- Zope - Zope lists

permissions for "propertyItems()"
by Heiko Stoermer 27 Aug '99

27 Aug '99

anyone know how I can find out which permissions I have go grant a user to let him view a document that implements a call to "propertyItems()"? -- Heiko Stoermer MIG Augsburg

1 0

Random Advert Generation
by wombat_harness＠hotmail.com 27 Aug '99

27 Aug '99

Hey everyone, I have just installed Zope 2 and imported my almost complete site into it. However all my random content generators have gone 'arghhhhhhhhh'. I followed the commands on www.zope.org for producing random content but this doesn't seem to work in zope 2. Here is the code i am using.  i have got a property called randomMontage set up of type 'token' with the values: montage1.jpg montage2.jpg montage3.jpg montage4.jpg montage5.jpg now on the old version, a random image would be pulled out and displayed but i am getting a '__CALL__' error now, Does anyone have any ideas, Cheers, Wom.

1 0

Re: [Zope] Idea: ZTest: Integrated Use-case based web site testing.
by David Kankiewicz 27 Aug '99

27 Aug '99

Having stayed current with all the email in most of the Zope mailing groups (about two months) it seems to be one tool that everyone sees as future/present solution, depending on ones needs. Most comment that it needs this or that, a Scheduler as come up before. Someone said make a web page and call the URL from an external program.. Its would be nice to have a simple interface to something like this and nicer have management without the need to access the underlying system. I, much like this person, have an over active imagination for zope products and would like to at least try my (very new python) hand at creating a Scheduler. (I know anyone could do it a lot faster then me if they had the time). As this case points out it could be used to do almost anything, from checking to see if everything is working (if someone implemented it) to ZCataloging everything not ZCatalogAware on a night/weekly basis (something any site might want to do). And every product Zope gets makes it that much easier to work within odd situations. So I'm going to get together the requirements, functionality, and see if I can implement and learn a little more about python in the process, and Zope for that matter. If someone wants to suggest use cases for a Scheduler or ZTest (I'll probably never get to programming it), I'll maintain the ideas and maybe someone will make use of them in the future. At 11:37 AM 8/27/99 +1000, Jay, Dylan wrote: >To much caffeine and some morning reading >(http://www.rational.com/products/rup/prodinfo/whitepapers/dynamic.jtmpl?doc >_key=101057) has given me a bit of a brainwave. >When building complex dynamic site testing is very important. I've created >use-cases and test plans for my web-site. I've investigated automatic site >testing tools a little bit but have found them a) expensive b) not very good >when it comes to testing dynamic content and getting past authentication. >What if a Zope site could test it self regularly and just mail the webmaster >when something is wrong? Sense authentication is internal and if ZTest was setup to take this into account it would be easy to test most objects/methods. > >So what I envisage is test-case objects that have a specification of a >certain scenario and expected results. It would be a series of URLs with >parameters to be given and a specification of what the requested page should >look like. Since this is internal to Zope then some information about what >is interesting and what is not can probably be leveraged. Such ZTestCase >objects can be created either by instrumenting the site and then the you >step through the pages or can be edited manually. Specified code could be >executed before and/or after each test case to set and reset test >conditions. ZTestRun objects can then be created which specify a whole range >of test-cases to run. I see this as manageable as long as the object being tested was not interacting with databases (unless that part of the system was redirected to the ZTest or rolled back to pretest condition or restricted to read only, that is above what I could do right now). > >Another idea I had involves the creation of a scheduler inside Zope so that >specified Zope code could be executed in the future or periodically. >Caching, indexing, retrieving external data, backup etc and of corse >scheduling running of test cases. > As I said up top, I'm going to look into this shortly.. >Unfortunately for me I seem to have way more ideas than I have time so I >don't know when I would get time to try implementing it so I thought I'd >bounce it out into the community to see what others thought. The Zope community really should have a more formal method of cataloging Ideas that are suggested and rate the interest and try to set some priorities, informal of course. Like in the ZCommerce mailing list, ideas flow fast! :) hope that continues to progress.. Thanks for the interest in and ideas for Zope, David Kankiewicz > >______________________________________________ >Zope maillist - Zope(a)zope.org >http://www.zope.org/mailman/listinfo/zope > >(To receive general Zope announcements, see: >http://www.zope.org/mailman/listinfo/zope-announce > >For developer-specific issues, zope-dev(a)zope.org - >http://www.zope.org/mailman/listinfo/zope-dev ) >

