| *** srichter has quit IRC | 00:35 | |
| *** batlogg_ has quit IRC | 00:38 | |
| *** tarek has joined #zope3-dev | 00:38 | |
| *** gumpa-afk has left #zope3-dev | 00:46 | |
| *** zbir has quit IRC | 00:47 | |
| *** flox has quit IRC | 00:55 | |
| *** rocky has joined #zope3-dev | 01:17 | |
| * mgedmin discovers that test.py allows multiple -s parameters and is happy | 01:22 | |
| *** tarek has quit IRC | 01:32 | |
| *** tarek has joined #zope3-dev | 01:33 | |
| *** pcardune has joined #zope3-dev | 01:36 | |
| *** mgedmin has quit IRC | 01:42 | |
| *** _sussi has left #zope3-dev | 01:49 | |
| *** yota has quit IRC | 01:53 | |
| *** nathany has joined #zope3-dev | 01:56 | |
| *** MJ has quit IRC | 01:56 | |
| graham | I have an object with an attribute which is a list of subobjects | 02:05 |
|---|---|---|
| graham | I can render those subobjects in a page template | 02:05 |
| graham | But now I need to write a python view which modifies them before they go to the page template | 02:06 |
| graham | I'm getting a ForbiddenAttribute when I try to access them | 02:06 |
| graham | Do I need to do something different, or have I just made a mistake? | 02:07 |
| *** roym has quit IRC | 02:10 | |
| graham | I think I've just made a mistake.... | 02:12 |
| *** tarek has quit IRC | 02:16 | |
| *** jhauser has quit IRC | 02:16 | |
| *** tarek has joined #zope3-dev | 02:19 | |
| *** J1m has left #zope3-dev | 02:30 | |
| *** rocky has quit IRC | 02:37 | |
| *** zbir has joined #zope3-dev | 02:45 | |
| *** pcardune has quit IRC | 02:47 | |
| *** nathany has quit IRC | 04:02 | |
| *** J1m has joined #zope3-dev | 04:04 | |
| *** J1m has quit IRC | 04:05 | |
| *** pcardune has joined #zope3-dev | 04:11 | |
| *** benji has quit IRC | 04:13 | |
| *** zbir has quit IRC | 04:18 | |
| *** oferw has joined #zope3-dev | 04:40 | |
| *** TrevorP has quit IRC | 05:18 | |
| *** oferw has quit IRC | 05:22 | |
| *** TrevorP has joined #zope3-dev | 05:23 | |
| *** TrevorP has quit IRC | 05:37 | |
| *** TrevorP has joined #zope3-dev | 05:38 | |
| *** pcardune_ has joined #zope3-dev | 06:34 | |
| *** alecm has quit IRC | 06:38 | |
| *** stub has joined #zope3-dev | 06:41 | |
| *** elbixio has joined #zope3-dev | 06:42 | |
| *** baijum has joined #zope3-dev | 06:42 | |
| *** TrevorP has quit IRC | 06:49 | |
| *** pcardune has quit IRC | 06:49 | |
| *** TrevorP has joined #zope3-dev | 06:58 | |
| *** d2m has quit IRC | 07:13 | |
| *** d2m has joined #zope3-dev | 07:30 | |
| *** tarek has quit IRC | 08:02 | |
| *** dobee has joined #zope3-dev | 08:02 | |
| *** eins has joined #zope3-dev | 08:03 | |
| *** Theuni has joined #zope3-dev | 08:11 | |
| *** dunny has quit IRC | 08:33 | |
| *** ChrisW has joined #zope3-dev | 08:38 | |
| *** dunny has joined #zope3-dev | 08:51 | |
| *** srichter has joined #zope3-dev | 08:54 | |
| *** ChrisW has quit IRC | 08:57 | |
| *** MJ has joined #zope3-dev | 09:01 | |
| *** elbixio has quit IRC | 09:01 | |
| *** romanofski has joined #zope3-dev | 09:04 | |
| romanofski | moin :) | 09:06 |
| *** hdima has joined #zope3-dev | 09:07 | |
| *** flox_ has joined #zope3-dev | 09:08 | |
| *** flox_ is now known as flox | 09:09 | |
| *** dunny has quit IRC | 09:10 | |
| *** batlogg has joined #zope3-dev | 09:15 | |
| *** flox has quit IRC | 09:28 | |
| *** TrevorP has quit IRC | 09:31 | |
| *** jhauser has joined #zope3-dev | 09:31 | |
| *** TrevorP has joined #zope3-dev | 09:37 | |
| *** yota has joined #zope3-dev | 10:01 | |
