| anguenot | hi there. | 00:02 |
|---|---|---|
| anguenot | I can't find how is it possible to increase the number of threads ? | 00:02 |
| srichter | zope.conf | 00:05 |
| *** natea|away is now known as natea | 00:06 | |
| *** sashav__ has joined #zope3-dev | 00:08 | |
| anguenot | what's the directive's id ? `threads` ? | 00:10 |
| *** sashav has quit IRC | 00:12 | |
| tarek | anguenot, would you mind bring me back a pack of oreos from the airport shop ? | 00:14 |
| * benji_york just figured out how to bribe tarek | 00:14 | |
| tarek | :) | 00:14 |
| srichter | anguenot: I dunno; I forgot | 00:15 |
| *** srichter has quit IRC | 00:17 | |
| *** srichter has joined #zope3-dev | 00:17 | |
| *** ChanServ sets mode: +o srichter | 00:19 | |
| anguenot | tarek, ok :) | 00:20 |
| tarek | yes !! | 00:20 |
| tarek | thx | 00:20 |
| *** J1m has joined #zope3-dev | 00:20 | |
| *** sashav_ has quit IRC | 00:20 | |
| tarek | (the black ones) | 00:21 |
| *** srichter has quit IRC | 00:21 | |
| *** srichter has joined #zope3-dev | 00:23 | |
| *** stub has joined #zope3-dev | 00:24 | |
| anguenot | `threads` is recognized (the directive doesn't raise any error) but I dont't see any effects on the number of threads | 00:25 |
| anguenot | can it be a bug ? | 00:25 |
| * benji_york goes the kitchen to get some milk and cookies | 00:25 | |
| *** ChanServ sets mode: +o srichter | 00:26 | |
| clueck | benji_york, I found yet another way: failsavelogin.html -- in its view class zope.app.security.browser.auth.HTTPBasicAuthenticationLogin the HTTPBasicAuth page is hardcoded instead of looking up a utility (the plugin I deactivated). | 00:32 |
| benji_york | what would the URL be? | 00:33 |
| clueck | yust localhost:8080/failsavelogin.html | 00:34 |
| benji_york | cool, didn't know about that | 00:35 |
| *** d2m has quit IRC | 00:40 | |
| benji_york | oh, and clueck (and for anyone watching) it's actually "failsafelogin.html" | 00:42 |
| clueck | yes, sorry | 00:43 |
| benji_york | NP, just wanted to make sure the spectators weren't misinformed :) | 00:44 |
| clueck | :) | 00:44 |
| *** BjornT has quit IRC | 00:49 | |
| *** BjornT has joined #zope3-dev | 00:49 | |
| *** BjornT has quit IRC | 00:57 | |
| *** BjornT has joined #zope3-dev | 00:58 | |
| *** zbir has quit IRC | 01:05 | |
| *** zagy_ has quit IRC | 01:19 | |
| *** sashav__ has quit IRC | 01:24 | |
| *** niemeyer has quit IRC | 01:26 | |
| *** jhauser_ has quit IRC | 01:42 | |
| *** fdrake has quit IRC | 01:54 | |
| *** anguenot has quit IRC | 02:00 | |
| *** J1m has quit IRC | 02:02 | |
| tarek | srichter, ayt ? | 02:09 |
| *** BjornT_ has joined #zope3-dev | 02:34 | |
| srichter | tarek: not really, I am about to go to dinner | 02:36 |
| srichter | I will be online tomorrow | 02:36 |
| tarek | ok srichter cya, have a good dinner | 02:40 |
| *** BjornT has quit IRC | 02:48 | |
| *** clueck has quit IRC | 03:02 | |
| *** newpers has joined #zope3-dev | 03:25 | |
| benji_york | tarek, are you around? | 03:31 |
| tarek | benji_york, yup | 03:37 |
| benji_york | hey | 03:37 |
| tarek | finished your cookies ? :) | 03:38 |
| benji_york | I just checked in some code inspired by your pystone performance decorator | 03:38 |
| benji_york | yes actually :) | 03:38 |
| tarek | cool | 03:38 |
| benji_york | part of it was a PystoneTimer object that I thought you might be interested in. | 03:38 |
| * tarek is looking at it | 03:39 | |
| *** yota has quit IRC | 03:40 | |
| tarek | pretty cool | 03:42 |
| tarek | maybe a tolerance 'constant' could be added | 03:43 |
| benji_york | I might try to find time to use the timer to implement the "top N" switch to the new test runner | 03:43 |
| benji_york | re: tolerance, for the current use case, I don't see a need (because in doctests you'd add your own tolerance when you compare against a constant | 03:44 |
| benji_york | by saying "< 2000" or whatever | 03:45 |
| tarek | sure, but i am still looking for that magic number, that would deal with the cpu variations | 03:46 |
| benji_york | you mean variations between different computers? | 03:47 |
| tarek | no i mean variations on the same computer that can cover the variations that can happen from one run to antoher | 03:48 |
| benji_york | right, for example if the machine is under light load when the pystone is calculated and then heavy load when something is tested | 03:49 |
