: I almost checked in a btk.cfg file to zopetoolkit/trunk a while back18:34
fdrakeArgh!  Did I really miss the Python 2.4 support vote?18:34
LeoRochaelfdrake: most everybody is still here18:35
Charlie_Xfdrake: postal votes still accepted18:35
regebrofdrake: Yeah. If you are -1 you are still outvoted, though. :)18:35
fdrake-1 for dropping 2.4 support.18:35
cwarner+1 for dropping18:35
fdrakeYeah, but for the record.18:35
regebrofdrake: Any arguments?18:35
* mgedmin is wavering between -0 and +018:35
Charlie_Xfdrake: interested in your reasoning.18:35
fdrakeI'm still using it for production applications that are still evolving.18:36
* Charlie_X likes mgedmin's decisiveness18:36
fdrakeI'd like to be able to track bug fixes without having to backport everything myself.18:36
cwarnerfdrake: evolving how nearly all decent modules are 2.5+18:36
regebroWhat's stopping you from using 2.5 or 2.6?18:37
fdrakecwarner: Not sure I can parse that.18:37
Charlie_Xfdrake are you in one of those environments where a more recent version of Python is no go?18:37
fdrakeregebro: Time.18:37
cwarnerright now I spend copious amounts of time backporting 2.5, 2.6+ modules if i really need them.. it's painful18:37
fdrakeCharlie_X: No.  It's a question of scheduling work.18:37
*** baijum has quit IRC18:37
cwarneryeah 2.4 is what.. 6 years old now?18:38
regebrofdrake: The forward breaks from 2.4 are quite small, except highly magic stuff that's in Zope 2 and zope.proxy and such.18:38
fdrakecwarner: I agree it's ancient.18:38
cwarnervery few people are still targetting 2.4.. it's 2.5+18:38
regebroShouldn't take that long, or?18:38
TresEquisfdrake: can you help keep 2.4 support viable?18:38
fdrakeI suspect so.18:38
TresEquise.g., run buildbot / nightly tests, and help track fixes?18:38
fdrakeAt least where specific issues come up.18:39
TresEquisI think we can leave the "don't intentionally break 2.4" policy in place for a while18:39
fdrakeI can't commit a buildbot, but I can help out with specific issues.18:39
TresEquisit would help if you can let us know more or less how long you need it18:39
hannoschfor how long? another three years?18:39
*** deux_ has quit IRC18:39
ccombI can leave the 2.4 buildbot, it's already there18:39
fdrakeTresEquis: That would be acceptable to me.18:39
TresEquisccomb: OK18:40
ccombmaybe just removing the mail alert18:40
cwarneryeah i'm with hannosch.. How long do we keep the floppy disk around?18:40
*** davisagli has quit IRC18:40
regebroI have no problem with keeping in 2.4 support until a ZTK 1.0, but after that I think it should be dropped.18:40
fdrakehannosch: I'm hoping for not all that long, but I'm not the one to schedule resources and deployments on this.18:40
regebro(And cleaned out in fact).18:40
TresEquisthat seems reasonable to me18:40
Charlie_XI'd be okay with that but I think we need a perspective for dropping Python 2.418:40
TresEquisassuming that ZTK 1.0 release is this quarter18:40
cwarnerif it is kept, people will keep relying on it. Sometimes you just need to make a clean break and move on18:41
TresEquisthen 1.0.x releases (if there are any) would keep 2.418:41
TresEquisbut the trunk would be clean18:41
Charlie_Xfdrake: would it help get resources if a date for dropping support is published?18:41
hannoschI still think we should officially drop 2.4 support and leave it to interested community members to keep it running18:41
fdrakeCharlie_X: Not sure.18:41
cwarnerhannosch: sounds logical18:41
TresEquishannosch: what about new language features18:41
hannoschTresEquis: I'm fine with the "do no harm" approach18:42
fdrakeFor the same reason as before: They aren't my resources to allocate, so I can't comment meaningfully.18:42
hannoschso not allowing new 2.5+ features18:42
