[Zope3-dev] Florent's O-R blog entry

[Gary Poster]

On Aug 24, 2005, at 6:27 AM, Martijn Faassen wrote:

> Hi there,
>
> A few contributions to this interesting discussion...
>
> [snip the Zope 3 catalog is not a hack, and clean and simple]
>
>
>> The catalog and index code is not a hack, and is in fact simple,   
>> effective and flexible.  Python is the query language, and the  
>> lack  of an optimizer is not a reason to go running to an RDBMS  
>> index.  The  catalog and index code could use polish and even  
>> alternate  implementations, but the BTrees, the core code, are  
>> fantastic tools.
>>
>
> I have had some opportunity to work with the Zope 3 catalog  
> recently, and I have a few comments. First of all, I agree with the  
> main idea that the Zope 3 catalog is not a hack, and is clean and  
> flexible. I believe the catalog should be invested in, as I think  
> it's cool.
>
> Now as to where I see areas where features are lacking in the Zope  
> 3 catalog:
>
> Underfeatured query API
> -----------------------
>
> I do think that currently the API to query it is woefully  
> underfeatured.
>
> I've tried to work on this problem and am sitting on some code that  
> just needs a bit of time to polish and release that allows a simple  
> query language on top of the catalog. It's just building up a tree  
> of python objects for queries, nothing special, but it is a lot  
> higher level than what's already there.

As I think I've said before, sounds like a good start. :-)

Another way of looking at this--or simply an additional feature on  
top of a query langauge--might be to make the IFBTree results easier  
to manipulate in an easier way.  The code in the zc sandbox for the  
extent (http://svn.zope.org/Sandbox/zc/catalog/extentcatalog.py) is a  
sketch of what I mean--following some of the set API, for instance.   
The reason for my interest in this is that we have very little code  
that uses the catalog to return objects--just IFBTree data  
structures.  Just working with the IFBTree data structures gives you  
a lot more flexibility for integration of catalog results with other  
data structures.

Casey Duncan had explored some very interesting ideas in his pypes  
project (http://cvs.zope.org/Packages/pypes/) for a query language,  
by the way, but his ambition is still largely unrealized, even though  
much of his query language work could be ported to Zope 3 without a  
huge amount of trouble.

Arguably, query optimization would be the feature that would make a  
given syntax win.

> Fast, easy batching/sorting
> ---------------------------
>
> I don't know how to do easy, efficient batching/sorting with the  
> catalog. I'd like to be able to query *just* a batch of objects,  
> sorted, for user interface purposes. There doesn't seem to be a  
> straightforward way to do this yet, and this is a very common use  
> case. The batching implementation sitting out there in  
> zope.bugtracker.batching is nice, but doesn't deal with the catalog.
>
> I think this should be fixable with a bit more infrastructure  
> though. Getting the right batch is just a query on an index, and  
> the result can be sorted afterwards, though there are tricky issues  
> getting the right batch *size*.

Since we usually work with IFBTree data structures until the very  
end, we get most of the benefits of batching.  Once you are done with  
your processing, you can simply wrap the result in a  
zope.app.catalog.catalog.ResultSet (or similar) and be good to go.   
This is why I think making the story for working directly with the  
BTree data structures easier might be a good way to go.

Sorting is hard to do efficiently, and easy to *think* you are making  
an optimization.  We are currently doing it "naively" (to the degree  
that using the very efficient Python sort is naive), and Jim refers  
to research that indicates that a good non-naive approach is not  
clear.  I can certainly imagine various approaches.  Carefully  
arranging your merges can actually result in a pre-sorted set.  We're  
not being that careful.

