Skip to content

Speakr is a personal, self-hosted web application designed for transcribing audio recordings

Python

Long file names shortened when stored; titles fit the column (#430)

Murtaza Nasir4a659b614acc
diff --git a/src/api/recordings.py b/src/api/recordings.py
index 8974c6d9..cc81ed3e 100644
--- a/src/api/recordings.py
+++ b/src/api/recordings.py
@@ -53,6 +53,7 @@ from src.services.llm import client, chat_client, call_llm_completion, call_chat
from src.services.embeddings import process_recording_chunks
from src.file_exporter import export_recording, mark_export_as_deleted
from src.utils.mime import served_media_mime, sandbox_media_response
+from src.utils.filenames import shorten_filename
from src.utils.ffprobe import get_codec_info, get_creation_date, get_duration, try_repair_malformed_webm, FFProbeError
from src.utils.audio_conversion import convert_if_needed
from src.services.storage import get_storage_service
@@ -2230,7 +2231,7 @@ def ingest_uploaded_recording(
response_data['idempotent_replay'] = True
return jsonify(response_data), success_status
- safe_filename = secure_filename(original_filename)
+ safe_filename = shorten_filename(secure_filename(original_filename))
storage = get_storage_service()
staging_dir = storage.get_staging_dir()
filepath = os.path.join(
diff --git a/src/file_monitor.py b/src/file_monitor.py
index 9529c9af..be11b279 100644
--- a/src/file_monitor.py
+++ b/src/file_monitor.py
@@ -19,6 +19,8 @@ from werkzeug.utils import secure_filename
from src.utils.ffprobe import get_codec_info, get_creation_date, get_duration, FFProbeError
from src.utils.ffmpeg_utils import FFmpegError, FFmpegNotFoundError
from src.utils.audio_conversion import convert_if_needed
+from src.utils.filenames import shorten_filename
+from src.utils.titles import placeholder_titles
# Video retention - when enabled, video files keep their video stream for playback
VIDEO_RETENTION = os.environ.get('VIDEO_RETENTION', 'false').lower() == 'true'
@@ -368,7 +370,7 @@ class FileMonitor:
raise_notice(user, quota_error, original_filename)
return
self._quota_skipped.discard(str(processing_path.with_name(original_filename)))
- safe_filename = secure_filename(original_filename)
+ safe_filename = shorten_filename(secure_filename(original_filename))
timestamp = datetime.now().strftime('%Y%m%d%H%M%S')
new_filename = f"auto_{timestamp}_{safe_filename}"
@@ -511,7 +513,7 @@ class FileMonitor:
recording = Recording(
audio_path=None,
original_filename=original_filename,
- title=f"Auto-processed - {original_filename}",
+ title=placeholder_titles(original_filename)[1],
file_size=file_size,
status='PENDING',
meeting_date=meeting_date,
diff --git a/src/services/storage/service.py b/src/services/storage/service.py
index f49243e1..151c0fd2 100644
--- a/src/services/storage/service.py
+++ b/src/services/storage/service.py
@@ -11,6 +11,8 @@ from typing import Iterator, Optional
from uuid import uuid4
import threading
+from src.utils.filenames import shorten_filename
+
from .factory import StorageSettings, build_local_backend, build_s3_backend, load_storage_settings_from_env
from .interfaces import AudioDeliveryResult, MaterializedFile, StorageLocator, StoredObject
from .locator import build_local_locator, parse_locator, relative_key_from_local_path
@@ -59,7 +61,7 @@ class StorageService:
now = now or datetime.utcnow()
base_name = os.path.basename(original_filename or 'recording.bin').replace('\\', '_').replace('/', '_')
safe_name = ''.join(c for c in base_name if c.isalnum() or c in (' ', '.', '-', '_')).strip() or 'recording.bin'
- safe_name = safe_name.replace(' ', '_')
+ safe_name = shorten_filename(safe_name.replace(' ', '_'))
