dojox.data.QueryReadStore
dojo.require("dojox.data.QueryReadStore");
defined in dojox/data/QueryReadStore.js
By default all the sorting is done serverside before the data is returned
which is the proper place to be doing it for really large datasets.
Examples
Example 1
// The parameter "query" contains the data that are sent to the server.
var store = new dojox.data.QueryReadStore({url:'/search.php'});
store.fetch({query:{name:'a'}, queryOptions:{ignoreCase:false}});
// Since "serverQuery" is given, it overrules and those data are
// sent to the server.
var store = new dojox.data.QueryReadStore({url:'/search.php'});
store.fetch({serverQuery:{name:'a'}, queryOptions:{ignoreCase:false}});
<div dojoType="dojox.data.QueryReadStore"
jsId="store2"
url="../tests/stores/QueryReadStore.php"
requestMethod="post"></div>
<div dojoType="dojox.grid.data.DojoData"
jsId="model2"
store="store2"
sortFields="[{attribute: 'name', descending: true}]"
rowsPerPage="30"></div>
<div dojoType="dojox.Grid" id="grid2"
model="model2"
structure="gridLayout"
style="height:300px; width:800px;"></div>
ConstructorsBack to top
The request contains the data as defined in the Read-API.
Additionally there is following keyword "serverQuery".
The *serverQuery* parameter, optional.
This parameter contains the data that will be sent to the server.
If this parameter is not given the parameter "query"'s
data are sent to the server. This is done for some reasons:
- to specify explicitly which data are sent to the server, they
might also be a mix of what is contained in "query", "queryOptions"
and the paging parameters "start" and "count" or may be even
completely different things.
- don't modify the request.query data, so the interface using this
store can rely on unmodified data, as the combobox dijit currently
does it, it compares if the query has changed
- request.query is required by the Read-API
I.e. the following examples might be sent via GET:
fetch({query:{name:"abc"}, queryOptions:{ignoreCase:true}})
the URL will become: /url.php?name=abc
fetch({serverQuery:{q:"abc", c:true}, query:{name:"abc"}, queryOptions:{ignoreCase:true}})
the URL will become: /url.php?q=abc&c=true
// The serverQuery-parameter has overruled the query-parameter
// but the query parameter stays untouched, but is not sent to the server!
// The serverQuery contains more data than the query, so they might differ!
FunctionsBack to top
See dojo.data.util.simpleFetch.fetch() this is just a copy and I adjusted
only the paging, since it happens on the server if doClientPaging is
false, thx to http://trac.dojotoolkit.org/ticket/4761 reporting this.
Would be nice to be able to use simpleFetch() to reduce copied code,
but i dont know how yet. Ideas please!
See dojo.data.api.Read.hasAttribute()
This function tests whether the item passed in is indeed a valid 'attribute' like type for the store.
It throws an error if item is not valid, so you can call it in every method that needs to
throw an error when item is invalid.
If the data from servers needs to be processed before it can be processed by this
store, then this function should be re-implemented in subclass. This default
implementation just return the data unchanged.
