mirror of
https://github.com/yusufipk/dikte.git
synced 2026-09-11 10:56:10 +00:00
Compare commits
4
Commits
v1.0.0
...
update-check
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
084c48151d | ||
|
|
6bf5c8b349 | ||
|
|
2927e15833 | ||
|
|
f55e201785 |
@@ -107,7 +107,10 @@ Meetings are not supported there yet; the details are in the
|
||||
two global shortcuts, whose keys are its two arguments, or the ones already in
|
||||
your settings when it is given none. `./scripts/update.sh` pulls and puts all of
|
||||
that back; `./scripts/uninstall.sh` takes it away again and leaves your settings
|
||||
and dictations alone unless you pass `--purge`.
|
||||
and dictations alone unless you pass `--purge`. Dikte looks at the releases page
|
||||
once a day and puts a line in the tray menu when a newer version is out, which
|
||||
opens the page rather than installing anything; the General tab turns that off
|
||||
or runs it on the spot.
|
||||
|
||||
Speech to text and cleanup each pick a provider in the settings window, and both
|
||||
run here by default, on models of your own. The cloud is the other option:
|
||||
@@ -130,6 +133,7 @@ set next to it.
|
||||
| Speak a command to an agent | Tray menu → *Ask Claude*, or `dikte ask` |
|
||||
| Start / end a meeting | Tray menu → *Record a meeting*, or `dikte meeting` |
|
||||
| Settings | Tray menu → *Settings*, or `dikte settings` |
|
||||
| Look for a newer version | General tab → *Check now*, or `dikte update` |
|
||||
| Reload after an update | Tray menu → *Restart*, or `dikte restart` |
|
||||
| Quit | Tray menu → *Quit*, or `dikte quit` |
|
||||
|
||||
@@ -240,6 +244,7 @@ api.py transcription and cleanup requests (stdlib only)
|
||||
cleanup.py who rewrites the transcript: OpenRouter, here, Claude or Codex
|
||||
ggml.py whisper.cpp and llama.cpp here: fetch, verify, keep serving
|
||||
hub.py what GitHub and Hugging Face have on offer today
|
||||
update.py whether a newer release is out, and the page it is on
|
||||
worker.py transcribe → clean up → clipboard → paste
|
||||
vad.py deciding whether a recording holds speech at all
|
||||
filetranscribe.py file transcription: ffmpeg, chunking, timestamps
|
||||
|
||||
+5
-1
@@ -104,7 +104,9 @@ toplantı kaydı henüz yok, ayrıntılar [Windows README](README.windows.md)'si
|
||||
başlatmayı ve iki global kısayolu kurar; tuşları iki argümanı, argüman
|
||||
verilmezse ayarlarında duranlar. `./scripts/update.sh` son sürümü çeker ve
|
||||
bunları yerine koyar; `./scripts/uninstall.sh` hepsini geri alır, `--purge`
|
||||
demedikçe ayarlarına ve diktelerine dokunmaz.
|
||||
demedikçe ayarlarına ve diktelerine dokunmaz. Dikte sürüm sayfasına günde bir
|
||||
kez bakar ve yeni sürüm çıkmışsa tepsi menüsüne bir satır koyar; o satır bir şey
|
||||
kurmaz, sayfayı açar. Genel sekmesi bu denetimi kapatır ya da anında çalıştırır.
|
||||
|
||||
Sesi yazıya çevirme ve temizleme, ayarlar penceresinde ayrı ayrı sağlayıcı
|
||||
seçer; ikisi de varsayılan olarak burada, kendi modellerinle çalışır. Bulutu
|
||||
@@ -127,6 +129,7 @@ yanındaki kutudan düşünme seviyesini de seçebilirsin.
|
||||
| Ajana sesle komut ver | Tepsi menüsü → *Claude'a sor*, ya da `dikte ask` |
|
||||
| Toplantıyı başlat / bitir | Tepsi menüsü → *Toplantı kaydet*, ya da `dikte meeting` |
|
||||
| Ayarlar | Tepsi menüsü → *Ayarlar*, ya da `dikte settings` |
|
||||
| Yeni sürüm var mı bak | Genel sekmesi → *Şimdi bak*, ya da `dikte update` |
|
||||
| Güncelleme sonrası yeniden yükle | Tepsi menüsü → *Yeniden başlat*, ya da `dikte restart` |
|
||||
| Çık | Tepsi menüsü → *Çık*, ya da `dikte quit` |
|
||||
|
||||
@@ -234,6 +237,7 @@ api.py transkript ve temizleme istekleri (yalnız stdlib)
|
||||
cleanup.py transkripti kim temizler: OpenRouter, burası, Claude ya da Codex
|
||||
ggml.py whisper.cpp ve llama.cpp'yi indirip burada çalıştırma
|
||||
hub.py GitHub ve Hugging Face'te bugün ne olduğu
|
||||
update.py yeni sürüm çıkmış mı, çıkmışsa hangi sayfada
|
||||
worker.py transkript → temizleme → pano → yapıştırma
|
||||
vad.py kayıtta gerçekten konuşma var mı kararı
|
||||
filetranscribe.py dosyadan transkript: ffmpeg, parçalama, zaman damgaları
|
||||
|
||||
+1
-1
@@ -10,4 +10,4 @@ business loading Qt to answer one question.
|
||||
# both the .dmg's Info.plist and the AppImage's file name are built from it. A
|
||||
# build off master rather than off a tag appends the commit to it, so that a
|
||||
# bug report from someone running "latest" names a commit.
|
||||
__version__ = "1.0.0"
|
||||
__version__ = "1.0.1"
|
||||
|
||||
+17
-4
@@ -58,10 +58,20 @@ def timestamp_model(provider, selected=""):
|
||||
return "openai/whisper-1" if provider == "openrouter" else "whisper-1"
|
||||
|
||||
|
||||
# What a gateway in front of the model answers of its own accord: the request
|
||||
# never reached the model, or the model was still working when the connection
|
||||
# was given up on. Trying again is the only thing that fixes any of them, and
|
||||
# with a long file it is worth the second try rather than losing the run.
|
||||
RETRY_STATUS = frozenset({408, 429, 500, 502, 503, 504})
|
||||
|
||||
|
||||
class ApiError(Exception):
|
||||
def __init__(self, message, status=None):
|
||||
def __init__(self, message, status=None, retryable=None):
|
||||
super().__init__(message)
|
||||
self.status = status
|
||||
# Anything not on that list is the request itself being wrong, and it
|
||||
# will be just as wrong the second time.
|
||||
self.retryable = status in RETRY_STATUS if retryable is None else retryable
|
||||
|
||||
|
||||
class Aborted(Exception):
|
||||
@@ -218,7 +228,7 @@ def explain(exc, service):
|
||||
if exc.status == 429:
|
||||
return ApiError(t("{service} is rate limiting you (HTTP 429). Try again in "
|
||||
"a moment.", service=service), exc.status)
|
||||
return ApiError(f"{service}: {exc}", exc.status)
|
||||
return ApiError(f"{service}: {exc}", exc.status, retryable=exc.retryable)
|
||||
|
||||
|
||||
def _request(url, data, headers, timeout=120, aborter=None):
|
||||
@@ -234,8 +244,11 @@ def _request(url, data, headers, timeout=120, aborter=None):
|
||||
# not the network failing. URLError is an OSError, so both land here.
|
||||
if aborter is not None and aborter.aborted:
|
||||
raise Aborted from None
|
||||
# A connection that dropped or timed out is the same bad minute as a
|
||||
# 502, so it is worth the same second try.
|
||||
raise ApiError(t("Could not connect: {reason}",
|
||||
reason=getattr(exc, "reason", exc))) from exc
|
||||
reason=getattr(exc, "reason", exc)),
|
||||
retryable=True) from exc
|
||||
except json.JSONDecodeError as exc:
|
||||
raise ApiError(t("Could not parse the response: {error}", error=exc)) from exc
|
||||
|
||||
@@ -318,7 +331,7 @@ def local_failure(service, server, exc):
|
||||
"""
|
||||
detail = server.error()
|
||||
return ApiError(f"{service}: {exc}" + (f" ({detail})" if detail else ""),
|
||||
exc.status)
|
||||
exc.status, retryable=exc.retryable)
|
||||
|
||||
|
||||
def _transcribe_request(target, audio_path, language, prompt, response_format,
|
||||
|
||||
+99
-2
@@ -33,8 +33,9 @@ if sys.platform == "darwin":
|
||||
os.environ.get("PATH", "")) if part
|
||||
)
|
||||
|
||||
from PyQt6.QtCore import QTimer, QElapsedTimer, QSocketNotifier # noqa: E402
|
||||
from PyQt6.QtGui import QAction, QIcon # noqa: E402
|
||||
from PyQt6.QtCore import (QObject, QTimer, QElapsedTimer, QSocketNotifier, # noqa: E402
|
||||
QUrl, pyqtSignal)
