| nederhoed_ | good night | 00:03 |
|---|---|---|
| *** nederhoed_ has quit IRC | 00:04 | |
| *** lunatik has joined #zope3-dev | 00:19 | |
| *** RaFromBRC has joined #zope3-dev | 00:23 | |
| *** lunatik has joined #zope3-dev | 00:27 | |
| *** RaFromBRC has quit IRC | 00:27 | |
| *** lunatik has quit IRC | 00:28 | |
| *** mexiKON has joined #zope3-dev | 00:31 | |
| *** bradb is now known as bradb-away | 00:37 | |
| *** mexiKON has quit IRC | 00:46 | |
| *** GaryPoster has quit IRC | 00:59 | |
| *** J1m has quit IRC | 01:05 | |
| *** gintas has quit IRC | 01:13 | |
| *** benji_york has quit IRC | 01:20 | |
| *** jhauser_ has quit IRC | 02:47 | |
| *** projekt01 has quit IRC | 03:05 | |
| *** epx has joined #zope3-dev | 03:44 | |
| *** projekt01 has joined #zope3-dev | 03:48 | |
| *** stub has joined #zope3-dev | 04:04 | |
| *** epx has quit IRC | 04:23 | |
| *** yota has quit IRC | 05:26 | |
| *** Volpe has joined #zope3-dev | 06:06 | |
| Volpe | Hi, I'm trying to being a Container object in Zope3 but when I try to add an instance of my object I get a 'system error' in the traceback in says "object has no attribute '_SampleContainer__data'" (my object inherits from BTreeContainer) | 06:08 |
| Volpe | has anyone got any suggestions? | 06:08 |
| Volpe | being = build* | 06:08 |
| Volpe | Hmm, I guess no ones around :\ | 06:11 |
| bob2 | the list is a good place to ask | 06:12 |
| bob2 | but I'd just be ripping code from srichter's book to begin wit | 06:12 |
| Volpe | I don't have his book (in the mail :D). I was having a play with it on my own for a while first. :\ | 06:13 |
| bob2 | it's on the website | 06:13 |
| bob2 | an older version, anyway | 06:13 |
| Volpe | it is? | 06:13 |
| bob2 | yes... | 06:13 |
| * Volpe feels stupid | 06:13 | |
| bob2 | from the documentation section | 06:14 |
| bob2 | I was surprised to see it on the shelf at Dymocks once, too | 06:14 |
| Volpe | is it in the zope 3 second? | 06:15 |
| Volpe | ah found it :D | 06:16 |
| Volpe | Is there a way to 'refresh' a product in zope 3? (like in zope 2) | 06:24 |
| *** Volpe has quit IRC | 07:08 | |
| *** sashav has quit IRC | 07:57 | |
| *** jhauser has joined #zope3-dev | 08:28 | |
| *** hdima has joined #zope3-dev | 08:57 | |
| *** hdima has quit IRC | 09:00 | |
| *** MJ has quit IRC | 09:00 | |
| *** tvon has quit IRC | 09:00 | |
| *** deo has quit IRC | 09:00 | |
| *** srichter has quit IRC | 09:00 | |
| *** Jim7J1AJH has quit IRC | 09:00 | |
| *** SteveA has quit IRC | 09:00 | |
| *** zagy has quit IRC | 09:00 | |
| *** Hellfried has quit IRC | 09:00 | |
| *** peaceman has quit IRC | 09:00 | |
| *** jhauser has quit IRC | 09:00 | |
| *** stub has quit IRC | 09:00 | |
| *** BjornT_ has quit IRC | 09:00 | |
| *** lucia12345 has quit IRC | 09:00 | |
| *** VladDrac has quit IRC | 09:00 | |
| *** genconc has quit IRC | 09:00 | |
| *** lunarosity has quit IRC | 09:00 | |
| *** d2m has quit IRC | 09:00 | |
| *** tanghus has quit IRC | 09:00 | |
| *** dman13 has quit IRC | 09:00 | |
| *** vinsci has quit IRC | 09:00 | |
| *** srichter has joined #zope3-dev | 09:00 | |
| *** vinsci has joined #zope3-dev | 09:00 | |
| *** VladDrac has joined #zope3-dev | 09:00 | |
| *** lucia12345 has joined #zope3-dev | 09:00 | |
| *** genconc has joined #zope3-dev | 09:00 | |
| *** peaceman has joined #zope3-dev | 09:00 | |
| *** dman13 has joined #zope3-dev | 09:00 | |
| *** tanghus has joined #zope3-dev | 09:00 | |
| *** d2m has joined #zope3-dev | 09:00 | |
| *** lunarosity has joined #zope3-dev | 09:00 | |
| *** BjornT_ has joined #zope3-dev | 09:00 | |
| *** stub has joined #zope3-dev | 09:00 | |
| *** jhauser has joined #zope3-dev | 09:00 | |
| *** Jim7J1AJH has joined #zope3-dev | 09:00 | |
| *** zagy has joined #zope3-dev | 09:00 | |
| *** SteveA has joined #zope3-dev | 09:00 | |
| *** Hellfried has joined #zope3-dev | 09:00 | |
| *** Jim7J1AJH has quit IRC | 09:00 | |
| *** hdima has joined #zope3-dev | 09:00 | |
| *** Jim7J1AJH has joined #zope3-dev | 09:00 | |
| *** tanghus has quit IRC | 09:00 | |
