[web2py] Re: caching variable data sample

[Pierre]

thanks Anthony...

looks like the cache select puzzle is now complete. :  )


however we still have a 'serious productivity lag here in Bangladesh'...


-- 
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+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[web2py] Re: caching variable data sample

[Anthony]

On Monday, September 18, 2017 at 10:09:51 AM UTC-4, Pierre wrote:
>
> I'd prefer the module option but I need to pass *time_expire* as a 
> variable (*db.atable.__variableName*), I don't think I can do that in a 
> module can I ?
>

You are not limited to using the decorator to cache the results of a 
function. Instead, you can just do something like this:

from mymodule import cache_this

myrows = cache.redis('myrows', cache_this, time_expire=myvariable)

Anthony

-- 
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+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[web2py] Re: caching variable data sample

[Pierre]

I'd prefer the module option but I need to pass *time_expire* as a variable 
(*db.table.__variableName*), I don't think I can do that in a module can I ?


On Friday, September 15, 2017 at 6:15:44 AM UTC-4, Pierre wrote:
>>
>> actually the real function is a little more complex since it returns rows 
>> plus other datas that don't require caching. I tried to move the cache_this 
>> function to a module (don't know if this is orthodox) but then i get a 
>> decorator error:
>>
>> @current.cache('data_sample',* time_expire=60*, 
>> cache_model=current.cache.redis)
>> attributeError cache object has not attribute redis
>>
>
> You don't necessarily have to put the function in a module -- it can go in 
> a model or even in the controller (if the function takes no arguments, 
> start its name with a double underscore to prevent it from being exposed as 
> an action). You can even define the function inside the controller function 
> that calls it if that's the only place it is needed.
>
> Anthony
>

-- 
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+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[web2py] Re: caching variable data sample

[Anthony]

Forgot about that. It's documented here: 
http://web2py.com/books/default/chapter/29/04/the-core?search=lazy_cache#Warning--Do-not-use-the-current-object-in-global-scope-in-a-module.
 
Yes, that would be the way to go if you want to move the function to a 
module.

Anthony

On Friday, September 15, 2017 at 8:15:21 AM UTC-4, Pierre wrote:
>
> the more digging the deeper one gets :
>
>
> https://groups.google.com/forum/?fromgroups#!searchin/web2py/cache$20in$20module/web2py/AZa5Boj3y3E/_BPMTdXwSaMJ
>
> and here is the *cache_this*  module version :
>
> from gluon.cache import lazy_cache
>
> @lazy_cache('data_sample', time_expire=60, cache_model='redis')
> def cache_this():
> ...
> rows = db(db.atable.id > 0).select(limitby=limitby, cacheable=True)
> return rows
>
>
> This seems to do the trick.  Is it 'kosher' ?
>

-- 
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+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[web2py] Re: caching variable data sample

[Anthony]

On Friday, September 15, 2017 at 6:15:44 AM UTC-4, Pierre wrote:
>
> actually the real function is a little more complex since it returns rows 
> plus other datas that don't require caching. I tried to move the cache_this 
> function to a module (don't know if this is orthodox) but then i get a 
> decorator error:
>
> @current.cache('data_sample', time_expire=60, 
> cache_model=current.cache.redis)
> attributeError cache object has not attribute redis
>

Hard to say what's going wrong without seeing your code. But you don't 
necessarily have to put the function in a module -- it can go in a model or 
even in the controller (if the function takes no arguments, start its name 
with a double underscore to prevent it from being exposed as an action). 
You can even define the function inside the controller function that calls 
it if that's the only place it is needed.

Anthony

-- 
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+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[web2py] Re: caching variable data sample

[Pierre]

the more digging the deeper one gets :

https://groups.google.com/forum/?fromgroups#!searchin/web2py/cache$20in$20module/web2py/AZa5Boj3y3E/_BPMTdXwSaMJ

and here is the *cache_this*  module version :

from gluon.cache import lazy_cache

@lazy_cache('data_sample', time_expire=60, cache_model='redis')
def cache_this():
...
rows = db(db.atable.id > 0).select(limitby=limitby, cacheable=True)
return rows


This seems to do the trick.  Is it 'kosher' ?