| *** oferw has joined #zope3-dev | 10:19 | |
| *** romanofski has quit IRC | 10:21 | |
| *** oferw has quit IRC | 10:22 | |
| *** pcardune_ has quit IRC | 10:27 | |
| *** oferw has joined #zope3-dev | 10:28 | |
| *** MJ has quit IRC | 10:31 | |
| *** TrevorP has quit IRC | 10:35 | |
| *** TrevorP has joined #zope3-dev | 10:36 | |
| *** flox_ has joined #zope3-dev | 10:41 | |
| *** flox_ is now known as flox | 10:41 | |
| *** dobee has quit IRC | 10:46 | |
| *** dobee has joined #zope3-dev | 10:49 | |
| *** dobee has joined #zope3-dev | 10:51 | |
| *** dunny has joined #zope3-dev | 10:54 | |
| *** MJ has joined #zope3-dev | 11:10 | |
| graham | How do I render an attribute in a pt replacing all the newlines with <br> tags? | 11:25 |
| graham | I'm pretty sure I need to do the processing in the view class | 11:25 |
| philiKON | in an attribute?!? | 11:25 |
| philiKON | yah, 'tis best done in python | 11:26 |
| graham | Is there a function for it? | 11:26 |
| philiKON | text.replace('\n', '<br>') :) | 11:26 |
| graham | Okay, is there then a function to escape everything else? | 11:27 |
| philiKON | hmm. cgi.escape? | 11:27 |
| graham | Okay. I wasn't sure whether that did enough. Thanks | 11:28 |
| SteveA | in Launchpad, we wrote a something/fmt:nl_to_br for this | 11:42 |
| *** volvox has joined #zope3-dev | 12:01 | |
| graham | Ah, that's quite nice. Thinking back, I remember you talking about extending TALES at EP2005! | 12:02 |
| *** harobed has joined #zope3-dev | 12:34 | |
| *** mkerrin has joined #zope3-dev | 12:44 | |
| *** dunny has quit IRC | 13:00 | |
| *** jinty has joined #zope3-dev | 13:04 | |
| *** ignas has joined #zope3-dev | 13:06 | |
| *** mexiKON has joined #zope3-dev | 13:16 | |
| *** philiKON has quit IRC | 13:26 | |
| *** eins has quit IRC | 14:12 | |
| *** eins has joined #zope3-dev | 14:19 | |
| *** baijum has quit IRC | 14:26 | |
| *** alga has joined #zope3-dev | 14:28 | |
| *** nathany has joined #zope3-dev | 14:29 | |
| *** jinty has quit IRC | 14:39 | |
| *** jhauser_ has joined #zope3-dev | 14:40 | |
| *** jhauser has quit IRC | 14:43 | |
| *** rock1 has joined #zope3-dev | 14:43 | |
| *** rock1 is now known as rocky | 14:44 | |
| *** mgedmin has joined #zope3-dev | 14:52 | |
| *** zbir has joined #zope3-dev | 14:57 | |
| *** rock1 has joined #zope3-dev | 14:57 | |
| *** rocky has quit IRC | 14:59 | |
| *** rock1 is now known as rocky | 14:59 | |
| *** alga has quit IRC | 15:09 | |
| *** zbir has quit IRC | 15:17 | |
| *** TrevorP has quit IRC | 15:24 | |
| *** TrevorP has joined #zope3-dev | 15:24 | |
| *** zbir has joined #zope3-dev | 15:32 | |
| *** oferw has quit IRC | 15:36 | |
| harobed | hi, where is http://www.zope.org/Wikis/DevSite/Projects/ComponentArchitecture/FrontPage/Zope3Book source code in SVN ? | 15:48 |
| harobed | isn't exists ? | 15:48 |
| *** zbir has quit IRC | 16:02 | |
| *** pcardune_ has joined #zope3-dev | 16:03 | |
| *** eins has quit IRC | 16:08 | |
| *** zbir has joined #zope3-dev | 16:15 | |
| volvox | harobed: I think the examples are in svn://svn.zope.org/repos/main/book/trunk | 16:16 |
| *** flox has quit IRC | 16:17 | |
| *** flox has joined #zope3-dev | 16:18 | |
| *** pcardune_ is now known as pcardune | 16:22 | |
| *** edar has joined #zope3-dev | 16:23 | |
| edar | Hi! | 16:23 |
| edar | Does anyone here know who wrote the untrusted python library? | 16:24 |
| edar | or is anyone here very familiar with it? | 16:24 |