| *** dman13 has quit IRC | 09:00 | |
| *** vinsci has quit IRC | 09:00 | |
| *** lunarosity has quit IRC | 09:00 | |
| *** d2m has quit IRC | 09:00 | |
| *** lunarosity has joined #zope3-dev | 09:00 | |
| *** peaceman has quit IRC | 09:00 | |
| *** peaceman has joined #zope3-dev | 09:00 | |
| *** lunarosity has quit IRC | 09:00 | |
| *** Hellfried has quit IRC | 09:00 | |
| *** zagy has quit IRC | 09:00 | |
| *** SteveA has quit IRC | 09:00 | |
| *** tvon has joined #zope3-dev | 09:01 | |
| *** lunarosity has joined #zope3-dev | 09:01 | |
| *** MJ has joined #zope3-dev | 09:01 | |
| *** Hellfried has joined #zope3-dev | 09:01 | |
| *** peaceman has quit IRC | 09:01 | |
| *** genconc has quit IRC | 09:01 | |
| *** stub has quit IRC | 09:01 | |
| *** BjornT_ has quit IRC | 09:01 | |
| *** jhauser has quit IRC | 09:01 | |
| *** lucia12345 has quit IRC | 09:01 | |
| *** VladDrac has quit IRC | 09:01 | |
| *** povbot has joined #zope3-dev | 09:01 | |
| *** tanghus has joined #zope3-dev | 09:01 | |
| *** zagy has joined #zope3-dev | 09:01 | |
| *** Hellfried has joined #zope3-dev | 09:01 | |
| *** d2m_ has joined #zope3-dev | 09:01 | |
| *** SteveA has joined #zope3-dev | 09:01 | |
| *** jhauser has joined #zope3-dev | 09:01 | |
| *** stub has joined #zope3-dev | 09:01 | |
| *** BjornT_ has joined #zope3-dev | 09:01 | |
| *** VladDrac has joined #zope3-dev | 09:01 | |
| *** lucia12345 has joined #zope3-dev | 09:01 | |
| *** genconc has joined #zope3-dev | 09:01 | |
| *** vinsci has joined #zope3-dev | 09:01 | |
| *** bob2 has joined #zope3-dev | 09:01 | |
| *** hazmat has joined #zope3-dev | 09:01 | |
| *** bradb-away has joined #zope3-dev | 09:01 | |
| *** d2m_ is now known as d2m | 09:01 | |
| *** andrew_ has joined #zope3-dev | 09:02 | |
| *** mp has joined #zope3-dev | 09:02 | |
| *** peaceman has joined #zope3-dev | 09:19 | |
| *** sashav has joined #zope3-dev | 09:27 | |
| *** sashav_ has joined #zope3-dev | 09:33 | |
| bob2 | holy god I hate subversion | 09:44 |
| bob2 | merging the testbrowser branch into 3.1 means having to find the branch point myself | 09:44 |
| *** sashav has quit IRC | 09:52 | |
| *** jhauser has quit IRC | 10:02 | |
| *** Aiste has joined #zope3-dev | 10:07 | |
| *** MJ has quit IRC | 10:09 | |
| *** philiKON has joined #zope3-dev | 10:22 | |
| *** BjornT_ has quit IRC | 10:34 | |
| *** BjornT has joined #zope3-dev | 10:34 | |
| *** jhauser has joined #zope3-dev | 10:37 | |
| *** MJ has joined #zope3-dev | 11:28 | |
| *** MJ has left #zope3-dev | 11:29 | |
| *** MJ has joined #zope3-dev | 11:30 | |
| *** tvon has quit IRC | 11:49 | |
| *** yota has joined #zope3-dev | 11:54 | |
| *** MrTopf has joined #zope3-dev | 11:59 | |
| bob2 | restarting to pick up changes to resources is a bit crap | 12:00 |
| *** antoine has joined #zope3-dev | 12:18 | |
| antoine | hello worls | 12:18 |
| antoine | i looking for a man how install imagemagick for dev c++ | 12:18 |
| antoine | who knows that | 12:18 |
| hdima | bob2: try 'resourceDirectory' | 12:21 |
| *** andrew_ is now known as andrew_m | 12:46 | |
| *** lunatik has joined #zope3-dev | 12:55 | |
| *** lunatik has quit IRC | 12:57 | |
| *** lunatik has joined #zope3-dev | 13:00 | |
| *** lunatik has left #zope3-dev | 13:03 | |
| *** danfairs has joined #zope3-dev | 13:37 | |
| *** tarek has joined #zope3-dev | 13:42 | |
| *** mkerrin has joined #zope3-dev | 14:03 | |
| *** Soulraven has joined #zope3-dev | 14:43 | |
| *** ignas has joined #zope3-dev | 14:47 | |
| *** antoine has quit IRC | 15:01 | |
| *** jajcus has joined #zope3-dev | 15:02 | |
| *** bskahan has joined #zope3-dev | 15:10 | |
| Soulraven | hmm is there any way to use plone on Z3 ? | 15:15 |
| philiKON | Soulraven, no | 15:22 |
| philiKON | Soulraven, but there is a way to use z3-style software in plone | 15:22 |
| philiKON | by using five | 15:22 |
| philiKON | five 1.0 is part of zope 2.8 | 15:23 |
| *** MrTopf has quit IRC | 15:32 | |
| *** sashav_ has quit IRC | 15:32 | |