| tarek | exacty | 03:49 |
| *** projekt01 has quit IRC | 03:49 | |
| benji_york | that's a difficult issue | 03:49 |
| tarek | hopefully pystone works pretty well to get a portable measure when you run it into other computer | 03:49 |
| tarek | yes, i am stil looking around for a better soltuion | 03:50 |
| benji_york | pystone is a good indicator of Zope's overall performance, but not neccesarily a good indicator of the performance of a particular piece of Python code | 03:50 |
| tarek | yes, anyway i think we have something pretty cool, as we can already prevent from big performance drops | 03:51 |
| benji_york | hey, I forgot we already had a --top-fifty switch to the test runner that shows you the slowest fify tests | 03:54 |
| tarek | oh cool | 03:55 |
| benji_york | but I don't know if the new test runner has it... checking | 03:55 |
| * tarek needs to look over the whole testing package | 03:55 | |
| tarek | I have to run (it's 3 am here) cya later benji_york | 04:03 |
| benji_york | later | 04:03 |
| *** benji_york has quit IRC | 04:12 | |
| *** tvon has joined #zope3-dev | 04:17 | |
| *** tvon has quit IRC | 04:19 | |
| *** horizon5 has quit IRC | 04:35 | |
| *** anguenot has joined #zope3-dev | 04:38 | |
| *** zbir has joined #zope3-dev | 04:41 | |
| *** zbir has quit IRC | 04:44 | |
| newpers | has anyone in here made a SlackBuild for zope3? if so, i'd like to take a look at it | 05:01 |
| *** natea has quit IRC | 05:02 | |
| *** natea has joined #zope3-dev | 05:03 | |
| *** natea has quit IRC | 06:03 | |
| *** natea has joined #zope3-dev | 06:04 | |
| *** stub has quit IRC | 07:08 | |
| *** dobee has joined #zope3-dev | 07:16 | |
| *** tarek has quit IRC | 07:22 | |
| *** stub has joined #zope3-dev | 07:24 | |
| *** philiKON has joined #zope3-dev | 07:33 | |
| *** d2m has joined #zope3-dev | 08:09 | |
| newpers | I'm looking at 3.1.0 final's configure file. it says ACCEPTABLE is used to "Order a list of "acceptable" python numbers (as returned by sys.version) below in "best" to "worst" order, not including the target version..." ACCEPTABLE="2.3.5 2.4.1" the TARGET="2.3.5" which is in the list, but you say not to include it there | 08:36 |
| *** SureshZ has left #zope3-dev | 08:37 | |
| philiKON | huh? | 08:37 |
| newpers | heh | 08:38 |
| newpers | i'm just saying that zope says not to do something, but it does it. (it's not a big deal at all, i'm just going over the configure file and noticed it) | 08:38 |
| newpers | unless i'm tired and am misreading this | 08:38 |
| newpers | which is very likely | 08:39 |
| newpers | anyways, it's ok to add 2.4.2 to the ACCEPTABLE list, right? | 08:39 |
| whit | I would think so | 08:39 |
| philiKON | not really necessary | 08:39 |
| philiKON | just do ./configure --with-python=.... | 08:39 |
| newpers | yeah | 08:39 |
| newpers | ok, thanks | 08:40 |
| philiKON | though 2.4.2 should also be acceptable if 2.4.1 is acceptable | 08:40 |
| philiKON | it doesn't find your python2.4.2 installation? | 08:40 |
| newpers | it says it's not acceptable | 08:40 |
| newpers | it's 2.4.2 | 08:40 |
| philiKON | that sucks | 08:40 |
| newpers | so i added it the list and it works | 08:40 |
| philiKON | zpkgsetup's mechanism should definitely also accept higher minor releases of acceptable versions | 08:41 |
| philiKON | newpers, maybe you can file a bug collector on this? | 08:41 |
| anguenot | philiKon, hi | 08:41 |
| philiKON | hi anguenot | 08:41 |
| anguenot | just checked your comment on the wiki | 08:42 |
| anguenot | happy you might like the .xpt extension :) | 08:42 |
| philiKON | i don't like it and i don't want to like it... i might have no choice, however... | 08:42 |
| philiKON | because i can't come up with a better idea | 08:42 |
| anguenot | We don't *have* too | 08:43 |
| anguenot | currently | 08:43 |
| anguenot | we have other solutions | 08:43 |
| philiKON | like? | 08:43 |
| anguenot | sniffing | 08:43 |
| philiKON | well | 08:43 |
| philiKON | - sniffing the <?xml ?> is bad, we all agree on that | 08:44 |
| *** MacYET has joined #zope3-dev | 08:44 | |
| philiKON | (IE troubles) | 08:44 |
| anguenot | actually check the branch it's implemented | 08:44 |
| philiKON | - sniffing whether NS declarations are there or not is also not a guarantee for XML mode or HTML mode | 08:44 |