4 3

RE: [Zope] Idea: ZTest: Integrated Use-case based web site testin g.
by Rob Page 27 Aug '99

27 Aug '99

> Once I was thinking about a scheduler object for Zope and actually > made the first steps and designed it around Tim O Malleys(?) eventloop > module. However I gave up when I considered autorization > details. Do you > have any good ideas on how to deal with authorization? What about giving the scheduler object Proxy Roles like you can for other documents..?? --Rob

1 0

RE: [Zope] Idea: ZTest: Integrated Use-case based web site testin g.
by Alexander Staubo 27 Aug '99

27 Aug '99

I agree that it would be nice -- I've been thinking of it earlier, then submitted to chron, because I'm lazy -- I'm just saying that we do have the tools, and running ZClient externally really isn't that different. Ok, I'm sold. I disagree slightly about your event object design though. Imho the event object should be similar to a DTML Method -- so the event "action" would in fact be DTML. The immediate benefit is that you can do anything -- you're not restricted to launching URLs, and you get the Zope context in the box. You could allow passing the calling event object to other parts of the site. If event objects descended from DTMLMethod, you could also use event objects just like DTML Methods and call them up -- manual execution if you will. The scheduler should support concurrency, not just queueing. In fact, all we need is a simple "max concurrent tasks" semaphore to regulate the thread count from the master thread. A scheduler also needs a logging/reporting system. A history API would be cool -- you could so <dtml-in "MyScheduler.pastEvents()"> and build your own schedule reports. Maybe all output from the event DTML should be attached to the history entry. And so on. Who writes this thing? This being an open source initiative(TM), I'm all for chipping in. Anyone want to join the Scheduler Task Force? -- Alexander Staubo http://www.mop.no/~alex/ "Give me an underground laboratory, half a dozen atom smashers and a beautiful girl in a diaphanous veil waiting to be turned into a chimpanzee, and I care not who writes the nation's laws." --S. J. Perelman > -----Original Message----- > From: Martijn Pieters [mailto:mj@antraciet.nl] > Sent: 27. august 1999 09:47 > To: Alexander Staubo > Subject: RE: [Zope] Idea: ZTest: Integrated Use-case based web site > testin g. > > > At 08:53 27/08/99 , you wrote: > >Aside: Do you really need this within Zope if you can run a ZClient > >script with chron? While a scheduler within Zope isn't a bad > idea, the > >infrastructure -- at least on Unix and NT -- is already there. > > I do see the use of a web-administrable scheduler for Zope. > This way users > without telnet to the server can schedule calls to Zope > objects as well. It > would also make cross-platform, easily transportable > scheduled applications > possible. Just imagine: You develop on an NT server, then > just copy over > Data.fs to a Sun server at the ISP, and all scheduled calls > are set up as > well.. No fuss, no hassle. > > A scheduler object would need: > * Permissions, so a site maintainer can determine who can > schedule calls. > * A separate thread to manage it's queue. > > Scheduled events would need: > * Interval settings (one time only, repeat, number of > repeats, repeat interval) > * URL to call > * Security settings > * Call parameters (including call type, GET or POST) > * Maybe a file field to upload? > > -- > Martijn Pieters, Web Developer > | Antraciet http://www.antraciet.nl > | Tel: +31-35-7502100 Fax: +31-35-7502111 > | mailto:mj@antraciet.nl http://www.antraciet.nl/~mj > | PGP: http://wwwkeys.nl.pgp.net:11371/pks/lookup?op=get&search=0xA8A32149 ------------------------------------------

3 2

RE: [Zope] Semi-persistent objects, volatile attributes
by Alexander Staubo 27 Aug '99

27 Aug '99

> Alexander Staubo wrote: > > > > Is it possible to have persistent objects with attributes > that are only > > persistent in memory? > > Yes, but it takes some work and the mechanism need to be formalized. > Would you want to specify this on a per-class or per-instance basis? That would be great. Personally? Per class. But wouldn't it be possible to do something like self._p_nodeactivate = 1 or was that being too naive? :) [snip] > > > > def my_method(self): > > self._v_foo = (hasattr(self, '_v_foo') and self._v_foo) or > > '<default>' > > This is one way. > > > ...ad tedium if you have lots of methods. > > Yup. > > > Does the persistence engine > > support an "onload" mechanism which you can override to > know when your > > instance has been loaded and needs to initialize itself? > > Yup, it's called __setstate__. __setstate__ is called when > the object state is loaded. You can overload __setstate__ > and initialize any volitile variables when you load your state: > > > class C: > def __setstate__(self, state): > C.inheritedAttribute('__setstate__')(self, state) > self._v_spam=..... Just what I was looking for. Yum, thanks. For future reference and posterity's sake, is this stuff documented anywhere except in the UML documentation? > Jim > > -- > Jim Fulton mailto:jim@digicool.com Python Powered! > Technical Director (888) 344-4332 http://www.python.org > Digital Creations http://www.digicool.com http://www.zope.org -- Alexander Staubo http://www.mop.no/~alex/ "Give me an underground laboratory, half a dozen atom smashers and a beautiful girl in a diaphanous veil waiting to be turned into a chimpanzee, and I care not who writes the nation's laws." --S. J. Perelman

