July 2001 - Zope-Dev

Re: [Zope-dev] BTrees and ConflictErrors.
by Anthony Baxter 31 Jul '01

31 Jul '01

>>> Chris McDonough wrote > I didn't test the conflict resolution code with ClientStorage. You may > want to try the same test without ZEO temporarily if you've got the > time... Ah. It's not so much a matter of time as "if I can't make it work like this, it's not going to be much use for our environment... :)" I will give it a go, tho, just to check. I'm curious - what sort of additional stuff is needed for conflict resolution in a ZEO environment? -- Anthony Baxter <anthony(a)interlink.com.au> It's never too late to have a happy childhood.

3 2

Globbing lexicon bug?
by sean.upton＠uniontrib.com 31 Jul '01

31 Jul '01

I'm having some ZCatalog search problems with wildcards in Zope 2.3.2. Suppose I am searching for a Volkswagen, which is commonly listed in Classified advertisements as 'VW' for brevity. I try to index text with the word VW, which works fine... my problem is that a search for: *vw* - Yields only ads with 'vw' or 'VW' in the middle of the string vw* - yeilds ads including ones containing a 2-char word 'VW' Why is this? Is it a bug in the lexicon digram processing in get()? My problem is that right now, I rewrite all queries from a search so that all search terms are wrapped in * on each side of the word, as a means of making the search friendly to the public... This won't work if there is a problem wrapping * around both sides of a 2 letter word, and I will have to switch my rewrite to putting a * only at the end, or ideally, figure out a way to fix the Lexicon code... Sean ========================= Sean Upton Senior Programmer/Analyst SignOnSanDiego.com The San Diego Union-Tribune 619.718.5241 sean.upton(a)uniontrib.com =========================

2 1

Change Request in TextIndex implementation
by Casey Duncan 31 Jul '01

31 Jul '01

I have a product that uses a Catalogs (just plain, not ZCatalogs) for indexing objects. It uses several Catalogs each of which share a Vocabulary. This was easy to do in 2.3, however, some code in 2.4 makes this a pain in the rear. Specifically line 204 of TextIndex.py, which tries to get the vocabulary lexicon from the parent ZCatalog like so: self._lexicon = self.aq_parent.aq_parent[self.vocabulary_id].getLexicon() In a world of implicit aquisition, this strikes me as very bad form. For this to work, the vocabulary must reside in the parent of the parent of the index, which must be a mapping object for it to find it. This is not the case with my implementation, in fact the vocabulary can be basically anywhere above the index. Of course, I could simply pass in the lexicon when the Index is instantiated. This is great and wonderful, except, I would like to support Zope 2.3 in as seemless a way possible. Writing crap like this bothers me (although I have little choice at this point methinks): try: index = TextIndex('spam', lexicon=eggs) except: index = 'TextIndex' Can't we change the above line to be something like this, which does not depend on a specific hierarchy? self._lexicon = self.aq_parent.getattr(self.vocabulary_id).getLexicon() That would make me very happy... -- | Casey Duncan | Kaivo, Inc. | cduncan(a)kaivo.com `------------------>

3 2

absolute_url(1) doesn't start with a /
by Steve Alexander 31 Jul '01

31 Jul '01

absolute_url(1) is supposed to give you the "physical path relative to a VirtualRoot". However, the returned value doesn't start with a "/", so it is not really a useful relative path. Is this a bug, or was it designed that way? (I suspect the answer might be "yes" and "yes" from my point of view :) ) Reference: /lib/python/OFS/Traversable.py -- Steve Alexander Software Engineer Cat-Box limited

3 2

Where does Acquisition.so come from?
by John Ziniti 31 Jul '01

31 Jul '01

I'm trying to get a fuller understanding of the way Zope works and the time has come to see what is going on inside Acquisition.aq_get(), but this is inside a compiled object, and I can't find anything resembling a source file in my Zope-2.3.3-src release. Am I just mistaken about which relese I downloaded? Where does Acquisition.so come from? Thanks in advance. John

1 0

Re: [Zope-dev] Change Request in TextIndex implementation
by Casey Duncan 31 Jul '01

31 Jul '01

Chris Withers wrote: > > Casey Duncan wrote: > > > > In a world of implicit aquisition, this strikes me as very bad form. For > > this to work, the vocabulary must reside in the parent of the parent of > > the index, which must be a mapping object for it to find it. This is not > > the case with my implementation, in fact the vocabulary can be basically > > anywhere above the index. > > Yeah, I put a bug in the tracker kindof related to this. You'll find that if you > add a vocabulary above the ZCatalogs in the acquisition path, you won't get to > choose which splitter you use since that choice requires an attribute to be > present that is defined in the __init__ method of ZCatalog. This also means that > you get the same efefct if you try and add a new vocabulary to an old ZCatalog > instance :-( All of the internal class interdependence bothers me. These should be written more generically then they have been and are... > > > Of course, I could simply pass in the lexicon when the Index is > > instantiated. This is great and wonderful, except, I would like to > > support Zope 2.3 in as seemless a way possible. Writing crap like this > > bothers me (although I have little choice at this point methinks): > > The PluginIndex stuff does seem a bit scrappy in places :-S The more I read of it, the more I concur. It is moving in the right direction, but there are definitely holes... [snip] > > Women have bad hair days, I have bad Zope days, this is one of them ;-) > Bad Zope! Bad Zope! Now go to your room!! 8^) -- | Casey Duncan | Kaivo, Inc. | cduncan(a)kaivo.com `------------------>