| anguenot | we agree | 08:44 |
| newpers | ok, so it works when i do --with-python=/usr/bin/python or when i add 2.4.2 to the ACCEPTABLE list in configure | 08:44 |
| gnosis | philiKON, I just got your book today, it looks like just what I need, thanks | 08:44 |
| newpers | let me know if you would like me to report it | 08:44 |
| philiKON | gnosis, you're welcome | 08:45 |
| newpers | I have it as well. I like it as well | 08:45 |
| newpers | heh | 08:45 |
| philiKON | gnosis, i'd welcome some feedback when you've read it | 08:45 |
| *** stub has quit IRC | 08:45 | |
| gnosis | philiKON, you bet | 08:45 |
| philiKON | newpers, yes, i would like you to report it | 08:45 |
| philiKON | in the collector | 08:45 |
| newpers | ok | 08:45 |
| philiKON | essentially, we shouldn't have to add every single acceptable python version to the configure file | 08:45 |
| gnosis | philiKON, as far as differences between 3.1 and 3.0, only services are the major difference, right? | 08:46 |
| philiKON | it should figure out that 2.4.2 is also acceptable if 2.4.1 is | 08:46 |
| philiKON | gnosis, services and PAU | 08:46 |
| gnosis | philiKON, thanks | 08:46 |
| philiKON | I'm the process of writing a "What's new in Zope 3.1 & 3.2" document that will compile all the changes since X3 3.0. | 08:47 |
| gnosis | philiKON, that would be great | 08:47 |
| philiKON | i'm making no promise on delivery yet | 08:47 |
| philiKON | i'm so busy with other things :( | 08:47 |
| gnosis | philiKON, it will take me awhile to get through it as is :) | 08:47 |
| newpers | philiKON, I added it: http://www.zope.org/Collectors/Zope3-dev/472 | 08:51 |
| philiKON | perfect, thanks | 08:52 |
| newpers | np | 08:52 |
| newpers | thank you | 08:53 |
| *** deo has quit IRC | 09:37 | |
| *** stub has joined #zope3-dev | 09:45 | |
| *** einheit has joined #zope3-dev | 10:13 | |
| *** mgedmin has joined #zope3-dev | 10:20 | |
| *** sashav has joined #zope3-dev | 10:20 | |
| *** BjornT_ has quit IRC | 10:29 | |
| *** stub has quit IRC | 10:40 | |
| *** BjornT has joined #zope3-dev | 10:47 | |
| *** whit is now known as whit|zZz | 10:51 | |
| *** newpers has quit IRC | 10:53 | |
| *** horizon5 has joined #zope3-dev | 10:57 | |
| *** tekNico has joined #zope3-dev | 11:49 | |
| *** tekNico has left #zope3-dev | 11:49 | |
| *** projekt01 has joined #zope3-dev | 11:52 | |
| d2m | hi projekt01, i have read your postings on the zope3 windows binary installer in zope3 mailing list; i'd like to talk about how we can organize that; please /msg me if you got some time | 11:57 |
| *** tarek has joined #zope3-dev | 11:59 | |
| projekt01 | d2m, hi | 12:00 |
| projekt01 | d2m, what do you think about my mail? | 12:01 |
| d2m | projekt01: i have not tried both of the installers, i have already found in SVN the part that cretaes the program groups etc. for the Zope2.8.x releases | 12:03 |
| projekt01 | I guess most part where we need are developed for zope2, plone or something else. | 12:04 |
| projekt01 | The main question is if we should offer a "by instance" installation on windows. | 12:04 |
| d2m | projekt01: this is another part i'd like to solve, atm there is only one zope-win install possible | 12:05 |
| d2m | projekt01: it should be possible to have more than one zope version installed | 12:06 |
| projekt01 | I build my installer for customer projects with BitRock InstallBuilder. | 12:06 |
| projekt01 | Yup | 12:06 |
| d2m | how can we work on this together ? | 12:07 |
| projekt01 | Perhaps we should write a proposal first and use the ideas from the ?? (there is a wiki page) | 12:07 |
| projekt01 | Then we can see how we can integrate this with zpkgtools | 12:07 |
| d2m | projekt01: ok, which page ? i created a page on zopewiki.org, btw | 12:08 |
| projekt01 | Ah, ok, that's the page. | 12:09 |
| d2m | projekt01: from what you know about the bitrock installer: is it possible to install more than 1 zope version with it ? | 12:09 |
| projekt01 | It's not a question of the install builder. It's a question if we install a running version or a base for use with mkzopeinstance. | 12:10 |
| projekt01 | The steps are: | 12:10 |
| projekt01 | bildout a working running version | 12:11 |
| projekt01 | package this with additional scripts | 12:11 |
| projekt01 | and build the installer with a predefined install builde configuration file | 12:11 |