> Missing powerful query concepts
> -------------------------------
>
> Certain powerful query concepts like joins, available in a  
> relational setting, are missing. I've already run into a scenario  
> where I wanted to someting like this: given a bunch of version  
> objects with field 'id', where multiple objects can have the same  
> 'id' to indicate they're versions of the same object, I want all  
> objects where field 'workflow_state' is 'PUBLISHED' unless there is  
> another object with the same id that have workflow_state 'NEW', in  
> which case I want that one'.
>
> I think joins would be a way to solve it, though I haven't figured  
> out the details, nor how to implement them efficiently on top of  
> the catalog. This kind of thing is where a relational database  
> makes life a lot simpler.

I guess that's taste.  I'd be happier with Python.

> Zope catalog benefits
> =====================
>
> Now as to benefits of using the ZODB instead of a relational  
> database. I've seen some mentioned already, and I think there are  
> more that haven't been mentioned yet. I realize that some of these  
> issues don't exist so much with 'transparent' maps like Ape, which  
> acts like the ZODB, though *if* a relational database is used by an  
> application I also think that those features will be used  
> (otherwise, why do it?), which will still reduce the portability to  
> non-Ape settings.

Right: my concern.

> Common development platform
> ---------------------------
>
> I've seen it mentioned elsewhere in this thread that the ZODB can  
> unify the development community, whereas O/R mapping technologies  
> (in particular those not transparent to the ZODB) run the risk of  
> scattering it. I think this is an interesting argument so I'd just  
> like to underline it.
>
> Ease of deployment
> ------------------
>
> Right now a Zope application is relatively difficult to deploy  
> compared to some other solutions like PHP, but, it's probably  
> easier to deploy than other solutions which require a relational  
> database backend. Now it might seem that 'enterprise' deployments  
> are big anyway, so we don't have to worry about making this harder,  
> but:
>
>   * enterprises will ask questions like "which relational database  
> does it support? we standardized on relational database system foo,  
> does it work with that?" We run the risk of having to say "no", or,  
> if "yes", we may run the risk of "oops, we cannot test this easily  
> with database system foo as we don't have it here."
>
>   * requiring a RDB for deployment makes it harder to market our  
> software, as it's harder to just download and install software into  
> your Zope to try it out. You need to set up a relational database  
> as well. I may be mistaken, but think Plone would be less popular  
> if a relational database would be *necessary* in order to play with  
> it.

While I agree with your general point, Ruby on Rails might call that  
assertion into question a bit.

>   * closely related, requiring a RDB for deployment makes it harder  
> to market our open source software to other developers. This ties  
> in closely to the argument above involving the risk of the  
> community fracturing. Even inside a company having more software  
> requirements like a RDB may hinder team development (where each  
> team member runs a separate instance of the software), as there's  
> simply that much more to set up and thus harder for someone to get  
> up to speed with a project.
>
> Testability
> -----------
>
> One point I also haven't seen mentioned yet is that I don't want to  
> have to have a relational database installed in order to run my  
> tests. The great thing about the ZODB and persistence is that it's  
> very transparent. Persistent instances are very very similar to non- 
> persistent ones.

Good points.

I'll add another.

Component system

Because of the Zope 3 component system, if we can use the current  
catalog interface, or invent another, to develop both a ZODB/BTree- 
based implementation and an RDBMS-based implementation, it's possible  
that users who wanted to choose the RDBMS strengths would be able to  
do so without dividing the user community.

> [snip blob support argument]
>
> I agree that the blob argument for RDB mapping is not convincing.  
> There are other solutions around and this is being improved rapidly.
>
> Anyway, all of the arguments against object/relational mapping  
> aside, I do think this is an interesting area to explore. You *do*  
> get a whole lot of power using a relational database, after all. I  
> myself am actually in two minds concerning very transparent ZODB- 
> style solutions like Ape or less transparent but more explicit uses  
> of O/R mappings like SQLObject. While the transparency has many  
> benefits mentioned before, the more straightforward mapping has the  
> benefits of simplicity, may map to relational databases more  
> easily, and may expose powerful relational features more  
> straightforwardly.

It's true.  I hope that an entire platform doesn't force the decision  
on its potential users, though.

Gary