| *** tvon has joined #zope3-dev | 00:25 | |
| *** MiUlEr has quit IRC | 00:29 | |
| *** JZ has quit IRC | 00:45 | |
| *** JZ has joined #zope3-dev | 00:47 | |
| *** timte has quit IRC | 00:58 | |
| *** dman13_ has joined #zope3-dev | 01:09 | |
| *** dman13 has quit IRC | 01:09 | |
| *** Aiste has quit IRC | 01:16 | |
| *** Aiste has joined #zope3-dev | 01:17 | |
| *** roym has quit IRC | 01:29 | |
| *** yota has quit IRC | 01:47 | |
| *** d2m has quit IRC | 02:02 | |
| *** deo has quit IRC | 02:21 | |
| *** Aiste has quit IRC | 02:34 | |
| C8N | notte | 02:36 |
|---|---|---|
| *** C8N has left #zope3-dev | 02:36 | |
| *** JZ has left #zope3-dev | 02:36 | |
| *** MiUlEr has joined #zope3-dev | 02:49 | |
| *** MiUlEr has quit IRC | 02:56 | |
| *** projekt01 has quit IRC | 03:00 | |
| *** MiUlEr has joined #zope3-dev | 03:42 | |
| *** tvon has quit IRC | 03:52 | |
| *** tvon has joined #zope3-dev | 04:06 | |
| *** MiUlEr has quit IRC | 05:35 | |
| *** vinsci|2 has joined #zope3-dev | 06:18 | |
| *** vinsci|2 is now known as vinsci | 06:19 | |
| *** tvon has quit IRC | 07:50 | |
| * bob2 creats his first utility | 08:29 | |
| *** zagy has joined #zope3-dev | 08:30 | |
| *** zagy has quit IRC | 08:40 | |
| *** zagy has joined #zope3-dev | 08:45 | |
| *** tav has quit IRC | 08:57 | |
| *** Theuni has joined #zope3-dev | 08:59 | |
| *** hdima has joined #zope3-dev | 09:00 | |
| *** tav has joined #zope3-dev | 09:05 | |
| *** timte has joined #zope3-dev | 09:34 | |
| *** tvon has joined #zope3-dev | 10:05 | |
| *** yota has joined #zope3-dev | 10:06 | |
| *** anguenot has quit IRC | 10:11 | |
| *** timte has quit IRC | 10:26 | |
| *** tav_ has joined #zope3-dev | 10:45 | |
| *** tav has quit IRC | 10:53 | |
| *** tav_ is now known as tav | 11:04 | |
| *** anguenot has joined #zope3-dev | 11:22 | |
| *** zagy has left #zope3-dev | 11:25 | |
| *** zagy has joined #zope3-dev | 11:25 | |
| *** Aiste has joined #zope3-dev | 11:35 | |
| *** sashav_ has joined #zope3-dev | 12:20 | |
| *** faassen has joined #zope3-dev | 12:37 | |
| *** sashav has joined #zope3-dev | 12:40 | |
| *** sashav_ has quit IRC | 12:41 | |
| *** mexiKON has joined #zope3-dev | 12:41 | |
| *** jinty has joined #zope3-dev | 12:43 | |
| *** jinty has quit IRC | 12:51 | |
| *** philiKON has quit IRC | 12:56 | |
| *** alga has joined #zope3-dev | 12:59 | |
| mexiKON | anguenot, just replied to your comment on issue 284 | 13:08 |
| *** projekt01 has joined #zope3-dev | 13:11 | |
| anguenot | checking | 13:17 |
| anguenot | ok it works as well | 13:18 |
| anguenot | the idea was to avoid the confusion at the first place I guess, so id the NotFoundError is deprecated why not | 13:18 |
| mexiKON | yeah | 13:19 |
| mexiKON | thanks for looking into this issue; i hadn't realized it still existed :) | 13:23 |
| anguenot | I'm checking them right now | 13:23 |
| anguenot | really want a 3.1 out :) | 13:23 |
| anguenot | Windows users in here ? Would be cool if someone could check if the behavior of #316 still occurs with the new testrunner | 13:25 |
| anguenot | http://www.zope.org/Collectors/Zope3-dev/316/ | 13:25 |
| anguenot | python bin/test -vpu --dir src\\zope\\ | 13:25 |
| anguenot | I guess this will tell us | 13:25 |
| mexiKON | i think srichter is having a bug day this friday too. won't probably be able to make it | 13:25 |
| anguenot | really ? | 13:26 |
| anguenot | cool | 13:26 |
| anguenot | I was writting a mail to propose one :) | 13:26 |
| anguenot | great ;) | 13:26 |
| mexiKON | hehe | 13:26 |
| mexiKON | yeah, he hasn't gotten aroudn to propose it yet | 13:26 |
| projekt01 | mexiKON, are you coming to the NeckarSprint? | 13:26 |
| mexiKON | projekt01, nope | 13:26 |
| mexiKON | unfortunately not | 13:27 |
| mexiKON | it's 2 weeks too late for me | 13:27 |
| projekt01 | :-( | 13:27 |
| mexiKON | yeh, uwe originally asked me to coach it | 13:27 |