| projekt01 | package/pack | 12:11 |
| d2m | projekt01: ok, btw the base install should install outside the python site-packages or (at least) install with different (user defined) package names in there | 12:13 |
| projekt01 | +1 on outside python | 12:14 |
| d2m | projekt01: ok -- will check zpkgtools and the installer now -- see you later | 12:15 |
| projekt01 | Good luck, it tooks me about a week to understand and fix zpkgtools last time. But now it seams to work on windows. | 12:16 |
| srichter | projekt01: so zpkgtools grows on you? :-) | 12:18 |
| projekt01 | Remember, zpkgtools will checkout a distutil distro and not a running zope version in your workspace. | 12:18 |
| projekt01 | No, just reported some svn command bugs on windows. | 12:18 |
| *** BjornT has quit IRC | 12:19 | |
| srichter | I see | 12:19 |
| projekt01 | srichter, are the twisted issues fixed on the trunk? | 12:32 |
| *** sashav has quit IRC | 12:42 | |
| *** sashav has joined #zope3-dev | 12:43 | |
| srichter | projekt01: at least the test failures are | 12:57 |
| srichter | projekt01: I think Michael still has some open SFTP issues | 12:58 |
| srichter | projekt01: I just finsihed the vielwet tests; I am going to write some motivation and design decisions next | 13:32 |
| *** yota has joined #zope3-dev | 13:41 | |
| tarek | hello yota | 13:44 |
| yota | hi tarek :) | 13:45 |
| MacYET | morning | 13:47 |
| tarek | yota, fyi, there's a python meeting in paris this afternoon, just to let you know if you're interested (they gather in #afpy) | 13:47 |
| *** einheit has quit IRC | 13:56 | |
| *** baiju has joined #zope3-dev | 13:59 | |
| *** baiju has quit IRC | 14:31 | |
| *** sashav has joined #zope3-dev | 15:07 | |
| *** zagy has joined #zope3-dev | 15:16 | |
| srichter | projekt01: okay roger (wow this rhymes), I sent a RFC to the mailing list concerning the content provider and viewlet code | 15:26 |
| *** benji_york has joined #zope3-dev | 15:27 | |
| *** sashav_ has joined #zope3-dev | 15:29 | |
| *** sashav has quit IRC | 15:30 | |
| *** mgedmin has quit IRC | 15:37 | |
| *** sashav_ has quit IRC | 15:38 | |
| projekt01 | srichter, Yeah, I'm wondering about the comments on this mail. | 15:41 |
| *** ignas has joined #zope3-dev | 15:50 | |
| *** zbir has joined #zope3-dev | 15:52 | |
| *** srichter has quit IRC | 15:53 | |
| *** mgedmin has joined #zope3-dev | 16:03 | |
| *** srichter has joined #zope3-dev | 16:09 | |
| *** ChanServ sets mode: +o srichter | 16:10 | |
| *** baiju has joined #zope3-dev | 16:12 | |
| *** sashav has joined #zope3-dev | 16:22 | |
| *** srichter has quit IRC | 16:26 | |
| *** baiju has quit IRC | 16:40 | |
| *** srichter has joined #zope3-dev | 16:42 | |
| *** ChanServ sets mode: +o srichter | 16:42 | |
| *** srichter has quit IRC | 16:45 | |
| *** srichter has joined #zope3-dev | 16:46 | |
| *** ChanServ sets mode: +o srichter | 17:10 | |
| * MacYET is away: Blicket auf zum Retterblick, Alle reuig Zarten, Euch zu sel'gem Glueck, Dankend umzuarten! | 17:21 | |
| srichter | benji_york: are you there? | 17:28 |
| benji_york | yep | 17:28 |
| benji_york | what's up srichter? | 17:31 |
| srichter | do you guys have a snapshot of your portlet implementation somewher? | 17:31 |
| benji_york | not publicly available | 17:31 |
| srichter | benji_york: or to answer some questions: did you implment JSR 168? | 17:31 |
| srichter | i.e. can your code provide portlets to Java portals? | 17:31 |
| benji_york | I'd be glad to answer some questions | 17:31 |
| benji_york | not directly, we thought that might be an eventual use case, we didn't do anything to prevent it | 17:32 |
| srichter | I am trying to figure out whether it would be useful to start an independent implementation now and get the best of both later or wait till I can have a peek at your code | 17:32 |
| srichter | are your portlets totally independent applications? | 17:33 |
| benji_york | If you're really interested, we might be able to open it sooner than later, I can bring this up next week at ZC | 17:33 |
| srichter | i.e. do they have their own URLs to send requests to | 17:33 |
| srichter | yeah, that would be good | 17:33 |
| srichter | I just do not want to solve solved problems | 17:34 |
| benji_york | they don't presently, but were designed so that would be easy | 17:34 |