1 0

managing properties dynamically
by Heiko Stoermer 27 Aug '99

27 Aug '99

another one from me... :-) I have a form that contains input fields for some values (say 'header' and 'text'). After submitting the form I have to manage_changeProperties() of an object that has the above properties, but also many more. Of course only the property values entered in the form are supposed to be changed. now comes the tricky bit: the method that is supposed to change properties does know the target object, but it does not yet know which properties are to be changed. These are defined dynamically in the preceding form (and there exist various of these). In theory the solution is pretty simple: <with target object> <in propertyItems() of the target object> <if there is a corresponding form field in REQUEST> --> manage_changeProperties(current property = REQUEST[current]) </if> </in> </with> The arrow indicates the "problem area": I have the name of the property (and thus the name of the input field) in a variable called "current". if if put  nothing happens. Zope simply runs over the code and does _nothing_. What goes wrong here, and how do I get it running? Thanks so far, Heiko -- Heiko Stoermer MIG Augsburg

1 0

Zope 2.0.0b5 released...
by Jim Fulton 27 Aug '99

27 Aug '99

Zope 2.0.0 beta 6 is available at: http://www.zope.org/Download/Releases/Zope-2.0.0b6 This release has a number of minor bug fixes. -- Jim Fulton mailto:jim@digicool.com Python Powered! Technical Director (888) 344-4332 http://www.python.org Digital Creations http://www.digicool.com http://www.zope.org Under US Code Title 47, Sec.227(b)(1)(C), Sec.227(a)(2)(B) This email address may not be added to any commercial mail list with out my permission. Violation of my privacy with advertising or SPAM will result in a suit for a MINIMUM of $500 damages/incident, $1500 for repeats.

2 2

Zope 2.0.0b6 released...
by Jim Fulton 27 Aug '99

27 Aug '99

Zope 2.0.0 beta 6 is available at: http://www.zope.org/Download/Releases/Zope-2.0.0b6 This release has a number of minor bug fixes. Jim -- Jim Fulton mailto:jim@digicool.com Python Powered! Technical Director (888) 344-4332 http://www.python.org Digital Creations http://www.digicool.com http://www.zope.org Under US Code Title 47, Sec.227(b)(1)(C), Sec.227(a)(2)(B) This email address may not be added to any commercial mail list with out my permission. Violation of my privacy with advertising or SPAM will result in a suit for a MINIMUM of $500 damages/incident, $1500 for repeats.

1 0

Semi-persistent objects, volatile attributes
by Alexander Staubo 27 Aug '99

27 Aug '99

Is it possible to have persistent objects with attributes that are only persistent in memory? Effectively, an object that is always cached, for volatile attributes to be preserved during the Zope process' execution. Is there a way for an object to load itself permanently into the cache, say? I figure I could store the attributes in the product module as part of a dictionary keyed by the object instance, but this feels very dirty. Speaking of volatile attributes, is there an idiom for handling these? Since they're not stored, and __init__() is only ever called when the object is initially created (not persistently loaded), it seems you have to verify their existence, something like: def my_method(self): self._v_foo = (hasattr(self, '_v_foo') and self._v_foo) or '<default>' ...ad tedium if you have lots of methods. Does the persistence engine support an "onload" mechanism which you can override to know when your instance has been loaded and needs to initialize itself? Pointers to appropriate documentation would be nice. I've been rummaging through the ZODB UML stuff and while it's enlightening, the documentation is also quite fragmented. -- Alexander Staubo http://www.mop.no/~alex/ "Give me an underground laboratory, half a dozen atom smashers and a beautiful girl in a diaphanous veil waiting to be turned into a chimpanzee, and I care not who writes the nation's laws." --S. J. Perelman

2 1