Struct TextStream 

pub struct TextStream { /* private fields */ }

A streaming answer, read one delta at a time.

let mut stream = ai.prompt("Tell me a story").stream().await?;
while let Some(delta) = stream.next().await? {
    print!("{delta}");
}

Implementations

impl TextStream

pub fn events(body: Body, decode: Decoder) -> TextStream

A stream over server-sent events.

pub fn lines(body: Body, decode: Decoder) -> TextStream

A stream over newline-delimited JSON.

pub fn scripted(deltas: impl IntoIterator<Item = String>) -> TextStream

A stream of deltas decided in advance, for fakes and tests.

pub fn measured( self, provider: &'static str, model: impl Into<String>, ) -> TextStream

Report an ai.call event when this stream runs out.

Only then are the token counts known: providers put them in the last frame, so a stream cannot be measured when it is opened.

pub async fn next(&mut self) -> Result<Option<String>>

The next piece of text, or None at the end of the answer.

Frames that carry only bookkeeping are consumed silently, so a caller never has to know what a message_delta is.

pub async fn collect(&mut self) -> Result<String>

Read the whole answer, for a caller who wanted streaming for the latency rather than for the display.

pub fn usage(&self) -> Usage

The tokens reported so far. Complete only once the stream has ended.

pub fn stop_reason(&self) -> Option<&StopReason>