| srichter | did you implement the concept of a portlet container literarely? | 17:34 |
| srichter | i.e. do you have a utility or some other container that aggregates portlets? | 17:35 |
| benji_york | yep, that's a slot | 17:35 |
| srichter | ok | 17:35 |
| benji_york | the user can add/remove portlets in slots, or drag them between slots | 17:35 |
| srichter | we moved away from using the name "slot" because of metal | 17:35 |
| srichter | benji_york: ok | 17:36 |
| srichter | I am not interested in the user experience that much yet ;-) | 17:36 |
| benji_york | :) | 17:36 |
| srichter | mmh, okay, I am really going to wait till you give me something ;-) | 17:37 |
| srichter | of course, I want to build portlets on top of the viewlet code | 17:37 |
| benji_york | right | 17:38 |
| srichter | also, I want portlets to be accessible like this: path/to/object/view.html/++portlet++/portletManagerName/portletName | 17:39 |
| benji_york | I'm watching a screencast of someone building a simple blog in Java, it is extreamly painful! | 17:39 |
| srichter | :-) | 17:39 |
| benji_york | yep, that's good | 17:39 |
| benji_york | oops, nope, not a good idea :) | 17:39 |
| benji_york | what happens when the portlet is moved to a different manager? | 17:40 |
| srichter | then it has a different meaning | 17:41 |
| srichter | we can decide to move the settings as well | 17:41 |
| srichter | the portlet URL does nto have to be canconical | 17:41 |
| benji_york | right, so what if some AJAX wants to access a portlet from the client, but the portlet has been moved, it gets the wrong one | 17:41 |
| srichter | ok, so that begs the question of use cases | 17:42 |
| srichter | I would say that a portlet is not moved and used at the same time | 17:42 |
| projekt01 | benji_york, we don't define the URL, the URL is given by its container, so you can move how much you like | 17:43 |
| benji_york | that assumes that only one person would ever be interacting with a portlet at a time | 17:43 |
| srichter | the use cases I have suggest that putting a portlet into a container is a development/configuration task | 17:43 |
| projekt01 | btw, hi together | 17:43 |
| benji_york | :) hi | 17:43 |
| srichter | benji_york: no, it does not | 17:43 |
| benji_york | what if it's moved to a different container? | 17:43 |
| projekt01 | the portlet? | 17:43 |
| srichter | benji_york: well, do you think users of a WEb site (not the site manager) can move portlets among containers? | 17:44 |
| benji_york | right, the portlet might be moved | 17:44 |
| srichter | Reading JSR 168 as far as I did never suggested sucha use case | 17:44 |
| benji_york | srichter, defininately | 17:44 |
| projekt01 | yes, we can move portlets | 17:44 |
| srichter | so the container is user dependent? | 17:45 |
| srichter | if so, can you give me a place where this is described? | 17:45 |
| srichter | with user I mean the end user, not anyone involved in generating the site | 17:45 |
| benji_york | srichter, if you're asking me: we have two different types of containers (slots), one is usually placeful and the other is specifict to a user | 17:46 |
| srichter | sh, I see | 17:47 |
| *** anguenot has quit IRC | 17:47 | |
| srichter | not that this is a problem for my URL case above | 17:47 |
| benji_york | the placeful type might have someone change it between viewings | 17:47 |
| srichter | ok | 17:48 |
| srichter | I see | 17:48 |
| srichter | that users can mve portlets between containers is definitely a use case I have not considered yet | 17:48 |
| srichter | that is a really tough one | 17:49 |
| srichter | I am pretty sure the Java world has not implemented this use case, but I could be wrong | 17:50 |
| benji_york | Gary came up with a good way of naming things so they can be referred to even if their containment (or order) changes | 17:50 |
| projekt01 | srichter, moving portlets is a user task where we can do via javascript. This implementation needs a "xy" container utility where is called via the PortletManager in our implementation. | 17:50 |
| benji_york | I think he described it in the subview README, let me see if it's there | 17:51 |
| srichter | projekt01: but that means that the portletmanager implementation looks very, very different from what the viewlet manager looks now | 17:51 |
