March 2002 - Zope-Dev

Call Profiler strangeness - lots of ZODB transactions
by Richard Jones 08 Apr '02

08 Apr '02

I'm investigating why our product ZODB is growing when it shouldn't. I've found that the undo log has entries in it every five minutes saying "Installed product CallProfiler by Zope". More infrequently (not regularly) we also have "Installed product ZOracleDA by Zope". The Call Profiler was designed specifically not to hit the ZODB - it's in-memory only with. Our setup consists of seven ZEO clients talking to the one ZEO server - the one that's getting all these hits in the ZODB. Does anyone else see this happening? Anyone have any ideas why we're seeing this behaviour? Richard

2 3

ZCatalog - hiding query results
by Igor Stroh 02 Apr '02

02 Apr '02

Hi all, I don't know if it's the right list to post to, but I have the following problem: I have several objects (documents, folders etc) that are accessible only by a certain user role, this objects are cataloged. Now if I query the catalog the brains of these objects are returned correctly, but to _all_ users that issue a query. That means, users that don't have the permission to "View" or "Access Content Information" can see the brains as well... I tried to filter the result set by converting the brains into real objects (brain.getObject) in an external method (I thought, this way I should be able to exclude unauthorized users by adding the "skip_unauthorized" to the dtml-in), but that doesn't work since there are "brains" that are actually NoBrainer instances... Does anyone have an idea of how I could solve this problem? Actually I thought this kind of "information hiding" is supported by basic ZCatalog machinery, but now it looks like I'll have to hack a workaround... Any help greatly appreciated. Greetings, Igor

5 4

ZCatalog: hiding search results from unauthorized users - hack
by Igor Stroh 02 Apr '02

02 Apr '02

Hi all, in case someone have same problem as me (see "ZCatalog - hiding query results" thread for more info) - I got a solution: - create a product that subclasses ZCatalog - in this product overwrite ZCatalog.getobject with def getobject(self, rid): """foo""" obj = self.aq_parent.unrestrictedTraverse(self.getpath(rid),\ restricted=1) return obj - create a python script in your catalog (e.g. "filterCat") with a parameter that reoresents the catalog itself ("brains" here): retval = [] for brain in brains: try: obj = container.getobject(brain.getRID()) retval.append(brain) except: pass return retval - adjust your catalog query reports, so that they call "filterCat(_['<catalogname>'])" instead of <catalogname> Now your users see only those hits in a query which apply to objects they are allowed to "View". greetings, Igor

3 3

[Zope Crash Solution] Positive Report
by Stefano Noferi 02 Apr '02

02 Apr '02

Hi Zopers! Thanks all for your time, especially to Matthew and Leonardo. My patched Python/Zope systems run well now in multi-threads mode. I tested with very hard Python Scripts too and my report is at all positive. My relief is great, like many others zopers, I think ; ) Thanks again and Happy Easter! Stefano ---------------------------------------------------------- Stefano Noferi n o z e S.r.l. Soluzioni Open-Source Via Caduti del Lavoro, 32 56122 Pisa (PI) - Italy Tel: +39 (0)50 533320 Fax: +39 (0)50 526604 Email: stefano(a)noze.it Web: http://www.noze.it ----------------------------------------------------------

3 3

Re: [Zope-dev] ZClass Constructor Cleanup for 2.6
by Didier Georgieff 02 Apr '02

02 Apr '02

On Wed, 2002-03-27 at 17:31, Casey Duncan wrote: > ZClasses have been overlooked enough. Nobody has even bothered > putting a ZCatalogPathAware available for them... I've done that locally (changing __init__ and the the py but had no chance to make it work with __existing__ ZClasses. Something was coded inside the zclass instance ;-< So i had to patch ZCatalog for getting the right behaviour for my existing products. So, even if ZClasses are not cooked completely, they allow guys like me to use zope, to promote zopet even if we are useless at good python programming ;-), and until Z3 exists. All my support for ZCatalogPathAware in 2.6. Maybe we should investigate : - how updating old CatalogAware ZClasses for taking care of that - what impact for already updated construtor scripts ? (shuild be none, but not sure) Didier.

5 4

Simplification via custom protocol handlers: cvs://server/project, zope://foo ?
by Craeg K Strong 01 Apr '02