| *** lucia12345 has quit IRC | 15:32 | |
| *** VladDrac has quit IRC | 15:32 | |
| *** stub has quit IRC | 15:32 | |
| *** genconc has quit IRC | 15:32 | |
| *** genconc has joined #zope3-dev | 15:32 | |
| *** stub has joined #zope3-dev | 15:32 | |
| *** MrTopf has joined #zope3-dev | 15:32 | |
| *** sashav has joined #zope3-dev | 15:33 | |
| *** lucia12345 has joined #zope3-dev | 15:33 | |
| *** VladDrac has joined #zope3-dev | 15:33 | |
| *** jajcus has left #zope3-dev | 15:34 | |
| *** bskahan has quit IRC | 15:39 | |
| *** lunarosity has quit IRC | 15:40 | |
| *** gintas has joined #zope3-dev | 15:41 | |
| *** gintas has quit IRC | 15:44 | |
| *** benji_york has joined #zope3-dev | 15:56 | |
| *** ChanServ sets mode: +o srichter | 15:57 | |
| *** mgedmin has joined #zope3-dev | 16:03 | |
| srichter | benji_york: I wonder whether the parsing of all forms and links could be done at once... | 16:08 |
| srichter | other than that I don't see obvious inefficiencies | 16:08 |
| srichter | do you have have profiling data of what takes long? | 16:08 |
| benji_york | srichter, no data at hand, I haven't kept any of it | 16:21 |
| benji_york | I'm not quite sure what you mean by "parsing of all forms and links could be done at once" | 16:21 |
| srichter | well, basically the links are retrieved using pullparser | 16:22 |
| srichter | and the forms using clientform | 16:22 |
| benji_york | right | 16:22 |
| *** J1m has joined #zope3-dev | 16:22 | |
| srichter | that means the HTML is parsed twice (if you use both) | 16:22 |
| benji_york | ahh | 16:23 |
| srichter | the lazy parsing you checked in yesterday helps a lot of course | 16:23 |
| srichter | btw, I tried the StrictFormParser the other day and it is slower than the standard one | 16:23 |
| benji_york | it might be possible, but remember that JJL wants all those packages to not depend on one another (and more than they already do), so any optimizations will have to be stand-alone | 16:24 |
| benji_york | re StrictFormParser: hmm, I don't think many people use mechanize in a cpu bound way, that's probably why it's not super-speedy | 16:25 |
| srichter | what you basically want is an HTML parser to which you pass events | 16:25 |
| benji_york | perhaps | 16:25 |
| srichter | so basically you have a generic parser and the link code and form code register a method that gets their info | 16:26 |
| srichter | on the other hand, I think we could decouple the parsing from the rest of the code and register our own parser | 16:26 |
| benji_york | yep, but like I said, we're somewhat at JJL's mercy about what changes he'll accept... | 16:27 |
| benji_york | don't get me wrong, he's been very plesant to work with, we just have to be consious of it | 16:27 |
| srichter | right | 16:27 |
| srichter | so we do it non-intrusively | 16:27 |
| srichter | we just create hooks, so we have the choice to provide a separate hook | 16:28 |
| benji_york | perhaps | 16:28 |
| srichter | since we are all doing XHTML, I am sure we can get a better XML parser going | 16:28 |
| srichter | I really need another 30-50% speedup if possible | 16:28 |
| benji_york | I wouldn't complain | 16:29 |
| benji_york | BRB, meeting | 16:29 |
| srichter | ok | 16:30 |
| *** dman13 has joined #zope3-dev | 16:35 | |
| *** bradb-away is now known as bradb | 16:36 | |
| benji_york | back | 16:45 |
| srichter | wow, quick meeting | 16:48 |
| *** stub has quit IRC | 16:59 | |
| *** projekt01 has joined #zope3-dev | 17:15 | |
| *** deo has joined #zope3-dev | 17:20 | |
| *** sashav has quit IRC | 17:22 | |
| *** hdima has quit IRC | 17:27 | |
| *** fdrake has joined #zope3-dev | 17:35 | |
| benji_york | srichter, well, it's supposed to be a 15 minute meeting, so we actually went a few minutes over :) | 17:38 |
| srichter | :-) | 17:39 |
| *** alga has joined #zope3-dev | 17:44 | |
| fdrake | is there an IRC channel for Zope 2 + Five? other than #zope? | 17:47 |
| mgedmin | wouldn't that make Zope Seven? | 17:48 |
| fdrake | mebbe | 17:48 |