1 0

Pluggable Indexes and Splitters
by Andy Dawkins 31 Jul '01

31 Jul '01

I have a small problem with Pluggable Indexes and Splitters which I think is an oversight of the current design implementation. The Details Bit: Zope 2.4.0 (final) Python 2.1 I can successfully create a new Splitter (called MySplitter) I add it to availablesplitters, put it in its own directory in PluginIndexes/TextIndex/Splitter/ Then I create a new vocabulary and MySplitter appear on the list of splitters. Fantastic thats great. Next I create my own index. This index is very similar to the TextIndex, so I copy the PluginIndexes, rename it, remove the indexes I don't want to modify (FieldIndex, KeywordIndex, PathIndex) and rename TextIndex to MyTextIndex I make the relevant changes to MyTextIndex, head back to the ZMI and add an index using MyTextIndex. All is fine. Here is a quick diagram of my current directory structure (for those who are easily lost) /2.4.0_base/lib/python/Product/PluginIndexes/TextIndex/Splitter/MySpliter - (MySplitter) /MyInstance/Products/MyTextIndex/MyTextIndex - (MyTextIndex) Now because MySplitter is completely useless when used with TextIndex and because other zope instance's use the 2.4.0_base that don't need MySplitter, I move MySplitter out of the /PluginIndexes/TextIndex/Splitter folder and into /MyTextIndex/Splitter folder. I return the availablesplitters attribute in /TextIndex/Splitter/__init__.py to what is should be and make the change to my own /MyTextIndex/Splitter/__init__.py Because MySplitter is the _only_ splitter that I want MyTextIndex to use I remove all other splitters from the /MyTextIndex/Splitter/ folder, and their relevant entries in availablesplitters. Head back to the ZMI, Remove the catalog and start again. New ZCatalog <click>, Create a vocabulary for me <click>, ok <click> New Vocabulary <click>, Name it 'MyVocabulary' <tap><tap><tap>, Select my splitter <STOP> My splitter isn't listed anymore. And thats where the problem lies. Deaper investigation reveals that inside /2.4.0_base/lib/python/Products/ZCatalog/Vocabulary.py Lies a single line which reads. from Products.PlugginIndexs.TextIndex.Vocabulary import * So it _always_ uses the vocabulary in TextIndex, which always uses the Splitters in TextIndex. Oooops, This isn't very friendly, How am i supposed to create my own Indexes with thier Own Splitters??????? 1) I could change this file, But it is in a base zope folder so it will screw all the Catalogs that use this base. 2) I could leave MySplitter in the /TextIndex/Splitter folder, But I don't want a normal TextIndex to use this splitter. Ideally I want PluggableIndexes to be able to have their own splitters. -Andy Dawkins

3 4

dtml-sqlvar, should it insert 'null' if variable is None ?
by Christian Theune 31 Jul '01

31 Jul '01

Hi. In my oppinion, dtml-sqlvar is supposed to quote and prepare date for sql database. Therefore I think it should change None values into null, as it does with the optional tag, when values are omitted, but not when they are None. Any comments on that? -- Christian Theune - ct(a)gocept.com gocept gmbh & co.kg - schalaunische strasse 6 - 06366 koethen/anhalt tel.+49 3496 3099112 - fax.+49 3496 3099118 mob. - 0178 48 33 981 reduce(lambda x,y:x+y,[chr(ord(x)^42) for x in 'zS^BED\nX_FOY\x0b'])

1 0

Re: [Zope-dev] BTrees and ConflictErrors.
by Anthony Baxter 31 Jul '01

31 Jul '01

>>> Chris McDonough wrote > Is this all via ZEO? Yep. ZEO all the way. Anthony > Anthony Baxter wrote: > > > > >>> Anthony Baxter wrote > > > I thought the newer BTrees code was meant to deal with conflicts from > > > two entries being added at the same time? > > > I'm using CoreSessionTracking 0.8 with an external data source on a > > > bsddb3Storage.Minimal storage mounted via ZEO (using the current Zope2.3 > > > code). If two sessions are created around the same time, I'm seeing > > > ConflictErrors - there doesn't seem to be any attempt to invoke the > > > resolveConflict code. > > > > Ok, things that I've found: > > > > Minimal doesn't even try to implement Conflict resolution - I've switched t o > > 'Full'. To no real avail, unfortunately - it's still breaking when I hammer > > it (10 clients all creating one new session per second). > > > > I've tried putting the SessionData object in a normal Data.fs - still no > > joy. I've also tried using CST's internal data container (RAM based). > > The ConflictErrors keep on flying. > > > > About the only thing I haven't tried yet is to try Zope2.4 - this is > > something of a pain, though. > > > > Has anyone got any other ideas? Am I just completely out of luck here? > > Do people have other suggestions for containers that can take a large > > number of new object create/deletes? > > > > Thanks, > > Anthony > > > > _______________________________________________ > > 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 ) > > -- > Chris McDonough Zope Corporation > http://www.zope.org http://www.zope.com > > """ Killing hundreds of birds with thousands of stones """ > -- Anthony Baxter <anthony(a)interlink.com.au> It's never too late to have a happy childhood.

2 1

Local classes inside python scripts
by Jay, Dylan 31 Jul '01

31 Jul '01

I'm trying to understand the security reasons behind the forbidding of local class definitions inside a pythonscript. It would be helpful to be able to define and use transient objects without having to create products. Anyone know why this is?

1 0