| benji_york | projekt01, it's not moving them freely on an x,y plane, it's reordering them in slots and moving them between slots (but an x,y container would be cool, too;) | 17:52 |
| srichter | Currently viewlets are registered for their manager, meaning they can depend on the manager API; they could not do this anymore | 17:52 |
| srichter | also, the concept of making portlets adapters, like we do now, is becoming a much harder issue | 17:53 |
| benji_york | if you're interested the basics are described in http://svn.zope.org/Zope3/branches/f12gsprint-widget/src/zope/subview/README.txt?rev=38721&view=markup | 17:53 |
| projekt01 | srichter, no, we need a additional component where manages the move part. But this is only a observer during the move process. | 17:53 |
| benji_york | it's about the names and prefixes, etc. | 17:53 |
| srichter | I would guess that if you can move a portlet from slot X to Y, they would at least need to share a common interface | 17:53 |
| projekt01 | don't forget about he typing patter we disdussed! | 17:53 |
| projekt01 | We can type viewlets during the move procces, so they belong to another manager. | 17:54 |
| srichter | don't forget that those configurations are on a user-by-user basis! | 17:54 |
| projekt01 | the marker is only a directly provided interface. | 17:55 |
| srichter | the problem is that the component architecture does not register adapters on a user-by-user basis | 17:55 |
| projekt01 | Ok, right ;-( | 17:55 |
| projekt01 | Why not ;-) | 17:55 |
| srichter | right, I have to think about this issue | 17:55 |
| srichter | (that will keep me busy the next weel) | 17:55 |
| srichter | I am still convinced that portlets should be adapters | 17:56 |
| srichter | one choice would be to not make portlets be based on viewlets | 17:56 |
| srichter | and register portlets more like regualr content providers | 17:56 |
| srichter | and make portlet managers more virtual containers | 17:56 |
| srichter | which work with the user-specific configuration data | 17:57 |
| srichter | ok, this is a good way to do it I think | 17:57 |
| srichter | the use cases for portlets and viewlets (as we have them now) are more different than I thought | 17:57 |
| projekt01 | but if portlets are named adapters and we can call this on a pre user basis if we call only the names where the user defined. | 17:58 |
| srichter | all this moving around of portlets only works well, if portlets are totally autonomous mini-applications | 17:58 |
| srichter | projekt01: right | 17:58 |
| benji_york | srichter, regarding differences: that's why we started working on subviews, there seems to be a basic model underlying all this (viewlets, portlets, CPS Skins portlets, etc.) | 17:58 |
| srichter | right, this is the reason we have content providers :-) | 17:59 |
| benji_york | content providers are basically things that produce HTML, right? | 18:00 |
| srichter | yes | 18:00 |
| projekt01 | Yes, I'm absolutly sure that content provider can be used a s a base for all concepts | 18:00 |
| srichter | yeah, I think we got content providers right | 18:00 |
| benji_york | I think there is actually a more "featureful" concept that is shared | 18:00 |
| srichter | I'll note that the true feature comes from the way they are registered | 18:01 |
| srichter | and how they can interact with TAL | 18:01 |
| benji_york | unfortunately I can't discuss this right now, have to feed the baby (I'm Mr. Mom this weekend :) | 18:02 |
| srichter | ok | 18:02 |
| srichter | see ya | 18:02 |
| * benji_york is away: I'm busy | 18:03 | |
| projekt01 | srichter, I think if we use the name of portlets we can render them in different manager as well on a pre user basis. Or not? | 18:04 |
| projekt01 | This means we need a utility where the different managers tell what names are used. | 18:05 |
| projekt01 | per user | 18:05 |
| srichter | yep, I agree | 18:05 |
| projekt01 | do you see other problems? | 18:05 |
| srichter | no | 18:06 |
| srichter | all I am saying is that the current viewlet implementation does not work for this....but that's ok | 18:06 |
| projekt01 | If you implement a different manager and use a utility where this manager tells what viewlet should be called, it's also possible. Or not? | 18:07 |
| srichter | right | 18:08 |
| projekt01 | I guess it would be a little bit overhead to support drag and drop over managers as a base concept | 18:08 |
