There are quite a few use cases to keep in mind here. It should be
easy for a developer to configure how often this framework should
poll a server for data:
1) at document initialization
- this would basically initialize the "collection" array from a
remote data source in a more elegant fashion than loading an
external script
- this method would typically be used with finite and hopefully
small collections with up to a couple thousand members, and
would lean on the user's browser to do the entire matching.
- this could conceivably be used with 2) below.
2) upon user-interaction
- this approach would be best suited for use cases similar to
"Google Suggest". Each time a user types data, there is a near-
inifite amount of possibilities.
- it would be nice to be able to configure the polling frequency: poll
at each character being typed ... poll every 2 characters being
typed ... poll each time a word is completed, denoted by a space,
or XXXX seconds of typing inactivity, or the tab key being hit, or
the return key being hit, etc etc. [insert your own suggestions].
as mentioned in 1), I can see use cases for usage of 1) only or 2)
only or both. It'll highly depend on the nature of the web
application, its user base, and the amount of data it surfaces to its
users.
I'd also like to spend some time in this framework coming-up with
sample implementations of server-side integrations, and concrete
applications of various http optimizations techniques, such as the
use of the HTTP 304 response, the setting of various caching
headers, and GZIP compression via Accept-Encoding/Response-
Encoding.
what am i forgetting? would anybody care to elaborate with more
use cases? configuration options?
Logged In: NO
Hey, I've found this code a couple of years ago, and thought it was great. But I had to hack it to work with ColdFusion where i was building the JavaScript arrays. It was a bit of a mess, and struggled with long lists...
When do you plan to add the httprequest support?