Queries
On this page, you can learn how to use Apollo to attach GraphQL query results to your Angular UI. This guide assumes some familiarity with GraphQL itself. You can read about GraphQL queries themselves in detail at graphql.org.
One of our core values is "it's just GraphQL." When using Apollo Client, you don't have to learn anything special about the query syntax, since everything is just standard GraphQL. Anything you can type into the GraphQL query IDE, you can also put into your Apollo Client code.
#
Basic QueriesWhen we are using a basic query, we can use the Apollo.watchQuery
method in a
very simple way. We simply need to parse our query into a GraphQL document using
the graphql-tag
library.
For instance, in GitHunt, we want to display the current user (if logged in) in
the Profile
component:
The watchQuery
method returns a QueryRef
object which has the valueChanges
property that is an Observable
.
We can see that the result object contains loading
, a Boolean indicating if
the query is "in-flight." The observable will only emit once when the query is
complete, and loading
will be set to false unless you set the watchQuery
parameters notifyOnNetworkStatusChange
to true. Once
the query has completed, it will also contain a data
object with
currentUser
, the field we've picked out in CurrentUserForProfile
.
We can expect the data.currentUser
to change as the logged-in-ness of the
client and what it knows about the current user changes over time. That
information is stored in Apollo Client's global cache, so if some other query
fetches new information about the current user, this component will update to
remain consistent.
It's also possible to fetch data only once. The query
method of Apollo
service returns an Observable
that also resolves with the same result as
above.
#
What is QueryRefAs you know, Apollo.query
method returns an Observable that emits a result,
just once. Apollo.watchQuery
also does the same, except it can emit multiple
results. (The GraphQL query itself is still only sent once, but the watchQuery
observable can also update if, for example, another query causes the object to
be updated within Apollo Client's global cache.)
So why doesn't Apollo.watchQuery
expose an Observable?
Apollo service and ApolloClient share pretty much the same API. It makes things easy to understand and use. No reason to change it.
In ApolloClient.watchQuery
returns an Observable, but not a standard one, it
contains many useful methods (like refetch()
) to manipulate the watched query.
A normal Observable, has only one method, subscribe()
.
To use that Apollo's Observable in RxJS, we would have to drop those methods. Since they are necessary to use Apollo to its full potential, we had to come up with a solution.
This is why we created QueryRef
.
The API of QueryRef
is very simple. It has the same methods as the Apollo
Observable we talked about. To subscribe to query results, you have to access its
valueChanges
property which exposes a clean RxJS Observable.
It's worth mentioning that QueryRef
accepts two generic types. More about that in
).
options
#
Providing watchQuery
and query
methods expect one argument, an object with options. If
you want to configure the query, you can provide any available option in the
same object where the query
key lives.
If your query takes variables, this is the place to pass them in:
#
Using with AsyncPipeIn Angular, the simplest way of displaying data that comes from Observable is to
put AsyncPipe
on top of the property inside the UI. You can also achieve this
with Apollo.
Note: Using async pipe more than once in your template will trigger the query for each pipe. To avoid this situation, subscribe to the data in the component, and display the data from the component's property.
An Observable returned by watchQuery().valueChanges
holds the actual result
under the data
field, so you can not directly access one of the properties of
that object.
This is why we created SelectPipe
. The only argument it receives is the name
of the property you want to get from data
.
The result of the query has this structure:
Without using SelectPipe
, you would get the whole object instead of only the
data.feed
.
#
Using with RxJSApollo
is compatible with RxJS by using same Observable so it can be used with
operators.
What's really interesting is that, because of this, you can avoid using SelectPipe
:
The map
operator we are using here is provided by the RxJS Observable
which
serves as the basis for the Observable
.
To be able to use the map
operator (and most others like switchMap
,
filter
, merge
, ...) these have to be explicitly imported as done in the
example: import {map} from 'rxjs/operators'
.