- Zope-Dev - Zope lists

RE: [Zope-dev] CST 0.8 Updater Fixed...
by Schmidt, Allen J. 27 Apr '01

27 Apr '01

Sorry to contradict you Chris, but it was actually Sgt. Shultz who stated that famous line. It's not often I can contribute so I thought I would give something back! <8^) Allen (active member of ZA - -Zopaholics Anonymous) -----Original Message----- From: Chris McDonough [mailto:chrism@digicool.com] Sent: Friday, April 27, 2001 12:21 PM To: Chris Withers; zope-dev(a)zope.org Subject: Re: [Zope-dev] CST 0.8 Updater Fixed... > I stopped the above by wrapping the code as follows: > <snip> Cool! > I wonder why it's necessary? Something to do with the weirdy mounted internal > session data container? As Colonel Klink says so well, "I see nussing!" _______________________________________________ Zope-Dev maillist - Zope-Dev(a)zope.org http://lists.zope.org/mailman/listinfo/zope-dev ** No cross posts or HTML encoding! ** (Related lists - http://lists.zope.org/mailman/listinfo/zope-announce http://lists.zope.org/mailman/listinfo/zope )

2 1

Updaters for ZCatalog and Core Session Tracking Available
by Chris Withers 27 Apr '01

27 Apr '01

Hi, I've done some Updaters to help with the move to the new ZCatalog and Core Session Tracking implementations. These updaters help you quickly update all objects of a particular type in a ZODB. The and information about them can be found at: http://www.zope.org/Members/chrisw/UpdateSupport cheers, Chris

2 1

Re: [Zope-dev] 27 million objects.
by Dieter Maurer 27 Apr '01

27 Apr '01

Erik Enge writes: > I've just finished adding a somewhat small number of objects: 5000. > For every 1000th object, the Data.fs seemed to grow to about 900MB; that's > when things started going slow, in a non-linear fashion (this is more a > hunch than something I payed much attention to). > > I paused the script (fancy Unix-command: "^Z") for every 1000th object, > packed the database (which shrunk to 19.5MB! Hmpf.) and restarted the > script (again, fancy Unix-command: "fg"). Then I was back to the same > speed as I initially had. Are the imported object "CatalogAware"? This would imply that each object's import rewrites part of the catalog indexes and lets your ZODB grow drastically. The old (pre 2.3.1) catalog implementation was know not to be very storage friendly. If a significant portion of the catalog indexes would be affected by imports, then you would see a quadratic storage increase. The new BTree implementation in 2.3.1 might reduce the problem. Dieter

4 5

makerequest.py
by Chris Withers 27 Apr '01

27 Apr '01

Hi, This is quite a useful little utility, but it creates a REQUEST object without any AUTHENTICATED_USER beign present. Should this be remedied? If so, how? I do have memories that REQUEST.AUTHENTICATED_USER has been deprecated. In which, case, how should you get hold of the current user's object and can this be made to work in a testing environment? Nirvana here would be not having to import Zope so that we can run tests while Zope is still up and running in a non-ZEO situation. Does ZUnit help with this? If so, how? cheers, Chris

2 1

ZCatalog comments
by Chris Withers 27 Apr '01

27 Apr '01

Michel Pelletier wrote: > > + There are some rules to consider when querying this method: > + > + - an empty query mapping (or a bogus REQUEST) returns all > + items in the > + catalog. Is it only me that finds this really irritating? Surely a blank query mapping should return nothing, not everything? (at the least it'd be nice if this was configurable ;-) > + Depending on the type of index you are querying, you may be > + able to provide more advanced search parameters that can > + specify range searches or wildcards. These features are > + documented in The Zope Book. Wildcards? Hmmm... that's enticing, where and how will they be supported? cheers, Chris

3 2

CST 0.8 Question
by Chris Withers 27 Apr '01

27 Apr '01

Hi Chris, While writing the CST 0.7 -> 0.8 updater, I noticed something weird in SessionIdManager.py. In constructSessionIdManager(), you call SessionIdManager.__init__ with a final parameter called 'rewritemode'. I don't know what that's meant to do, but all I can see is that it results in a call to self.turnOn() when you create the manager. The name would suggest that something else should probably happen, but maybe rewritemode is just a relic from a previous version? Hope you can shed some light, cheers, Chris

