docs: fix spelling `clonable` -> `cloneable` (#191)
parent
e7fcd56b12
commit
28a13232c0
|
@ -56,14 +56,14 @@ macro_rules! log_retry {
|
|||
///
|
||||
/// This middleware always errors when given requests with streaming bodies, before even executing
|
||||
/// the request. When this happens you'll get an [`Error::Middleware`] with the message
|
||||
/// 'Request object is not clonable. Are you passing a streaming body?'.
|
||||
/// 'Request object is not cloneable. Are you passing a streaming body?'.
|
||||
///
|
||||
/// Some workaround suggestions:
|
||||
/// * If you can fit the data in memory, you can instead build static request bodies e.g. with
|
||||
/// `Body`'s `From<String>` or `From<Bytes>` implementations.
|
||||
/// * You can wrap this middleware in a custom one which skips retries for streaming requests.
|
||||
/// * You can write a custom retry middleware that builds new streaming requests from the data
|
||||
/// source directly, avoiding the issue of streaming requests not being clonable.
|
||||
/// source directly, avoiding the issue of streaming requests not being cloneable.
|
||||
pub struct RetryTransientMiddleware<
|
||||
T: RetryPolicy + Send + Sync + 'static,
|
||||
R: RetryableStrategy + Send + Sync + 'static = DefaultRetryableStrategy,
|
||||
|
@ -149,7 +149,8 @@ where
|
|||
// since the byte abstraction is a shared pointer over a buffer.
|
||||
let duplicate_request = req.try_clone().ok_or_else(|| {
|
||||
Error::Middleware(anyhow!(
|
||||
"Request object is not clonable. Are you passing a streaming body?".to_string()
|
||||
"Request object is not cloneable. Are you passing a streaming body?"
|
||||
.to_string()
|
||||
))
|
||||
})?;
|
||||
|
||||
|
|
Loading…
Reference in New Issue