| *** baldtrol has joined #zope3-dev | 16:26 | |
| baldtrol | when an @action is executed in a form.Form, does the view get re __init__'d once the post action is handled? | 16:31 |
| SteveA | yes | 16:35 |
| SteveA | it is a new web request | 16:35 |
| SteveA | entirely independent | 16:35 |
| baldtrol | ok, i thought so | 16:35 |
| baldtrol | it seemed like it should, but i'm troubleshooting a bit of behavior i don't understand, and wanted to rule that out | 16:35 |
| baldtrol | thanks :) | 16:36 |
| *** J1m has joined #zope3-dev | 16:44 | |
| *** RaFromBRC has joined #zope3-dev | 16:50 | |
| edar | Hi Jim :) | 16:57 |
| edar | J1m: are you there? | 16:58 |
| *** hdima has quit IRC | 16:59 | |
| edar | I need to know who wrote the untrusted python interpreter for zope3 | 16:59 |
| J1m | Yes, I did. | 17:02 |
| *** jinty has joined #zope3-dev | 17:08 | |
| edar | Cool! | 17:08 |
| edar | I have some questions :) | 17:08 |
| edar | I'm trying to write a doctest quiz product, where I create a doctestquiz object and people can submit their solutions | 17:09 |
| edar | The doctestquiz then, should test the solutions, however, I don't trust the solutions so I'm trying to use doctests and untrustedpython together | 17:10 |
| SteveA | edar: there's no good solution for someone putting an infinite loop in the doctest | 17:11 |
| J1m | right | 17:12 |
| pcardune | you could put time constraints on how long the code should take to run | 17:12 |
| pcardune | although i suppose that still isn't a *good* solution | 17:12 |
| J1m | You never want to accept code from people you don't trust, at least to some degree. | 17:12 |
| edar | Well, let's say those people are students, and I do trust them to some extent | 17:13 |
| SteveA | if you run the code in a separate server instance... I mean like using xen or vmware | 17:13 |
| J1m | and assume you are willing to have your server seize up from time to time... | 17:13 |
| SteveA | then you cna constrain the amount of time it can run for, and be somewhat assured it won't do bad things to the rest of your environment | 17:13 |
| SteveA | on some OSes you could use process limits / quotas / ulimit etc. to restrict the resources... but you open yourself to local exploits there a lot. I wouldn't do it. | 17:14 |
| edar | hmmm however the tests are browser based, as in you click something to make the test happen | 17:14 |
| edar | if someone makes and infinite loop and clicks test ... the page will never refresh, or it shouldn't | 17:14 |
| edar | so the user will close the browser eventually, which will terminate the process | 17:14 |
| SteveA | I'd look more at having a way to run the code on the client side | 17:15 |
| SteveA | not on the server side | 17:15 |
| J1m | closing the browser will notterminate the process. | 17:15 |
| edar | I thought it would ... uh oh | 17:15 |
| J1m | They could also enter expressions that use vast amounts of memory. puttng your server into swap death. | 17:15 |
| J1m | That | 17:16 |
| J1m | That's a good idea -- running in the client | 17:16 |
| SteveA | you can use ulimits to get limit the memory usage on your server process | 17:16 |
| SteveA | still, there are all sorts of issues running this on the server | 17:16 |
| edar | However running it on the client means the user can do whatever he wants with the doctest | 17:17 |
| edar | thus he can make him/her self pass all the time | 17:17 |
| SteveA | you said you trusted them! | 17:17 |