|
||||
from PyQt6.QtGui import QAction, QDesktopServices, QIcon # noqa: E402
|
||||
from PyQt6.QtNetwork import QLocalServer, QLocalSocket # noqa: E402
|
||||
from PyQt6.QtWidgets import QApplication, QMenu, QSystemTrayIcon # noqa: E402
|
||||
|
||||
@@ -44,11 +45,13 @@ from . import cli # noqa: E402
|
||||
from . import config as cfg # noqa: E402
|
||||
from . import ggml # noqa: E402
|
||||
from . import hotkey # noqa: E402
|
||||
from . import hub # noqa: E402
|
||||
from . import i18n # noqa: E402
|
||||
from . import integrate # noqa: E402
|
||||
from . import ipc # noqa: E402
|
||||
from . import meeting # noqa: E402
|
||||
from . import trayicon # noqa: E402
|
||||
from . import update # noqa: E402
|
||||
from .i18n import t # noqa: E402
|
||||
from .meeting import MeetingPipeline # noqa: E402
|
||||
from .overlay import Overlay # noqa: E402
|
||||
@@ -77,6 +80,37 @@ ECHO_MS = 2000
|
||||
# short enough not to sit in the corner for the rest of the hour.
|
||||
PEEK_MS = 12000
|
||||
|
||||
# When the releases page is looked at, and how often it is thought about after
|
||||
# that. The delay is there so that a check never shares the first seconds of a
|
||||
# start with the model being loaded and the desktop drawing the tray; the
|
||||
# interval is not the interval between checks, which update.py holds at a day,
|
||||
# but how often that clock is read, so that a machine left running for a week
|
||||
# still asks once a day rather than once a boot.
|
||||
UPDATE_DELAY_MS = 20000
|
||||
UPDATE_POLL_MS = 3 * 3600 * 1000
|
||||
|
||||
|
||||
class UpdateCheck(QObject):
|
||||
"""One look at the releases page, off the interface thread.
|
||||
|
||||
An object of its own because the application is not one: a plain thread
|
||||
cannot touch a widget, and a signal is the only way back onto the thread
|
||||
that may.
|
||||
"""
|
||||
|
||||
# The newer release, or None when there is nothing to say, and the reason
|
||||
# nothing could be found out instead.
|
||||
done = pyqtSignal(object, str)
|
||||
|
||||
def start(self):
|
||||
def work():
|
||||
try:
|
||||
self.done.emit(update.check(), "")
|
||||
except hub.HubError as exc:
|
||||
self.done.emit(None, str(exc))
|
||||
|
||||
threading.Thread(target=work, daemon=True).start()
|
||||
|
||||
|
||||
class Dikte:
|
||||
def __init__(self, app):
|
||||
@@ -159,6 +193,17 @@ class Dikte:
|
||||
self.meeting_ticker.setInterval(500)
|
||||
self.meeting_ticker.timeout.connect(self._meeting_tick)
|
||||
|
||||
# What the last check found, read from disk rather than asked for, so
|
||||
# that a tray built in the next line already knows to say so.
|
||||
self.update_release = update.pending()
|
||||
self.updates = UpdateCheck()
|
||||
self.updates.done.connect(self._on_update_checked)
|
||||
self.update_ticker = QTimer()
|
||||
self.update_ticker.setInterval(UPDATE_POLL_MS)
|
||||
self.update_ticker.timeout.connect(self._look_for_update)
|
||||
self.update_ticker.start()
|
||||
QTimer.singleShot(UPDATE_DELAY_MS, self._look_for_update)
|
||||
|
||||
self.tray = QSystemTrayIcon()
|
||||
self._apply_settings()
|
||||
self.tray.show()
|
||||
@@ -211,6 +256,11 @@ class Dikte:
|
||||
self.menu.addAction(self.meeting_cancel_action)
|
||||
self.menu.addSeparator()
|
||||
|
||||
# Named in _refresh_update, and hidden until a check has found one.
|
||||
self.update_action = QAction("", self.menu)
|
||||
self.update_action.triggered.connect(self.open_release_page)
|
||||
self.menu.addAction(self.update_action)
|
||||
|
||||
self.settings_action = QAction(t("Settings…"), self.menu)
|
||||
self.settings_action.triggered.connect(self.open_settings)
|
||||
self.menu.addAction(self.settings_action)
|
||||
@@ -227,6 +277,7 @@ class Dikte:
|
||||
self.tray.setContextMenu(self.menu)
|
||||
self.tray.setToolTip(t("Dikte: ready"))
|
||||
self.tray.activated.connect(self._tray_clicked)
|
||||
self._refresh_update()
|
||||
self._set_icon("audio-input-microphone")
|
||||
|
||||
def _tray_clicked(self, reason):
|
||||
@@ -901,12 +952,58 @@ class Dikte:
|
||||
if len(message) > len(first_line):
|
||||
self.tray.showMessage("Dikte", message, QSystemTrayIcon.MessageIcon.Warning, 8000)
|
||||
|
||||
# ---- updates ----------------------------------------------------------
|
||||
|
||||
def _look_for_update(self):
|
||||
"""The timer. update.py decides whether this is a request or a memory."""
|
||||
if not self.conf["update_check"]:
|
||||
return
|
||||
self.updates.start()
|
||||
|
||||
def _on_update_checked(self, release, error):
|
||||
if error:
|
||||
# Nobody asked for this, so nobody is waiting to be told it failed.
|
||||
# A machine that is offline, or a GitHub that is rate-limiting the
|
||||
# address, is not a thing to interrupt a dictation about.
|
||||
print(f"dikte: update check: {error}", file=sys.stderr)
|
||||
return
|
||||
if release is None:
|
||||
return
|
||||
self._found_update(release)
|
||||
# Once per version. A check that runs every day must not be a
|
||||
# notification every day for an update somebody has decided to skip.
|
||||
if update.announced() != release.version:
|
||||
update.mark_announced(release.version)
|
||||
self.tray.showMessage(
|
||||
"Dikte",
|
||||
t("Dikte {version} is out. The tray menu has the release page.",
|
||||
version=release.version),
|
||||
QSystemTrayIcon.MessageIcon.Information, 8000,
|
||||
)
|
||||
|
||||
def _found_update(self, release):
|
||||
self.update_release = release
|
||||
self._refresh_update()
|
||||
|
||||
def _refresh_update(self):
|
||||
release = self.update_release
|
||||
self.update_action.setVisible(release is not None)
|
||||
if release is not None:
|
||||
self.update_action.setText(
|
||||
t("Dikte {version} is out…", version=release.version))
|
||||
|
||||
def open_release_page(self):
|
||||
release = self.update_release
|
||||
QDesktopServices.openUrl(
|
||||
QUrl(release.url if release is not None else update.RELEASES_PAGE))
|
||||
|
||||
# ---- settings ---------------------------------------------------------
|
||||
|
||||
def open_settings(self):
|
||||
if self.settings_window is None:
|
||||
self.settings_window = SettingsWindow(self.conf, self.meetings)
|
||||
self.settings_window.applied.connect(self._apply_settings)
|
||||
self.settings_window.update_found.connect(self._found_update)
|
||||
self.settings_window.finished.connect(self._settings_closed)
|
||||
self.settings_window.show()
|
||||
self.settings_window.raise_()
|
||||
|
||||
@@ -20,6 +20,7 @@ import signal
|
||||
import subprocess
|
||||
import sys
|
||||
import time
|
||||
import webbrowser
|
||||
|
||||
from PyQt6.QtCore import QCoreApplication, QTimer
|
||||
|
||||
@@ -30,10 +31,12 @@ from . import cleanup
|
||||
from . import config as cfg
|
||||
from . import filetranscribe
|
||||
from . import hotkey
|
||||
from . import hub
|
||||
from . import ipc
|
||||
from . import integrate
|
||||
from . import meeting
|
||||
from . import paste
|
||||
from . import update
|
||||
from . import __version__
|
||||
|
||||
NOT_RUNNING = 3
|
||||
@@ -797,6 +800,33 @@ def cmd_integrate(opts):
|
||||
f"{verb}:\n{listing}" if paths else "Nothing to change.")
|
||||
|
||||
|
||||
def cmd_update(opts):
|
||||
"""Whether a newer Dikte has been released, and where it is.
|
||||
|
||||
It looks and nothing more: what to do about the answer is a download page,
|
||||
because the AppImage, the disk image, the Windows setup and a checkout are
|
||||
four different installations and only their owner knows which one this is.
|
||||
"""
|
||||
try:
|
||||
release = update.latest(refresh=True)
|
||||
except hub.HubError as exc:
|
||||
return fail(opts, exc)
|
||||
# Written down even when there is nothing new, so that the application does
|
||||
# not go and ask the same question an hour later.
|
||||
update.remember(release)
|
||||
waiting = update.newer(release.version)
|
||||
payload = {"ok": True, "current": __version__, "latest": release.version,
|
||||
"update": waiting, "url": release.url}
|
||||
if not waiting:
|
||||
return out(opts, payload,
|
||||
f"Dikte {__version__} is the newest release.")
