Guidedog / Documentation 0.2.0

fetch¶

Types

fetch.Error :: enum u8¶

None¶

Unsupported¶: not http, https, or a file; or no libcurl on this platform.

Missing¶: no such file.

Network¶: the connection or transfer failed.

Timeout¶

Status¶: the server answered with a status other than 2xx.

Too_Large¶

fetch.Options :: struct¶

Options bound one get. The zero value is the defaults.

timeout_ms: int¶: the whole transfer; 0: 30 seconds; below 0: none.

max_bytes: int¶: 0: 64 MiB.

user_agent: string¶

fetch.Result :: struct¶

Result is what get read, or why it could not. Its data, url, and message are allocated with the allocator given to get, and the caller owns them: destroy frees them.

data: []u8¶

url: string¶: where the data came from, after redirects.

status: int¶: the HTTP status, or 0 for files.

error: Error ¶

message: string¶: what went wrong, for a report.

Procedures

fetch.destroy :: proc(r: ^Result, allocator := context.allocator)¶: destroy frees what get allocated for r, with the same allocator, and clears r.

fetch.get :: proc(location: string, options: Options, allocator := context.allocator) -> Result ¶: get reads a URL or path, allocating the result with allocator; the caller owns it and frees it with destroy. It fails with an Error in the result, never a panic: a file that is missing, a transfer that fails, times out, answers with a status other than 2xx, or grows beyond max_bytes. It blocks until the transfer ends; threads may call it at once.

Foreign imports

foreign import fetch.curl "system:curl"¶: Availability: Darwin, Linux, FreeBSD, OpenBSD, NetBSD, WASI, JS, Orca, Freestanding.