| SteveA | if you don't trust the students to mark themselves properly, why would you trust them not to crack your server? | 17:18 |
| edar | to some extent ^^ | 17:18 |
| edar | Hold on .... | 17:19 |
| edar | I need to get the instigator here, Jeff Elkner | 17:19 |
| SteveA | on an unrelated issue, someone (who was it...?) was talking about using distributed revision control systems to combat plagarism | 17:20 |
| SteveA | if you get your students to do their work using fine-grained checkins to a VCS, then you can inspect the flow of their work later | 17:20 |
| pcardune | that's an interesting concept | 17:21 |
| SteveA | and see that there is a process of creation going on, rather than things just appearing from nowhere | 17:21 |
| pcardune | as a student I would hate it, it's like a teacher breathing down your neck 24/7 and they will always know when you did everything at the last minute | 17:22 |
| flox | hello. i try to learn some good features of Zope3 for "presentation" components. | 17:23 |
| flox | what about 'zope.formlib'? | 17:23 |
| flox | is it a good start for 'input templates'? | 17:24 |
| flox | and 'zope.viewlet' for display templates? | 17:24 |
| *** jelkner has joined #zope3-dev | 17:25 | |
| edar | there is the real culprit >_< | 17:25 |
| baldtrol | flox: formlib is the place to go for designing any kind of form. | 17:26 |
| jelkner | j1m: hi jim | 17:26 |
| J1m | Hi | 17:26 |
| baldtrol | viewlet has a bit of a different use case. it's for reusable views, not core to the representation of the current context, but desirable to be displayed there. | 17:26 |
| jelkner | seems like i've started a controversy | 17:27 |
| jelkner | i want doctest quiz | 17:27 |
| jelkner | they tell me you are trying to talk them out of it | 17:27 |
| flox | baldtrol: for 'input templates' we should use 'zope.formlib' (except for content components where we use ZCML <browser:editform>) | 17:31 |
| baldtrol | flox: that's my opinion, yessir. form.Form, form.EditForm and page.Page (not formlib proper, but related) are, in my opinion, the best way to build presentation logic | 17:32 |
| Theuni | is there any elegant way to register a browser:page for multiple interfaces (but not *) without repeating the whole zcml-statement for every interface? | 17:33 |
| baldtrol | theuni: i don't know, but i hope someone says *yes* ;) | 17:33 |
| Theuni | hmm. i can of course make a marker interface that i use as an interface that all the interfaces i want to associate extend | 17:33 |
| Theuni | i can also make english sentences that are more complicated than they need to be | 17:33 |
| baldtrol | you should rephrase everything to have yoda-syntax | 17:34 |
| *** benji has joined #zope3-dev | 17:34 | |
| Theuni | rephrasing everything you must | 17:35 |
| baldtrol | ye-eaa-es. | 17:36 |
| Theuni | yoda: very dark the other side is | 17:36 |
| Theuni | obi wan: shut up. eat your toast. | 17:36 |
| baldtrol | hah! | 17:37 |
| baldtrol | nice | 17:37 |
| flox | baldtrol: and 'zope.viewlet' is to display some re-usable views inside defined regions of the browser. | 17:38 |
| baldtrol | flox: that's what i've used it for in the past. you can even embed formlib forms in viewlets, so long as you cleverly swap out the context for actions first. | 17:38 |
| baldtrol | credit to srichter for giving me tips on how to do that too ;) | 17:39 |
| *** jelkner has quit IRC | 17:39 | |
| *** RaFromBRC is now known as RaFromBRC|brb | 17:39 | |