| fdrake | i can't count that high :-) | 17:49 |
| yota | fdrake: #z3-base ? | 17:50 |
| fdrake | ah, thanks! | 17:51 |
| *** GaryPoster has joined #zope3-dev | 17:53 | |
| GaryPoster | srichter: AYT? | 17:54 |
| srichter | yeah | 17:54 |
| GaryPoster | cool. I'd like to make it possible to instantiate a preference with a specific principal | 17:54 |
| GaryPoster | sorry preference group | 17:54 |
| GaryPoster | (This is on head, of course) | 17:55 |
| srichter | oh, instead of globally? | 17:55 |
| GaryPoster | the use case is that I'd like to be able to let some users, e.g. admins view/change users prefs | 17:56 |
| srichter | I see | 17:56 |
| srichter | I have not used prefs in anything else than apidoc | 17:57 |
| srichter | so feel free to change anything around | 17:57 |
| GaryPoster | OK, awesome. thanks :-) | 17:57 |
| *** zagy has quit IRC | 17:59 | |
| *** alga has quit IRC | 18:00 | |
| *** faassen has joined #zope3-dev | 18:05 | |
| *** tvon has joined #zope3-dev | 18:14 | |
| *** FarcePest has joined #zope3-dev | 18:15 | |
| *** danfairs has left #zope3-dev | 18:21 | |
| projekt01 | GaryPoster, +1, I like to use preferences in one or two month on a productive site. | 18:24 |
| projekt01 | GaryPoster, would be nice to have pereferences in the next release | 18:25 |
| projekt01 | fdrake, ayt? | 18:26 |
| *** bskahan has joined #zope3-dev | 18:27 | |
| *** sashav has joined #zope3-dev | 18:27 | |
| *** MrTopf has quit IRC | 18:29 | |
| GaryPoster | projekt01: For now, I think I may write this up somewhere instead: the security changes are more than I have time to worry with ATM. Also, a month later I still have a review of your two text widgets on my plate. :-( It will still be awhile. :-( Do you have any interest in contributing the FCK work to either Zope 3 or the FCK project, btw? | 18:30 |
| fdrake | projekt01, yes | 18:32 |
| projekt01 | GaryPoster, ok for preferences, perhaps I can find time if I need this. | 18:32 |
| projekt01 | FCK editor, yes this would be Ok. | 18:33 |
| GaryPoster | cool | 18:33 |
| projekt01 | GaryPoster, but we use a special implementation for the filemanager part where you can select images and links. | 18:33 |
| GaryPoster | yes, I remember | 18:33 |
| GaryPoster | If we can make the pertinent parts pluggable (or even port the filemanager part) that would be ideal. | 18:34 |
| mgedmin | tal:content="python:unicode_object.replace('foo', 'bar')" gives me a ForbiddenAttribute error that I didn't expect | 18:34 |
| mgedmin | uh, that's because it wasn't unicode_object, but a function_object | 18:35 |
| mgedmin | never mind | 18:35 |
| projekt01 | GaryPoster, it's all decoupled already and separated in different packages | 18:35 |
| GaryPoster | cool | 18:35 |
| projekt01 | I thkn we just use a marker interface on objects. This marker tells you that there is a adapter for the filemanager | 18:36 |
| projekt01 | thnk/think | 18:36 |
| GaryPoster | hm, sounds good | 18:36 |
| projekt01 | I'm going to vacation next week, are you in hurry with this? | 18:37 |
| GaryPoster | projekt01: no, but thank you. Probably won't have real need till October | 18:37 |
| GaryPoster | Have a good vacation! :-) | 18:38 |
| projekt01 | What is the prefered location for this package? the z3 trunk? | 18:38 |
| projekt01 | Ok, then I can work on this at the next sprint ;-) | 18:38 |
| projekt01 | fdrake, Ok, then another question, | 18:40 |
| projekt01 | fdrake, is it possible to change the output of finddeps.py and provide the correct order? | 18:40 |
| srichter | benji_york: wow, rewriting pullparser for libxml2 with xpath support would be so easy :-) | 18:40 |
| philiKON | srichter, you mean lxml :) | 18:42 |
| * fdrake is trying to remember finddeps.py again... | 18:42 | |
| srichter | no, I mean libxml2 in general | 18:42 |
| srichter | I have not looked at lxml yet, since it will require me to install more libs | 18:42 |
| projekt01 | fdrake, utilities/finddeps.py | 18:43 |
| fdrake | yeah, I found the impl | 18:43 |
| fdrake | what do you mean "correct order"? | 18:44 |