2 1

27 million objects.
by Erik Enge 27 Apr '01

27 Apr '01

Hi, Michael. Sorry for taking so long getting back to you. The programmer solving our problems with the post codes has solved it in a different way than what I would've done (his method is way superior), so we're not ending up adding all addresses as Zope Objects. Therefore, I don't have any benchmark tests available. We are going to transfer some 10GB of data at a later stage though (within a month), and that could result in some tests being done - if so, I'll send you an email. :-)

6 16

Trying to set roles/permissions programmatically - Now I have a headache!
by Danny William Adair 27 Apr '01

27 Apr '01

Hello! I want to set the roles of a permission and turn acquisition of a permission (mapping) on and off programmatically. And of course I want to get a permission by its name. This looks definitely more complex than simply changing an object's property! I took a look at the management screens(Role.py -> manage_changePermissions). The aX and pYrZ mapping for the *checkboxes* (=sole purpose) gave me a lot of confusion, I first thought this was Zope internal naming :-). Then I took a look at Permission.py (setRoles). But I haven't dealt with this before, and it all looks a little weird. I managed to write a couple of functions that give me roles to permissions and permissions to roles or check if a permission setting has been acquired and so on, but I'm stuck with *setting* permissions with a function of my own. a) Could this be correct? # roles as list: acquisition will be turned on roles=['Anonymous','Manager'] # roles as tuple: acquisition will be turned off roles=tuple(roles) # set as specified some_permission.setRoles(roles) b) How can I get a hold of some_permission when I know its name? This does not work: def setRolesForPermission(ob, permission, roles, setacquired): permission_trans = '_' + replace(permission, ' ', '_') + '_Permission' p = getattr(ob, permission_trans) # this does *not* give me the permission itself! if not setacquired: roles=tuple(roles) p.setRoles(roles) It must have something to do with acquisition :-) If it is turned on for that permission, getattr() won't find the attribute. If it's turned off, it returns a tuple. Where's the permission instance itself? I see that ac_inherited_permissions(1) gives me a list of permission names, but I don't know how to access a specific permission directly. This *did* work: def permissionAcquired(ob, permission): # Tell me if acquisition has been turned on # (There's also some other way, checking if the returned is of type List or Tuple, I think I saw that somewhere in Role.py) permission_trans = '_' + replace(permission, ' ', '_') + '_Permission' return (not hasattr(ob.aq_base, permission_trans)) Help! Thank you so much in advance, Danny

1 0

Re: a plea for help (again) (tav)
by Tim McLaughlin 27 Apr '01

27 Apr '01

Tav, I don't see why you need to reimplement __repr__ in the first place. I believe that what yours is trying to do is the default implementation on a class. (somebody tell me if I am wrong) But this probably is not the fundamental issue. TIm

1 0

a plea for help (again)
by tav 27 Apr '01

27 Apr '01

sorry for sending this to the list again, but i would _really_ appreciate some help with the matter. thanks =) ---- heyas i created a zope product called, XnetPage, which is basically a folderish dtml document/zwikipage the source for the product can be seen at http://mu2.espnow.com/~tav/XnetPage.py it's been a bumpy ride, but everything's gone well so far. however, i have run into a problem which is beyond me. it is with regards to error messages. whenever, the text of the xnetpage is rendered, and there's an error... lets say that i call <dtml-var hmz> and hmz, hasn't been defined... instead of displaying a NameError, or whatever the appropriate error should be, it gives me a TypeError i even installed ZDebug, but, it's of no help here :/ in fact, the error message is over-ridden by the one in HTTPResponse.py i have included a sample error traceback below, and hope that someone can enlighten me as to what's wrong. most of the code in the product is from the zwiki product, and some of the voodoo magic is beyond me ;p thanks in advance ---- File /home/zope/k1/lib/python/DocumentTemplate/DT_String.py, line 538, in __call__ (Object: test_page) File /home/zope/k1/lib/python/Products/ZDebug/DTMLDebug.py, line 259, in debug_render_blocks File /home/zope/k1/lib/python/Products/ZDebug/DTMLDebug.py, line 228, in debugException File /home/zope/k1/lib/python/Products/ZDebug/DTMLDebug.py, line 155, in debugWithNamespace File /home/zope/k1/lib/python/Products/ZDebug/DTMLDebug.py, line 74, in explodeNamespaceStack File /home/zope/k1/lib/python/ZPublisher/HTTPRequest.py, line 935, in __str__ File /home/zope/k1/lib/python/Products/XnetPage/XnetPage.py, line 811, in __repr__ (Object: test_page TypeError: call of non-function (type string) ---- -- best regards, tav tav(a)espnow.com

2 2