September 2005 - Zope-Dev

RE: [Zope-dev] Re: Z SQL Caching
by Dale Hirt 26 Sep '05

26 Sep '05

Thank you, Martijn. That makes it very clear. :) -----Original Message----- From: zope-dev-bounces(a)zope.org [mailto:zope-dev-bounces@zope.org]On Behalf Of Martijn Pieters Sent: Monday, September 26, 2005 11:49 AM To: zope-dev(a)zope.org Subject: [Zope-dev] Re: Z SQL Caching Dale Hirt wrote: > I have a Z SQL method that is currently returning around 25000 rows. > Is there a way to see if it is pulling that data from the cache or is > doing a refresh? > > Alternatively, and maybe this will answer the first question, the Z SQL > method is run essentially twice, each time with different parameters. > Does it cache each instance, or does it cache the first, then when it's > run again with different parameters, kill the cache and re-run > everything? A ZSQLMethod cache (if configured with a cache duration greater than 0) is keyed on the rendered SQL. So, if different parameters lead to a different SQL query to be sent to the database, then a different key is used. Caches are kept for the duration of the configured cache duration, and a given cache key will not influence earlier caches of different keys. The SQL is taken quite literally here. If you use the following template: SELECT * FROM foo WHERE <dtml-sqltest bar type="nb"> Then different values for 'bar' will result in different SQL queries and thus different cache keys. If the same method is called with the same value for 'bar' within the cache duration, an earlier cached result will be returned instead of querying the database. Another value for 'bar' in between those calls will not kill the cache. Martijn Pieters _______________________________________________ Zope-Dev maillist - Zope-Dev(a)zope.org http://mail.zope.org/mailman/listinfo/zope-dev ** No cross posts or HTML encoding! ** (Related lists - http://mail.zope.org/mailman/listinfo/zope-announce http://mail.zope.org/mailman/listinfo/zope )

1 0

Z SQL Caching
by Dale Hirt 26 Sep '05

26 Sep '05

Hi all, I have a Z SQL method that is currently returning around 25000 rows. Is there a way to see if it is pulling that data from the cache or is doing a refresh? Alternatively, and maybe this will answer the first question, the Z SQL method is run essentially twice, each time with different parameters. Does it cache each instance, or does it cache the first, then when it's run again with different parameters, kill the cache and re-run everything? Thanks, Dale Hirt

2 1

[Fwd: Re: [Zope-dev] Problems adding many objects]
by Morten W. Petersen 26 Sep '05

26 Sep '05

Oops, one for the list as well. -------- Original Message -------- Subject: Re: [Zope-dev] Problems adding many objects Date: Mon, 26 Sep 2005 13:34:50 +0200 From: Morten W. Petersen <morten(a)nidelven-it.no> To: Chris McDonough <chrism(a)plope.com> References: <4336AF98.3030606(a)nidelven-it.no> <249B923B-2427-4A01-B656-56300CC7EE88(a)plope.com> <4337D9A8.70107(a)nidelven-it.no> <5806920C-3A04-42EE-A2D5-487F630A10CB(a)plope.com> > It doesn't say "returning issue, None"? My guess is that the if > request is not none branch is called and it goes that way. But there's > no way to know for sure without running it. It says 'returning issue <issue at xxxxx>' > This is a good place to use the Python debugger... as the first line > of this method, insert "import pdb; pdb.set_trace()", then start up > Zope in the foreground (runzope or zopectl fg), then cause the method > to be invoked through a browser. > > On the console you will see a pdb prompt. You can step through the > code by using "n". To print the value of a variable type "p > <variablename>", to step in to a function call press "s". When you're > finished, press "c" to continue or "q" to quit. This will be the > fastest way to see what's happening in this method. Yep, I'll have a go at this later. Thanks for describing the debug process. :) -Morten -- Regards, Morten Tlf: +47 45 44 00 69 Blog: http://www.blogologue.com

