AMR GraphQL Querying

This topic tells you how to connect to the GraphQL playground and how to query each supported data model. For information about the AMR data models, see AMR Data Models.

Connecting to AMR GraphQL

There are two ways you can perform GraphQL queries:

Using the GraphQL playground
Using cURL

VMware recommends enabling ingress. The Supply Chain Security Tools for Tanzu – Store and Artifact Metadata Repository (AMR) packages share the same ingress configuration. For information about enabling ingress, see Ingress support for Supply Chain Security Tools - Store.

Retrieving the AMR GraphQL Access Token

When you access the AMR GraphQL by using the GraphQL playground or cURL, you must retrieve the access token.

To fetch the token, run:

kubectl -n metadata-store get secret amr-graphql-view-token -o json | jq -r ".data.token" | base64 -d

Connecting to AMR GraphQL playground

To connect to the AMR GraphQL playground when you enabled ingress, visit https://amr-graphql.INGRESS-DOMAIN/play.

Where INGRESS-DOMAIN is the domain of the ingress you want to use.

In the Headers tab at the bottom of the query window, add a JSON block containing the following authentication header:

{
  "Authorization": "Bearer ACCESS-TOKEN"
}

Where ACCESS-TOKEN is the AMR GraphQL access token.

You can use this to write and execute your own GraphQL queries to fetch data from the AMR.

Connecting to AMR GraphQL through cURL

To connect to the AMR GraphQL using cURL when you enabled ingress, you first need the AMR GraphQL access token and its CA certificate.

To fetch the AMR GraphQL CA certificate:

kubectl get secret ingress-cert -n metadata-store -o json | jq -r '.data."ca.crt"' | base64 -d > /tmp/graphql-ca.crt

After the token and certificate are retrieved, you can use cURL to perform GraphQL queries by using the https://amr-graphql.INGRESS-DOMAIN/query endpoint.

Where INGRESS-DOMAIN is the domain of the ingress you want to use.

For example:

curl "https://amr-graphql.<ingress-domain>/query" \
  --cacert /tmp/graphql-ca.crt \
  -H "Authorization: Bearer ACCESS-TOKEN" \
  -H 'accept: application/json' \
  -H 'content-type: application/json' \
  --data-raw '{"query":"query getAppAcceleratorRuns { appAcceleratorRuns(first: 250){ nodes { guid name namespace timestamp } pageInfo{ endCursor hasNextPage } } }"}' | jq .

Where:

ACCESS-TOKEN is the AMR GraphQL access token
/tmp/graphql-ca.crt is the file location containing the AMR GraphQL CA certificate

You can use this to write and execute your own GraphQL queries to fetch data from the AMR.

This section uses cURL to query the AMR GraphQL endpoint, but you can use other similar tools to access the endpoint and provide them with the AMR GraphQL access token and CA certificate.

Querying for locations (alpha)

This section tells you about GraphQL query arguments, and the list of fields available for Location.

Locations query arguments

You can specify the following supported arguments when querying for Location. They are all optional.

query: expects an object that specifies additional arguments used to query. The following arguments are supported in this query object:
- reference: string UID representing the location, as a String value. A location’s reference is automatically set to be the kube-system namespace UID by the AMR. It is not configurable by the user.
For example:
```
locations(query:{reference: "f4f63c6a10ed7cfb06dfb03c2b2d6a9d5bbe95931c71f8cb346fd2284b1e5d82"})
```
- labels: specify labels the location must contain. Expects a list of labels. A label has key and value fields of String type, at least one of those fields must be provided when querying with labels.
Note
The result returns locations that have all given labels (i.e. this is an AND operation).

For example:
```
locations(query:{ labels:[{value: "run"},
                          {key: "env"},
                          {key: "region", value: "east1b"} ]
})
```

Locations fields

You can choose the following felids to return in the GraphQL query. You must specify at least one field.

reference: string UID representing the location
labels: labels associated with the location, has the following fields
- key: key of the label
- value: value of the label

Sample locations queries

Query for location with the reference "test-cluster". The query results return information about the reference and labels showing both key and value.
```
query getLocationByReference {
locations(query:{reference: "test-cluster"}) {
  reference
  labels {
    key
    value
  }
}
}
```
Query for locations that contain two labels, one with key "env" and one {key: "region", value: "east1b"}. The query results return information about reference only.
```
query getLocationByLabel {
locations(query:{labels:[{key: "env"},
                          {key: "region", value: "east1b"} ]
}) {
  reference
}
}
```

Querying for AppAcceleratorRuns (alpha)

This section tells you about GraphQL query arguments, and lists the fields available for AppAcceleratorRuns and AppAcceleratorFragments.

AppAcceleratorRuns query arguments

(Optional) You can specify the following supported arguments when querying for AppAcceleratorRuns.

query: expects an object that specifies additional arguments to query. The following arguments are supported in this query object:
guid: UID identifying the run, as a String value. Each AppAcceleratorRun is automatically assigned a UID.

For example:
```
appAcceleratorRuns(query:{guid: "d2934b09-5d4c-45da-8eb1-e464f218454e"})
```
source: string representing the client used to run the accelerator. Supported values include TAP-GUI, VSCODE, and INTELLIJ.

For example:
```
appAcceleratorRuns(query:{source: "TAP-GUI"})
```
username: string representing the user name of the person who runs the accelerator, as captured by the client UI. For example:
```
appAcceleratorRuns(query:{username: "homer.simpson"})
```
namespace and name: strings representing the accelerator that was used to create an application. For example:
```
appAcceleratorRuns(query:{name: "tanzu-java-web-app"})
```
appAcceleratorRepoURL, appAcceleratorRevision, and appAcceleratorSubpath: actual location in VCS (Version Control System) about the accelerator sources used. For example:
```
appAcceleratorRuns(query:{
appAcceleratorRepoURL: "https://github.com/vmware-tanzu/application-accelerator-samples.git",
appAcceleratorRevision: "v1.6"
})
```
timestamp: string representation of the exact time the accelerator ran. You can query for runs that happened before or after a particular instant: For example:
```
appAcceleratorRuns(query: {timestamp: {after: "2023-10-11T13:40:46.952Z"}})
```

AppAcceleratorRuns fields

You can choose the following fields to return in the GraphQL query. See the section above for details about those fields. You must specify at least one field.

guid: UID identifying the run
source: string representing the client used to run the accelerator
username: string representing the user name of the person who ran the accelerator
namespace and name: strings representing the accelerator which was used to create an application
appAcceleratorRepoURL, appAcceleratorRevision, and appAcceleratorSubpath: actual location in VCS of the sources of the accelerator used
appAcceleratorSource: VCS information of the sources of the accelerator used, but navigable as a Commit
timestamp: the exact time the accelerator was run
appAcceleratorFragments: a one-to-many container of nodes representing the fragment versions used in each AppAcceleratorRun. Those fragment nodes share many of the fields with AppAcceleratorRun, with the same semantics but applied to the particular fragment. Those include:
- namespace and name: strings representing the identity of the fragment
- appAcceleratorFragmentSourceRepoURL , appAcceleratorFragmentSourceRevision, and appAcceleratorFragmentSourceSubpath: actual location in VCS of the sources of the fragment used
- appAcceleratorFragmentSource: VCS information of the sources of the fragment, but navigable as a Commit

Sample Application Accelerator queries

Get the list of all Application Accelerator runs, with the fragments used for each.

query getAllAcceleratorRuns {
  appAcceleratorRuns {
    nodes {
      name
      appAcceleratorFragments {
        nodes {
          name
        }
      }
    }
  }
}

Additional resources

AMR Data Models