fdrakeA published drop-dead date makes sense as a way to approach it, though.18:42
regebroOK, how about this:18:43
TresEquisfdrake: would you be OK with tracking ZTK 1.0?  There are likely to be much bigger changes on the horizon after that18:43
*** goschtl has quit IRC18:43
regebroWe keep Python 2.4 support for ZTK, but we don't require of most developers to support it.18:43
regebroThose who need it, ie fdrake, or are interested, like me, try to keep the 2.4 buildbot clean.18:44
fdrakeTresEquis: I think so.  ZTK 1.1+ should be safe for dropping 2.4 support.18:44
TresEquisworks for me18:44
ccombZTK 1.0 is not supposed to have the very latest changes18:44
hannoschhhm. ok. whatever ZTK version BlueBream is going to use, we can support 2.418:44
Charlie_XOkay. Supporting Python 2.4 in ZTK 1.0 as legacy costs nothing. A definite date may help project managers plan migration.18:44
TresEquislets get ZTK 1.0 out the door ASAP18:44
hannoschI think we'll just label some version set 1.0, which is going to be shared between the next Grok and BB releases18:45
ccombis there a way to sync versions found in z2, bb and grok, and call it ZTK 1.0?18:45
ccombit should not be very different18:46
hannoschccomb: you, me and Jan-Wijbrand can figure that out18:46
ccombwas python 2.4 already dropped for zope 2.12?18:47
Charlie_Xfdrake: Python 2.4 will not be supported beyond ZTK 1.0 or 2010-12-31. Whichever comes earlier.18:47
hannoschyes. 2.12 only supports 2.6+18:47
fdrakeCharlie_X: That seems fair to me.18:47
hannoschccomb: 2.12 is actually really incompatible with 2.4 on 64bit. there's hard breakage18:48
fdrakeIf 2010-12-31 comes first, it's unlikely I'll care about the ZTK.  :-)18:48
* Charlie_X notes that Python 2.5 is the next version for the chop18:48
fdrakePython 2.5 I don't care about at all.  :-)18:49
cwarner2.4 is just extreme legacy in the grand scheme of things..18:49
hannoschI think it makes sense to aim for two 2.x releases at the same time in general. once 2.7.1 comes out, we can talk about dropping 2.518:49
regebroYeah, but Python 2.6 breaks so little, that 2.5 support is easy.18:49
regebroUnless we want to start doing __future__ imports. :)18:49
*** MJ has quit IRC18:49
cwarnerit's like trying to support a car built in 1930 with fuel injection.. at some point it just gets silly18:49
Charlie_Xregebro: I think we've done well avoiding the shiny new toys so far18:50
Charlie_X@decorator1 @decorator2 @decorator3 def simple(): return "hello world"18:50
ccombnow it's much more interesting spending time to support 3.1 than to keep 2.418:51
*** thetet has quit IRC18:51
ccombwas that the last topic?18:51
hannoschwe are long overtime :)18:52
Charlie_Xccomb: as noted by many Python developers - Python 3.x comes with a whole new set of warts and few compelling reasons as things stand.18:52
Charlie_XWe went into extra time and had a silver goal to keep Python 2.4 for ZTK18:52
Charlie_XI'll post a summary to the list.18:52
Charlie_XThank you for attention and same time next week.18:53
hannoschCharlie_X: cool. thx!18:53
TresEquisregebro: usnig with is the coolest new 2.6 feature18:53
TresEquishannosch: I think the pytz pin has broken zope2's installability in a virtualenv18:54
VqCharlie_X: what new warts?18:54
* Vq <- somewhat python3 ignorant18:54
hannoschTresEquis: is that because the earlier pytz versions aren't reachable from PyPI?18:54
TresEquiswe get conflicting requirements18:55
TresEquisI'm trying now in a fresh virtualenv to double check18:56
hannoschTresEquis: which index are you using?18:56
Charlie_XVq - string handling is still a problem18:56
Charlie_XApparently that is a real problem in real application development. And with no performance improvements no one is very keen on what can be extensive refactoring.18:58