| *** sashav has quit IRC | 18:10 | |
| *** sashav has joined #zope3-dev | 18:10 | |
| projekt01 | Perhaps we have to think about the relation from viewlets and the directly provided interface. Perhaps we have to use a naming (named adapter) | 18:10 |
| projekt01 | So we can easier handle pre user basis settings (even for global registration) if we support a concept via preferences (store adapter names for users) | 18:11 |
| srichter | we already do | 18:12 |
| srichter | right | 18:12 |
| projekt01 | Do we not mark the viewlet for a manager with the viewletType anymore? | 18:12 |
| *** natea has quit IRC | 18:13 | |
| benji_york | WERRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRr | 18:13 |
| *** BjornT has joined #zope3-dev | 18:13 | |
| benji_york | mbdfe47aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa3bhht4y8565!111k | 18:14 |
| *** natea has joined #zope3-dev | 18:14 | |
| srichter | projekt01: no | 18:15 |
| projekt01 | Ok, I'll take a look at this part tonight and think about the portlet-move-case | 18:15 |
| srichter | cool | 18:17 |
| projekt01 | I guess this part depends on the "portlet box state" part and not on the portlet instance itself. So I guess portlets are totaly independend here. Only the Box around a portlet has to know about all this. | 18:17 |
| philiKON | benji_york, are you ok? | 18:18 |
| projekt01 | What we really need is a very flexible portlet box | 18:18 |
| projekt01 | srichter, he is feeding his baby on the table. | 18:18 |
| mgedmin | I didn't know babies made such scary sounds :-) | 18:19 |
| projekt01 | srichter, btw, I changed my wfmc implementation. Now I use roles for map participants to principals. This makes it possible to send notification mails. | 18:20 |
| projekt01 | mgedmin, I know it, I have two of them at home ;-) | 18:21 |
| benji_york | philiKON, the 4 year old apparently wanted to talk on IRC | 18:31 |
| *** MJ has quit IRC | 18:33 | |
| *** MJ has joined #zope3-dev | 18:34 | |
| philiKON | heh | 18:35 |
| *** philiKON has quit IRC | 18:48 | |
| srichter | ok, I am going to implement devmode now; any objections? | 19:06 |
| srichter | in the release, should the development mode be turned on or off by default | 19:48 |
| benji_york | srichter, it depends on our target audience; I doubt the current releases would be useful for anyone but developers, so I'd say it should be on | 19:49 |
| srichter | right, I agree; okay on it will be for now | 19:50 |
| *** tarek has quit IRC | 19:53 | |
| *** roym has joined #zope3-dev | 20:08 | |
| roym | folks, how do I run a doctest from the command line? The "-vpu" options to test.py is the closest I can find to this, but doesn't work... | 20:09 |
| *** mgedmin has quit IRC | 20:10 | |
| srichter | roym: you have to setup a tests.py for the test | 20:13 |
| srichter | roym: the source code is full of examples | 20:14 |
| srichter | see zope.app.preference.tests for example | 20:14 |
| roym | I must be doing something wrong: | 20:18 |
| roym | cd /usr/local/Zope3/src/zope/app/preference | 20:18 |
| roym | python2.3 tests.py | 20:18 |
| roym | > Traceback (most recent call last): | 20:18 |
| roym | > File "tests.py", line 40, in ? | 20:18 |
| roym | > unittest.main(default='test_suite') | 20:18 |
| roym | > TypeError: __init__() got an unexpected keyword argument 'default' | 20:18 |
| srichter | oops, it should be defaultTest=... | 20:20 |
| srichter | not default | 20:20 |
| srichter | we usually do not catch those problems, since we are using the test runner | 20:20 |
| roym | I tried that.. it now comes back with *many* errors of the form: | 20:21 |
| roym | Failed example: | 20:21 |
| roym | settings.skin = 'Basic' | 20:21 |
| roym | Exception raised: | 20:21 |
| roym | Traceback (most recent call last): | 20:21 |
| roym | File "/usr/local/Zope3/src/zope/testing/doctest.py", line 1256, in __run | 20:21 |
| roym | compileflags, 1) in test.globs | 20:21 |
| roym | ... | 20:21 |
| roym | raise ComponentLookupError(interface, name) | 20:22 |
| roym | ComponentLookupError: (<InterfaceClass zope.app.principalannotation.interfaces.IPrincipalAnnotationUtility>, '') | 20:22 |
| roym | 20:22 | |
| roym | Maybe there is some configuration that I need? | 20:22 |
| *** newpers has joined #zope3-dev | 20:23 | |
| srichter | what is the first error you get? | 20:28 |