-- 
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+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[web2py] Re: caching variable data sample

[Pierre]

very True: productivity rules and I am far below modern 
standards(more or less the 'Bangladesh level'):)

Le jeudi 14 septembre 2017 16:55:29 UTC+2, Anthony a écrit :
>
>  
> Maybe you should have asked a week ago. ;-)
>
> Anthony
>

-- 
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+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[web2py] Re: caching variable data sample

[Pierre]

actually the real function is a little more complex since it returns rows 
plus other datas that don't require caching. I tried to move the cache_this 
function to a module (don't know if this is orthodox) but then i get a 
decorator error:

@current.cache('data_sample', time_expire=60, 
cache_model=current.cache.redis)
attributeError cache object has not attribute redis

Le jeudi 14 septembre 2017 16:49:49 UTC+2, Anthony a écrit :
>
> Looks like you want to cache the results of the entire function -- so why 
> not do that:
>
> @cache(some_key, time_expire=60, cache_model=cache.ram)
> def cache_this():
> ...
> rows = db(db.atable.id > 0).select(..., cacheable=True)
> return dict(rows=rows)
>
> Just be sure to set cacheable=True, which will make the entire Rows object 
> cacheable.
>
> Anthony
>
>
>

-- 
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+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[web2py] Re: caching variable data sample

[Anthony]

On Thursday, September 14, 2017 at 9:33:42 AM UTC-4, Pierre wrote:
>
> the cache select mechanism is full of mystery:
>
> given book example code:
>
> def cache_db_select(): 
> logs = db().select(db.log.ALL, cache=(cache.ram, 60)) 
> return dict(logs=logs)
>
>
> what happens to next cache_db_select call after the 60 seconds has elapsed 
> ?
> Does it overwrite the previously cached rows ?
>

It is just using the web2py caching mechanism, which is described here: 
http://web2py.com/books/default/chapter/29/04/the-core#cache. So, yes, if 
the time has expired, the old cached value will be replaced with a fresh 
value.
 

> One need to know a key in order to clear a specific cached value 
> associated with that key. When performing a cache select no key is being 
> passed so how to clear a specific cached select item (for instance logs) ?
>

If you need to manually clear cached select, you should probably not use 
the above mechanism, as the key used is part of the internal implementation 
(it is the md5 has of the database URI plus the raw SQL query). Instead, 
you should set cacheable=True in the .select() call, and then just cache 
the resulting Rows object using cache.ram() with whatever key you like.
 

> this gives me nerve crisesIt should be simple to do a simple thing 
> however one week digging this and I didn't move 1 inch ahead
>
 
Maybe you should have asked a week ago. ;-)

Anthony

-- 
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+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[web2py] Re: caching variable data sample

[Anthony]

Looks like you want to cache the results of the entire function -- so why 
not do that:

@cache(some_key, time_expire=60, cache_model=cache.ram)
def cache_this():
...
rows = db(db.atable.id > 0).select(..., cacheable=True)
return dict(rows=rows)

Just be sure to set cacheable=True, which will make the entire Rows object 
cacheable.

Anthony

On Wednesday, September 13, 2017 at 4:09:23 PM UTC-4, Pierre wrote:
>
> purpose is to keep a single renewable pseudo-random data sample live 
>
> the w2p books describes a situation of a constant query :
> http://web2py.com/books/default/chapter/29/06/the-database-abstraction-layer?search=caching+selects#Caching-selects
> def cache_db_select(): 
> logs = db().select(db.log.ALL, cache=(cache.ram, 60)) 
> return dict(logs=logs)
>
> now suppose every call a different query like here :
>
> def cache_this():
> nget = somevalue
> kount = db(db.atable.id > 0).count() 
> offset = randint(0, kount - nget)
> limitby = (offset, offset + nget)
> rows = db(db.atable.id > 0).select(limitby=limitby, cache=(cache.redis, 
> 60))
> return dict(rows=rows)
>
>
>
> this accumulates cached 'material' and returns a different rows on every call
> how do I make *cache_this* behave like the book example ?
>
> I could cache kount and offset variables as well :
>
> kount = db(db.atable.id > 0).count(cache=(cache.redis, 60)) 
> offset = cache.redis('offset', lambda: randint(0, kount-nget), 
> time_expire=60) 
>
>
> but i suppose this would lead to a synchronization problem and thus possible 
> duplicate cached select
>
>

-- 
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+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[web2py] Re: caching variable data sample

[Pierre]

the cache select mechanism is full of mystery:

given book example code:

def cache_db_select(): 
logs = db().select(db.log.ALL, cache=(cache.ram, 60)) 
return dict(logs=logs)


what happens to next cache_db_select call after the 60 seconds has elapsed ?
Does it overwrite the previously cached rows ? 

One need to know a key in order to clear a specific cached value associated 
with that key. When performing a cache select no key is being passed so how 
to clear a specific cached select item (for instance logs) ?

this gives me nerve crisesIt should be simple to do a simple thing 
however one week digging this and I didn't move 1 inch ahead

-- 
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+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.