Re: [web2py] Re: Auth Decorator / Model Question.
Great, thanks. I spent the first few days having web2py create our database tables from our models and now we're moving on to actual coding. I am LOVING this framework! Kevin Cackler Tech Daddies 501-205-1512 http://www.techdaddies.com On 9/4/2012 3:18 PM, Anthony wrote: In my limited experience with web2py, I'm thinking I should store the code in a model file and have that code save the store info to a web2py session, but I wanted to get a second opinion before continuing down that path. Yes, that's exactly right. Anthony -- --
Re: [web2py] Re: Auth Decorator / Model Question.
Every store has a unique URL (So store.mysite.com) which will be the key we look up on. This is why the session trick will work. Because we are storing the store data for store.site.com in session.store.mysite.com so every store would have a unique session. I guess the real question is - Should we just cache this query result or should we store it in a session? Which way is preferred for Python / web2py development? Kevin Cackler Tech Daddies 501-205-1512 http://www.techdaddies.com On 9/4/2012 3:22 PM, Niphlod wrote: you can certainly do STORE_DETAILS = db(db.stores.user_id == auth.user_id).select() in models. You'd have the variable STORE_DETAILS available in all controllers and every time a user loads a page the data will be refreshed. In order to reduce the db pressure, you can STORE_DETAILS = db(db.stores.user_id == auth.user_id).select(cache=(cache.ram, 60)) Doing so, the 2nd select will be fired only if more than 60 seconds passed from the 1st (i.e. a new page requested by the same user within 60 seconds will be fetched from the cache and not from the db) What you are doing in php works for web2py also: if you are positive that once a user is logged-in he would get the same stores forever (so it's not necessary to fetch the data every time you load the page), you can cache it with a high number or simply store the store details in session, i.e. if not session.store_details: #so it will be fetched one time only, if no store_details key is found on session store_details = db(db.stores.user_id == auth.user_id).select() Then you'd have to access this data as session.store_details That's all if I got it correctly: if I didn't understand please post more details. -- --
Re: [web2py] Re: Auth Decorator / Model Question.
cache.ram or even better (memcached) is preferred! if you use session and your sessions are stored on filesystem, if you have too much data it will be hard to load on each request. DATA = cache.ram(request.http_host, lambda: db(..).select(cacheable=True), 86400) # keeps for 24 hours. Now, on every place where data is changed you can call cache.ram.clear(regex=None) to reset that cache. --
Re: [web2py] Re: Auth Decorator / Model Question.
Note, sessions are specific to individual users (clients). If you put something in session.store.mysite on a given request, it will only be available to that specific user, not to anyone who goes to the store.mysite.com URL. If you need data accessible to all users store-wide, you should put it in the cache. Anthony On Tuesday, September 4, 2012 4:26:02 PM UTC-4, Kevin C wrote: Every store has a unique URL (So store.mysite.com) which will be the key we look up on. This is why the session trick will work. Because we are storing the store data for store.site.com in session.store.mysite.com so every store would have a unique session. I guess the real question is - Should we just cache this query result or should we store it in a session? Which way is preferred for Python / web2py development? Kevin Cackler Tech Daddies 501-205-1512 http://www.techdaddies.com On 9/4/2012 3:22 PM, Niphlod wrote: you can certainly do STORE_DETAILS = db(db.stores.user_id == auth.user_id).select() in models. You'd have the variable STORE_DETAILS available in all controllers and every time a user loads a page the data will be refreshed. In order to reduce the db pressure, you can STORE_DETAILS = db(db.stores.user_id == auth.user_id).select(cache=(cache.ram, 60)) Doing so, the 2nd select will be fired only if more than 60 seconds passed from the 1st (i.e. a new page requested by the same user within 60 seconds will be fetched from the cache and not from the db) What you are doing in php works for web2py also: if you are positive that once a user is logged-in he would get the same stores forever (so it's not necessary to fetch the data every time you load the page), you can cache it with a high number or simply store the store details in session, i.e. if not session.store_details: #so it will be fetched one time only, if no store_details key is found on session store_details = db(db.stores.user_id == auth.user_id).select() Then you'd have to access this data as session.store_details That's all if I got it correctly: if I didn't understand please post more details. -- --