lisppaste6TresEquis pasted "Zope2 install broken in virtualenv" at
TresEquishannosch: the toilet19:00
hannoschTresEquis: right. but that would "break" on mechanize 0.2 as well19:00
VqCharlie_X: that is a new wart?19:00
hannoschTresEquis: automatically installing everything from the toilet is just not possible19:01
TresEquismgedmin: PyPI19:01
mgedminisn't it evil to explicitly require old versions of pytz?19:01
* mgedmin was huh'ing at the pasted traceback19:01
hannoschmgedmin: no. our software isn't compatible19:01
mgedminisn't it evil of pytz to break backwards-compat?19:02
TresEquismgedmin: Stuart added a BBB-incompatible change19:02
TresEquiswhich breaks Z2's DateTime19:02
mgedminpytz carries timezone info. timezones change all the time.  using old pytz is a bug19:02
hannoschwe cannot support the latest on PyPi every single minute somebody releases a new package19:02
TresEquishannosch: we should be *trying* to, though ;)19:02
mgedminoh dear, what a mess :(19:03
hannoschin this case doing bin/easy_install pytz==2010b, bin/easy_install Zope2 should work19:03
hannoschTresEquis: sure. I don't know enough about timezones to fix any of this19:03
*** J1m has left #zope19:03
TresEquismkzopeinst then blows up on missing zope.password  :(19:03
hannoscheh. then you got the wrong mkzopeinst I guess19:04
hannoschthe one from Zope2 doesn't use zope.password19:04
hannoschscripts with the same name overwriting each other19:04
TresEquisI'm redoing the virtualenv with -i
TresEquisas noted in the docs:
hannoschthey do that silently without a warning. I think that's why buildout defaults to ignoring scripts19:05
TresEquismgedmin: maybe you can help us lobby Stuart for a fix:
*** Kabz|4D has quit IRC19:09
*** Theuni1 has quit IRC19:13
seppo14has anyone else experienced problems with Transmit 4 on MacOS & ZOPE FTP access not uploading files the right way?19:16
mgedminTresEquis, from those three bug reports I get the impression that somebody needs to sit down and think about what's the right thing here and which of the packages needs to be fixed19:20
TresEquisfixing it in DateTime would require an ugly straddle19:21
TresEquisunless we pinned pytz >= current19:21
mgedmincould you paste/link to the test failures in lp#572715?19:21
mgedminincidentally, anybody using lucid?19:22
TresEquisnot yet19:22
TresEquisplanning to upgrade all the machines soon19:23
* mgedmin is getting lots of /home/mg/.buildout/eggs/setuptools-0.6c11-py2.6.egg/setuptools/command/ UserWarning: Unbuilt egg for setuptools [unknown version] (/usr/lib/python2.6/dist-packages)19:23
mgedminwhich don't seem to break anything, other than my calm19:23
*** neo|4D has quit IRC19:23
hannoschif setuptools gives you a rather long warning text, it's usually unimportant. if you only get very short messages you should be careful. at least that's my experience :)19:24
hannoschthere's an inverse helpfulness vs. criticality relationship19:25
*** davisagli has joined #zope19:26
mgedminok, I can reproduce 8 DateTime failures with pytz2010h19:27
TresEquisI just added a recipe to the LP bug for reproducing it19:27
LeoRochaelTresEquis: what do you mean by "would require an ugly straddle"?19:29
*** JaRoel|4_ has quit IRC19:29
TresEquisLeoRochael:  if you try to make DateTime compatible with both older and newer pytz's, it has to "straddle" to API change19:29
TresEquiswhich is tough, because the spelling of the API didn't change at all, only the returned result19:30
LeoRochaelTresEquis: ok, what would be the drawback of just fixing DateTime and pinning to newer versions?19:31
LeoRochael... pinning pytz to newer versions...19:31
TresEquisLeoRochael: that might be the best way out19:32
hannoschI'm fine with that. document the API change and release a new feature release of DateTime19:33
