Best practices for Lazy Tables

[Jim S]

Hi

I need to take advantage of the performance boost available using lazy_table=True.

My problem is that I have lots of table definitions in db.py that look like this:

productSequenceTag = db.define_table('productSequenceTag',
                                     Field('productSequenceTagId', 'id', readable=False),
                                     Field('productId', db.product, label='ProductId',
                                           requires=IS_IN_DB(db, 'product.id', '%(productNumber)s - %(productName)s',
                                                             zero='..')),
                                     Field('sequenceTagId', db.sequenceTag, label='Sequence Tag', ondelete='RESTRICT'),
                                     plural='Seq Tags')

There isn't much on the subject in the book, but there are a number of good threads here in the group.  However, it is really hard (for me at least) to sift through all this info and know definitively what is the best way to define a table and make sure it is lazy.

Has anyone successfully implemented lazy_tables on a large db project?  I've got 155 tables in my application.

I really need to find a way to make this load quicker.  Any help would really be appreciated.

-Jim

[Richard Vézina]

Jim, 

As far as I can see and from which I know about lazy_table you already lazy_table compliant...

What have been said about it application or not once flag is set to true was that you must not define your validator outside model definition (you actually do that in your example)...

You can't do :

db.define_table('productSequenceTag',
                                     Field('productSequenceTagId', 'id', readable=False),

                                     Field('productId', db.product, label='ProductId'),

                                     Field('sequenceTagId', db.sequenceTag, label='Sequence Tag', ondelete='RESTRICT'),
                                     plural='Seq Tags')

db.productSequenceTag.productId.requires = IS_IN_DB(db, 'product.id', '%(productNumber)s - %(productName)s',

                                                             zero='..')

But you don't do that actually...

Richard

--
Resources:
- http://web2py.com
- http://web2py.com/book (Documentation)
- http://github.com/web2py/web2py (Source code)
- https://code.google.com/p/web2py/issues/list (Report Issues)
---
You received this message because you are subscribed to the Google Groups "web2py-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to web2py+un...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Richard Vézina]

From performance stand point, I read that avoiding function definition in models files improve a lot...

So, considering that you may already had reach the optimization provide by lazy_table, you may have look and make sure you didn't have defined function or class in web2py models files somewhere...

Richard

[Niphlod]

response.toolbar() holds a pretty nice view of what's lazy and what's not, so what's the issue ?

The deal is ideally going throughout the whole model and NEVER see ANY db.tablename, i.e. like the part in red

[Anthony]

Though note that you can use db.tablename inside a function definition (including a lambda function), as it will not be evaluated until the function is actually called. So, you can do:

Field('productId', 'reference product', represent=lambda id, r: db.product(id).name)

Also, if you have 155 tables, make sure you aren't defining all 155 on every request (lazy or not). Instead, use conditional models or move the table definitions to functions/classes within modules and import and define only where needed.

Anthony

[Vinicius Assef]

Dont't define a table the way you did.

I.e, don't expect a return object from db.define_table() when creating a lazy table because it doesn't exist, actually.

About performance, I got good results simply setting "migrate=False" and creating modules, instead of creating global functions in models.



---- On Thu, 05 Feb 2015 20:04:47 -0300 Jim S wrote ----

[Jim Steil]

Vinicius - I'd be interested in learning more about your method.  Can you point me to some reading or examples?

-Jim

    

You received this message because you are subscribed to a topic in the Google Groups "web2py-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/web2py/XipyU9kQ2Qw/unsubscribe.
To unsubscribe from this group and all its topics, send an email to web2py+un...@googlegroups.com.

[Vinicius Assef]

---- On Sun, 08 Feb 2015 18:06:39 -0300 Jim Steil wrote ----

>Vinicius - I'd be interested in learning more about your method.  Can you point me to some reading or examples?

Which one?

a) Don't expect a returning object from db.define_table().
b) "migrate=False".
c) Creating modules?

[Jim Steil]

Creating modules.

[Richard Vézina]

Jim search model-less on the mailing-list...

Richard

On Mon, Feb 9, 2015 at 1:45 PM, Jim Steil <ato....@gmail.com> wrote:

Creating modules.

[Richard Vézina]

Here : 

https://groups.google.com/forum/#!topic/web2py/O07KnnuBUEk

You have to be carefull following this path... You may lost many web2py goodiess...

Richard

[Vinicius Assef]

It's easy, in fact.

Instead of having functions defined in models/ directory, you move them to modules/ dir. Then, you need to import them when you need them, in your controllers and in your models.

Be aware of how to access the Web2py API from modules [0].

[0] http://web2py.com/books/default/chapter/29/04/the-core#API


---- On Mon, 09 Feb 2015 15:45:13 -0300 Jim Steil wrote ----