| mexiKON | i would've liked to do it | 13:27 |
| projekt01 | did you see your article in the newest (I forgot).. about z3. I think it's the already published one, right? | 13:29 |
| mexiKON | you mean iX? | 13:29 |
| projekt01 | Could be | 13:29 |
| mexiKON | or linux magazine? | 13:29 |
| projekt01 | It's a dark blue cover | 13:29 |
| mexiKON | ummm | 13:30 |
| * mexiKON looks | 13:30 | |
| projekt01 | Yeah, it's IX see, http://www.heise.de/ix/ | 13:30 |
| projekt01 | http://www.heise.de/ix/inhalt.shtml | 13:30 |
| mexiKON | yup | 13:31 |
| projekt01 | Content Management komponentenorientiert: Zope X3.0 | 13:31 |
| mexiKON | they took ages to publish this article | 13:31 |
| projekt01 | ;-) | 13:31 |
| projekt01 | X3.0 ;-) | 13:31 |
| mexiKON | yup | 13:31 |
| mexiKON | anguenot, as for #316, i don't think the trunk is using the new test runner yet | 13:32 |
| mexiKON | and it's questionable whether 3.1 will | 13:32 |
| mexiKON | (as far as i understand J1m, it won't) | 13:32 |
| anguenot | right | 13:32 |
| mexiKON | i never had problems running X3.0 tests on windows back when i wrote the book | 13:33 |
| mexiKON | but maybe i just didn't use this particular setup/set of options/etc. | 13:33 |
| anguenot | would be cool if someone could simply test | 13:33 |
| * mexiKON doesn't feel like installing Windows on his mac again | 13:34 | |
| anguenot | me neither :) | 13:34 |
| anguenot | Don't have a mac though | 13:35 |
| anguenot | but still ;) | 13:35 |
| mexiKON | VirtualPC is so slow... | 13:35 |
| mexiKON | anguenot, you could write an email to the zope3-dev list. there's bound to be some windows users that know how to make a checkout of the trunk and run the tests in the described manner | 13:35 |
| anguenot | ok doing this | 13:36 |
| wiggy | I can do a test on friday if nobody has done it before then | 13:36 |
| *** mkerrin has joined #zope3-dev | 13:39 | |
| *** Aiste has quit IRC | 13:43 | |
| *** Aiste has joined #zope3-dev | 13:44 | |
| *** mgedmin has joined #zope3-dev | 14:32 | |
| *** mkerrin has quit IRC | 14:35 | |
| *** tvon has quit IRC | 14:39 | |
| *** tvon has joined #zope3-dev | 14:57 | |
| *** SteveA has joined #zope3-dev | 15:04 | |
| *** __gotchaway has joined #zope3-dev | 15:07 | |
| *** __gotchaway is now known as __gotcha | 15:07 | |
| *** SteveA has quit IRC | 15:12 | |
| *** Aiste has quit IRC | 15:15 | |
| *** regebro has joined #zope3-dev | 15:18 | |
| *** tvon has quit IRC | 15:21 | |
| *** niemeyer has joined #zope3-dev | 15:22 | |
| *** SteveA has joined #zope3-dev | 15:30 | |
| *** tvon has joined #zope3-dev | 15:32 | |
| *** bskahan has joined #zope3-dev | 15:33 | |
| *** tarek has joined #zope3-dev | 15:37 | |
| *** SteveA_ has joined #zope3-dev | 15:37 | |
| *** tav has quit IRC | 15:40 | |
| *** stub has joined #zope3-dev | 15:42 | |
| *** __gotcha has quit IRC | 15:45 | |
| *** __gotcha has joined #zope3-dev | 15:45 | |
| *** MrTopf has joined #zope3-dev | 15:48 | |
| *** SteveA has quit IRC | 15:50 | |
| *** tvon has quit IRC | 15:51 | |
| *** tvon has joined #zope3-dev | 15:53 | |
| *** __gotcha has quit IRC | 15:55 | |
| *** __gotcha has joined #zope3-dev | 15:56 | |
| *** MiUlEr has joined #zope3-dev | 15:58 | |
| *** gintas has joined #zope3-dev | 16:04 | |
| *** horizon5 has joined #zope3-dev | 16:18 | |
| *** bradb-afk is now known as bradb | 16:25 | |
| *** sashav has quit IRC | 16:35 | |
| mgedmin | does anybody else think ILocalErrorReportingUtility is weird? | 16:48 |
| mgedmin | it has a getProperties method that returns a dict, and a setProperties method that takes keyword arguments -- instead of three simple schema fields | 16:49 |
| srichter | mgedmin: I doubt anyone really ever looked at it, other than the original author, which was SteveA :-) | 16:50 |
| bob2 | but it's EXTENSIBLE | 16:50 |
