| *** povbot has joined #zope3-dev | 14:15 | |
| *** J1m has quit IRC | 14:40 | |
| *** ignas__ has joined #zope3-dev | 14:48 | |
| bob2 | hm, where can I find how to use fssync? | 14:54 |
|---|---|---|
| *** ignas has quit IRC | 14:59 | |
| *** philiKON has joined #zope3-dev | 15:06 | |
| * andrew_m is away: I'm busy | 15:07 | |
| bob2 | hm, was it you, philiKON who pointed me at fssync? | 15:13 |
| philiKON | dunno | 15:21 |
| *** tarek_ has joined #zope3-dev | 15:22 | |
| *** robrecord has joined #zope3-dev | 15:35 | |
| *** tarek_ has quit IRC | 15:35 | |
| *** tarek_ has joined #zope3-dev | 15:35 | |
| *** tarek__ has joined #zope3-dev | 15:35 | |
| *** tarek_ has quit IRC | 15:35 | |
| *** tarek_ has joined #zope3-dev | 15:35 | |
| mgedmin | probably me | 15:43 |
| mgedmin | it's in zope.app.fssync | 15:43 |
| mgedmin | and also in zope.fssync | 15:44 |
| mgedmin | I've never used it | 15:44 |
| mgedmin | and I think it doesn't work now | 15:44 |
| mgedmin | but it was a great idea nevertheless | 15:44 |
| bob2 | haha | 15:45 |
| robrecord | i'm so excited about zope 3. | 15:49 |
| robrecord | i can hardly contain myself | 15:49 |
| robrecord | excuse me | 15:49 |
| SteveA | you can't contain yourself. | 15:59 |
| SteveA | there's an event that guarentees that invariant | 15:59 |
| SteveA | you'd be circular otherwise | 16:00 |
| *** tarek_ is now known as tarek | 16:07 | |
| *** tarek has joined #zope3-dev | 16:07 | |
| *** SureshZ has joined #zope3-dev | 16:20 | |
| faassen | is it possible to have local security policies? | 16:25 |
| faassen | i.e. a site specific security policy possible? | 16:27 |
| faassen | def setSecurityPolicy(aSecurityPolicy): | 16:29 |
| faassen | """Set the system default security policy, and return the previous | 16:29 |
| faassen | value. | 16:29 |
| faassen | This method should only be called by system startup code. | 16:29 |
| faassen | It should never, for example, be called during a web request. | 16:29 |
| faassen | """ | 16:29 |
| faassen | does this mean that it's not possible to have a local security policy? | 16:29 |
| mgedmin | you can have a global security policy that knows how to delegate to local ones | 16:30 |
| * mgedmin is trying to quote J1m from memory | 16:31 | |
| faassen | yeah, but that'd be hard to install. | 16:31 |
| faassen | as we'd have to ask people to edit their zcml. | 16:31 |
| faassen | hm.. | 16:31 |
| faassen | maybe I'm wrong. | 16:31 |
| faassen | no, I'm not wrong. | 16:32 |
| faassen | it's unfortunate that people have to edit their zcml to install this smarter security policy. | 16:32 |
| mgedmin | zope 3 is not ready as an application server that can host multiple unrelated widely different applications, imho | 16:32 |
| faassen | right. | 16:33 |
| faassen | that's our feeling too. | 16:33 |
| faassen | so the next question is..how do you do this schoolbell trickery that makes it the zope root? | 16:33 |
| mgedmin | faassen, http://source.schooltool.org/viewcvs/trunk/schoolbell/src/schoolbell/app/main.py?rev=4082&view=markup | 16:35 |
| SteveA | i think it is... provided you rip out the default publication | 16:35 |
| SteveA | and add an api to get at the zodb root from elsewhere, as a thread local | 16:35 |
| faassen | SteveA: you mean you think it's ready? not security policy-wise, right? as that's a global variable. | 16:36 |
| SteveA | that smartly starts zodb connections only when needed | 16:36 |
| mgedmin | waah | 16:36 |
| SteveA | and replace the security policy with a more general one | 16:36 |
| faassen | SteveA: right. | 16:36 |
| mgedmin | /@@view.html with <form action="."> submits the form to / | 16:36 |
| SteveA | i talked with tres about security policies at EP | 16:36 |
| mgedmin | /view.html with <form action="."> submits the form to /view.html | 16:36 |
| SteveA | use form action="" | 16:36 |
| faassen | SteveA: I imagine something like that might be more sane, as then at least sites are more fundamentally independent. | 16:36 |
| SteveA | yes | 16:36 |
| SteveA | i am totally infavour of doing all of this | 16:37 |
| mgedmin | no, I'm mistaken | 16:37 |
| ignas__ | my bad :/ | 16:37 |
| SteveA | sadly i have no time to do it | 16:37 |
| SteveA | ignas__: i think you mean "I'm bad." ;-) | 16:37 |
| *** jinty has joined #zope3-dev | 16:37 | |
| faassen | SteveA: revolution is near!! | 16:37 |
| * faassen looks around to see whether he sees Jim. | 16:37 | |
