66b3b7e4c8
Replaces the `.transcript.txt` sidecar with a structured `.json` file for recording metadata. This change consolidates transcript text, duration, and other potential metadata into a single, extensible JSON object. This also refactors the `TranscriptState` enum to better represent the on-disk state (absence of file means pending) and the in-memory representation. The `Transcript` enum now specifically models the terminal outcomes of the transcriber (`Silent` or `Content`). The commit includes updates to documentation, data structures, path handling, and various tests to align with the new metadata format.
137 lines
5.3 KiB
Rust
137 lines
5.3 KiB
Rust
//! Filesystem-fixture builders for case directories, recordings,
|
|
//! oneliner state files, and close markers.
|
|
//!
|
|
//! All functions use synchronous `std::fs` (tests run inside
|
|
//! `#[tokio::test]` but these writes are short and their output is
|
|
//! consumed sequentially, so there is no win from async I/O here). A
|
|
//! few of the oneliner_api tests use `tokio::fs` directly because they
|
|
//! need to run inside the same async flow; that's fine — the two
|
|
//! coexist.
|
|
|
|
use std::path::{Path, PathBuf};
|
|
use std::time::{Duration, SystemTime};
|
|
|
|
use doctate_common::oneliners::OnelinerState;
|
|
use doctate_common::{ONELINER_FILENAME, RECORDING_META_SUFFIX, RecordingMeta, Transcript};
|
|
use doctate_server::paths::CLOSE_MARKER;
|
|
use filetime::FileTime;
|
|
|
|
/// Create `<data_path>/<slug>/<case_id>/` and return its path. This is
|
|
/// the only "has to be on disk for the route to see it" fixture —
|
|
/// recordings and derived artefacts layer on top.
|
|
pub fn seed_case(data_path: &Path, slug: &str, case_id: &str) -> PathBuf {
|
|
let dir = data_path.join(slug).join(case_id);
|
|
std::fs::create_dir_all(&dir).unwrap();
|
|
dir
|
|
}
|
|
|
|
/// Write `<stem>.m4a` with placeholder bytes, plus optionally a
|
|
/// `<stem>.json` sidecar that mirrors the transcribe outcome.
|
|
///
|
|
/// `transcript`:
|
|
/// - `None` → no sidecar written (recording stays in `Pending` state).
|
|
/// - `Some(text)` with whitespace-only `text` → sidecar with
|
|
/// `Transcript::Silent` (matches the old `transcript.txt = ""` rule).
|
|
/// - `Some(text)` non-empty → sidecar with `Transcript::Content`.
|
|
///
|
|
/// Returns the audio filename (no path) so the caller can feed it back
|
|
/// to the delete endpoint.
|
|
pub fn seed_recording(case_dir: &Path, stem: &str, transcript: Option<&str>) -> String {
|
|
let filename = format!("{stem}.m4a");
|
|
std::fs::write(case_dir.join(&filename), b"audio-bytes").unwrap();
|
|
if let Some(text) = transcript {
|
|
seed_recording_meta(case_dir, stem, transcript_for(text), None);
|
|
}
|
|
filename
|
|
}
|
|
|
|
/// Like [`seed_recording`] but the sidecar carries a real duration
|
|
/// (`duration_seconds: Some(42)`). Used by tests that assert the
|
|
/// duration is rendered or cleaned up.
|
|
pub fn seed_recording_with_sidecars(case_dir: &Path, stem: &str, transcript: &str) -> String {
|
|
let filename = format!("{stem}.m4a");
|
|
std::fs::write(case_dir.join(&filename), b"audio-bytes").unwrap();
|
|
seed_recording_meta(case_dir, stem, transcript_for(transcript), Some(42));
|
|
filename
|
|
}
|
|
|
|
/// Write a fully-specified `<stem>.json` next to an already-seeded
|
|
/// `<stem>.m4a`. Use when a test needs precise control over the
|
|
/// transcript variant or duration; the higher-level helpers above
|
|
/// cover the common cases.
|
|
pub fn seed_recording_meta(
|
|
case_dir: &Path,
|
|
stem: &str,
|
|
transcript: Transcript,
|
|
duration_seconds: Option<u32>,
|
|
) {
|
|
let meta = RecordingMeta {
|
|
transcript,
|
|
duration_seconds,
|
|
};
|
|
let bytes = serde_json::to_vec(&meta).unwrap();
|
|
std::fs::write(
|
|
case_dir.join(format!("{stem}.{RECORDING_META_SUFFIX}")),
|
|
bytes,
|
|
)
|
|
.unwrap();
|
|
}
|
|
|
|
/// Map a free-text transcript to the `Transcript` variant the worker
|
|
/// would have produced for it: whitespace-only → `Silent`, otherwise
|
|
/// `Content`. Mirrors the old `transcript.txt = "" ⇒ Silent` rule so
|
|
/// existing tests keep their meaning.
|
|
fn transcript_for(text: &str) -> Transcript {
|
|
if text.trim().is_empty() {
|
|
Transcript::Silent
|
|
} else {
|
|
Transcript::Content {
|
|
text: text.to_owned(),
|
|
}
|
|
}
|
|
}
|
|
|
|
/// Write an m4a with an artificial `mtime` in the past (relative to
|
|
/// now). Used by the retention-sweep tests to simulate "recorded N days
|
|
/// ago" without a clock abstraction.
|
|
pub fn seed_recording_with_age(case_dir: &Path, stem: &str, age: Duration) {
|
|
let filename = format!("{stem}.m4a");
|
|
let path = case_dir.join(&filename);
|
|
std::fs::write(&path, b"audio").unwrap();
|
|
let target = SystemTime::now() - age;
|
|
filetime::set_file_mtime(&path, FileTime::from_system_time(target)).unwrap();
|
|
}
|
|
|
|
/// Persist `OnelinerState::Ready { text }` with a fixed `generated_at`.
|
|
/// Used by UI-rendering tests that care about the displayed text, not
|
|
/// the timestamp.
|
|
pub fn seed_oneliner_ready(case_dir: &Path, text: &str) {
|
|
let state = OnelinerState::Ready {
|
|
text: text.to_owned(),
|
|
generated_at: "2026-04-18T10:00:00Z".into(),
|
|
};
|
|
seed_oneliner_state(case_dir, &state);
|
|
}
|
|
|
|
/// Persist an arbitrary [`OnelinerState`] as `oneliner.json`.
|
|
pub fn seed_oneliner_state(case_dir: &Path, state: &OnelinerState) {
|
|
let bytes = serde_json::to_vec(state).unwrap();
|
|
std::fs::write(case_dir.join(ONELINER_FILENAME), bytes).unwrap();
|
|
}
|
|
|
|
/// Write a `.closed` marker with a caller-chosen `closed_at` string.
|
|
/// Tests that need a specific age pass a historical RFC3339 stamp; a
|
|
/// helper for that is [`past_rfc3339`].
|
|
pub fn write_closed_marker(case_dir: &Path, closed_at: &str) {
|
|
let json = format!(r#"{{"closed_at":"{closed_at}"}}"#);
|
|
std::fs::write(case_dir.join(CLOSE_MARKER), json).unwrap();
|
|
}
|
|
|
|
/// Format "now - age" as RFC3339 UTC. Used to mint historical
|
|
/// `closed_at` strings for retention-purge tests.
|
|
pub fn past_rfc3339(age: Duration) -> String {
|
|
use time::format_description::well_known::Rfc3339;
|
|
let t = time::OffsetDateTime::now_utc() - time::Duration::seconds(age.as_secs() as i64);
|
|
t.format(&Rfc3339).unwrap()
|
|
}
|