| bob2 | that does sound dodge | 16:50 |
| mgedmin | waaah, and the inteface doesn't match the implementation | 16:51 |
| srichter | mgedmin: feel free to fix it :-) | 16:51 |
| mgedmin | the interface mentions getProperties()['copy_to_logfile'], but the implementation uses 'copy_to_zlog' | 16:51 |
| *** tav has joined #zope3-dev | 16:51 | |
| bob2 | does zlog even exist in zope3? | 16:51 |
| srichter | :-( | 16:52 |
| srichter | mgedmin: oh, look at where this API is used | 16:52 |
| srichter | I think there are use cases in the publisher somewhere | 16:52 |
| *** hdima has quit IRC | 16:59 | |
| SteveA_ | i didnt' write the ILocalErrorReportingUtility | 17:04 |
| SteveA_ | i may have touched in when doing some refactoring in the publisher | 17:04 |
| srichter | mmh, ok; I would have swarn it was you :-) | 17:05 |
| SteveA_ | that is a scary name for the utility | 17:05 |
| * SteveA_ opposes "local" utilities, as currently understood | 17:05 | |
| srichter | I see | 17:06 |
| SteveA_ | my point is, an error reporting utility cannot store data persistently | 17:09 |
| SteveA_ | (if it is sane) | 17:09 |
| SteveA_ | so, its being "local" in terms of persistent in the zodb, is not relevant | 17:09 |
| SteveA_ | it could be that you want it for only a particular "site" | 17:10 |
| SteveA_ | but i say that's yagni at this point in zope3's development | 17:10 |
| srichter | that's true, though then a lot of other stuff can be removed | 17:10 |
| SteveA_ | we should say "yagni" more often | 17:15 |
| *** efge has joined #zope3-dev | 17:16 | |
| srichter | Jim wants to simplify the local component stuff for 3.2; maybe it's worth brining it up | 17:17 |
| *** Theuni has quit IRC | 17:26 | |
| *** J1m has joined #zope3-dev | 17:27 | |
| *** regebro has quit IRC | 17:37 | |
| SteveA_ | J1m: has anything happened on multiple ZODBs in zope3 ? | 17:58 |
| J1m | yes | 17:59 |
| SteveA_ | can i read about it? | 17:59 |
| J1m | You can specify multiple databases in your zope.conf file. | 17:59 |
| SteveA_ | i had a proposal of a different way to do local utilities | 17:59 |
| J1m | Each is registered as an IDatabase utility. | 18:00 |
| SteveA_ | and when i last talked with you about it, you said to talk again about it once the multiple zodbs issue had been handled | 18:00 |
| SteveA_ | okay | 18:00 |
| J1m | The one named "main" is used by ZopePublication. | 18:00 |
| SteveA_ | so, from a global utility, i could get a thread local database connection to "main" | 18:00 |
| J1m | All of the databases are arranged in a multidatabase | 18:00 |
| SteveA_ | okay | 18:01 |
| J1m | You can read about multidatabases in some zodb doc tests. | 18:01 |
| SteveA_ | can i get the "current connection to 'main' being used by the current thread, or a new connection if there is no current one" ? | 18:01 |
| J1m | No, there's nothing registering a thread-local database connection. | 18:01 |
| J1m | You obiosly *could* do so though. | 18:02 |
| SteveA_ | so, my hypothetical utility would, at present, need to do its own "get a connection" logic | 18:02 |
| J1m | Right and this has some potential problems. | 18:02 |
| SteveA_ | if there were such a "thread-local database connection" facility, the publisher could use it | 18:02 |
| SteveA_ | and so could global utilities that want zodb-based state | 18:02 |
| J1m | The publication you mean. | 18:02 |
| SteveA_ | yes, the publication | 18:03 |
| J1m | Possibly, or the publication might set it up. | 18:03 |
| SteveA_ | wbat are the potential problems you have in mind? | 18:03 |
| SteveA_ | s/wbat/what/ | 18:04 |
| J1m | Well, it's generally not good to mix multiple connections to the same database in a single computation. | 18:04 |
| SteveA_ | that's why i'm proposing "get the 'current' connection to that database, if there is one, or a new one otherwise" | 18:04 |