|
||||
if opts.open:
|
||||
webbrowser.open(release.url)
|
||||
return out(opts, payload,
|
||||
f"Dikte {release.version} is out; this is {__version__}.\n"
|
||||
f"{release.url}")
|
||||
|
||||
|
||||
def cmd_status(opts):
|
||||
reply = ipc.send("status")
|
||||
if reply is None:
|
||||
@@ -1098,6 +1128,11 @@ def build_parser():
|
||||
integrated.set_defaults(func=cmd_integrate)
|
||||
|
||||
# --- the application --------------------------------------------------
|
||||
updates = leaf(subs, "update", "whether a newer Dikte has been released")
|
||||
updates.add_argument("--open", action="store_true",
|
||||
help="open the release page in a browser")
|
||||
updates.set_defaults(func=cmd_update)
|
||||
|
||||
leaf(subs, "status", "what it is doing right now").set_defaults(func=cmd_status)
|
||||
for name, help_text in (("settings", "open the settings window"),
|
||||
("restart", "reload the running instance"),
|
||||
|
||||
@@ -457,6 +457,9 @@ DEFAULTS = {
|
||||
"overlay_corner": "bottom-left",
|
||||
"keep_audio": False,
|
||||
"history_limit": 200,
|
||||
# A look at the releases page once a day, and nothing more than a look:
|
||||
# what is found opens a browser, never an installer.
|
||||
"update_check": True,
|
||||
"file_timestamps": False,
|
||||
"file_cleanup": True,
|
||||
"file_cleanup_prompt": "", # empty -> language-specific default
|
||||
|
||||
+95
-33
@@ -1,15 +1,19 @@
|
||||
"""Transcribe an existing audio/video file with the same models.
|
||||
|
||||
ffmpeg converts whatever comes in to 16 kHz mono WAV, and for a hosted API to
|
||||
mp3 on top of that. The upload limit is the only reason a file is ever cut up,
|
||||
and uncompressed audio reaches it after ten minutes where mp3 takes an hour.
|
||||
mp3 on top of that. Two things decide where a file is cut up: the upload limit,
|
||||
which uncompressed audio reaches after ten minutes where mp3 takes an hour, and
|
||||
the clock. An hour of audio in one request is minutes of work at the other end,
|
||||
and the gateway in front of the model hangs up long before the answer comes
|
||||
back, which arrives here as a 502 with the whole chunk lost. So a chunk is also
|
||||
capped at MAX_CHUNK_SECONDS however small it is on disk.
|
||||
|
||||
That is worth the encoder, because a cut is not free. Whisper hears in thirty
|
||||
second windows and decides for itself where one cue ends and the next begins; a
|
||||
chunk that starts in the middle of a sentence can come back as one cue per
|
||||
window, twenty seconds of text at a time, for the whole rest of the chunk. So
|
||||
the file is cut as rarely as the limit allows, what is cut overlaps, and
|
||||
stitch() drops the half that was heard twice.
|
||||
A cut is not free, which is what the encoder buys and why nothing is cut more
|
||||
finely than that. Whisper hears in thirty second windows and decides for itself
|
||||
where one cue ends and the next begins; a chunk that starts in the middle of a
|
||||
sentence can come back as one cue per window, twenty seconds of text at a time,
|
||||
for the whole rest of the chunk. So what is cut overlaps, and stitch() drops
|
||||
the half that was heard twice.
|
||||
"""
|
||||
|
||||
import contextlib
|
||||
@@ -19,6 +23,7 @@ import shutil
|
||||
import subprocess
|
||||
import tempfile
|
||||
import threading
|
||||
import time
|
||||
import wave
|
||||
|
||||
from PyQt6.QtCore import QObject, pyqtSignal
|
||||
@@ -29,10 +34,14 @@ from . import ggml
|
||||
from .i18n import t
|
||||
|
||||
UPLOAD_LIMIT = 24 * 1024 * 1024 # the APIs take 25 MB; leave the form its room
|
||||
MAX_CHUNK_SECONDS = 900 # as much audio as a hosted request can outlive
|
||||
MP3_BITRATE = "48k" # mono speech at 16 kHz: whisper hears nothing less
|
||||
OVERLAP_SECONDS = 30 # a whisper window: how far back a chunk starts
|
||||
WAV_CHUNK_SECONDS = 600 # 19 MB, for the caller that uploads the WAV itself
|
||||
CLEANUP_CHUNK_CHARS = 12000 # keep each cleanup call comfortably small
|
||||
HOSTED_TIMEOUT = 600 # a quarter hour of audio, with room for the upload
|
||||
RETRIES = 3 # how many times one chunk is asked for in all
|
||||
RETRY_WAIT = 5 # seconds before the second try, doubled after that
|
||||
RATE = 16000
|
||||
MIN_SUBTITLE_SECONDS = 1.5 # how long a cue with no end time of its own stays up
|
||||
|
||||
@@ -86,9 +95,40 @@ class FileTranscriber(QObject):
|
||||
def _check(self):
|
||||
self._abort.check()
|
||||
|
||||
def _wait(self, seconds):
|
||||
"""Sleep on it, with the Stop button still able to get through."""
|
||||
deadline = time.monotonic() + seconds
|
||||
while time.monotonic() < deadline:
|
||||
self._check()
|
||||
time.sleep(0.25)
|
||||
self._check()
|
||||
|
||||
def _attempt(self, call, stage):
|
||||
"""`call`, asked again when what failed was the network rather than us.
|
||||
|
||||
One chunk is a quarter hour of audio that took a minute to encode and a
|
||||
minute to upload, so a gateway having a bad moment is worth waiting out
|
||||
rather than throwing the run away over. `stage` is what the status line
|
||||
said before the failure, put back once the wait is over.
|
||||
"""
|
||||
for attempt in range(1, RETRIES + 1):
|
||||
self._check()
|
||||
try:
|
||||
return call()
|
||||
except api.ApiError as exc:
|
||||
if attempt == RETRIES or not exc.retryable:
|
||||
raise
|
||||
self.progress.emit(t(
|
||||
"{error} Trying again ({attempt}/{total})…",
|
||||
error=exc, attempt=attempt + 1, total=RETRIES))
|
||||
self._wait(RETRY_WAIT * 2 ** (attempt - 1))
|
||||
self.progress.emit(stage)
|
||||
|
||||
def _work(self, path, timestamps, do_cleanup):
|
||||
conf = self.conf
|
||||
workdir = None
|
||||
pieces = []
|
||||
segments = []
|
||||
try:
|
||||
if not shutil.which("ffmpeg"):
|
||||
raise api.ApiError(t("ffmpeg not found. Install it to transcribe files."))
|
||||
@@ -104,39 +144,36 @@ class FileTranscriber(QObject):
|
||||
if len(chunks) > 1:
|
||||
self.progress.emit(t("Splitting into {count} chunks…", count=len(chunks)))
|
||||
|
||||
pieces = []
|
||||
segments = []
|
||||
for index, (chunk_path, offset) in enumerate(chunks, start=1):
|
||||
self._check()
|
||||
self.progress.emit(
|
||||
t("Transcribing chunk {index}/{count}…",
|
||||
stage = (t("Transcribing chunk {index}/{count}…",
|
||||
index=index, count=len(chunks))
|
||||
if len(chunks) > 1 else t("Transcribing…")
|
||||
)
|
||||
if len(chunks) > 1 else t("Transcribing…"))
|
||||
self.progress.emit(stage)
|
||||
if timestamps:
|
||||
heard = self._attempt(lambda: api.transcribe_segments(
|
||||
target,
|
||||
chunk_path,
|
||||
language=conf["language"],
|
||||
prompt=conf["transcribe_prompt"],
|
||||
timeout=HOSTED_TIMEOUT,
|
||||
aborter=self._abort,
|
||||
), stage)
|
||||
segments = stitch(segments, [
|
||||
(start + offset, end + offset, line)
|
||||
for start, end, line in api.transcribe_segments(
|
||||
target,
|
||||
chunk_path,
|
||||
language=conf["language"],
|
||||
prompt=conf["transcribe_prompt"],
|
||||
aborter=self._abort,
|
||||
)
|
||||
for start, end, line in heard
|
||||
])
|
||||
else:
|
||||
pieces.append(api.transcribe(
|
||||
pieces.append(self._attempt(lambda: api.transcribe(
|
||||
target,
|
||||
chunk_path,
|
||||
language=conf["language"],
|
||||
prompt=conf["transcribe_prompt"],
|
||||
timeout=HOSTED_TIMEOUT,
|
||||
aborter=self._abort,
|
||||
))
|
||||
), stage))
|
||||
|
||||
if timestamps:
|
||||
pieces = [f"[{format_timestamp(start)}] {line}"
|
||||
for start, _, line in segments]
|
||||
text = "\n".join(pieces) if timestamps else " ".join(pieces)
|
||||
text = _joined(pieces, segments, timestamps)
|
||||
|
||||
if do_cleanup and text:
|
||||
self._check()
|
||||
@@ -148,6 +185,15 @@ class FileTranscriber(QObject):
|
||||
except Cancelled:
|
||||
self.progress.emit(t("Stopped."))