| Theuni | hmm. J1m, btw: do you remember why "ISomething + Something" was choosen over "Something + SomethingImpl" as the standard for naming interfaces and classes? | 17:40 |
| flox | ok, i'll concentrate on these: zope.formlib and zope.publisher (BrowserPage) | 17:41 |
| flox | then i will look at zope.viewlet | 17:41 |
| flox | thks, baldtrol | 17:41 |
| ignas | Theuni: err i'd guess because SomethingImpl is more verbose and kind of ugly? and there is a usecase of "I have this lot of classes and want to make them Web enabled using Zope3" | 17:42 |
| Theuni | i see the second reasoning | 17:42 |
| J1m | jelkner, I'm not trying very hard. I just said they shouldn't do it. Actually, StevaA said they shouldn't do it and I agree. | 17:42 |
| Theuni | a friend of mine who happens to know java was a bit surprised by our naming scheme | 17:42 |
| *** flox is now known as flox|away | 17:42 | |
| mexiKON | Theuni, the question is whether you're interface centric or not | 17:43 |
| Theuni | because the interface is supposed to specify a "something", not a "isomething" whereas the implementation is just a "detail" (as far as i understood that reasoning) | 17:43 |
| mexiKON | Theuni, also, lots of other systems tend to use the ISomething form for interfaces | 17:43 |
| mexiKON | Theuni, right. | 17:43 |
| J1m | Theuni, I never considered the second option, but if I did, I would reject it because I thin kit's ugly. | 17:43 |
| Theuni | ic | 17:43 |
| J1m | IFoo is pretty standard in some circles, and it stand out pretty well. | 17:43 |
| mexiKON | yeah, Impl is ugly | 17:43 |
| mexiKON | afaik, .NET uses IFoo | 17:44 |
| Theuni | mexiKON: for which flavor does the "whether you're interface centric" argument stand? | 17:44 |
| mexiKON | and (i could be wrong) some java projects do too | 17:44 |
| * rocky adds his two cents | 17:44 | |
| rocky | IFoo sucks imho, but its a necessary evil | 17:44 |
| * Theuni takes 0.04 cents. someone has to cover up. | 17:44 | |
| Theuni | bzz. s/0.04/4/ | 17:45 |
| mexiKON | Theuni, interface centric == Foo + SomeFoo; impl centric: Foo + IFoo | 17:45 |
| rocky | in java, there are language constructs for interfaces, so its never confusing whether you're dealing with an interface versus a class | 17:45 |
| J1m | I certainly don't like imbedding types in names as a rule. But I made an exception for interfaces. | 17:45 |
| rocky | python doesn't have that luxury | 17:45 |
| J1m | There's nothing saying you have to follow this pattern. | 17:45 |
| mexiKON | true | 17:45 |
| rocky | well... here's the problem | 17:45 |
| Theuni | right. i was just wondering how we came up with it. i'm perfectly fine using it. :) | 17:46 |
| J1m | rocky, right, in Java, classes and interfaces aren't first-class objects. | 17:46 |
| rocky | if some people stop using the IFoo convention, it will get *very* confusing for other people looking at the code right now | 17:46 |
| J1m | That's one way Java sucks. :) | 17:46 |
| rocky | its my recommendation that zope identify that IFoo is the "preferred" convention in order to keep things a little sane | 17:46 |
| J1m | In Python Interfaces are first-class objects and classes are almost first-class objects. | 17:46 |
| rocky | i hate it.. but it will certainly reduce confusion | 17:46 |
| J1m | Agreed | 17:47 |
| Theuni | apropos. I think I didn't say "I love Zope 3" today, yet. | 17:48 |
| *** edar has left #zope3-dev | 17:49 | |
| rocky | hehe | 17:49 |