| SteveA_ | a component would manage this | 18:04 |
| J1m | For example, if the utility held regerences to objects, directly or indirectly, you could end up with cross-connection references, which will cause errors. | 18:05 |
| SteveA_ | yes | 18:05 |
| J1m | Yes, I agree | 18:05 |
| SteveA_ | now, if the 'main' database has a place called 'utility state' for utilities to use, keyed by an id... | 18:06 |
| J1m | It would be even nicer, in some ways if the utilities could just be persistent themselves. | 18:06 |
| *** SureshZ has joined #zope3-dev | 18:06 | |
| SteveA_ | i disagree | 18:06 |
| J1m | But that would require a change to utility-lookup semantics. | 18:07 |
| J1m | ok :) I disagree with your disagreement and raise you 2 arguments. ;) | 18:07 |
| SteveA_ | i think it is easier to explain and understand to have standard utility lookup, and have code in a utility's implementation to get whatever state it needs | 18:07 |
| SteveA_ | whether from the zodb, or from an RDB or from disk... | 18:07 |
| J1m | It depends on how complicated one made setting up persistent utilities. | 18:08 |
| J1m | I certainly don't object to your approach. | 18:09 |
| SteveA_ | okay. thanks for the chat. i must get back to my launchpad-in-brazil meetings | 18:10 |
| J1m | are you in Brazil? | 18:10 |
| SteveA_ | yes | 18:10 |
| SteveA_ | sao carlos | 18:10 |
| SteveA_ | async open source's offices | 18:10 |
| J1m | cool | 18:10 |
| *** SureshZ has quit IRC | 18:24 | |
| srichter | J1m: it would be really nice if we could somehow describe the workflowRelevantData with a schema | 18:24 |
| srichter | also, I think there is a lot of potential to use schemas for tools as well | 18:25 |
| *** bskahan has quit IRC | 18:25 | |
| srichter | I think one of the simple use cases is: | 18:25 |
| srichter | a tool modifies a workflow-relevant data variable | 18:26 |
| *** bskahan has joined #zope3-dev | 18:26 | |
| J1m | Well, xpdl allows wfrd to be described by an XML Schema, for all the good that would do anyone. ;) | 18:26 |
| srichter | LOL :-) | 18:26 |
| srichter | maybe we could hook just the XML Schema URI up to a Python interface | 18:27 |
| J1m | I would never go through this much trouble myself. | 18:27 |
| J1m | I never bother to specify type information when defining data in xpdl. | 18:28 |
| J1m | Pretty much everything is a string. | 18:28 |
| anguenot | But you may use external references with XPDL | 18:28 |
| srichter | I did, but I did not make use of it, oifc ourse | 18:28 |
| anguenot | I've to check the scope though | 18:29 |
| srichter | J1m: intersting, I used complex objects in wf relevant data as well | 18:29 |
| srichter | anguenot: what I propose would be optional of course | 18:29 |
| anguenot | I like the idea stephan | 18:29 |
| srichter | the reason I have this desire is that I wan to autogenerate some screens | 18:29 |
| J1m | You can do that without dragging xpdl into it. | 18:30 |
| srichter | right now I do a lot of the workitem screens manually by creating a schema and using the zope.app.form.utility functions | 18:30 |
| srichter | (btw, I am brainstorming here; I have no clear idea what I am talking about ;-) | 18:31 |
| anguenot | :) | 18:31 |
| anguenot | BTW | 18:31 |
| anguenot | I have xpdlcore bend to wfmc within Nuxeo's svn waiting for the 3.1 branch :) | 18:31 |
| srichter | great; you can check it in Friday evening | 18:32 |
| anguenot | If you have xpdl files you want to test again you're very welcome | 18:32 |
| srichter | I have the feeling that it will not take us long on Friday with the outstanding bugs | 18:32 |
| anguenot | I'm currently on the #296 :) | 18:32 |
| srichter | yipee! :-) | 18:32 |
| anguenot | I'd like to see the 3.1 as RC :) | 18:33 |
| anguenot | http://svn.nuxeo.org/trac/pub/browser/z3lab/wfmc/trunk/ | 18:33 |
| anguenot | here is the zope.wfmc using xpdlcore | 18:33 |
