April 2001 - Zope-Dev

Re: [Zope-dev] How do I call an HTMLFile in context provided by a path
by Tres Seaver 20 Apr '01

20 Apr '01

Itai Tavor <itai(a)optusnet.com.au> wrote: > Dieter Maurer wrote: > > >Itai Tavor writes: > > > P.S Now that aq_parent is working... why can't I do > > > absolute_url() on DTMLFiles :-( > >"absolute_url" is provided by the class "OFS.Traverable". > >Apparently, "DTMLFile" is not derived from it. > > > > > >Dieter > > Well, obviously... the question is why nobody saw the need to add > Traversable to DTMLFiles... DTMLFiles are supposed to function as *methods* of their classes; methods *don't* have their own 'absolute_url', but return the url of the object on which they are called. DTMLMethods have exactly the same semantics: <dtml-var absolute_url> returns the folder/context object's URL, by design. Tres. -- =============================================================== Tres Seaver tseaver(a)digicool.com Digital Creations "Zope Dealers" http://www.zope.org

2 1

Introducing a New Concept on Advanced Garbage Treatment Process
by zwj 20 Apr '01

20 Apr '01

Introducing a new concept on advanced garbage treatment process, and licensing patents For details, please reference the web site: http://zwj5382.3322.net http://zwj5382.163.net The inventor would like to assign exclusive patent license to anyone who is interested in patented advanced garbage treatment process technology. For patent license, please contact the inventor via E-mail: zwj88zww(a)371.net

1 0

ZSQL and Normalized databases (or why ZSQL sucks)
by The Doctor What 20 Apr '01

20 Apr '01

Does any one have an example of ZSQL being used witha normalized database? Or is ZSQL just useless? Near as I can tell, between: * Broken type marshalling * Loosing the variable between the form and dtml-if * Inability to handle table.field names for variables * And enough flexibility to work around the above problems This makes ZSQL extreamely nasty, and impossible to use with a normalized database. What's the point? ZSQL sucks, how do I talk to the DB directly? Yeah, I keep almost getting this to be useful. But damn it if I don't keep hitting a brick wall. And there are no complete examples or demos that I can find to load up. Irritatedly yours, DocWhat -- I don't want to achieve immortality through my work... I want to achieve it through not dying. -- Woody Allen The Doctor What: A really hip dude http://docwhat.gerf.org/ docwhat(a)gerf.org KF6VNC

5 10

Zope development firm in Canada
by John Stevenson, idrc.ca 19 Apr '01

19 Apr '01

Hello Zope-Dev - We are looking for a firm that can take on a large Zope developement project. We need to hire a Canadian firm to do most or all of the work. Please contact me at the email addres below. Thanks - John Stevenson -- Web Strategist International Development Research Centre, Ottawa johns(a)internet.idrc.ca http://www.idrc.ca

1 0

how to add to the pythonscript allowed import list?
by R. David Murray 19 Apr '01

19 Apr '01

I've got a little Product that does some init hacks. One of the things I want to do is expose a couple of python fuctions such that they can be imported into pythonscripts. After much spelunking in the mailing list and the PythonMethods wiki on zope.org, I *think* that what I need to do is something like this: ---------------------- from AccessControl import ModuleSecurityInfo security = ModuleSecurityInfo() security.declarePublic('SignedEditions') security.declarePublic('stripCardNum','verifyCardNumber') from cccheck import stripCardNum, verifyCardNumber security.apply(globals()) ---------------------- Now, that SignedEditions one is my attempt to solve the following error message when I attempt to do 'from SignedEditions import stripCardNum': Error Type: ImportError Error Value: import of "SignedEditions" is unauthorized However, it does not solve the problem. Hmm. I just noticed that I forgot to prefix that with "Products.". Which would seem to make that error message a bug, since SignedEditions shouldn't exist in the import path. If I do from Products.SignedEditions import stripCardNum then ZDebug tells me: Unauthorized: Access denied for <module 'Products.SignedEditions' from '/usr/local/zope/sites/signededitions/Products/SignedEditions/__init__.py'> because its container, <module 'Products' from '/usr/local/zope/Zope231b1/lib/python/Products/__init__.pyc'>, has no security assertions. What do I need to do to assert that it is OK to import from the SignedEditions product? The wiki does not seem to address this question at all, though it implies that it is possible, since it *does* talk about the above assertions to declare things *inside* the product as importable. (Oh, BTW, I tried changing 'security' to "ZopeSecurity", but that didn't seem to change the behavior). --RDM