| * rocky is spending a little time lately trying to convince plone developers that formlib is ready :/ | 17:50 | |
| mgedmin | "classes are almost first-class objects"? why "almost"? | 17:51 |
| graham | I think J1m meant interfaces are almost... | 17:54 |
| mexiKON | no, interfaces are first class objects | 17:55 |
| graham | Okay, I'll shut up | 17:55 |
| *** ChanServ sets mode: +o srichter | 17:56 | |
| *** alga has joined #zope3-dev | 17:57 | |
| mgedmin | what's second class about classes then? | 17:57 |
| mexiKON | i don't really know. i thought they were first class now | 17:58 |
| mgedmin | so did I | 17:58 |
| J1m | classes don't have their own attributes. | 17:59 |
| J1m | Their attribute namespace is all mixed up with their instances' | 17:59 |
| mexiKON | hmmm | 18:00 |
| J1m | So, for example could couldn't write a picklable class. | 18:00 |
| *** RaFromBRC|brb is now known as RaFromBRC|arb | 18:00 | |
| *** RaFromBRC|arb is now known as RaFromBRC | 18:00 | |
| J1m | I don't have time atm for a deeper explanation. Ask SteveA. :) | 18:00 |
| mexiKON | J1m, ah, right | 18:00 |
| mexiKON | no that already makes sense to me | 18:00 |
| *** flox|away has quit IRC | 18:01 | |
| mgedmin | hmm | 18:01 |
| *** alecm has joined #zope3-dev | 18:02 | |
| mgedmin | SteveA is away for a stroll and a lunch | 18:02 |
| harobed | hi, what is the roadmap about Zope 3 web site (http://www.modscape.com/zope/v3) ? | 18:03 |
| mexiKON | looks nice. never heard about it before, though | 18:04 |
| harobed | thread is here : http://mail.zope.org/pipermail/zope-web/2006-March/003767.html | 18:04 |
| mexiKON | harobed, you should ask the relevant ZF committee, namely Martijn Faassen & co. | 18:06 |
| harobed | ok | 18:06 |
| d2m | i think the design is used for the zopefoundation.org website | 18:09 |
| rocky | anyone here know if it'd be somehow possible to have the named template lookup be dependent on the current zodb context? right now it seems to only adapt the view | 18:35 |
| *** rock1 has joined #zope3-dev | 18:43 | |
| *** rocky has quit IRC | 18:44 | |
| *** rock1 is now known as rocky | 18:44 | |
| mexiKON | rocky, that's how named templates work. they adapt the thing they're attached to | 18:45 |
| rocky | right, which are views | 18:46 |
| mexiKON | yup | 18:47 |
| rocky | and of course we can't reliably expect looking up local utilities to work in zope2 without passing a real zodb context | 18:48 |
| *** stub has quit IRC | 18:49 | |
| *** BjornT has quit IRC | 18:49 | |
| mgedmin | could someone explain the point of named templates to me? | 18:52 |
| mgedmin | if I have to override a view class, then it is pretty simple for me to override the template attribute with a different ViewPageTemplateFile, should I desire so | 18:52 |
| rocky | mgedmin: yep | 18:52 |
| rocky | thats it | 18:52 |
| mexiKON | mgedmin, nobody says you *have* to use it :) | 18:53 |
| *** alecm has quit IRC | 18:53 | |
| rocky | its gonna be a life saver for plone | 18:53 |
| mgedmin | but some people use them, and I want to know what I'm missing | 18:54 |
| *** alecm has joined #zope3-dev | 18:54 | |
| mgedmin | can skins override named templates for views? | 18:54 |
| rocky | not afaik | 18:54 |
| rocky | there's nothing extraordinary about them | 18:54 |
| rocky | they're useful sometimes, sometimes they're not | 18:54 |
| mgedmin | can I get an example? | 18:56 |
| mgedmin | when are they useful? | 18:56 |