|
||||
except (api.ApiError, OSError, subprocess.SubprocessError, wave.Error) as exc:
|
||||
# An hour of a long file already heard is not worth throwing away
|
||||
# because the chunk after it failed, or because cleanup did. Hand
|
||||
# over what there is, and say in the same breath where it stops.
|
||||
partial = _joined(pieces, segments, timestamps)
|
||||
if partial:
|
||||
self.finished.emit(partial, segments)
|
||||
self.failed.emit(t("{error} The transcript up to there is below.",
|
||||
error=exc))
|
||||
else:
|
||||
self.failed.emit(str(exc))
|
||||
finally:
|
||||
self._local = None
|
||||
@@ -181,12 +227,21 @@ class FileTranscriber(QObject):
|
||||
self._local = ggml.llm if cleanup.provider(conf) == "local" else None
|
||||
prompt = conf.cleanup_prompt(with_timestamps=timestamps, subtitles=True)
|
||||
out = []
|
||||
stage = t("Cleaning up…")
|
||||
for block in split_text(text, timestamps):
|
||||
self._check()
|
||||
out.append(cleanup.run(block, conf, prompt, aborter=self._abort))
|
||||
out.append(self._attempt(
|
||||
lambda: cleanup.run(block, conf, prompt, aborter=self._abort), stage))
|
||||
return ("\n" if timestamps else "\n\n").join(out)
|
||||
|
||||
|
||||
def _joined(pieces, segments, timestamps):
|
||||
"""The transcript as one string, out of whichever of the two is holding it."""
|
||||
if timestamps:
|
||||
pieces = [f"[{format_timestamp(start)}] {line}" for start, _, line in segments]
|
||||
return "\n".join(pieces) if timestamps else " ".join(pieces)
|
||||
|
||||
|
||||
def format_timestamp(seconds):
|
||||
seconds = int(seconds)
|
||||
hours, rest = divmod(seconds, 3600)
|
||||
@@ -309,13 +364,20 @@ def wav_seconds(wav_path):
|
||||
def chunk_seconds(path, duration):
|
||||
"""How many seconds of this audio fit in one request, or 0 when all of it does.
|
||||
|
||||
Measured rather than worked out: what an encoder makes of an hour of speech
|
||||
depends on the speech, and the file on disk is the only honest answer.
|
||||
Whichever of the two limits bites first. How much fits under the upload
|
||||
limit is measured rather than worked out: what an encoder makes of an hour
|
||||
of speech depends on the speech, and the file on disk is the only honest
|
||||
answer. The other limit is MAX_CHUNK_SECONDS, and it is the one that catches
|
||||
a long file at this bitrate: an hour and a half of mp3 is two chunks by size
|
||||
and one of them is an hour of audio in a single request, which no hosted
|
||||
gateway stays on the line for.
|
||||
"""
|
||||
size = os.path.getsize(path)
|
||||
if size <= UPLOAD_LIMIT or duration <= 0:
|
||||
if duration <= 0:
|
||||
return 0.0
|
||||
return max(60.0, duration * UPLOAD_LIMIT / size * 0.95)
|
||||
size = os.path.getsize(path)
|
||||
fits = duration * UPLOAD_LIMIT / size * 0.95 if size > UPLOAD_LIMIT else duration
|
||||
seconds = max(60.0, min(fits, MAX_CHUNK_SECONDS))
|
||||
return 0.0 if seconds >= duration else seconds
|
||||
|
||||
|
||||
def split_wav(wav_path, workdir, seconds=WAV_CHUNK_SECONDS, overlap=OVERLAP_SECONDS):
|
||||
|
||||
@@ -136,6 +136,22 @@ def release(repo, tag="latest", refresh=False):
|
||||
return data.get("tag_name") or tag, assets
|
||||
|
||||
|
||||
def newest_release(repo, refresh=False):
|
||||
"""(tag, page, published) for the newest release of a repository.
|
||||
|
||||
release() above is for taking a file out of one and insists on there being
|
||||
files to take; this is for the number, which a release with nothing
|
||||
attached answers just as well. GitHub keeps prereleases out of "latest" on
|
||||
its own, which is what leaves the nightly build off this answer.
|
||||
"""
|
||||
data = _fetch(f"gh-newest-{repo}",
|
||||
f"{GITHUB_API}/repos/{repo}/releases/latest", refresh=refresh)
|
||||
if not isinstance(data, dict) or not data.get("tag_name"):
|
||||
raise HubError(t("{repo} has published no release.", repo=repo))
|
||||
return (data["tag_name"], data.get("html_url") or "",
|
||||
data.get("published_at") or "")
|
||||
|
||||
|
||||
def files(repo, revision="main", refresh=False):
|
||||
"""[Item] for every file in a Hugging Face repository.
|
||||
|
||||
|
||||
@@ -192,6 +192,25 @@ TR = {
|
||||
"Silence threshold": "Sessizlik eşiği",
|
||||
"Keep audio files ({path})": "Ses kayıtlarını sakla ({path})",
|
||||
|
||||
# --- updates --------------------------------------------------------
|
||||
"Updates": "Güncelleme",
|
||||
"Look for a newer version once a day": "Günde bir kez yeni sürüm var mı diye bak",
|
||||
"Dikte only looks. What it finds opens the release page in your browser; "
|
||||
"it downloads and installs nothing by itself.":
|
||||
"Dikte yalnızca bakar. Bulduğu şey tarayıcında sürüm sayfasını açar; "
|
||||
"kendi başına hiçbir şey indirmez ve kurmaz.",
|
||||
"Check now": "Şimdi bak",
|
||||
"Looking…": "Bakılıyor…",
|
||||
"Open the release page": "Sürüm sayfasını aç",
|
||||
"This is Dikte {version}.": "Buradaki sürüm Dikte {version}.",
|
||||
"Dikte {version} is the newest release.": "En yeni sürüm zaten bu: Dikte {version}.",
|
||||
"Dikte {version} is out; this is {current}.":
|
||||
"Dikte {version} çıkmış; buradaki sürüm {current}.",
|
||||
"Dikte {version} is out…": "Dikte {version} çıkmış…",
|
||||
"Dikte {version} is out. The tray menu has the release page.":
|
||||
"Dikte {version} çıkmış. Sürüm sayfası tepsi menüsünde.",
|
||||
"{repo} has published no release.": "{repo} için yayımlanmış sürüm yok.",
|
||||
|
||||
# --- settings: api --------------------------------------------------
|
||||
"Keys": "Anahtarlar",
|
||||
"Speech to text": "Sesi yazıya çevirme",
|
||||
@@ -297,6 +316,10 @@ TR = {
|
||||
"Converting audio…": "Ses dönüştürülüyor…",
|
||||
"Splitting into {count} chunks…": "{count} parçaya bölünüyor…",
|
||||
"Transcribing chunk {index}/{count}…": "{index}/{count} parça yazıya çevriliyor…",
|
||||
"{error} Trying again ({attempt}/{total})…":
|
||||
"{error} Yeniden deneniyor ({attempt}/{total})…",
|
||||
"{error} The transcript up to there is below.":
|
||||
"{error} Oraya kadar çevrilen metin aşağıda.",
|
||||
"Done: {chars} characters.": "Bitti: {chars} karakter.",
|
||||
"Stopped.": "Durduruldu.",
|
||||
"Failed: {error}": "Başarısız: {error}",
|
||||
|
||||
@@ -13,6 +13,7 @@ from PyQt6.QtWidgets import (
|
||||
QPushButton, QScrollArea, QSpinBox, QTabWidget, QVBoxLayout, QWidget,
|
||||
)
|
||||
|
||||
from . import __version__
|
||||
from . import api
|
||||
from . import assistant
|
||||
from . import audio
|
||||
@@ -21,9 +22,11 @@ from . import config as cfg
|
||||
from . import filetranscribe
|
||||
from . import ggml
|
||||
from . import hotkey
|
||||
from . import hub
|
||||
from . import ipc
|
||||
from . import meeting
|
||||
from . import paste
|
||||
from . import update
|
||||
from .filetranscribe import FileTranscriber
|
||||
from .i18n import t
|
||||
|
||||
@@ -512,11 +515,16 @@ class LocalModelBox(QGroupBox):
|
||||
|
||||
class SettingsWindow(QDialog):
|
||||
applied = pyqtSignal()
|
||||
# A newer release this window's own check found, so that the tray icon
|
||||
# hears about it from here rather than waiting for its own next check.
|
||||
update_found = pyqtSignal(object)
|
||||
|
||||
_models_loaded = pyqtSignal(list, str)
|
||||
_transcribe_models_loaded = pyqtSignal(list, str)
|
||||
# Which key was tested, whether it worked, and what to write under it.
|
||||
_test_done = pyqtSignal(str, bool, str)
|
||||
# The release that was found, or None, and what went wrong instead.
|
||||
_update_checked = pyqtSignal(object, str)
|
||||
|
||||
def __init__(self, conf, meetings=None, parent=None):
|
||||
super().__init__(parent)
|
||||
@@ -533,6 +541,9 @@ class SettingsWindow(QDialog):
|
||||
self._key_fields = {}
|
||||
self._testers = {}
|
||||
self._shown_provider = ""
|
||||
# Where "Open the release page" goes: the release itself once a check
|
||||
# has named one, and the page that redirects to the newest until then.
|
||||
self._release_url = update.RELEASES_PAGE
|
||||
self.transcriber = FileTranscriber(conf, self)
|
||||
self.setWindowTitle(t("Dikte Settings"))
|
||||
|
||||
@@ -567,6 +578,7 @@ class SettingsWindow(QDialog):
|
||||
self._models_loaded.connect(self._on_models_loaded)
|
||||
self._transcribe_models_loaded.connect(self._on_transcribe_models_loaded)
|
||||
self._test_done.connect(self._on_test_done)
|
||||
self._update_checked.connect(self._on_update_checked)
|
||||
self.transcriber.progress.connect(self._on_file_progress)
|
||||
self.transcriber.finished.connect(self._on_file_finished)
|
||||
self.transcriber.failed.connect(self._on_file_failed)
|
||||
@@ -698,6 +710,23 @@ class SettingsWindow(QDialog):
|
||||
t("Keep audio files ({path})", path=str(cfg.RECORDINGS_DIR))
|
||||
)
|
||||
form.addRow("", self.keep_audio)
|
||||
|
||||
self.update_check = QCheckBox(t("Look for a newer version once a day"))
|
||||
self.update_check.setToolTip(
|
||||
t("Dikte only looks. What it finds opens the release page in your "
|
||||
"browser; it downloads and installs nothing by itself.")