| projekt01 | fdrake, is the order in the DEPENDENCY.cfg relevant? Where reported by finddeps.py? | 18:44 |
| projekt01 | the order of the imports in DEPENDENCY.cfg | 18:44 |
| fdrake | no, DEPENDENCIES.cfg represents a set, not a list | 18:45 |
| projekt01 | Ah, cool then I don't have to order them. | 18:45 |
| fdrake | any ordering is for human readers only | 18:45 |
| fdrake | right | 18:45 |
| projekt01 | btw, I just added all SETUP.cfg for the tiks subpackages and everything works again. | 18:46 |
| fdrake | cool! | 18:47 |
| fdrake | i expect you only needed SETUP.cfg for C extensions | 18:47 |
| benji_york | srichter, i doubt JJL would go for that :( | 18:47 |
| projekt01 | fdrake, No, after your changes we need it for generate the package-includes ;-) | 18:48 |
| fdrake | ah, that's right | 18:48 |
| *** Soulraven has quit IRC | 18:48 | |
| fdrake | we've been using the site.zcml to include what we want instead of using package-includes/ | 18:50 |
| projekt01 | Yes, but what do you reference? The old package-includes is gone? You have to build it with build_ext? Or do I get something wrong? | 18:52 |
| srichter | benji_york: I know, but we only need th hook for it | 18:52 |
| fdrake | we use <include package="what.we.need"/> | 18:52 |
| fdrake | instead of <include package="zope.app"/> | 18:53 |
| benji_york | srichter, ah, ok, so you'd add a hook to use a different parser, ok, that might pass | 18:53 |
| fdrake | and that package includes zope.app and whatever else it requires | 18:53 |
| fdrake | so package-includes/ is left for optional things that aren't relevant to the core application, should a customer be so bold. | 18:54 |
| fdrake | so an instance contains data, zope.conf, and just enough configuration to load the application | 18:55 |
| fdrake | the application is "assembled" in a package specific to the project | 18:55 |
| projekt01 | Ok, how do you assemble them? By hand? | 18:56 |
| fdrake | everything is in code, and reproducible | 18:56 |
| fdrake | the generations framework has an "install" phase that we use | 18:56 |
| projekt01 | generations framework? | 18:57 |
| fdrake | zope.app.generations | 18:58 |
| *** sidnei has joined #zope3-dev | 18:59 | |
| projekt01 | fdrake, How do you "assamble" your workspace for a new project? | 18:59 |
| fdrake | we create a new Subversion project that includes a build script, and tell it what versions of what packages to use | 19:00 |
| fdrake | then it loads those | 19:00 |
| fdrake | project specific code lives in the new Subversion project | 19:01 |
| projekt01 | Ah, Ok, you setup a repository first where contains something like tags of apps in other repositories | 19:01 |
| fdrake | when Zope starts, the initial generation installs itself into the database using the phase | 19:01 |
| fdrake | yes | 19:01 |
| projekt01 | The management of this repos is done by a script | 19:01 |
| fdrake | using externals, tarballs, etc. | 19:02 |
| fdrake | the script only control "building out" | 19:02 |
| fdrake | there are recipes for different software | 19:02 |
| projekt01 | You mean the script will checkout and setup your local workspace? | 19:02 |
| fdrake | configuration files say what to build in what order | 19:02 |
| fdrake | the process is 1) checkout project, 2) run build script | 19:03 |
| fdrake | the build script gets the other projects and builds them | 19:03 |
| fdrake | it also knows how to update them for things in revision control repos | 19:04 |
| *** sidnei has left #zope3-dev | 19:04 | |
| projekt01 | Yeah, that sounds good. Will this become a addon for zpkg? Somthing like a dev-zpkg ;-) | 19:07 |
| projekt01 | ;-) | 19:07 |
| fdrake | it's not using zpkg at this time; zpkg really just drives distutils | 19:08 |
| fdrake | this needs to deal with external C libraries and the like as well | 19:08 |
| benji_york | projekt01, there has been some thought about having the buildout use zpkg metadata, but we haven't figured out exactly what and how yet | 19:09 |