2 3

Opening the Zope repository...
by Brian Lloyd 19 Apr '01

19 Apr '01

Hi folks. Feedback would be much appreciated on this, especially any potential issues we may have missed: http://dev.zope.org/Wikis/DevSite/Proposals/OpenTheZopeRepository Brian Lloyd brian(a)digicool.com Software Engineer 540.371.6909 Digital Creations http://www.digicool.com

2 1

Re: [Zope] CatalogAware does not work?
by Casey Duncan 19 Apr '01

19 Apr '01

John Morton wrote: > > > > Why couldn't CatalogAware do this? If nobody has a good reason, I might > > push forth a patch to CatalogAware to implement this. > > A simple case for the present flexibility is that you don't need to reindex > for every property change, just the ones that are actually indexed. > Agreed, CatalogAware should ideally check this to make sure reindexing is necessary. > A more complicated case would be an operation that changes several instances > at once. It's expensive to reindex for every instance you change, so it would > be useful to pass a flag to the ZClass's change method to tell it not to do a > reindex, and reindex the lot once the operation over all the instances is > completed. I have actually done this for a Product I am developing. It allows you to accumulate property changes and reindex only once. Supposedly, the Catalog is more efficient in this regard anyway as of 2.3. > I guess that you could build this kind of functionality into CatalogAware, > but I'm generally in favour of having flexible and optional rather than > automatic reindexing, and just adding documentation to discuss some of the > different cases. > > You could hack Property Sheet Interface to add a call to reindex if the > ZClass is catalog aware. I would not want to make PropertyManager directly aware of CatalogAware if I could get away with it. Perhaps the way to deal with this is to have PropertyManager call a hookable method afterChangeProperties or some such, passing a list of properties changed. Then your ZClass could simply call reindex_object if it feels it is necessary. That would give full disclosure and behavior control back to your specific implementation. But it would simultaneously cover property changes made in the management screens and everywhere else. I am moving this discussion over to zope-dev to see if anyone else there has any ideas. -- | Casey Duncan | Kaivo, Inc. | cduncan(a)kaivo.com `------------------>

1 1

SmartObjects - Developers needed
by Stephan Richter 19 Apr '01

19 Apr '01

Hello everyone, SmartObjects (at http://demo.iuveno-net.de/iuveno/Products/SmartObjects) is not longer only an idea but we are going into some serious design discussions and plan to have our first release at the beginning of June (see release schedule). Since we want to make the framework as flexible and user-friendly as possible, we would love to get input on the design and hear about new ideas. Furthermore, we need developers, who would like to help!!! The project is totally Open Source. So, if you are interested, go to http://demo.iuveno-net.de/iuveno/Products/SmartObjects and look around. If you like what you see, just sign up for the mailing list and join our discussion. Regards, Stephan -- Stephan Richter CBU - Physics and Chemistry Student Web2k - Web Design/Development & Technical Project Management

1 0

External transaction integration bug?
by Randall F. Kern 19 Apr '01

19 Apr '01

I may just be missing something obvious here, but it seems like there is a hole in ZODB.Transaction.Transaction.commit and Shared.DC.ZRDB.TM.TM, that can cause external transactions (those that use the TM mixin class, like psycopg) to be abandoned (never get committed or rolled back). Let's say somewhere around line 300 in Transaction.py (the call to j.commit(o, self)) we get an exception. Furthermore, let's say the TM derived database object has already been committed (the only effect of which is to move the DB into the jars mapping, since TM.tpc_begin() and TM.commit() both do nothing). The exception dumps us down to about line 353, where we call _p_jar.abort() on all the uncommitted objects (note: it's important that the database connection isn't in this list, because TM.abort() will rollback the transaction. the way we keep the db out of this list is by having it appear in objects _before_ the object that failed the commit). Next we reach line ~366, where we should "unwind TPC for the jars that began it". What this means is calling tpc_abort() on each jar from the objects that were already committed (which includes the database). However, TM.tpc_abort() does nothing, leaving the external database transaction open. Does this make sense? If this makes sense (i.e. seems bad :), does catching tpc_abort() in TM and calling TM.abort() seem like a valid fix? Thanks, -Randy

2 1

ZCatalog.indexes()
by Chris Withers 19 Apr '01

19 Apr '01

Hi, I read in the interfaces wiki (I think :-S) that the indexes() method of ZCatalog is being deprecated. Is this still accurate? If so, how _should_ I be asking a ZCatalog for a list of it's indexes? I'd also like to know their types if possible :-) cheers, Chris

2 1