A <select> defaulting to "Default system audio", populated from the
enumerated render devices — picking one overrides which device
WhispAssist loopback-captures from instead of always following
Windows' system default.
list_audio_devices runs the (blocking, COM-based) enumeration off the
async runtime via spawn_blocking, same pattern as import_pst_core.
start_recording now passes settings.audio_output_device through to
WasapiCapture::start instead of always capturing the default device.
Device::get_id() string; None = system default (unchanged behavior).
#[serde(default)] so an existing settings.json without this key still
deserializes.
AudioCapture::start now takes an optional device id (Device::get_id())
instead of always resolving the system default render device.
open_capture_session/find_render_device resolve the configured device,
falling back to the system default if it's no longer present (same
degrade-gracefully spirit as the existing mid-recording reconnect,
which now retries the *same* selection rather than switching to
whatever's currently default).
Adds list_render_devices() (new AudioDeviceInfo) for the Settings
picker — enumeration was already fully supported by the wasapi crate,
just unused until now.
Generate tags mirrors Generate summary: reads the transcript (the same
build_prompt assembly, non-streamed) and suggests 1-8 tags, merged
into the working tag list rather than replacing it outright so a
manually-added tag never silently disappears.
Replaces the old comma-separated text input with GitHub-topic-style
chips: typing a comma commits everything before it as its own chip
immediately, each chip is removable via its 'x', and clicking a chip's
label filters the meeting list to that tag (meetings.filterByTag).
Writable $derived instead of local $state, so clicking a tag chip
elsewhere (the Tags panel) is reflected here too, not just changes
made directly in this dropdown.
Same mechanism as the sidebar's tag dropdown (load with a tag filter,
drop search mode), just callable from anywhere a tag chip is clicked —
not only the dropdown itself.
Rounded pill using the existing --accent-soft/--accent tokens.
Clicking the label filters the meeting list to this tag; an optional
'x' removes it from whatever list rendered it (the component doesn't
know or care what "remove" means to its caller).
Non-streaming — a 1-8 tag reply is short enough that a second
token-stream event isn't worth wiring up. Implemented for Ollama and
the OpenAI-compatible path (reusing summarize()'s SSE plumbing, just
accumulating instead of forwarding to a TokenSink); the Anthropic stub
mirrors its existing "not built yet" summarize() error.
parse_tags/sanitize_tag/strip_ordinal_prefix handle a model that
ignores "comma-separated, nothing else" (numbered lists, newlines,
quotes) and normalize everything to lowercase, hyphenated, chip-safe
tags, capped at 8 and deduped.
Nothing in the UI displayed or let you change a meeting's title
outside the sidebar list — recordings default to "Untitled meeting"
with no way to fix that. Adds an inline-editable header; blur/change
commits the rename.
attachEvent and renameMeeting can both change a meeting's title now, so
both refresh the list (not just the selected-detail view) to keep the
sidebar in sync.
rename_meeting backs a user-editable title (T2.2) — recordings had no
way to change their default "Untitled meeting" name from the UI at all.
attach_meeting_to_event now also copies the linked event's subject onto
the meeting's title when it has one: linking is meant to say "this
recording is that meeting," so the title should follow.
A real mailbox import is thousands of events — searching/filtering
belongs where the user is actually tying a meeting to one, not just
the general browse list in Settings. Defaults the date filter to the
selected meeting's own recording date, since that's almost always the
event being linked.
Prefills the file path from settings once loaded (without clobbering
an in-progress browse/edit) and persists it after a successful import,
so the user doesn't have to re-browse to the same file every launch.
The new checkbox just flips pst_auto_sync, which lib.rs's one-shot
startup pass reads.
Mirrors the existing sync-job-resume pattern right above it: runs once
at startup if pst_auto_sync is on and a path is remembered, no idle
timer (NFR-RES-1). Re-import is dedup'd by (source, raw_uid), so this
just catches up on new/changed events since last launch.
import_pst_core takes &AppHandle/&dyn Store directly instead of the
State<AppState> extractor, so the startup auto-sync pass (lib.rs) can
run the same import logic without going through a Tauri command.
Backing fields for remembering the imported .pst path across sessions
and an opt-in startup auto-resync (T6.2). #[serde(default)] so an
existing settings.json without these keys still deserializes.
expand_rrule reused dtstart's raw UTC time-of-day for every occurrence,
so a weekly meeting spanning a DST transition (e.g. created in winter
CST, recurring into summer CDT) drifted an hour once displayed in the
machine's local timezone.
Rewrites occurrence generation on chrono's Local/NaiveDate (already a
transitive dependency via sqlx, now promoted to direct) instead of
hand-rolled epoch-day arithmetic: each occurrence keeps dtstart's local
wall-clock hour/minute, re-resolving the UTC offset per occurrence date
so DST is applied correctly for that specific day. civil_from_days/
weekday_from_epoch_day/is_leap_year/days_in_month are no longer needed
(chrono's NaiveDate replaces them) and are removed.
RRULE was previously ignored entirely, so a recurring meeting only ever
stored its first occurrence — invisible to any later date filter/lookup.
Adds RRULE parsing + occurrence expansion (DAILY/WEEKLY/MONTHLY/YEARLY,
INTERVAL/COUNT/UNTIL/BYDAY/BYMONTHDAY/BYMONTH) covering every pattern
found in a real 7.2GB mailbox (ADR-0008), each occurrence stored as its
own row keyed by "{uid}@{ymd}" so re-import dedup still applies.
Also fixes a real parser bug found via that mailbox: this libpst build
joins multi-value BYDAY with `;` instead of RFC 5545's `,`
(BYDAY=MO;TU;WE;TH;FR), which previously made RRULE parsing bail out
silently for any event with more than one weekday.
Indefinite rules (no COUNT/UNTIL — only YEARLY holidays in practice) are
capped at 10 years/500 occurrences; ponytail-flagged as the ceiling to
raise if a real series needs more.
readpst writes some failure messages to stdout rather than stderr, so a
failing run with empty stderr surfaced as a blank "parse failed: " error
with no diagnostic content.
Tauri commands reject with a {kind, message} object, not a native Error,
so `e instanceof Error ? e.message : String(e)` always fell through to
String(e) -> "[object Object]", hiding the actual readpst failure reason.
iamdoubz