*** Kabz|4D has joined #zope19:33
mgedminTresEquis, was it this change that broke things: ?19:33
mgedminlooks like not19:34
mgedminr212 through r214 look like the ones19:36
mgedminlaunchpad is sloooooow :(19:37
hannoschtry to access bitbucket and you know what slowness / unavailability means ;)19:37
mgedminwhat on earth?19:41
fdrakemgedmin: I've been using Lucid for a few days now.19:41
mgedminso, DateTime('2002/5/2 8:00am US/Eastern')._micros differs by 4 hours across different pytz versions19:42
*** huajie has quit IRC19:43
*** sunew has joined #zope19:43
mgedminthe 399-line DateTime._parse_args method should be taken out and shot19:43
hannoschwell, the idea was always to drop DateTime as a whole19:44
hannoschit just never happens19:44
mgedminwaaah why is buildout not like virtualenv?19:45
mgedminno bin/python, no bin/easy_install19:45
hannoschbin/easy_install would be against its design principle19:45
mgedminwhich is bogus19:45
mgedminI want my ad-hoc development tools without modifying the whole buildout19:46
hannoschthen you wrap the buildout in a virtualenv19:46
mgedminI could agree that things installed with bin/easy_install should *not* be available for import inside scripts produced by bin/buildout19:46
LeoRochaelmgedmin: bin/python is easy enough, just use the "interpreter" param to the zc.recipe.egg part19:46
mgedminincidentally, bug?  versions = versions inside [test] sections, where it is ignored19:47
mgedminanybody with a pristine checkout of DateTime want to fix that?19:47
mgedminLeoRochael, easy, but painful -- edit file, re-run buildout, wait, remember to revert, etc.19:47
hannoschmgedmin: fixed in svn19:47
mgedminLeoRochael, not to mention packages like DateTime which don't use zc.recipe.egg19:48
seppo14i guess this has been asked before - but is there a way to parse XML data inside a web-python-script inside ZOPE2 without use of external methods?19:49
LeoRochaelmgedmin: use your own .cfg that extends buildout.cfg19:49
mgedminoh thank you for that ingenious suggestion19:49
* mgedmin just irritated because of all this yak shaving, ignore his whines please19:50
hannoschmgedmin: there's a bin/python now if you svn up19:50
hannoschand run buildout of course19:50
TresEquisseppo14: not really19:55
seppo14TresEquis: thanks. damn.19:55
mgedmin_calcIndependentSecondEtc('US/Eastern', 3197779200 0.0) produces different results19:55
mgedminincidentally, buildout rules19:55
mgedminbin/py1 uses pytz2010b, bin/py2 uses pytz2010h19:56
LeoRochaelmgedmin: that's a quick change of mind...19:56
TresEquisif you have a toehold on the filesystem, you can make the 'xml.dom.minidom' module importable in PythonScripts19:56
mgedminno need to rerun buildout to switch versions, or have two directory trees19:56
hannoschmgedmin: see, it's not all bad :)19:56
mgedmin_tzoffset(tz, 0.0) returns different values19:57
seppo14TresEquis: how that?19:57
*** TomBlockley has quit IRC19:57
LeoRochaelmgedmin: that's nice of course, though I'd still like to be able to define 'versions =' on separate zc.recipe.egg parts, just like I can define 'python ='19:57
seppo14TresEquis: and i'm aware of that beeing a security "override"… but in my case this would be the easiest solution to my problem19:57
LeoRochaeladded a wishlist bug against zc.buildout, for that, but it's being duly ignored...19:57
* mgedmin used eggs = DateTime {newline} pytz == 2010x19:57
TresEquisseppo14: see
*** evilbungle has quit IRC19:58
mgedminTresEquis, what's happening is that the bare "try:/except:" trips over and catches whatever error that's happening in info()19:59
TresEquismgedmin: hmm?19:59
TresEquisin DateTIme?19:59
*** hannosch has quit IRC20:00
mgedminwith new pytz version, DateTime.DateTime._tzoffset('timezone name', 0) always returns 0 because of some exception that's silently swallowed20:00