1 0

Problems adding many objects
by Morten W. Petersen 26 Sep '05

26 Sep '05

Hi, I've created a script that imports XML into the ZODB, using regular TTW methods - and all is fine, except for when I import large files, with many objects. After about 288 objects, the import fails when the method that adds an object in the ZODB returns None, instead of the object it added. From what I can tell, the added object is not None before it is returned, but the name object, set via the expression object = self.manage_add_issue(...) is None. So it looks like somewhere between 'return issue_' in manage_add_issue and the expression above, the object is lost. Any ideas what could be happening here? I've managed to work around it for now, but it is interesting nonetheless. Regards, Morten

2 3

Repository for python developers
by Mikhail Kashkin 23 Sep '05

23 Sep '05

Moscow, Russia, September 23th, 2005 We are happy to spread good news about `Key Solutions' <http://keysolutions.ru/>`_ new initiative. We have created a new public repository for python developers. The ulitimate goal of this project is to unite companies and people under the umbrella of Russian-speaking-python (Zope/Plone/Zope3/Twisted/etc) open source programmers community, so that we could jointly push forward new programs and technologies in Russia. The repository powered by `subversion <http://subversion.tigris.org/>`_. **Links**: - `Press release in russian <http://keysolutions.ru/News/subversion>`_ - `Новость на русском <http://keysolutions.ru/News/subversion>`_ - `Subversion <https://svn.keysolutions.ru/community/>`_ About Key Solutions -------------------- Key Solutions is an open source company with primary focus in development of corporative CRM Internet/intranet systems and web-communities. The company promotes and supports Zope, Plone, Asterisk platforms in Russia. To learn more about Key Solutions visit `eng.keysolutions.ru <http://eng.keysolutions.ru/>`_ and `keysolutions.ru <http://keysolutions.ru/>`_ -- Mikhail Kashkin, Key Solutions (http://keysolutions.ru/) Director Zope/Asterisk/Plone - Solutions/Consulting/Support Plone на русском http://plone.org.ru/ Plone Foundation Member (http://plone.org/foundation/members/)

1 0

Weird IndexableObjectWrapper + TopicIndex + aq_acquired interactions.
by Lennart Regebro 21 Sep '05

21 Sep '05

You know you are in trouble when you need to read the c-code of Acquisition to understand a problem. :) This took me half the day to figure out, and in fact, I still don't fully understand it. So I need help with this. CPS has an index, called cps_filter_sets. It's a TopicIndex, and TopicIndexes can have expressions as filters. This one has: "getattr(o, 'portal_type', None) not in ('Section', 'Workspace')" Where o is the object that are being indexed. But, this being an expression, it's protected, and therefore, a getattr becomes a guarded_gettatr. guarded_getattr does it's security check by doing: validate = SecurityManagement.getSecurityManager().validate aq_acquire(inst, name, aq_validate, validate) Where in this case inst will be the object that is being indexed. aq_acquire will, if the first parameter is not an AcquisitionWrapper, and the third parameter is not None, wrap the object. Now, in our case, the object is an IndexableObjectWrapper, wrapping an Acquisition wrapped object. So, aq_acquire will Acquicision wrap the IndexableObjectWrapper, with the result that the object being used now has no context! Then, it passes this to validate, who in turn passes it to allowed, who check that the object has the users user folder in it's context. And it hasn't, because it has no context. *blam* You get an AuthorizedError, and the object does not get indexed. OK, there are several ways to fix this, one being to not use getattr. But there is still something missing from my understandning of all this: - Not every object has this problem! Why, I have no idea. I had his problem in two products, and in ONE of them, it disappeared when I introduced the otherwise useless PropertyManager as a baseclass. The other product already has propertymanager as a baseclass! Any insight into this would be appreciated. //Lennart

2 5

Re: [Zope-dev] Re: [Zope-Checkins] SVN: Zope/trunk/lib/python/Products/ZCatalog/ProgressHandler.py added FilelogHandler
by Chris Withers 21 Sep '05

21 Sep '05

Andreas Jung wrote: > This is basically a backport of ongoing work where we can't and want use > the Zope logger. How come? Chris -- Simplistix - Content Management, Zope & Python Consulting - http://www.simplistix.co.uk

2 1

Re: [Zope-dev] Re: [Zope-Checkins] SVN: Zope/trunk/lib/python/Products/ZCatalog/ProgressHandler.py added FilelogHandler
by Chris Withers 21 Sep '05