|
||||
)
|
||||
form.addRow(t("Updates"), self.update_check)
|
||||
|
||||
self.update_status = WrappedLabel("")
|
||||
self.update_page = QPushButton(t("Open the release page"))
|
||||
self.update_page.clicked.connect(
|
||||
lambda: QDesktopServices.openUrl(QUrl(self._release_url))
|
||||
)
|
||||
self.update_now = QPushButton(t("Check now"))
|
||||
self.update_now.clicked.connect(self._check_for_update)
|
||||
form.addRow("", self._row(self.update_status, self.update_page,
|
||||
self.update_now))
|
||||
return page
|
||||
|
||||
def _api_tab(self):
|
||||
@@ -1568,6 +1597,8 @@ class SettingsWindow(QDialog):
|
||||
self.silence_db.setValue(int(conf["silence_db"]))
|
||||
self.filter_hallucinations.setChecked(conf["filter_hallucinations"])
|
||||
self.keep_audio.setChecked(conf["keep_audio"])
|
||||
self.update_check.setChecked(conf["update_check"])
|
||||
self._show_update(update.pending())
|
||||
|
||||
for name, who in cfg.TRANSCRIBERS.items():
|
||||
self._key_fields[name].setText(conf[who.key])
|
||||
@@ -1661,6 +1692,7 @@ class SettingsWindow(QDialog):
|
||||
conf["silence_db"] = float(self.silence_db.value())
|
||||
conf["filter_hallucinations"] = self.filter_hallucinations.isChecked()
|
||||
conf["keep_audio"] = self.keep_audio.isChecked()
|
||||
conf["update_check"] = self.update_check.isChecked()
|
||||
|
||||
provider = self.transcribe_provider.currentData() or "local"
|
||||
if provider in self._models:
|
||||
@@ -1891,6 +1923,45 @@ class SettingsWindow(QDialog):
|
||||
button.setEnabled(True)
|
||||
answer.setText(("✓ " if ok else "✗ ") + message)
|
||||
|
||||
# ---- updates ---------------------------------------------------------
|
||||
|
||||
def _check_for_update(self):
|
||||
"""The button, which asks GitHub whatever the daily clock says."""
|
||||
self.update_now.setEnabled(False)
|
||||
self.update_status.setText(t("Looking…"))
|
||||
|
||||
def work():
|
||||
try:
|
||||
self._update_checked.emit(update.check(force=True), "")
|
||||
except hub.HubError as exc:
|
||||
self._update_checked.emit(None, str(exc))
|
||||
|
||||
threading.Thread(target=work, daemon=True).start()
|
||||
|
||||
def _on_update_checked(self, release, error):
|
||||
self.update_now.setEnabled(True)
|
||||
if error:
|
||||
self.update_status.setText(error)
|
||||
return
|
||||
self._show_update(release, asked=True)
|
||||
if release is not None:
|
||||
self.update_found.emit(release)
|
||||
|
||||
def _show_update(self, release, asked=False):
|
||||
"""What the line under the checkbox says, and whether the page button
|
||||
is on it. `release` is None when this build is the newest one, and
|
||||
`asked` is what tells "nothing new" from "nobody has looked yet"."""
|
||||
self.update_page.setVisible(release is not None)
|
||||
if release is None:
|
||||
self.update_status.setText(
|
||||
t("Dikte {version} is the newest release.", version=__version__)
|
||||
if asked else t("This is Dikte {version}.", version=__version__))
|
||||
return
|
||||
self._release_url = release.url
|
||||
self.update_status.setText(
|
||||
t("Dikte {version} is out; this is {current}.",
|
||||
version=release.version, current=__version__))
|
||||
|
||||
# ---- audio file ------------------------------------------------------
|
||||
|
||||
def _choose_file(self):
|
||||
|
||||
+158
@@ -0,0 +1,158 @@
|
||||
"""Whether a newer Dikte has been published, and where to get it.
|
||||
|
||||
GitHub is asked for the newest release, its number is held against the one this
|
||||
build carries, and that is where it stops. Nothing is downloaded and nothing is
|
||||
replaced. The four downloads are installed in four different ways, and three of
|
||||
those belong to the platform rather than to Dikte: a Mac bundle is dragged into
|
||||
Applications and cannot rewrite itself while it is running, the Windows setup
|
||||
is an installer with an uninstall entry of its own, an AppImage is a single
|
||||
file kept wherever its owner keeps it, and a checkout is updated with git. A
|
||||
program that guessed at all four would be wrong on at least one of them, and
|
||||
being wrong there means an installation somebody has to repair by hand. So the
|
||||
answer ends in a browser, on the release page, where the same download that was
|
||||
installed the first time is waiting.
|
||||
|
||||
The clock is kept in a file of its own rather than in the settings. A check
|
||||
runs while the settings window may be open, and a background write into
|
||||
config.json is exactly what would overwrite a setting somebody is in the middle
|
||||
of changing.
|
||||
|
||||
Nothing here imports Qt or the rest of the application: `dikte update` at a
|
||||
terminal and the timer behind the tray icon ask the same three questions of the
|
||||
same module.
|
||||
"""
|
||||
|
||||
import collections
|
||||
import itertools
|
||||
import json
|
||||
import time
|
||||
|
||||
from . import __version__
|
||||
from . import hub
|
||||
from . import paths
|
||||
|
||||
REPO = "yusufipk/dikte"
|
||||
# Where somebody is sent. GitHub redirects this to whatever the newest release
|
||||
# is, so it stays right without anybody writing a number into it.
|
||||
RELEASES_PAGE = f"https://github.com/{REPO}/releases/latest"
|
||||
|
||||
# Once a day. A release happens every few weeks at best, and a question nobody
|
||||
# is waiting on is not one to ask GitHub on every start.
|
||||
INTERVAL = 24 * 3600
|
||||
|
||||
# When the last check was, what it found, and which version has already been
|
||||
# announced. In the data directory rather than the config one: it is not a
|
||||
# setting, nobody edits it, and losing it costs one extra request.
|
||||
STATE_FILE = paths.DATA_DIR / "update.json"
|
||||
|
||||
Release = collections.namedtuple("Release", "version url published")
|
||||
|
||||
|
||||
def _numbers(version):
|
||||
"""(1, 0, 2) for "v1.0.2", "1.0.2" and "1.0.2-dev.abc1234" alike.
|
||||
|
||||
Empty for anything that does not start with a number, which is what a tag
|
||||
naming something other than a version comes back as.
|
||||
"""
|
||||
number = str(version or "").strip().lstrip("vV").split("-")[0].split("+")[0]
|
||||
parts = []
|
||||
for piece in number.split("."):
|
||||
digits = "".join(itertools.takewhile(str.isdigit, piece))
|
||||
if not digits:
|
||||
break
|
||||
parts.append(int(digits))
|
||||
return tuple((parts + [0, 0, 0])[:3]) if parts else ()
|
||||
|
||||
|
||||
def newer(there, here=""):
|
||||
"""Whether the release numbered `there` is one this build has not got.
|
||||
|
||||
Only the numbers are compared, and what follows them is dropped. A build
|
||||
off master carries the released number with its commit after it
|
||||
(1.0.1-dev.abc1234), and that build is ahead of 1.0.1 rather than behind
|
||||
it; read as a version suffix it would be behind, and every nightly would be
|
||||
told to update to the release it was already past.
|
||||
"""
|
||||
theirs = _numbers(there)
|
||||
return bool(theirs) and theirs > _numbers(here or __version__)
|
||||
|
||||
|
||||
def state():
|
||||
"""What the last check wrote down; empty when there has never been one."""