| srichter | ok, cool | 18:33 |
| anguenot | the only file I changed is the xpdl.py | 18:33 |
| anguenot | Jim's tests are passing | 18:33 |
| anguenot | I see your XML schema stuffs at xpdlcore layer | 18:34 |
| anguenot | generating some Python objects for Jim to be happy ;) | 18:34 |
| srichter | :-) | 18:34 |
| anguenot | at zope.wfmc layer | 18:34 |
| *** __gotcha is now known as __gotchaway | 18:35 | |
| srichter | anguenot: have you developed WFs with zope.wfmc already? | 18:36 |
| anguenot | Just focused on xpdl right now | 18:36 |
| srichter | ok | 18:36 |
| anguenot | playing with Jawe for testing | 18:36 |
| anguenot | but definitely waiting for it :) | 18:36 |
| anguenot | We'll have to talk about lxml inclusion as well after Friday | 18:37 |
| anguenot | xpdlcore uses lxml | 18:37 |
| srichter | I would like to share some of my experiences with the code with others that have used it to see whether I got it right | 18:37 |
| srichter | and maybe find better patterns | 18:37 |
| anguenot | of course | 18:37 |
| srichter | I hope faassen will check it in to the core | 18:38 |
| anguenot | hope too :) | 18:38 |
| anguenot | faasen:are you here ? We are talking about your baby ;) | 18:39 |
| srichter | it's already approved and all | 18:39 |
| srichter | I really would like to use it for zope.configuration too, so that we can finally implement removal of directives using XPath | 18:40 |
| srichter | anguenot: btw, do you have any preferred topic for the neckar sprint? | 18:41 |
| srichter | (there will be 22 participants, so we should get a lot done ... yipee) | 18:42 |
| anguenot | workflows if you're there ;) | 18:45 |
| anguenot | and if you feel like | 18:45 |
| anguenot | Kidding I know you will be there | 18:46 |
| anguenot | <wink> | 18:46 |
| srichter | :-) | 18:47 |
| srichter | ok, I put this on the list | 18:48 |
| anguenot | great | 18:48 |
| srichter | I really want to have proposals for everything we want to do at the sprint | 18:48 |
| MrTopf | bah, how organized ;-) | 18:48 |
| *** d2m has joined #zope3-dev | 18:48 | |
| srichter | I think projekt01 will be interested in that too | 18:48 |
| anguenot | cool | 18:48 |
| srichter | MrTopf: ;-) With 3.2 we will return to proposal-based development | 18:49 |
| mexiKON | +1e24 | 18:49 |
| MrTopf | that sounds good :-) | 18:49 |
| anguenot | This is a goog thing | 18:49 |
| srichter | (that reminds me that I need to write a Twisted integration proposal with mkerrins help) | 18:50 |
| anguenot | And the twisted world coming on 3.2 !!! | 18:50 |
| mexiKON | weeee | 18:50 |
| srichter | itamar and james have worked hard to accomedate us | 18:51 |
| srichter | especially, they fixed their FTP server code :-) | 18:51 |
| anguenot | You in contact with the twisted folks ? | 18:51 |
| srichter | about half of the core developers live in the boston area | 18:51 |
| srichter | so I have met with Itamar for the inital integration | 18:51 |
| anguenot | cool | 18:52 |
| srichter | but I hope I can make it to one of their hacker meetings at some point and drain some know-how | 18:52 |
| anguenot | What do they think about Zope3 ? I mean do they plan to use Zope3 once Zope3 will work against twisted ? | 18:53 |
| srichter | I dunno; remember that they also have nevow | 18:53 |
| srichter | itamar pushes to include at least more Zope stuff | 18:53 |
| srichter | but I think awareness on both sides will increase because of the integration | 18:54 |
| anguenot | I guess we need to persuade them as well ;) | 18:54 |
| srichter | itamar recently started to write his own config language, so I really hope we can convince him to use zope.configuration | 18:54 |
| srichter | right :-) | 18:54 |
| srichter | they are also developing their own object persistence, because the ZODB lacked some features they needed | 18:55 |