mgedminI can't see that exception easily, DateTime *loves* nested bare except's20:00
mgedminValueError: Not naive datetime (tzinfo is already set)20:01
mgedminin   File "/home/mg/src/DateTime/src/DateTime/", line 224, in info20:01
TresEquisFixing DateTime to have less stupid exception handling would work20:01
mgedminwhere it is calling     offset = normalized_tzinfo.utcoffset(dt)20:01
mgedminfull traceback:
mgedminoh, right, I don't have tags for pytz :/20:03
*** TresEquis has quit IRC20:07
mgedminTresEquis, is a patch that makes the test suite pass20:07
mgedminwhy'd you have to run away?20:07
*** Kabz|4D has quit IRC20:10
mgedminpatch attached to
mgedminsince I have no personal investment in this issue, I'll leave it here20:11
LeoRochaelmgedmin: looks short enough..20:13
*** mgedmin has quit IRC20:16
*** emrojo has quit IRC20:18
Charlie_Xseppo14: you can use the allow_module and allow_class statements to allow access to things from PythonScripts. This can work well with simple modules and classes but quickly gets difficult to work with.20:23
seppo14Charlie_X:  thanks, I'll look into that20:23
LeoRochaelanyone knows if mr.developer can work with bzr branches?20:24
* Charlie_X senses déjà vu20:24
LeoRochaelCharlie_X: lemme guess, people want it, but nobody has bothered yet20:25
Charlie_XNo, I saw the question earlier. I have no idea whether it does or not. I think Sidnei and Gary mentioned it on the ML a bit.20:25
*** maiky has joined #zope20:26
*** sim_sim has quit IRC20:26
sidneiLeoRochael, Tres has a patch for it20:26
*** Hypergraphe has joined #zope20:27
LeoRochaelgreat, he's just disconnected20:27
*** vigith has joined #zope20:28
sidneiLeoRochael, one sec20:29
*** davisagli_ has joined #zope20:30
*** RaFromBRC has joined #zope20:30
LeoRochaelsidnei: thx20:30
sidneiexciting times. a patch to add support for bzr hosted in git. :)20:30
*** davisagli has quit IRC20:30
*** davisagli_ is now known as davisagli20:30
* Charlie_X gives sidnei a safety helmet and protective gloves20:32
* Charlie_X starts throwing things as sidnei20:32
*** pepeu has quit IRC20:38
*** RaFromBRC has quit IRC20:39
*** sunew has quit IRC20:39
*** RaFromBRC has joined #zope20:39
*** mgedmin has joined #zope20:40
*** igbun has quit IRC20:44
*** eperez has quit IRC20:45
*** KurtB has joined #zope20:50
*** mcdonc has quit IRC20:51
*** dayne has quit IRC20:58
*** dayne has joined #zope21:00
*** igbun has joined #zope21:06
*** TKtiddle has joined #zope21:23
*** Charlie_X has quit IRC21:25
*** vigith has quit IRC21:29
*** agroszer has quit IRC21:46
*** dayne has quit IRC21:48
*** mcdonc has joined #zope21:53
*** igbun has quit IRC21:55
*** seppo14 has quit IRC21:58
*** mcdonc_ has joined #zope21:59
*** mcdonc has quit IRC21:59
*** alvaro is now known as Guest640522:13
*** fdrake has left #zope22:15
*** JaRoel|4D has joined #zope22:16
*** teix has quit IRC22:19
*** nebo has joined #zope22:25
*** dayne has joined #zope22:28
*** daMaestro has quit IRC22:47
*** maiky has quit IRC22:56
*** daMaestro has joined #zope23:00
*** mcdonc_ has quit IRC23:07
*** mcdonc_ has joined #zope23:07
*** giampaolo has quit IRC23:15
*** nebo has quit IRC23:16
*** mcdonc__ has joined #zope23:25
*** mcdonc_ has quit IRC23:26
*** TKtiddle has quit IRC23:27
*** mcdonc__ has quit IRC23:27
*** mcdonc__ has joined #zope23:28
*** mcdonc__ has quit IRC23:28
*** zagy has quit IRC23:32
*** zagy has joined #zope23:33
*** tarek has joined #zope23:34
*** zagy has quit IRC23:37
*** shen-long has joined #zope23:41
*** grahal_ has quit IRC23:47
*** TresEquis has joined #zope23:52
*** lisppaste6 has quit IRC23:55
*** lisppaste6 has joined #zope23:55

Generated by 2.15.1 by Marius Gedminas - find it at!