|
||||
try:
|
||||
stored = json.loads(STATE_FILE.read_text(encoding="utf-8"))
|
||||
except (OSError, ValueError):
|
||||
return {}
|
||||
return stored if isinstance(stored, dict) else {}
|
||||
|
||||
|
||||
def _store(**changes):
|
||||
stored = state()
|
||||
stored.update(changes)
|
||||
try:
|
||||
STATE_FILE.parent.mkdir(parents=True, exist_ok=True)
|
||||
STATE_FILE.write_text(json.dumps(stored), encoding="utf-8")
|
||||
except OSError:
|
||||
pass # a check that cannot be written down still happened
|
||||
return stored
|
||||
|
||||
|
||||
def due(now=0):
|
||||
"""Whether a day has gone by since the last time anybody asked."""
|
||||
return (now or time.time()) - float(state().get("checked") or 0) >= INTERVAL
|
||||
|
||||
|
||||
def latest(refresh=False):
|
||||
"""The newest published release, asked for outright. Raises HubError."""
|
||||
tag, url, published = hub.newest_release(REPO, refresh=refresh)
|
||||
return Release(tag.lstrip("vV"), url or RELEASES_PAGE, published)
|
||||
|
||||
|
||||
def remember(release):
|
||||
"""Write down that a check has just happened, and what it found."""
|
||||
_store(checked=time.time(), version=release.version, url=release.url,
|
||||
published=release.published)
|
||||
|
||||
|
||||
def pending():
|
||||
"""The newer release the last check found, without asking anybody.
|
||||
|
||||
What the tray icon is built from: the answer has to be there the moment it
|
||||
appears, and a request on the way to the screen is a request nobody has
|
||||
time for.
|
||||
"""
|
||||
stored = state()
|
||||
version = stored.get("version") or ""
|
||||
if not newer(version):
|
||||
return None
|
||||
return Release(version, stored.get("url") or RELEASES_PAGE,
|
||||
stored.get("published") or "")
|
||||
|
||||
|
||||
def check(force=False):
|
||||
"""A newer release, or None when there is nothing to say.
|
||||
|
||||
The scheduled half: it asks only when a day has gone by, and answers from
|
||||
what the last check found in between. `force` is the button in Settings and
|
||||
the command line, which ask whatever the clock says.
|
||||
|
||||
The clock here is the only throttle. Once it has decided to ask, it asks
|
||||
for real rather than reading hub.py's few hours of cache, which is there to
|
||||
keep a settings window from fetching the same model list twice in an
|
||||
evening and would only ever answer this with something it already knew.
|
||||
"""
|
||||
if not force and not due():
|
||||
return pending()
|
||||
release = latest(refresh=True)
|
||||
remember(release)
|
||||
return release if newer(release.version) else None
|
||||
|
||||
|
||||
def announced():
|
||||
"""The version somebody has already been shown a notification about."""
|
||||
return state().get("announced") or ""
|
||||
|
||||
|
||||
def mark_announced(version):
|
||||
"""Said once. A daily check must not be a daily interruption."""
|
||||
_store(announced=version)
|
||||
+4
-1
@@ -25,6 +25,7 @@ from unittest import mock
|
||||
from dikte import assistant
|
||||
from dikte import config as cfg
|
||||
from dikte import i18n
|
||||
from dikte import update
|
||||
|
||||
# What the application is, rather than what it does: PipeWire, wl-clipboard,
|
||||
# ydotool, KDE's shortcut file, /dev/input. A port to another desktop replaces
|
||||
@@ -86,8 +87,10 @@ class DikteTest(unittest.TestCase):
|
||||
MEETINGS_FILE=data_dir / "meetings.jsonl",
|
||||
)
|
||||
# Resolved from cfg.DATA_DIR when assistant was imported, so it needs
|
||||
# moving on its own.
|
||||
# moving on its own. The same goes for where the update check writes
|
||||
# down when it last ran.
|
||||
self.patch_attr(assistant, "SESSION_FILE", data_dir / "assistant.json")
|
||||
self.patch_attr(update, "STATE_FILE", data_dir / "update.json")
|
||||
|
||||
i18n.set_language("en")
|
||||
self.addCleanup(i18n.set_language, "en")
|
||||
|
||||
@@ -79,6 +79,33 @@ class Explain(DikteTest):
|
||||
def test_the_status_is_carried_through(self):
|
||||
self.assertEqual(self.error(429).status, 429)
|
||||
|
||||
def test_so_is_whether_it_is_worth_asking_again(self):
|
||||
self.assertTrue(self.error(502).retryable)
|
||||
self.assertFalse(self.error(401).retryable)
|
||||
|
||||
|
||||
class Retryable(unittest.TestCase):
|
||||
"""Which failures a second try can fix, and which will fail the same way."""
|
||||
|
||||
def test_a_gateway_that_gave_up_waiting(self):
|
||||
for status in (408, 429, 500, 502, 503, 504):
|
||||
with self.subTest(status=status):
|
||||
self.assertTrue(api.ApiError("x", status).retryable)
|
||||
|
||||
def test_a_request_that_was_wrong(self):
|
||||
for status in (400, 401, 402, 403, 404, 413, 422):
|
||||
with self.subTest(status=status):
|
||||
self.assertFalse(api.ApiError("x", status).retryable)
|
||||
|
||||
def test_an_error_of_our_own_is_not_the_network(self):
|
||||
self.assertFalse(api.ApiError("Transcript came back empty.").retryable)
|
||||
|
||||
def test_a_connection_that_dropped_is_worth_a_second_try(self):
|
||||
with fake_urlopen(url_error("connection reset")):
|
||||
with self.assertRaises(api.ApiError) as caught:
|
||||
api._request("https://example.test", b"{}", {})
|
||||
self.assertTrue(caught.exception.retryable)
|
||||
|
||||
|
||||
class ExtractError(unittest.TestCase):
|
||||
def test_the_usual_shape(self):
|
||||
|
||||
+67
-1
@@ -10,6 +10,8 @@ import contextlib
|
||||
import io
|
||||
import json
|
||||
import unittest
|
||||
import webbrowser
|
||||
from typing import ClassVar
|
||||
from unittest import mock
|
||||
|
||||
from dikte import audio
|
||||
@@ -17,9 +19,11 @@ from dikte import cli
|
||||
from dikte import config as cfg
|
||||
from dikte import ggml
|
||||
from dikte import hotkey
|
||||
from dikte import hub
|
||||
from dikte import ipc
|
||||
from dikte import paste
|
||||
from tests.support import DikteTest, fake_urlopen, only_these_tools
|
||||
from dikte import update
|
||||
from tests.support import DikteTest, fake_urlopen, only_these_tools, url_error
|
||||
|
||||
|
||||
class Options:
|
||||
@@ -420,6 +424,68 @@ class Providers(DikteTest):
|
||||
self.assertIn("Groq", out)
|
||||
|
||||
|
||||
class Updates(DikteTest):
|
||||
"""`dikte update` looks, says what it found, and installs nothing."""
|
||||
|
||||
RELEASE: ClassVar[dict] = {
|
||||
"tag_name": "v9.9.9",
|
||||
"html_url": "https://github.com/yusufipk/dikte/releases/tag/v9.9.9",
|
||||
}
|
||||
|
||||
def setUp(self):
|
||||
super().setUp()
|
||||
self.patch_attr(hub, "CACHE_DIR", self.path("cache"))
|
||||
# Nothing here may reach a browser, whatever the answer turns out to be.
|
||||
self.opened = []
|
||||
self.patch_attr(webbrowser, "open", self.opened.append)
|
||||
|
||||
def run_update(self, reply, **values):
|
||||
with fake_urlopen(reply), captured() as (out, err):
|
||||
code = cli.cmd_update(Options(open=False, **values))
|
||||
return code, out.getvalue(), err.getvalue()
|
||||
|
||||
def test_a_newer_release_is_named_with_its_page(self):
|
||||
code, out, _ = self.run_update(self.RELEASE)
|
||||
self.assertEqual(code, 0)
|
||||
self.assertIn("9.9.9", out)
|
||||
self.assertIn(self.RELEASE["html_url"], out)
|
||||
|
||||
def test_this_build_being_the_newest_is_not_a_failure(self):
|
||||
code, out, _ = self.run_update({"tag_name": f"v{cli.__version__}"})
|
||||
self.assertEqual(code, 0)
|
||||
self.assertIn("newest", out)
|
||||
|
||||
def test_the_json_answer_says_both_numbers(self):
|
||||
code, out, _ = self.run_update(self.RELEASE, json=True)
|
||||
answer = json.loads(out)
|
||||
self.assertTrue(answer["update"])
|
||||
self.assertEqual(answer["latest"], "9.9.9")
|
||||
self.assertEqual(answer["current"], cli.__version__)
|
||||
|
||||
def test_github_being_unreachable_is_a_failure_with_a_reason(self):
|
||||
with fake_urlopen(url_error("no route to host")), captured() as (_, err):
|
||||
code = cli.cmd_update(Options(open=False))
|
||||
self.assertEqual(code, 1)
|
||||
self.assertIn("api.github.com", err.getvalue())
|
||||
|
||||
def test_the_browser_is_opened_only_when_asked_and_only_when_there_is_one(self):
|
||||
self.run_update(self.RELEASE)
|
||||
self.assertEqual(self.opened, [])
|
||||
with fake_urlopen({"tag_name": f"v{cli.__version__}"}), captured():
|
||||
cli.cmd_update(Options(open=True))
|
||||
self.assertEqual(self.opened, [])
|
||||
with fake_urlopen(self.RELEASE), captured():
|
||||
cli.cmd_update(Options(open=True))
|
||||
self.assertEqual(self.opened, [self.RELEASE["html_url"]])
|
||||
|
||||
def test_the_answer_is_written_down_for_the_application(self):
|
||||
"""A check at a terminal is a check; the tray must not go and ask the
|
||||
same question an hour later."""