| rocky | mgedmin: well, for us with plone we can now override pageform.pt (the default template used for formlib) with our own pageform.pt that "fits" plone simply by providing our own named template | 18:57 |
| mgedmin | with zcml overrides? | 18:58 |
| *** BjornT has joined #zope3-dev | 18:58 | |
| *** MJ has quit IRC | 18:59 | |
| rocky | mgedmin: for us no overrides are necessary since we can adapt zope2/five's BrowserView base class since thats used for all <browser:page> views | 18:59 |
| *** volvox has quit IRC | 18:59 | |
| *** bradb has quit IRC | 19:03 | |
| *** baldtrol has quit IRC | 19:04 | |
| *** dobee has quit IRC | 19:18 | |
| *** dobee has joined #zope3-dev | 19:21 | |
| *** alecm has quit IRC | 19:22 | |
| *** projekt01 has joined #zope3-dev | 19:26 | |
| *** MJ has joined #zope3-dev | 19:30 | |
| *** harobed has quit IRC | 19:33 | |
| *** projekt01 has left #zope3-dev | 19:39 | |
| *** nathany has quit IRC | 19:41 | |
| *** ignas has quit IRC | 19:49 | |
| *** alecm has joined #zope3-dev | 19:51 | |
| *** Theuni has quit IRC | 19:57 | |
| *** flox has joined #zope3-dev | 20:06 | |
| *** flox is now known as flox|away | 20:06 | |
| *** volvox has joined #zope3-dev | 20:19 | |
| *** dobee has quit IRC | 20:28 | |
| SteveA | what's a named template? | 20:38 |
| mgedmin | zope/formlib/namedtemplate.txt | 20:39 |
| *** volvox has quit IRC | 20:41 | |
| *** volvox has joined #zope3-dev | 20:41 | |
| *** volvox has quit IRC | 20:47 | |
| *** roym has joined #zope3-dev | 21:09 | |
| *** jinty has quit IRC | 21:12 | |
| *** mkerrin has quit IRC | 21:17 | |
| roym | I got the latest trunk, did a mkzopeinstance, and then bin/runzope. | 21:23 |
| roym | I get: | 21:23 |
| roym | from zope.app.twisted.main import main | 21:23 |
| roym | ImportError: No module named app.twisted.main | 21:23 |
| roym | Am I missing something? | 21:23 |
| mgedmin | I can reproduce that | 21:26 |
| roym | It looks like main.py disappeared from twisted/ | 21:26 |
| benji | roym: did you do a bare "make"? | 21:27 |
| roym | yes | 21:27 |
| benji | hmm, something bad is going on then | 21:27 |
| *** ChrisW has joined #zope3-dev | 21:27 | |
| mgedmin | roym: bin/runzope in the instance has a strange SOFTWARE_HOME | 21:32 |
| mgedmin | /usr/lib/python2.4/site-packages | 21:32 |
| *** rocky has quit IRC | 21:33 | |
| *** dunny has joined #zope3-dev | 21:33 | |
| mgedmin | mkzopeinstance.py does 'import zope' and then looks into zope.__file__ | 21:33 |
| mgedmin | on my system this finds the 'zope' package that comes from the python-zope-interface Ubuntu package | 21:34 |
| *** RaFromBRC is now known as RaFromBRC|away | 21:42 | |
| roym | mgedmin: regardless, the import points to a non-existent package (zope.app.twisted.main). Does that mean that the twisted guys have changed their dir structure? Or is main.py supposed to be generated by the make? | 21:43 |
| mgedmin | no, it means that mkzopeinstance found the wrong zope package | 21:44 |
| pcardune | roym: I came across this problem a lot. A quick fix is to just hack whichever script you want to use. (runzope, zopectl) and specify the software home there | 21:47 |
| *** gumpa has joined #zope3-dev | 21:47 | |
| roym | pcardune: thanks - will do that until the dust settles on this. | 21:47 |
| mgedmin | is there a collector issue for this? | 21:48 |
| *** oferw has joined #zope3-dev | 21:48 | |
| roym | Can someone point me to a safe version # that doesn't have this issue? | 21:51 |
| roym | I am getting another error: | 21:52 |