| anguenot | argh | 18:55 |
| srichter | in general, the twisted people are were Zope 2 was: they develop too much themselves instead of integrating into existing code and improve that | 18:56 |
| *** mkerrin has joined #zope3-dev | 19:01 | |
| *** bska|mobile has joined #zope3-dev | 19:07 | |
| *** bskahan has quit IRC | 19:10 | |
| *** MrTopf has quit IRC | 19:11 | |
| *** bradb is now known as bradb-lunch | 19:15 | |
| *** SureshZ has joined #zope3-dev | 19:22 | |
| mgedmin | what is the purpose of ISystemErrorView? | 19:30 |
| mgedmin | ah, now I get it | 19:32 |
| mexiKON | To say "A system error has ocurred" :) | 19:32 |
| mgedmin | actually, to log the traceback to the SiteError logger | 19:39 |
| mgedmin | that's why there were no tracebacks in schooltool's log file -- our custom error view did not provide this interface | 19:39 |
| *** SureshZ has quit IRC | 19:40 | |
| mexiKON | ah, right, you meant the *interface* for the view | 19:40 |
| *** SureshZ has joined #zope3-dev | 19:41 | |
| *** tav has quit IRC | 19:43 | |
| *** regebro has joined #zope3-dev | 19:48 | |
| *** bradb-lunch is now known as bradb | 19:49 | |
| faassen | srichter, anguenot: I don't understand this idea of checking lxml into the core. What do you do with twisted, is that checked into the core? | 19:54 |
| faassen | srichter, anguenot: I mean, making it a dependency of the core, fine with me. I guess this is like a vendor import or something? I haven't figured out how svn treats those yet. | 19:55 |
| faassen | srichter, anguenot: to make it part of the core we'd need to adjust the build environment so it does the needed pyrex thingies, or decouple it from pyrex. | 19:55 |
| srichter | with twisted we are gonna do either: a vendor import of a branch created for us or a copy | 19:55 |
| faassen | srichter, anguenot: important to me is that Zope 3 inclusion isn't going to cause a lot of complication for me. | 19:55 |
| srichter | I guess we should do a vendor import here too; but it would be nice, if you could license the code under ZPL 2.1 for us :-) | 19:56 |
| faassen | srichter, anguenot: I release versions of lxml, and then they can be included in Zope 3 and i'm willing to help fiddle with the build scripts and do the occasional resynching, but it's developed in its own svn. | 19:56 |
| efge | I don't see the point of vendor imports. | 19:56 |
| efge | It's just another dependency. | 19:57 |
| faassen | srichter: well, me licensing it ZPL != lxml being half-owned by Zope corporation. | 19:57 |
| efge | Just say Zope 3.foo requires lxml 1.bar, and that's it. | 19:57 |
| srichter | yes | 19:57 |
| faassen | srichter: it's BSD. I find this "we want this ZPL" rather frustrating. | 19:57 |
| srichter | faassen: I know | 19:57 |
| srichter | well, it is a matter of not having a million licenses around | 19:58 |
| faassen | then dump the ZPL and just make it BSD. :) | 19:58 |
| faassen | be part of the solution. :) | 19:58 |
| J1m | The advantage of a vendor import is that you can make local changes if you need to. | 19:59 |
| faassen | anyway, basically I'll license it any which way, but I am being difficult as I want to be a guinea pig on this with lxml. | 19:59 |
| J1m | In general you have a little more control and it makes installation a little easier. | 19:59 |
| faassen | I want to make it easier for Zope 3 to be dependent on non-core Python libraries. | 19:59 |
| srichter | look, not my decesion, nor am I willing to discuss the issue, because I have no clue | 19:59 |
| faassen | J1m: you'd be taking on the role of a linux distribution. | 19:59 |
| J1m | I don't know what you mean by that. | 20:00 |
| faassen | J1m: you'd patch lxml and pass the patch upstream. | 20:00 |
| J1m | If necessary | 20:00 |
| J1m | I'm mostly after increased convenience for developers. | 20:00 |