|
||||
self.run_update(self.RELEASE)
|
||||
self.assertEqual(update.state()["version"], "9.9.9")
|
||||
self.assertFalse(update.due())
|
||||
|
||||
|
||||
class Doctor(DikteTest):
|
||||
"""One pass over everything the settings window checks behind its buttons."""
|
||||
|
||||
|
||||
@@ -214,10 +214,22 @@ class ChunkSeconds(DikteTest):
|
||||
self.assertEqual(ft.chunk_seconds(self.file(1024), 600), 0.0)
|
||||
|
||||
def test_a_file_over_the_limit_is_cut_by_what_it_measured(self):
|
||||
# Twice the limit over an hour, so a little under half an hour fits.
|
||||
seconds = ft.chunk_seconds(self.file(ft.UPLOAD_LIMIT * 2), 3600)
|
||||
self.assertGreater(seconds, 1500)
|
||||
self.assertLess(seconds, 1800)
|
||||
# Twice the limit over twenty minutes, so a little under ten fits.
|
||||
seconds = ft.chunk_seconds(self.file(ft.UPLOAD_LIMIT * 2), 1200)
|
||||
self.assertGreater(seconds, 500)
|
||||
self.assertLess(seconds, 600)
|
||||
|
||||
def test_a_chunk_is_never_more_audio_than_a_request_can_outlive(self):
|
||||
"""An hour in one request is a 502 from the gateway, whatever it weighs."""
|
||||
self.assertEqual(ft.chunk_seconds(self.file(ft.UPLOAD_LIMIT * 2), 3600),
|
||||
ft.MAX_CHUNK_SECONDS)
|
||||
|
||||
def test_a_small_file_that_is_still_hours_long_is_cut_on_the_clock(self):
|
||||
self.assertEqual(ft.chunk_seconds(self.file(1024), 7200),
|
||||
ft.MAX_CHUNK_SECONDS)
|
||||
|
||||
def test_a_file_short_enough_on_both_counts_is_not_cut(self):
|
||||
self.assertEqual(ft.chunk_seconds(self.file(1024), ft.MAX_CHUNK_SECONDS), 0.0)
|
||||
|
||||
def test_a_file_with_no_length_is_left_whole(self):
|
||||
self.assertEqual(ft.chunk_seconds(self.file(ft.UPLOAD_LIMIT * 2), 0), 0.0)
|
||||
@@ -378,6 +390,58 @@ class Transcriber(DikteTest):
|
||||
worker.stop()
|
||||
self.assertTrue(worker._abort.aborted)
|
||||
|
||||
def test_a_chunk_is_given_longer_to_answer_than_a_dictation(self):
|
||||
"""A quarter hour of audio is not a sentence: the default would cut it off."""
|
||||
worker = ft.FileTranscriber(self.conf)
|
||||
with mock.patch.object(ft, "_to_wav", side_effect=lambda *a: self.source), \
|
||||
mock.patch.object(ft, "_to_mp3",
|
||||
side_effect=lambda path, *a, **k: path), \
|
||||
mock.patch.object(ft.shutil, "which", return_value="/usr/bin/ffmpeg"), \
|
||||
mock.patch.object(api, "transcribe", return_value="text") as call:
|
||||
worker._work(self.source, False, False)
|
||||
self.assertEqual(call.call_args.kwargs["timeout"], ft.HOSTED_TIMEOUT)
|
||||
|
||||
def test_a_gateway_having_a_bad_moment_is_asked_again(self):
|
||||
with mock.patch.object(ft.FileTranscriber, "_wait"):
|
||||
done, failures, progress, _ = self.run_chain(
|
||||
fail=[api.ApiError("HTTP 502: timeout", 502), "raw text"])
|
||||
self.assertEqual(failures, [])
|
||||
self.assertEqual(done[0][0], "raw text")
|
||||
self.assertTrue(any("Trying again" in message for message in progress))
|
||||
|
||||
def test_a_rejected_key_is_not_asked_again(self):
|
||||
"""Trying again with the same key is only a slower way to fail."""
|
||||
call = mock.Mock(side_effect=api.ApiError("rejected the API key", 401))
|
||||
with mock.patch.object(ft.FileTranscriber, "_wait"):
|
||||
_, failures, _, _ = self.run_chain(fail=call)
|
||||
self.assertEqual(call.call_count, 1)
|
||||
self.assertIn("rejected", failures[0])
|
||||
|
||||
def test_a_chunk_is_given_up_on_after_the_last_try(self):
|
||||
call = mock.Mock(side_effect=api.ApiError("HTTP 502: timeout", 502))
|
||||
with mock.patch.object(ft.FileTranscriber, "_wait"):
|
||||
_, failures, _, _ = self.run_chain(fail=call)
|
||||
self.assertEqual(call.call_count, ft.RETRIES)
|
||||
self.assertIn("502", failures[0])
|
||||
|
||||
def test_what_was_heard_before_the_failure_is_still_handed_over(self):
|
||||
"""An hour already transcribed is not thrown away over the chunk after it."""
|
||||
boom = api.ApiError("HTTP 502: timeout", 502)
|
||||
with mock.patch.object(ft.FileTranscriber, "_wait"), \
|
||||
mock.patch.object(ft.FileTranscriber, "_chunks",
|
||||
side_effect=lambda wav, *a: [(wav, 0.0), (wav, 10.0)]):
|
||||
done, failures, _, _ = self.run_chain(
|
||||
fail=["first half"] + [boom] * ft.RETRIES)
|
||||
self.assertEqual(done[0][0], "first half")
|
||||
self.assertIn("502", failures[0])
|
||||
|
||||
def test_nothing_heard_at_all_is_a_plain_failure(self):
|
||||
call = mock.Mock(side_effect=api.ApiError("rejected the API key", 401))
|
||||
with mock.patch.object(ft.FileTranscriber, "_wait"):
|
||||
done, failures, _, _ = self.run_chain(fail=call)
|
||||
self.assertEqual(done, [])
|
||||
self.assertEqual(failures[0], "rejected the API key")
|
||||
|
||||
def test_a_second_start_while_one_is_running_is_ignored(self):
|
||||
worker = ft.FileTranscriber(self.conf)
|
||||
worker._thread = mock.Mock(is_alive=lambda: True)
|
||||
|
||||
@@ -25,6 +25,7 @@ from dikte import ipc
|
||||
from dikte import overlay as overlay_module
|
||||
from dikte import paste
|
||||
from dikte import settings_ui
|
||||
from dikte import update
|
||||
from tests.support import DikteTest, only_these_tools
|
||||
|
||||
# One application for the whole run; Qt allows no second one.
|
||||
@@ -103,6 +104,7 @@ CHANGED = {
|
||||
"pause_shortcut": "Meta+P",
|
||||
"evdev_hotkey": True,
|
||||
"history_limit": 50,
|
||||
"update_check": False,
|
||||
}
|
||||
|
||||
|
||||
@@ -252,6 +254,31 @@ class Settings(DikteTest):
|
||||
self.assertEqual(shown, [provider])
|
||||
self.assertFalse(box.isHidden())
|
||||
|
||||
def test_the_update_line_names_the_version_that_is_running(self):
|
||||
window = self.window(cfg.Config())
|
||||
self.assertIn(settings_ui.__version__, window.update_status.text())
|
||||
# Nothing to open until a check has found something to open.
|
||||
self.assertTrue(window.update_page.isHidden())
|
||||
|
||||
def test_a_newer_release_puts_the_page_button_on_screen(self):
|
||||
window = self.window(cfg.Config())
|
||||
told = []
|
||||
window.update_found.connect(told.append)
|
||||
release = update.Release("9.9.9", "https://example.invalid/9.9.9", "")
|
||||
window._on_update_checked(release, "")
|
||||
self.assertIn("9.9.9", window.update_status.text())
|
||||
self.assertFalse(window.update_page.isHidden())
|
||||
self.assertEqual(window._release_url, release.url)
|
||||
# And the tray hears about it from here rather than waiting a day.
|
||||
self.assertEqual(told, [release])
|
||||
|
||||
def test_a_check_that_failed_says_why_and_hands_the_button_back(self):
|
||||
window = self.window(cfg.Config())
|
||||
window.update_now.setEnabled(False)
|
||||
window._on_update_checked(None, "api.github.com answered HTTP 403.")
|
||||
self.assertIn("403", window.update_status.text())
|
||||
self.assertTrue(window.update_now.isEnabled())
|
||||
|
||||
def test_the_settings_the_window_does_not_show_are_left_alone(self):
|
||||
"""A tab nobody wrote must not reset what the command line set."""