01 Apr '02

Hello: I am the author of the CVSFile product, available at http://www.zope.org/Members/arielpartners/CVSFile I am thinking about implementing a new design, and wanted to get some feedback from the list first (perhaps others have tried this approach?) Here is a description of our problem: ExternalFile is a Zope product that behaves like a DTMLDocument/method/etc but points to a file outside of ZODB for its content (sort of like a symbolic link) CVSFile inherits from ExternalFile, where the content file is stored in a CVS sandbox. It features buttons to do updates, commits, etc. So far, so good. The problem is that we are working on a third Zope product whose instances represent XML documents that automatically render themselves to HTML (or whatever) when referred to. You edit the XML content using ZMI or whatever, but when you navigate there, you see the resulting web page/PDF/etc. We are going to call it something along the lines of XSLTFile because it uses XSLT to do the conversion. Here's the issue: the XML file that is the source of the transformation may or may not reside inside the ZODB. For example, we store most of our XML documents in CVS. But if revision control is not important (such as for one-off or throw-away documents) we keep them in ZODB. How did we do this? Well we had XSLTFile inherit from CVSFile. This is bad. What happens when we switch to some other revision control tool (like clearcase, subversion, sourcesafe)? Either you have multiple classes like "CVS-XSLTFile, Clearcase-XSLTFile, etc." or a single class that knows about every possible version control system. yuk. Clearly, inheritance and mixins are not the answer. Instead we want delegation. How to do this? One idea is to register a custom protocol handler with python, using urllib2. See: http://www.python.org/doc/current/lib/module-urllib2.html install_opener() I am imagining the following. a new XSLTFile instance is created, pointing to a URL for its content. Here are some examples of URLs: cvs://myserver/myproject/my/file#VERSION-10 zope://foobar http://www.foo.bar/myfile In this way, the XSLTFile instance knows nothing about CVS or any other revision control system. Instead, it only knows about a URL, and the CVS protocol handler would know about how to retrieve files from CVS. We could create our own scheme where a query parameter would specify the sandbox name or something. The Zope protocol handler idea is interesting, and I haven't thought this all the way through yet. Could <dtml-var blat> be thought of as referring to "zope://blat" where the Zope protocol is the default and therefore omitted? The idea of unifying all object references through URIs is intriguing to me. (for more info, see REpresentational State Transfer: http://conveyor.com/RESTwiki/moin.cgi ) Comments? --Craeg

3 3

Re: [Zope] Having to keep typing "./start &"
by Lennart Regebro 31 Mar '02

31 Mar '02

From: "Douglas Carnall" <dougie(a)carnall.org> > If I am right, then this is an FAQ. Each time I answer it, I plaintively > wonder aloud why the -D flag would be the default, but I am generally > ignored. One reason is that without it, the server prints out nothing at all at startup, which is confusing to new users. To solve this two things need to be changed: 1. There need to be a possibility to set the error severity just like with stupid_log_file, even for printouts to stdoutput. Maybe the stupid_log_file could be made to be the standard way of logging, and it could log to stdout if no stupid_log_file was set? 2. The default level for stupid_log_file should be to print INFO and higher messages. Today the default is only to print PROBLEM and higher.

3 4

Re: [Zope-dev] Medusa 0.5.1 maint release, planned for inclusion in Zope?
by Jeff Kowalczyk 30 Mar '02

30 Mar '02

[Chris] Does Zope 3 use Medusa at all? [Stephan] No. Shane rewrote all of the HTTP Server and I am writing right now the FTP server (which copies some of the Medusa code, but much is cleaned). We added interfaces and made it more component-based. [Jeff] That's good news about an interface refactoring, it was a little confusing researching the present customized version. My interest in (the Medusa revision) was to keep alive the idea of setting up hooks for developers to build support for other protocols (IMAP, NNTP, Jabber, SMTP, SMS etc.) in a standard, configurable way in Zope3. Will your rewritten components form on a general-purpose server framework like Medusa? Ideally, the support would be as easily added as other 3rd party Zope3 product/components, inheriting from the base components you guys are writing now. The topic kind of thudded when I brought it up last week, but I think it is important to keep an open mind about it while Zope3 is still brewing. Interfaces to external mail and other servers is only part of it, since that makes workflow programming on the messages much less direct, and reduces the cross-platform flexibility of Zope deployments, where most extensibility is done in Python. Lots of *small* businesses need what MS Exchange offers, but can't afford it nor safely maintain its staggering complexity. As these python protocol implmentations crop up, I'd like to see Zope3 offer those writers a mature, cross-platform back-end to store their messages in, if they are willing to implement a few well-defined interfaces.

1 0

OpenSSH configuration between ZEO clients & storage server
by Eric Roby 30 Mar '02

30 Mar '02

Anyone had any experience trying to secure the transactions between ZEO clients and a storage server??? Our shop is already using OpenSSH, I have read some introductory information about OpenSSH. It is just not clear to me the level of effort required to implement this or if implemented, will the additional overhead further exasperate the potential for unresolved conflict errors??? Any thoughts... Eric

8 15

Re: [Zope-dev] zope CVS on cygwin (will sum)
by Didier Georgieff 30 Mar '02

30 Mar '02

On 30 Mar 2002, at 19:10, "Chris McDonough" <chrism(a)zope.com> wrote: > But now i have a problem with ./start wich seems to be related to > some rebase/adress for python under cgwin according to > > http://aspn.activestate.com/ASPN/Mail/Message/python-list/951044 This was the solution. I'll make a how-to as promised. Again thanks Chris.

1 0