Skip to content

Commit

Permalink
fix docs typo in client.md (#1160)
Browse files Browse the repository at this point in the history
`HTTP.Pool(; max::Int)` -> `HTTP.Pool(max::Int)`
  • Loading branch information
haakon-e authored Mar 14, 2024
1 parent e27493d commit 57101ba
Showing 1 changed file with 1 addition and 1 deletion.
2 changes: 1 addition & 1 deletion docs/src/client.md
Original file line number Diff line number Diff line change
Expand Up @@ -93,7 +93,7 @@ When a connection is attempted to a remote host, sometimes the connection is una

#### `pool`

Many remote web services/APIs have rate limits or throttling in place to avoid bad actors from abusing their service. They may prevent too many requests over a time period or they may prevent too many connections being simultaneously open from the same client. By default, when `HTTP.request` opens a remote connection, it remembers the exact host:port combination and will keep the connection open to be reused by subsequent requests to the same host:port. The `pool` keyword argument specifies a specific `HTTP.Pool` object to be used for controlling the maximum number of concurrent connections allowed to be happening across the pool. It's constructed via `HTTP.Pool(; max::Int)`. Requests attempted when the maximum is already hit will block until previous requests finish. The `idle_timeout` keyword argument can be passed to `HTTP.request` to control how long it's been since a connection was lasted used in order to be considered 'valid'; otherwise, "stale" connections will be discarded.
Many remote web services/APIs have rate limits or throttling in place to avoid bad actors from abusing their service. They may prevent too many requests over a time period or they may prevent too many connections being simultaneously open from the same client. By default, when `HTTP.request` opens a remote connection, it remembers the exact host:port combination and will keep the connection open to be reused by subsequent requests to the same host:port. The `pool` keyword argument specifies a specific `HTTP.Pool` object to be used for controlling the maximum number of concurrent connections allowed to be happening across the pool. It's constructed via `HTTP.Pool(max::Int)`. Requests attempted when the maximum is already hit will block until previous requests finish. The `idle_timeout` keyword argument can be passed to `HTTP.request` to control how long it's been since a connection was lasted used in order to be considered 'valid'; otherwise, "stale" connections will be discarded.

#### `readtimeout`

Expand Down

0 comments on commit 57101ba

Please sign in to comment.