| SteveA | faassen the apostate | 16:38 |
| faassen | SteveA: darn, you have no time for revolutions either? how are we going to revolution then? | 16:38 |
| philiKON | faassen, i gues syou have to do your own little mini revolution | 16:38 |
| faassen | philiKON: hey, I'm actually sneaking in all kinds of mini revolutions, didn't you notice? :) | 16:38 |
| philiKON | faassen, btw, i released five 1.0.2 | 16:38 |
| SteveA | faassen: i'll have to take a revolution sabbatical | 16:39 |
| faassen | philiKON: I noticed, cool. JW looked at the change log, I haven't had the time yet. | 16:39 |
| ignas__ | SteveA, are you sure that an empty string a valid URI accorting to the spec ? | 16:39 |
| *** ignas__ is now known as ignas | 16:39 | |
| philiKON | faassen, five 1.1 isn't far away as far as i'm concerned. when are you going to write the adding tests? ;) | 16:39 |
| faassen | SteveA: yeah, that'd be cool. And then you ask Canonical to fund me too and I'll revolution along with you. :) | 16:39 |
| SteveA | i think "sabatical" means "i don't get paid for it" | 16:39 |
| faassen | philiKON: I really don't have time the coming 2 weeks. | 16:39 |
| philiKON | oh well | 16:40 |
| philiKON | there can always be a 1.1.1 :) | 16:40 |
| faassen | SteveA: shoot, can't we motivate this as 'fundamental infrastructural thingymabobby which will really help the bla-adapter-interaction-pattern-factory thingy? | 16:40 |
| faassen | mgedmin: you don't override the security policy in schoolbell, right? | 16:43 |
| mgedmin | no | 16:44 |
| faassen | I'm worrying about its interaction with, say, functional testing. | 16:45 |
| robrecord | SteveA - your observations are acute - the event could be something to do with zope 3 | 16:49 |
| *** bradb-away is now known as bradb | 16:51 | |
| jinty | hoi, I have some debian related questions about zope3: | 16:55 |
| jinty | 1. what ususally happens with outside libs that are imported into the subversion tree? | 16:56 |
| jinty | for example pytz, because there is a pytz package in debian, and the ideal would be for the zope package just to depend on that. | 16:56 |
| jinty | but we don't want to break stuff | 16:57 |
| *** mkerrin has joined #zope3-dev | 16:59 | |
| bob2 | so, say I want to provide a configuration file for an app | 17:14 |
| bob2 | hrm, I think srichter's book covers this, actually | 17:15 |
| *** hdima has quit IRC | 17:20 | |
| *** timte has quit IRC | 17:25 | |
| *** sashav has quit IRC | 17:36 | |
| *** robrecord has quit IRC | 18:00 | |
| *** stub has quit IRC | 18:12 | |
| *** elbixio has joined #zope3-dev | 18:16 | |
| elbixio | Hi everybody. I've never used the ZODB, and I was wondering: Is it possible to modify the ZODB of an instance from an external Python application while the instance is running? | 18:20 |
| mgedmin | yes and no | 18:22 |
| mgedmin | ZEO is one way to do it | 18:22 |
| elbixio | :O | 18:23 |
| elbixio | I have a Python app that runs periodically using cron. It extracts data from a website and store it in text files | 18:24 |
| elbixio | I was thinking of using Zope3 to publish the extracted data | 18:24 |
| elbixio | That's why I wanted to store the data in the ZODB, because I thought it would be easier to access the data from Zope | 18:25 |
| philiKON | z3 doesn't need data to be in the zodb anymore | 18:27 |
| philiKON | but you can use zeo to access zodb data from multiple python sessions (e.g. a zope instance and a python app) | 18:28 |
| elbixio | :O | 18:28 |
| *** j-w has quit IRC | 18:29 | |
| tim2p | hi, are there any examples of using the zope 2 -> 3 interface bridge code? | 18:38 |
| tim2p | i found this <http://cvs.zope.org/CMF/CMFDefault/bridge.zcml?rev=HEAD&content-type=text/vnd.viewcvs-markup> | 18:41 |
| tim2p | but it's not clear to me where i can import those new zope 3 interfaces from | 18:41 |
| *** elbixio has quit IRC | 18:43 | |
| *** elbixio has joined #zope3-dev | 18:44 | |
| *** mp has quit IRC | 18:47 | |
| VladDrac | tim2p: zope2.8/five/zope3? | 19:08 |
| tim2p | VladDrac: i'm on 2.8 and trying to bridge a CMF interface | 19:10 |
| philiKON | tim2p, you can import the z3 style interfaces from the module you specify | 19:13 |
| philiKON | <bridge zope2="..." package="my.package" name="IZ3Interface" /> | 19:13 |
| philiKON | from my.package import IZ3Interface | 19:14 |
| tim2p | philiKON: oh, that does make sense now I see it :-) | 19:14 |
