Struct SpokenDraft 

pub struct SpokenDraft {
    pub lines: Vec<String>,
}

A draft as it would be read out loud — every argument, nothing summarised.

The reviewable object of a message is the message, and that rule does not change when the reviewer is listening instead of looking. What changes is that a listener cannot skim: they hear it once, in order, at speaking speed, so the only safe spoken offer is one that utters the whole thing. A paraphrase read aloud is not a smaller review, it is a different document — and the field it leaves out (one more address on the to line) is exactly the field an injection would add.

So this is DraftView’s three buckets spoken in reading order, and it inherits that type’s guarantee: every argument key appears, with a test on it. The only thing it decides is wording.

Fields

lines: Vec<String>

The draft, in speakable lines. Concatenate with pauses between.

Implementations

impl SpokenDraft

pub fn chars(&self) -> usize

How much speech this is. Characters rather than words because that is what a TTS leg is actually handed, and the two are proportional at any rate worth caring about.

pub fn text(&self) -> String

Trait Implementations

impl Debug for SpokenDraft

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for SpokenDraft

fn default() -> SpokenDraft

Returns the “default value” for a type.

impl Eq for SpokenDraft

impl PartialEq for SpokenDraft

fn eq(&self, other: &SpokenDraft) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl StructuralPartialEq for SpokenDraft