Retries

Sometimes API requests fail because of intermediate problems between the client and the server. These can be retried.

Default

By default, these server-side errors (i.e. 5xx) are retried with an exponential-backoff strategy. This entails the client waits progressively longer and longer when it encounters the same retriable error.

If the server response specifies a "Retry-After" header it will take precedence over any delays.

Customize the backoff strategy

The backoff strategy is customizable by providing a simple class when initializing the client.

For instance, in this example we wait 10 seconds every time we get an error we can retry:

import gcapi
import httpx

class RetryAfterTenSeconds(gcapi.retries.BaseRetryStrategy):
    def get_delay(self, latest_response: httpx.Response):
        return 10

token = 'my-personal-api-token'
client = gcapi.Client(
            token=token,
            retry_strategy=RetryAfterTenSeconds,
        )

Retries

Sometimes API requests fail in because of intermediate problems between the client and the server. These can be retried.

Default

By default, these server-side errors (i.e. 5xx) are retried with an exponential-backoff strategy. This entails the client waits progressively longer and longer when it encounters the same retriable error.

If the server response specifies a "Retry-After" header it will take precedence over any delays.

Customize the backoff strategy

The backoff strategy is customizable by providing a simple class when initializing the client.

For instance, in this example we wait 10 seconds everytime we get an error we can retry:

import gcapi
import httpx

class RetryAfterTenSeconds(gcapi.retries.BaseRetryStrategy):
    def get_delay(self, latest_response: httpx.Response):
        return 10

token = 'my-personal-api-token'
client = gcapi.Client(
            token=token,
            retry_strategy=RetryAfterTenSeconds,
        )