| fdrake | we'll be looking at ways to incorporate zpkg so we can feed additional information into this, but haven't had time yet | 19:09 |
| benji_york | yeah, what fred said | 19:09 |
| benji_york | :) | 19:09 |
| projekt01 | I think it will be nice to support the "setup project workspace and commit" part in the future with additional tools or scripts. | 19:11 |
| *** tarek has quit IRC | 19:11 | |
| fdrake | perhaps; I suspect that could be as little as "svn copy svn+ssh://.../emptyProject svn+ssh://.../myNewProject" though :-) | 19:12 |
| projekt01 | I think about a file in xml or conf format where you can easy describe this checkouts, versions etc. for a project. | 19:16 |
| projekt01 | And a application where does the checkouts by the information in this file. | 19:16 |
| benji_york | that seems outside the scope of zpkg (what little I know about it) | 19:17 |
| *** mexiKON has joined #zope3-dev | 19:17 | |
| *** regebro has joined #zope3-dev | 19:18 | |
| fdrake | that's definately beyond what zpkg was intended to do | 19:18 |
| fdrake | and similar to what's done in the configuration file for Benji's buildout script | 19:18 |
| fdrake | it's a INI file, not an XML file, so we can even tell what it says :-) | 19:19 |
| projekt01 | Benji_york, Yes, it has not much to do with zpkg. I describe more a "project relevant workspace assamble tool" | 19:19 |
| projekt01 | Is this script available? | 19:19 |
| benji_york | not presently | 19:20 |
| projekt01 | Ok | 19:20 |
| *** bradb is now known as bradb-lunch | 19:21 | |
| fdrake | now, one thing that would make it particularly interesting to release the build script to the community would be that we could more easily share build recipes | 19:22 |
| benji_york | true | 19:22 |
| fdrake | everyone could contribute recipes if we shared the tool that used them | 19:22 |
| fdrake | something to mention to mgmt if/when there's time to discuss open-sourcing the buildout stuff | 19:23 |
| projekt01 | Yeah, I think there is defently some need for this in bigger projects. | 19:24 |
| *** philiKON has quit IRC | 19:24 | |
| fdrake | indeed; we've been using variants on the idea for years, but I think the current incarnation is a lot nicer to deal with | 19:28 |
| fdrake | we were using a system based on "make" for a long time, and that was painful (IMO), and certainly didn't work on Windws | 19:29 |
| fdrake | the new system does, which is nice | 19:29 |
| projekt01 | +1 from a windows user ;-) | 19:29 |
| fdrake | poor guy :-) | 19:30 |
| projekt01 | Yes, poor of time to learn this emac or vim stuff ;-) | 19:31 |
| fdrake | heh. learning emacs/xemacs/vim is easier than making notepad a good working environment :-) | 19:32 |
| benji_york | speaking as a lover of vim, and a recent (about a year) Linux convert, I will point out that there are some very good non-vim, non-emacs editors for Linux (and some are cross platform so you can use them on Windows too) | 19:35 |
| fdrake | yeah, there are other editors that are full-featured | 19:36 |
| projekt01 | Yup, but then we don't have a developer how is running the tests if I switch to linux ;-) | 19:37 |
| benji_york | just run both Linux and Windows, that's what I prefer | 19:39 |
| *** MJ has quit IRC | 19:40 | |
| *** lunatik has joined #zope3-dev | 19:49 | |
| *** lunatik has left #zope3-dev | 19:52 | |
| *** faassen has quit IRC | 19:58 | |
| *** stub has joined #zope3-dev | 20:24 | |
| *** MJ has joined #zope3-dev | 20:25 | |
| *** bradb-lunch is now known as bradb | 20:36 | |
| *** d2m has quit IRC | 20:37 | |
| srichter | benji_york: are you there? | 20:45 |
| srichter | I think mechanize does a mistake by searching for all links | 20:45 |
| srichter | at least for our use case we usually look up only one | 20:46 |
| benji_york | yep | 20:46 |
| benji_york | ok, does that look easy to fix? | 20:46 |
| srichter | so I am going to try to reimplement testbrowser's getLink not supporting regex but use xpath | 20:46 |