| faassen | J1m: if a distro already has a lxml version, or if upstream says 'no, I don't want this patch', what do you do? | 20:00 |
| efge | On an actively developed product like lxml where you could even have checkin rights, you can make local changes directly upstream if you need to :) | 20:01 |
| J1m | In general, it's a bad idea to depend on a vendor supplied version of things that change frquently. | 20:01 |
| faassen | J1m: yup, so depending on something needs to be carefully considered. | 20:01 |
| faassen | J1m: you don't want to depend on something that moves quickly. | 20:02 |
| J1m | You should on;y make local changes if you are willing to maintain them or if you are confident that you will be able to feed them back. | 20:02 |
| *** SureshZ has quit IRC | 20:02 | |
| *** zagy has quit IRC | 20:02 | |
| J1m | But we do want to depend on lxml, even though it changes quickly, but we want to control that dependence. | 20:02 |
| *** SureshZ has joined #zope3-dev | 20:03 | |
| faassen | J1m: why not depend on one fixed version instead? | 20:03 |
| J1m | That's an option, but the fixed version we want is likely to change from zope version to zope version. | 20:04 |
| J1m | Or from branch to branch. | 20:04 |
| *** Aiste has joined #zope3-dev | 20:04 | |
| faassen | J1m: anyway, I think you're going to make life hard for release management and linux distributors if you maintain your own patches to dependencies. sort of like it's not a good idea to make your own patches in Python itself, right? | 20:04 |
| J1m | For that reason, I'd tend to have a local version per workspace. | 20:05 |
| *** dman13_ has quit IRC | 20:05 | |
| faassen | J1m: anyway, we'll figure it out when the time comes. :) | 20:05 |
| J1m | sounds good :) | 20:05 |
| J1m | I do wish we had a lean mean xml machine in Zope 3 now. | 20:06 |
| faassen | J1m: but experimentally I'm going to pretend lxml is developed by a non-Zope developer, as I want Zope to be able to adopt those libraries. :) | 20:06 |
| faassen | J1m: what do you mean with lean and mean? | 20:06 |
| J1m | Experience with selenium confirms my suspicion that xpath would make our test assertions much cleaner. | 20:07 |
| *** bska|mobile has quit IRC | 20:07 | |
| J1m | I mean simpler, fast, easy to install, non-invasive. | 20:07 |
| J1m | stable | 20:07 |
| faassen | I should just release lxml 1.0 in the coming half year or so. :) | 20:08 |
| J1m | BTW, Benji York has created a mechanize-based add-on to the functional testing framework that makes functional tests selenionic. :) | 20:09 |
| J1m | I expect we'll be releasing it one of these days after we've had a chance to use it for a while and work the kinks out of it. | 20:10 |
| faassen | cool. | 20:10 |
| J1m | It makes functional tests much easier to write. | 20:10 |
| faassen | J1m: by the way, are you interested in a simple but powerful query engine? | 20:10 |
| J1m | But it would really benefit from the ability to use xpath in assertions. | 20:10 |
| J1m | uh, sure, for what? :) | 20:11 |
| faassen | J1m: I built one recently on top of the catalog. it's simple as it doesn't attempt to optimize much, but it's pretty powerful in the queries you can construct. | 20:11 |
| faassen | J1m: for the catalog, sorry. | 20:11 |
| faassen | J1m: well, multi-catalog queries. | 20:11 |
| J1m | sure | 20:11 |
| faassen | and, or, field equals, inrange, inset, things like that. | 20:11 |
| J1m | cool | 20:12 |
| faassen | I just need to clean it up and write tests for it. | 20:12 |
| J1m | k | 20:12 |
| faassen | I have tests, but I want to add a few doctests. | 20:12 |
| *** SureshZ has quit IRC | 20:24 | |
| *** SureshZ has joined #zope3-dev | 20:25 | |
| *** alienoid has joined #zope3-dev | 20:31 | |
| *** bskahan has joined #zope3-dev | 20:32 | |
| *** faassen has quit IRC | 20:37 | |
| *** alienoid has quit IRC | 20:40 | |