|
||||
self.write_config({"silence_db": -42.0, "speech_margin_db": 15.0,
|
||||
|
||||
@@ -0,0 +1,175 @@
|
||||
"""Whether a newer release is one worth telling somebody about.
|
||||
|
||||
Two things carry the weight here. A version is compared by its numbers alone,
|
||||
because a build off master carries the released number with its commit after
|
||||
it and is ahead of that release rather than behind it. And the clock lives in a
|
||||
file, so a day of asking nobody has to survive a restart.
|
||||
"""
|
||||
|
||||
import json
|
||||
import time
|
||||
|
||||
from dikte import hub
|
||||
from dikte import update
|
||||
from tests.support import DikteTest, fake_urlopen, url_error
|
||||
|
||||
RELEASE = {
|
||||
"tag_name": "v1.4.0",
|
||||
"html_url": "https://github.com/yusufipk/dikte/releases/tag/v1.4.0",
|
||||
"published_at": "2026-08-01T10:00:00Z",
|
||||
}
|
||||
|
||||
|
||||
class Numbers(DikteTest):
|
||||
def test_a_tag_and_a_bare_number_read_the_same(self):
|
||||
self.assertEqual(update._numbers("v1.4.0"), (1, 4, 0))
|
||||
self.assertEqual(update._numbers("1.4.0"), (1, 4, 0))
|
||||
|
||||
def test_a_short_number_is_filled_out(self):
|
||||
self.assertEqual(update._numbers("2"), (2, 0, 0))
|
||||
self.assertEqual(update._numbers("2.1"), (2, 1, 0))
|
||||
|
||||
def test_what_follows_the_number_is_dropped(self):
|
||||
self.assertEqual(update._numbers("1.0.1-dev.abc1234"), (1, 0, 1))
|
||||
self.assertEqual(update._numbers("1.0.1+build7"), (1, 0, 1))
|
||||
|
||||
def test_something_that_is_not_a_version_is_no_version(self):
|
||||
self.assertEqual(update._numbers("latest"), ())
|
||||
self.assertEqual(update._numbers(""), ())
|
||||
self.assertEqual(update._numbers(None), ())
|
||||
|
||||
|
||||
class Newer(DikteTest):
|
||||
def test_a_higher_number_is_newer(self):
|
||||
self.assertTrue(update.newer("1.4.0", "1.3.9"))
|
||||
self.assertTrue(update.newer("v2.0.0", "1.9.9"))
|
||||
|
||||
def test_the_same_number_is_not(self):
|
||||
self.assertFalse(update.newer("1.4.0", "1.4.0"))
|
||||
self.assertFalse(update.newer("1.3.0", "1.4.0"))
|
||||
|
||||
def test_a_build_off_master_is_ahead_of_the_release_it_names(self):
|
||||
"""1.0.1-dev.abc1234 was built after 1.0.1 went out, not before it.
|
||||
Read as a version suffix it would be older, and every nightly would be
|
||||
told to go back to the release it had already passed."""
|
||||
self.assertFalse(update.newer("1.0.1", "1.0.1-dev.abc1234"))
|
||||
self.assertTrue(update.newer("1.0.2", "1.0.1-dev.abc1234"))
|
||||
|
||||
def test_a_tag_that_is_not_a_version_is_never_newer(self):
|
||||
self.assertFalse(update.newer("nightly", "1.0.0"))
|
||||
|
||||
|
||||
class Asking(DikteTest):
|
||||
def setUp(self):
|
||||
super().setUp()
|
||||
self.patch_attr(hub, "CACHE_DIR", self.path("cache"))
|
||||
self.patch_attr(update, "__version__", "1.0.0")
|
||||
|
||||
def test_the_newest_release_comes_back_with_its_page(self):
|
||||
with fake_urlopen(RELEASE) as calls:
|
||||
release = update.latest()
|
||||
self.assertEqual(release.version, "1.4.0")
|
||||
self.assertEqual(release.url, RELEASE["html_url"])
|
||||
self.assertEqual(
|
||||
calls[0].full_url,
|
||||
"https://api.github.com/repos/yusufipk/dikte/releases/latest")
|
||||
|
||||
def test_a_release_with_no_page_falls_back_to_the_redirect(self):
|
||||
with fake_urlopen({"tag_name": "v1.4.0"}):
|
||||
release = update.latest()
|
||||
self.assertEqual(release.url, update.RELEASES_PAGE)
|
||||
|
||||
def test_a_repository_with_no_release_is_an_error(self):
|
||||
with fake_urlopen({"message": "Not Found"}):
|
||||
with self.assertRaises(hub.HubError):
|
||||
update.latest()
|
||||
|
||||
def test_a_check_answers_with_the_newer_release(self):
|
||||
with fake_urlopen(RELEASE):
|
||||
release = update.check()
|
||||
self.assertEqual(release.version, "1.4.0")
|
||||
|
||||
def test_a_check_that_finds_nothing_new_answers_with_nothing(self):
|
||||
self.patch_attr(update, "__version__", "1.4.0")
|
||||
with fake_urlopen(RELEASE):
|
||||
self.assertIsNone(update.check())
|
||||
|
||||
def test_a_second_check_the_same_day_asks_nobody(self):
|
||||
with fake_urlopen(RELEASE) as calls:
|
||||
update.check()
|
||||
release = update.check()
|
||||
self.assertEqual(len(calls), 1)
|
||||
# And still says what the first one found, since it is still true.
|
||||
self.assertEqual(release.version, "1.4.0")
|
||||
|
||||
def test_a_day_later_it_asks_again(self):
|
||||
with fake_urlopen(RELEASE) as calls:
|
||||
update.check()
|
||||
update._store(checked=time.time() - update.INTERVAL - 60)
|
||||
update.check()
|
||||
self.assertEqual(len(calls), 2)
|
||||
|
||||
def test_the_button_asks_whatever_the_clock_says(self):
|
||||
with fake_urlopen(RELEASE) as calls:
|
||||
update.check()
|
||||
update.check(force=True)
|
||||
self.assertEqual(len(calls), 2)
|
||||
|
||||
def test_a_check_that_cannot_reach_github_says_so(self):
|
||||
with fake_urlopen(url_error("no route to host")):
|
||||
with self.assertRaises(hub.HubError):
|
||||
update.check()
|
||||
|
||||
|
||||
class Remembering(DikteTest):
|
||||
def setUp(self):
|
||||
super().setUp()
|
||||
self.patch_attr(hub, "CACHE_DIR", self.path("cache"))
|
||||
self.patch_attr(update, "__version__", "1.0.0")
|
||||
|
||||
def test_what_the_last_check_found_survives_a_restart(self):
|
||||
with fake_urlopen(RELEASE):
|
||||
update.check()
|
||||
release = update.pending()
|
||||
self.assertEqual(release.version, "1.4.0")
|
||||
self.assertEqual(release.url, RELEASE["html_url"])
|
||||
|
||||
def test_nothing_was_ever_checked(self):
|
||||
self.assertIsNone(update.pending())
|
||||
self.assertEqual(update.state(), {})
|
||||
self.assertTrue(update.due())
|
||||
|
||||
def test_a_release_that_is_no_longer_newer_is_not_pending(self):
|
||||
"""The state file outlives the build that wrote it: an update that was
|
||||
found and then installed must not still be waiting afterwards."""
|
||||
update._store(version="1.4.0")
|
||||
self.patch_attr(update, "__version__", "1.4.0")
|
||||
self.assertIsNone(update.pending())
|
||||
|
||||
def test_a_version_is_announced_once(self):
|
||||
self.assertEqual(update.announced(), "")
|
||||
update.mark_announced("1.4.0")
|
||||
self.assertEqual(update.announced(), "1.4.0")
|
||||
|
||||
def test_a_state_file_that_is_rubbish_is_no_state_at_all(self):
|
||||
update.STATE_FILE.parent.mkdir(parents=True, exist_ok=True)
|
||||
update.STATE_FILE.write_text("half a {", encoding="utf-8")
|
||||
self.assertEqual(update.state(), {})
|
||||
self.assertIsNone(update.pending())
|
||||
|
||||
def test_a_state_file_that_cannot_be_written_is_not_a_failure(self):
|
||||
self.patch_attr(update, "STATE_FILE",
|
||||
self.path("nope") / "deeper" / "update.json")
|
||||
self.path("nope").write_text("a file where a directory would go")
|
||||
with fake_urlopen(RELEASE):
|
||||
release = update.check()
|
||||
self.assertEqual(release.version, "1.4.0")
|
||||
|
||||
def test_the_clock_is_kept_out_of_the_settings(self):
|
||||
"""A background check writes while the settings window may be open, and
|
||||
a write into config.json there would undo whatever it holds."""
|
||||
with fake_urlopen(RELEASE):
|
||||
update.check()
|
||||
stored = json.loads(update.STATE_FILE.read_text(encoding="utf-8"))
|
||||
self.assertEqual(stored["version"], "1.4.0")
|
||||
self.assertGreater(stored["checked"], 0)
|
||||
Reference in New Issue
Block a user