| srichter | it's only easy if pullparser returns an iterator and not a list | 20:47 |
| srichter | it does return an iterator | 20:47 |
| srichter | so you could fix it on the mechanize level | 20:48 |
| srichter | but I also want to see how much faster libxml2 and XPath will be | 20:48 |
| srichter | if it faster, then I can start refactor mechanize accordingly | 20:48 |
| *** regebro has quit IRC | 20:49 | |
| benji_york | ok | 20:49 |
| srichter | I just started developing some high-level HTML analyzation code for ST | 20:50 |
| srichter | and the first thing I did was to support XPath queries :-) | 20:51 |
| srichter | no need for big print statements anymore | 20:51 |
| fdrake | analyzation? "analysis", perhaps? | 20:51 |
| fdrake | :-) | 20:51 |
| benji_york | hmm, interesting... we've been thinking about that for a while, still not sure what form it should take (xpath, xslt, etc.) | 20:51 |
| srichter | xpath for selecting the elements you are interested in reviewing | 20:52 |
| srichter | it is a very short syntax and thus very suitable for tests | 20:52 |
| *** stub has quit IRC | 20:52 | |
| benji_york | srichter, yep, we've used it for that in Selenium tests, but the interesting question is what should the xpath query *return* | 20:56 |
| srichter | mine returns the serialized version of all found nodes | 20:58 |
| srichter | I cannot return libxml2 nodes easily, since I have to free the document | 20:59 |
| srichter | I'll wait for lxml to return nodes | 20:59 |
| hazmat | srichter, you can programtically use selenium as well in unit tests in the same manner as mechanize | 21:02 |
| srichter | but selenium requires a real browser, no? | 21:05 |
| srichter | I don't want to depend on any real browser | 21:05 |
| srichter | and I have heard selenium is very slow | 21:05 |
| fdrake | selenium is slow, since the browser loads all the "extra" stuff for pages (style sheets, images, etc) | 21:08 |
| fdrake | however, it lets you test in a real browser | 21:08 |
| fdrake | which means things like JavaScript are interpreted and tested | 21:09 |
| fdrake | that can be useful if you use JavaScript at all | 21:09 |
| srichter | benji_york: okay, I have to get ready for hiking; I look at it some more over the weekend; we can definitely optimize mechbrowser's find_link by only looking for the first hit | 21:09 |
| srichter | fdrake: sure, the UI people can do that | 21:09 |
| srichter | but I just need to ensure system integrety | 21:09 |
| srichter | and some sensible UI | 21:10 |
| fdrake | must be nice not to be your own UI people :-) | 21:12 |
| benji_york | srichter, ok have a good time | 21:13 |
| GaryPoster | srichter: in case this affects you, John J. Lee hopes to integrate our patches this weekend | 21:13 |
| *** RaFromBRC has joined #zope3-dev | 21:18 | |
| *** gintas has joined #zope3-dev | 21:39 | |
| *** RaFromBR1 has joined #zope3-dev | 21:48 | |
| *** gintas has quit IRC | 21:48 | |
| *** RaFromBRC has quit IRC | 21:49 | |
| *** RaFromBR1 is now known as RaFromBRC | 21:49 | |
| *** alienoid has joined #zope3-dev | 21:58 | |
| *** RaFromBRC has quit IRC | 21:59 | |
| *** MrTopf has joined #zope3-dev | 22:00 | |
| *** mkerrin has quit IRC | 22:04 | |
| *** alienoid has quit IRC | 22:17 | |
| *** _projekt01 has joined #zope3-dev | 22:52 | |
| *** projekt01 has quit IRC | 22:52 | |
| *** Aiste has quit IRC | 22:54 | |
| srichter | benji_york: ok, I still have some time now; :-) I just made the link lookup lazy and it decreased my test time from 16.2 secs to 13.6 secs (16%) | 22:55 |
| srichter | I clean it all up a bit and then send you the diff | 22:55 |
| *** mgedmin has quit IRC | 22:59 | |
| *** benji_york has quit IRC | 23:01 | |
| *** BjornT has quit IRC | 23:03 | |
| GaryPoster | srichter: send me the diff, if you could | 23:09 |
| GaryPoster | (that is, send it to gary, rather than or in addition to benji) | 23:11 |
| srichter | gary at zope dot com? | 23:11 |
| GaryPoster | yup | 23:11 |
| srichter | ok | 23:11 |
| srichter | I still get a failure now, so I have to see what happened here | 23:11 |
