gcf-fetch

Cloud Functions that fetch data from public APIs and store it in Google Cloud Storage.

• Runtime: Go 1.20
• Gen: Cloud Functions (2nd gen)

Features

• Fetch data is stored in a path based on the API's URL. ( like ghq )
  • Therefore, even if you get various API data, you do not need to manage the bucket's path.
• Fetch data is managed by Object Versioning.
• GCS price are optimized by Object Lifecycle Management.
  • The storage class is set to change from Standard to Coldline after 7 days from object creation and from Coldline to Archive after 30 days (easy to change).
• zl (zap based logger) for logging by severity level.
  • Logs in JSON format so it can check the element contents in detail with Cloud Logging.
  • You can also check the CloudEvent that triggered Functions.

Cloud Logging's log example (CloudEvent)

{
  "insertId": "xxxxxxxxxxxxxxxxxxxxxx",
  "jsonPayload": {
    "timestamp": "2022-06-12T00:45:17.427119741Z",
    "function": "github.com/nkmr-jp/gcf-fetch.parseEvent",
    "cloudEventContext": "Context Attributes,\n  specversion: 1.0\n  type: google.cloud.pubsub.topic.v1.messagePublished\n  source: //pubsub.googleapis.com/projects/[your project id]/topics/fetch-topic\n  id: xxxxxxxxxxx\n  time: 2022-06-12T00:45:14.378Z\n  datacontenttype: application/json\n",
    "cloudEventData": {
      "subscription": "projects/[your project id]/subscriptions/eventarc-asia-northeast1-fetch-xxxxx-sub-xxx",
      "message": {
        "publishTime": "2022-06-12T00:45:14.378Z",
        "messageId": "4863463195745766",
        "data": "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
      }
    },
    "caller": "https://github.com/nkmr-jp/gcf-fetch/blob/v1.0.0/fetch.go#L89",
    "version": "v1.0.0",
    "message": "CLOUD_EVENT_RECEIVED"
  },
  "resource": {
    "type": "cloud_run_revision",
    "labels": {
      "service_name": "fetch",
      "project_id": "[your project id]",
      "configuration_name": "fetch",
      "revision_name": "fetch-xxxx-xiv",
      "location": "asia-northeast1"
    }
  },
  "timestamp": "2022-06-12T00:45:17.427272Z",
  "severity": "INFO",
  "labels": {
    "goog-managed-by": "cloudfunctions",
    "instanceId": "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
  },
  "logName": "projects/[your project id]/logs/run.googleapis.com%2Fstderr",
  "receiveTimestamp": "2022-06-12T00:45:17.670946465Z"
}

Prepare

If you haven't already, install and set up the Cloud SDK.

Quick Start

Create GCP resources

make init

Run test

make test

Deploy to google cloud functions

make deploy

Send pub/sub event

make send URL="https://api.github.com/users/github"

Open resources in console

make open

Usage Example

The api data saved in GCS can be used for various purposes such as data analysis and machine learning by loading it into BigQuery. Of course, it can also be used in applications.

You can also run gcf-fetch periodically to collect data such as public APIs. Public APIs can be found, for example, in the Repository below.

List of public APIs.
GitHub - public-apis/public-apis: A collective list of free APIs

See

• https://github.com/GoogleCloudPlatform/golang-samples/tree/main/functions/functionsv2
• https://cloud.google.com/functions/docs/2nd-gen/getting-started#pubsub