| roym | ZopeXMLConfigurationError: File ".../etc/securitypolicy.zcml", line 11.2 | 21:52 |
| roym | ConfigurationError: ('Unknown directive', u'http://namespaces.zope.org/zope', u'role') | 21:52 |
| * mgedmin doesn't use mkzopeinstance | 21:52 | |
| roym | mgedmin: what do you do instead? | 21:53 |
| mgedmin | just run z3.py directly in the checkout | 21:53 |
| mgedmin | but wait, didn't I see this conversation already | 21:53 |
| mgedmin | here on this channel? | 21:53 |
| mgedmin | I think someone knowledgeable said mkzopeinstance wasn't supposed to work in a checkout | 21:54 |
| *** nathany has joined #zope3-dev | 21:54 | |
| roym | It has worked until very recently for me. | 21:54 |
| mgedmin | you are supposed to run 'make install' (or python setup.py install) and then run mkzopeinstance from the installation dir | 21:54 |
| mgedmin | maybe | 21:55 |
| mgedmin | don't trust my memory | 21:55 |
| roym | the "bare" make does exactly that.. | 21:55 |
| roym | ie: make install | 21:55 |
| mgedmin | um, I don't think so | 21:56 |
| mgedmin | it runs setup.py build_ext | 21:56 |
| mgedmin | ... and install_data into . | 21:56 |
| mgedmin | hmm | 21:56 |
| mgedmin | ok, I should stop talking about things I do not understand ;) | 21:56 |
| *** dunny has quit IRC | 21:56 | |
| *** rocky has joined #zope3-dev | 22:00 | |
| *** nathany has quit IRC | 22:08 | |
| *** tarek has joined #zope3-dev | 22:19 | |
| *** alga has quit IRC | 22:24 | |
| roym | Is there a way to undo the installation done by "/usr/bin/python install.py install". | 22:25 |
| roym | It put everything in my site-packages dir and I wanted it elsewhere. | 22:25 |
| roym | ie: Zope-3.3.0b1 | 22:25 |
| *** mgedmin has quit IRC | 22:25 | |
| benji | roym: there's a command for that, it's called "rm" :) | 22:26 |
| roym | I appreciate the humor... :) | 22:26 |
| roym | however, its mixed up with a lot of debian stuff. | 22:26 |
| *** tiredbones has quit IRC | 22:27 | |
| *** tiredbones has joined #zope3-dev | 22:28 | |
| roym | In case someone else needs to do this.. here is a way to get a list | 22:36 |
| roym | of the files installed, and then maybe do a cleanout... | 22:36 |
| roym | /usr/bin/python install.py install --root /usr/local/Zope3.tmp | 22:36 |
| roym | cd /usr/local/Zope3.tmp; find usr -type f | 22:36 |
| *** rocky is now known as rocky|away | 22:41 | |
| *** mgedmin has joined #zope3-dev | 22:53 | |
| *** tarek has quit IRC | 23:01 | |
| *** tarek has joined #zope3-dev | 23:01 | |
| *** alecm is now known as alecm|lunch | 23:04 | |
| *** RaFromBRC|away is now known as RaFromBRC | 23:07 | |
| *** ChrisW has left #zope3-dev | 23:14 | |
| *** gumpa has quit IRC | 23:14 | |
| *** russf has quit IRC | 23:17 | |
| *** Theuni has joined #zope3-dev | 23:17 | |
| *** Bernulf has joined #zope3-dev | 23:21 | |
| pcardune | J1m: I see you moved zwiki at revision 69045 "Moving zwiki to separate project" where is this separate project? | 23:37 |
| J1m | zwiki | 23:38 |
| J1m | It is now a top-level project | 23:38 |
| J1m | are you aware of why? | 23:38 |
| pcardune | nope, i'm not | 23:39 |
| J1m | It has the same security hole as the one we just fixed in z2 | 23:39 |
| pcardune | oh, alright then | 23:39 |
| J1m | I mentioned this on the zope3-dev list | 23:39 |
| pcardune | (I just subscribed) (again) | 23:40 |
| *** jhauser_ has quit IRC | 23:46 | |
| *** jhauser has joined #zope3-dev | 23:57 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!