| *** MrTopf_ has joined #zope3-dev | 23:11 | |
| GaryPoster | no prob. When you can. John J. Lee said he wanted to do the mergs this weekend. | 23:12 |
| srichter | right, that's the reason I want to get it to you, but it might not happen | 23:13 |
| srichter | I wish mechanize would have tests | 23:13 |
| *** MrTopf has quit IRC | 23:13 | |
| GaryPoster | hm, that's not good that it doesn't. timeline: if you get it to me by 6--that is, in an hour and 45 minutes--that should still give me enough time to get it to him this evening. Earlier would be better, but 16% sounds nice. :-) | 23:15 |
| *** ignas has quit IRC | 23:16 | |
| srichter | GaryPoster: I have to leave for my trip in 30 mins | 23:17 |
| GaryPoster | :-) ok | 23:17 |
| srichter | so you either have it then or not till Sunday | 23:17 |
| GaryPoster | ok. If Sunday, probably want to save it up with maybe some other accumulated changes for John later. | 23:18 |
| GaryPoster | (if he merges this weekend as he hopes) | 23:18 |
| srichter | is there a way to destroy an iterator? | 23:21 |
| GaryPoster | you mean for garbage collection? | 23:22 |
| srichter | yeah | 23:24 |
| srichter | I basically have to destory the iterator if I found the item I was looking for | 23:24 |
| srichter | (at least that's what I think causes the failure) | 23:24 |
| *** bskahan has quit IRC | 23:25 | |
| fdrake | just discard it | 23:25 |
| GaryPoster | del isn't working? | 23:25 |
| srichter | mmh | 23:25 |
| srichter | its really strange | 23:25 |
| srichter | anyways, I might have a different solution | 23:26 |
| GaryPoster | BTW, you know we need to stick to Py 2.0.1, right? At least for ClientForm | 23:26 |
| fdrake | perhaps something else is hanging on to the iterator? discarding should be sufficient | 23:26 |
| srichter | fdrake: yes | 23:26 |
| GaryPoster | Don't remember the target for the other packages | 23:26 |
| srichter | the iterator requires a file | 23:26 |
| srichter | GaryPoster: I am working on mechanize | 23:27 |
| GaryPoster | OK, and the mechanize target is...uh, whenever iterators came around? | 23:27 |
| srichter | 2.2? | 23:27 |
| GaryPoster | Sounds right, but I'd have to look it up, not sure | 23:28 |
| srichter | fdrake: I create an iterator that iterates through a file; when I first find the link I return it | 23:28 |
| srichter | then I call this same iterator again | 23:28 |
| srichter | looking for a link | 23:29 |
| srichter | this same link is not found anymore | 23:29 |
| srichter | I dunno why | 23:29 |
| fdrake | you create one iterator and iterate twice? | 23:31 |
| srichter | no, I call the function that returns the iterator twice | 23:32 |
| fdrake | with the same open file object? | 23:32 |
| srichter | yes | 23:32 |
| srichter | actually I am having a generator | 23:32 |
| srichter | (sorry, I always mix the two up) | 23:32 |
| fdrake | the iterator would start at the current position in the file unless you seek(0) the file | 23:33 |
| srichter | oh, never mind | 23:33 |
| srichter | some other thing goes wrong | 23:33 |
| srichter | right, we always seek to 0 | 23:33 |
| srichter | actually, I found the bugger | 23:35 |
| srichter | GaryPoster: pullparser uses iterators | 23:35 |
| srichter | so I think we are fine | 23:35 |
| GaryPoster | I looked up mechanize: it is 2.2. ClientForm is 2.0. Didn't look/don't know the others. They are all separate | 23:36 |
| srichter | GaryPoster: would it be okay for you if I check in the improvement to SVN? | 23:36 |
| GaryPoster | Yes, that would be fine | 23:36 |
| srichter | we have a private copy of mechnize anyways | 23:36 |
| GaryPoster | Right | 23:36 |
| *** tvon is now known as TomCruise | 23:44 | |
| *** TomCruise is now known as tvon | 23:45 | |
| srichter | ok, now I am really gone | 23:45 |
| srichter | have a great weekend | 23:45 |
| GaryPoster | ok, have fun. You too! | 23:45 |
| *** d2m has joined #zope3-dev | 23:59 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!