21 Sep '05

Andreas Jung wrote: > This is basically a backport of ongoing work where we can't and want use > the Zope logger. How come? Chris -- Simplistix - Content Management, Zope & Python Consulting - http://www.simplistix.co.uk

1 0

Re: [Zope-Checkins] SVN: Zope/trunk/lib/python/Products/ZCatalog/ProgressHandler.py added FilelogHandler
by Chris Withers 21 Sep '05

21 Sep '05

Is there any way this could use normal python loggers? logger = getLogger('progress') ? (and then with a more generic section in zope.conf for having more loggers configured...) Andreas Jung wrote: > Log message for revision 38482: > added FilelogHandler > > > Changed: > U Zope/trunk/lib/python/Products/ZCatalog/ProgressHandler.py > > -=- > Modified: Zope/trunk/lib/python/Products/ZCatalog/ProgressHandler.py > =================================================================== > --- Zope/trunk/lib/python/Products/ZCatalog/ProgressHandler.py 2005-09-15 13:22:20 UTC (rev 38481) > +++ Zope/trunk/lib/python/Products/ZCatalog/ProgressHandler.py 2005-09-16 08:29:14 UTC (rev 38482) > @@ -95,3 +95,16 @@ > def output(self, text): > LOG(self._ident, INFO, text) > > + > +class FilelogHandler(StdoutHandler) > + """ Use a custom file for logging """ > + > + __implements__ = IProgressHandler > + > + def __init__(self, filename, steps=100): > + StdoutHandler.__init__(self, steps) > + self.filename = filename > + > + def output(self, text): > + open(self.filename, 'a').write(text + '\n') > + > > _______________________________________________ > Zope-Checkins maillist - Zope-Checkins(a)zope.org > http://mail.zope.org/mailman/listinfo/zope-checkins > -- Simplistix - Content Management, Zope & Python Consulting - http://www.simplistix.co.uk

1 0

RE: [Zope-dev] Installing Python Modules Under Zope On Windows
by Jason R. Coombs 16 Sep '05

16 Sep '05

You can trick the installer to using a particular instance of Python with an easy registry hack. I've done this a couple of times to use Python version specific binary installers to my Plone install. The key is in HKLM\Software\Python\PythonCore\2.x\InstallPath. The value is a REG_SZ of the fully-qualified path of the python root for that version of Python. You can change that value, run the installer(s), then change it back (at least I do, for system integrity). Jason R. Coombs -----Original Message----- From: zope-dev-bounces(a)zope.org [mailto:zope-dev-bounces@zope.org] On Behalf Of tom(a)tompurl.com Sent: 16 September, 2005 09:21 To: zope-dev(a)zope.org Subject: [Zope-dev] Installing Python Modules Under Zope On Windows I have a hardware constraint where I have to run Zope on Windows. This has worked pretty well for me because Zope 2.7 used my "system" instance of Python 2.3.x. At least, that's my assumption since runzope.bat pointed at the system instance of Python. This setup was nice because it made it very easy to install Python modules (like PIL). When I ran the module installation executable, it would find my system instance of Python in the registry and install the module properly. With Zope 2.8.1, the "default" setting seems to be using the embedded Python interpreter that came with Zope. This is a problem because the embedded versions of Python do not appear in the registry, making it impossible to install a lot of Python modules on Windows. I thought that the following script (http://effbot.org/zone/python-register.htm) would allow me to add embedded versions of Python to my registry. However, you can't add more than one major version of a python release (2.2, 2.3, etc) to the registry. So if my system version of Python is the 2.3.x branch, and I want to use an embedded 2.3.x version of Python with Zope, then I'm out of luck if I want to install many Python modules with it. I can still edit runzope.bat so that it points at the system instance of Python, but I don't know if this is a best practice. Has anyone had any luck installing Python modules in an embedded version of Python on Windows? Is there a better way to work around this than referencing the system version of Python in the runzope.bat file? Thanks in advance for any help! Tom Purl _______________________________________________ Zope-Dev maillist - Zope-Dev(a)zope.org http://mail.zope.org/mailman/listinfo/zope-dev ** No cross posts or HTML encoding! ** (Related lists - http://mail.zope.org/mailman/listinfo/zope-announce http://mail.zope.org/mailman/listinfo/zope )

1 0