| srichter | btw, you should really run the tests using: Zope3$ python test.py -vpu1 --dir src/zope/app/preference tests.py README | 20:28 |
| roym | getting closer: | 20:33 |
| roym | python2.3 /usr/local/Zope3/test.py -vpu1 \ | 20:33 |
| roym | --dir /usr/local/Zope3/src/zope/app/preference tests.py README | 20:33 |
| roym | 20:33 | |
| roym | Running UNIT tests at level 1 | 20:33 |
| roym | Running UNIT tests from /usr/local/Zope3-trunk/src/zope/app/preference | 20:33 |
| roym | 20:33 | |
| roym | Error in test Invalid Test (.local.Zope3.src.zope.app.preference.tests) | 20:33 |
| roym | Traceback (most recent call last): | 20:33 |
| roym | File "/usr/local/Zope3-trunk/src/zope/app/testing/test.py", line 620, in get_suite | 20:33 |
| roym | mod = package_import(modname) | 20:33 |
| roym | File "/usr/local/Zope3-trunk/src/zope/app/testing/test.py", line 596, in package_import | 20:33 |
| roym | __import__(modname) | 20:33 |
| roym | ValueError: Empty module name | 20:33 |
| srichter | you should really use Python 2.4 | 20:34 |
| srichter | it is required on the trunk | 20:34 |
| roym | but it yields the same error ! | 20:34 |
| roym | I did try that as well... | 20:34 |
| srichter | is your current working directory /usr/local/Zope3? | 20:35 |
| roym | aha - that was it... thanks! | 20:35 |
| roym | Running UNIT tests from /usr/local/Zope3-trunk | 20:35 |
| roym | 1/46 ( 2.2%): /usr/local/Zope3-trunk/src/zope/app/preference/README.txt---------------------------------------------------------------------- | 20:35 |
| roym | Ran 46 tests in 0.135s | 20:35 |
| roym | 20:35 | |
| roym | OK | 20:35 |
| roym | I just assumed that the --dir option behaved from elsewhere! | 20:36 |
| *** sashav has quit IRC | 20:48 | |
| *** jinty has joined #zope3-dev | 21:10 | |
| headgoat | for everyone who was trying to help me get Zope 3 installed (earlier in the week), I thank you, and I am pleased to report that it now works. I succeeded in getting it installed on FC4. Some of the compile warnings, were probably just that - not fatal. I believe that I was getting stuck on the 'make check' step, because a test was failing - but hey, I think that was a bad test! | 21:13 |
| *** ignas has quit IRC | 21:26 | |
| benji_york | headgoat, if you think you have a spurious test failure, you should file a bug so someone can look at it | 21:27 |
| benji_york | congratulations, by the way | 21:27 |
| *** jinty has quit IRC | 21:32 | |
| headgoat | thanks, I'm going to do a little more testing... I don't want to report something that's just a dolt-newbie error. | 21:48 |
| headgoat | Is there a Products directory in Zope3 like there is in Zope2.8? (I set up an instance on Zope3, and no Products directory was created. Can I just create one, and put Plone in it?) | 21:50 |
| *** horizon5 has quit IRC | 21:50 | |
| *** horizon5 has joined #zope3-dev | 21:57 | |
| *** tvon has joined #zope3-dev | 21:58 | |
| d2m | headgoat: products are packages now and go anywhee in the python path, plone does not run with zope3 | 21:59 |
| headgoat | d2m: thanks for the info! saved me a lot of headaches | 22:05 |
| d2m | headgoat: preferred folders to put your packages in: INSTANCE/lib/python or BASE/lib/python/ (or with a SVN checkout it is BASE/src/) | 22:05 |
| headgoat | d2m: are there plans to get plone to run under zope3 (as far as you know)? | 22:06 |
| headgoat | d2m: I am not passionate about zope3, but would just prefer to learn the latest-and-greatest. | 22:06 |
| d2m | headgoat: they are just starting to use the zoep3 features provided through Five | 22:06 |
| headgoat | d2m: thx | 22:08 |
| d2m | headgoat: http://www.goldeggstack.org/one/Zope3views is a good description | 22:09 |
| *** newpers has quit IRC | 22:10 | |
| *** jinty has joined #zope3-dev | 22:23 | |
| *** roym has quit IRC | 22:25 | |
| *** whit|zZz is now known as whit | 22:45 | |
| *** suse-joe has joined #zope3-dev | 23:12 | |
| *** sashav has joined #zope3-dev | 23:17 | |
| *** sysfault has joined #zope3-dev | 23:18 | |
| *** sysfault has left #zope3-dev | 23:18 | |
| *** horizon5 has joined #zope3-dev | 23:24 | |
| *** headgoat has quit IRC | 23:29 | |
| *** _projekt01 has joined #zope3-dev | 23:33 | |
| *** projekt01 has quit IRC | 23:33 | |
| *** sashav has quit IRC | 23:37 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!