| philiKON | btw, the future way will be the other way around | 19:14 |
| philiKON | you write z3 interfaces | 19:14 |
| philiKON | and bridge them to z2 ones if you still need z2 ones around | 19:15 |
| philiKON | zope 2.9 will have this feature | 19:15 |
| philiKON | maybe five 1.1 too | 19:15 |
| tim2p | philiKON: does that mean that all zope core and cmf interfaces will have their source adjusted to be proper zope.interface.Interface derived? | 19:17 |
| philiKON | zope 2.9 core interfaces will be z3 style | 19:17 |
| philiKON | CMF 1.6 or whatever it's called too, i guess | 19:17 |
| philiKON | i'm not a cmf developer | 19:17 |
| tim2p | philiKON: sure. thanks | 19:18 |
| philiKON | but it just doesn't make sense anymore to write z2 interfaces if all you do is bridge them to z3 anyway | 19:18 |
| d2m | philiKON: just posted: http://www.zope.org/Products/CMF/docs/requirements/CMF20Roadmap | 19:18 |
| philiKON | weeeeeeeew | 19:18 |
| tim2p | i agree, although I'd be worried if i weren't able to bridge legacy interfaces to the new sort (say from third party products) | 19:18 |
| philiKON | right, you'll still be able to bridge old stuff | 19:19 |
| tim2p | cool, then i'm happy :-) | 19:20 |
| *** SureshZ has quit IRC | 19:22 | |
| *** gintas has quit IRC | 19:47 | |
| *** jinty has quit IRC | 20:05 | |
| *** bradb is now known as bradb-lunch | 20:19 | |
| philiKON | what's wrong with zope-cvs@zope.org? anyone know? | 20:34 |
| d2m | dont know, should forward to webmaster@zope.org | 20:36 |
| d2m | philiKON: sorry, wrong answer | 20:38 |
| *** elbixio has quit IRC | 20:43 | |
| philiKON | d2m? | 20:48 |
| d2m | there was a support mail address for cvs and svn webrepository which is now linked to webmaster@zope.org | 20:49 |
| *** faassen has quit IRC | 20:50 | |
| *** RaFromBRC has joined #zope3-dev | 20:55 | |
| *** tim2p has quit IRC | 20:57 | |
| *** bradb-lunch is now known as bradb | 21:10 | |
| *** zagy has quit IRC | 21:14 | |
| *** projekt01 has joined #zope3-dev | 21:15 | |
| *** SureshZ has joined #zope3-dev | 21:18 | |
| *** jinty has joined #zope3-dev | 21:23 | |
| *** Pupeno- is now known as Pupeno | 21:23 | |
| *** regebro has quit IRC | 21:25 | |
| *** mkerrin has quit IRC | 21:25 | |
| *** J1m has joined #zope3-dev | 21:51 | |
| *** ignas has quit IRC | 22:26 | |
| philiKON | J1m, i'm experiencing a very weird bug regarding the testing of security policies | 22:34 |
| philiKON | got a minute? | 22:34 |
| J1m | k | 22:34 |
| philiKON | so, placelesssetup calls newInteraction | 22:35 |
| philiKON | so, before I call placelsssetup.setUp(), I set my own security policy | 22:35 |
| J1m | k | 22:35 |
| philiKON | >>> from zope.security.management import setSecurityPolicy | 22:35 |
| philiKON | >>> from Products.Five.security import FiveSecurityPolicy | 22:35 |
| philiKON | >>> oldpolicy = setSecurityPolicy(FiveSecurityPolicy) | 22:35 |
| philiKON | getSecurityPolicy() returns FiveSecurityPolicy after those lines, as it should | 22:35 |
| philiKON | now I do placelesssetup.setUp() | 22:36 |
| J1m | I bet placelesssetup resets the security policy | 22:36 |
| J1m | I'm sure it does | 22:36 |
| philiKON | if it does so, it doesn't do it through setSecurityPolicy | 22:36 |
| philiKON | i couldn't find where it would | 22:37 |
| J1m | See _clear in zope.security.management | 22:37 |
| J1m | after calling setUp, call | 22:38 |
| J1m | endInteraction | 22:38 |
| J1m | setSecurityPolicy | 22:38 |
| J1m | newInteraction | 22:38 |
| philiKON | right, i've got that working | 22:38 |
| philiKON | it just feels weird | 22:38 |
| philiKON | i see z.s.management._clear | 22:38 |
| philiKON | but it's a cleanup | 22:38 |
| J1m | I don't know why | 22:38 |
| philiKON | right? | 22:38 |
| J1m | right | 22:38 |
| philiKON | so, why would placelesssetup.setUp() call _clear? | 22:39 |
| J1m | to make sure it has a known starting point | 22:39 |
| philiKON | hrm | 22:39 |
| philiKON | ok | 22:39 |
| philiKON | thanks | 22:39 |
| *** timte has joined #zope3-dev | 23:32 | |
| projekt01 | Pupeno, ayt? | 23:36 |
| *** replicant is now known as hazmat | 23:44 | |
| *** ChanServ sets mode: +o hazmat | 23:44 | |
| *** mgedmin_ has joined #zope3-dev | 23:49 | |
| *** mgedmin has quit IRC | 23:49 | |
| strichter | J1m: btw, I am back :-) | 23:57 |
| *** strichter is now known as srichter | 23:57 | |
| *** ChanServ sets mode: +o srichter | 23:57 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!