| *** alienoid_ has joined #zope3-dev | 20:41 | |
| alienoid_ | /msg nickserv link alienoid nhfcnyjdfy | 20:41 |
| wiggy | nice | 20:42 |
| alienoid_ | oops | 20:42 |
| *** dman13 has joined #zope3-dev | 20:44 | |
| *** MrTopf has joined #zope3-dev | 20:45 | |
| *** SureshZ has quit IRC | 20:46 | |
| *** suresh has joined #zope3-dev | 20:47 | |
| *** alienoid_ has quit IRC | 20:50 | |
| *** alienoid_ has joined #zope3-dev | 20:51 | |
| *** alienoid_ is now known as alienoid | 20:51 | |
| *** alienoid is now known as alienoid_ | 20:53 | |
| *** alienoid_ is now known as alienoid | 20:56 | |
| *** _projekt01 has joined #zope3-dev | 21:00 | |
| *** gintas has quit IRC | 21:04 | |
| *** anguenot has quit IRC | 21:05 | |
| *** suresh has quit IRC | 21:07 | |
| *** tav has joined #zope3-dev | 21:07 | |
| *** suresh has joined #zope3-dev | 21:08 | |
| *** regebro has quit IRC | 21:12 | |
| *** projekt01 has quit IRC | 21:12 | |
| *** bskahan has quit IRC | 21:18 | |
| *** tav has quit IRC | 21:18 | |
| *** tav has joined #zope3-dev | 21:19 | |
| *** SteveA_ has quit IRC | 21:24 | |
| *** SteveA__ has joined #zope3-dev | 21:24 | |
| *** SteveA__ is now known as SteveA | 21:25 | |
| *** suresh has quit IRC | 21:35 | |
| *** suresh has joined #zope3-dev | 21:35 | |
| *** SteveA has quit IRC | 21:36 | |
| *** mkerrin has quit IRC | 21:40 | |
| *** SteveA has joined #zope3-dev | 21:43 | |
| *** MrTopf_ has joined #zope3-dev | 21:46 | |
| *** hazmat has joined #zope3-dev | 21:46 | |
| *** d2m has quit IRC | 21:52 | |
| *** MrTopf has quit IRC | 21:56 | |
| *** suresh has quit IRC | 21:58 | |
| *** suresh has joined #zope3-dev | 22:05 | |
| *** timte has joined #zope3-dev | 22:12 | |
| *** Aiste has quit IRC | 22:18 | |
| *** alga has quit IRC | 22:19 | |
| *** mgedmin has quit IRC | 22:21 | |
| *** timte has quit IRC | 22:21 | |
| *** alienoid has quit IRC | 22:24 | |
| *** SureshZ has joined #zope3-dev | 22:31 | |
| *** alienoid has joined #zope3-dev | 22:31 | |
| *** SteveA is now known as sabdl1 | 22:33 | |
| *** sabdl1 is now known as sabdf1 | 22:33 | |
| *** sabdf1 is now known as stevea | 22:34 | |
| *** suresh has quit IRC | 22:34 | |
| *** stevea is now known as SteveA | 22:34 | |
| *** tarek has quit IRC | 22:37 | |
| *** SteveA has quit IRC | 22:41 | |
| *** SteveA_ has joined #zope3-dev | 22:41 | |
| *** suresh has joined #zope3-dev | 22:44 | |
| *** bradb is now known as bradb-brb | 22:49 | |
| *** SureshZ has quit IRC | 22:56 | |
| *** MrTopf_ has quit IRC | 22:56 | |
| J1m | SteveA_, ayt? | 22:58 |
| *** bskahan has joined #zope3-dev | 22:58 | |
| *** suresh has quit IRC | 23:00 | |
| *** suresh has joined #zope3-dev | 23:01 | |
| SteveA_ | hi J1m | 23:02 |
| J1m | I'm just trying out an alternative approach to connection.add for dealing with situations in which we need the _p_oid or database of an object before it has been saved the first time. | 23:05 |
| J1m | I'm doing a transaction.savepoint. | 23:05 |
| SteveA_ | i have a break from meetings in 25 minutes | 23:06 |
| SteveA_ | i can think about it then | 23:06 |
| J1m | Just thought you'd be interested. | 23:06 |
| J1m | I'll let you know how it went when you come back. | 23:06 |
| SteveA_ | ok, cool | 23:07 |
| *** bradb-brb is now known as bradb | 23:14 | |
| *** bskahan has quit IRC | 23:19 | |
| *** yota has quit IRC | 23:26 | |
| *** stub has quit IRC | 23:26 | |
| *** suresh has quit IRC | 23:27 | |
| *** suresh has joined #zope3-dev | 23:27 | |
| *** stub has joined #zope3-dev | 23:28 | |
| *** stub has quit IRC | 23:37 | |
| *** stub has joined #zope3-dev | 23:38 | |
| *** SteveA_ has quit IRC | 23:41 | |
| *** SteveA_ has joined #zope3-dev | 23:42 | |
| *** suresh has quit IRC | 23:53 | |
| *** suresh has joined #zope3-dev | 23:53 | |
| *** bskahan has joined #zope3-dev | 23:55 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!