ts = now.strftime('%Y%m%d%H%M%S')
rec_part = str(recording_id) if recording_id is not None else f"tmp-{uuid4().hex[:8]}"
prefix = (self.settings.key_prefix or 'recordings').strip('/').replace('\\', '/')
diff --git a/src/utils/filenames.py b/src/utils/filenames.py
new file mode 100644
index 00000000..7f4a09a9
--- /dev/null
+++ b/src/utils/filenames.py
@@ -0,0 +1,23 @@
+"""Length of stored file names (#430).
+
+A file system allows 255 bytes in one name. Speakr adds a timestamp and an id
+to an uploaded file's name when it stores the file, so a name of 208 or more
+characters failed with a 500. The stored name is shortened; the original
+name is kept in Recording.original_filename for display.
+"""
+
+import os
+
+MAX_NAME_BYTES = 150
+
+
+def shorten_filename(name, max_bytes=MAX_NAME_BYTES):
+ """``name`` cut to ``max_bytes`` of UTF-8, keeping its extension."""
+ if len(name.encode('utf-8')) <= max_bytes:
+ return name
+ stem, ext = os.path.splitext(name)
+ if len(ext.encode('utf-8')) > 20:
+ ext = ''
+ budget = max_bytes - len(ext.encode('utf-8'))
+ stem = stem.encode('utf-8')[:budget].decode('utf-8', 'ignore')
+ return stem + ext
diff --git a/src/utils/titles.py b/src/utils/titles.py
index 18f81c6d..db08943d 100644
--- a/src/utils/titles.py
+++ b/src/utils/titles.py
@@ -15,16 +15,20 @@ normal upload instead of being silently skipped.
"""
+# Length of Recording.title; PostgreSQL refuses a longer value (#430)
+TITLE_MAX_LENGTH = 200
+
+
def placeholder_title(original_filename):
"""The default placeholder title for an upload with no user-supplied title."""
- return f"Recording - {original_filename}"
+ return f"Recording - {original_filename}"[:TITLE_MAX_LENGTH]
def placeholder_titles(original_filename):
"""All placeholder titles the AI title generator may overwrite."""
return (
- f"Recording - {original_filename}",
- f"Auto-processed - {original_filename}",
+ f"Recording - {original_filename}"[:TITLE_MAX_LENGTH],
+ f"Auto-processed - {original_filename}"[:TITLE_MAX_LENGTH],
)
@@ -36,7 +40,7 @@ def resolve_upload_title(user_title, original_filename):
recognises, so AI title generation runs.
"""
if user_title and user_title.strip():
- return user_title.strip()
+ return user_title.strip()[:TITLE_MAX_LENGTH]
return placeholder_title(original_filename)
diff --git a/tests/test_cov_recordings_write.py b/tests/test_cov_recordings_write.py
index 42303a7d..6d55f59a 100644
--- a/tests/test_cov_recordings_write.py
+++ b/tests/test_cov_recordings_write.py
@@ -219,6 +219,23 @@ def test_upload_with_html_extension_is_not_stored_as_html():
_cleanup(rec, user)
+def test_upload_with_a_long_file_name():
+ """A name of 208 or more characters failed with a 500 (#430)."""
+ long_name = "n" * 250 + ".mp3"
+ with app.app_context():
+ user = _mk_user("up_long")
+ client = app.test_client()
+ _login(client, user)
+ staging = os.path.join(app.config["UPLOAD_FOLDER"], f"stg_{uuid.uuid4().hex[:6]}")
+ with _upload_mocks(staging):
+ resp = _do_upload(client, filename=long_name)
+ assert resp.status_code == 202, resp.data
+ rec = _latest_rec(user)
+ assert rec.original_filename == long_name
+ assert len(rec.title) <= 200
+ _cleanup(rec, user)
+
+
def test_upload_default_title_is_placeholder():
with app.app_context():
user = _mk_user("up_title")
diff --git a/tests/test_long_filenames.py b/tests/test_long_filenames.py
new file mode 100644
index 00000000..14d86c12
--- /dev/null
+++ b/tests/test_long_filenames.py
@@ -0,0 +1,41 @@
+"""Uploads with long file names (#430).
+
+Speakr adds a timestamp and an id to an uploaded file's name when it stores
+the file, so a name of 208 or more characters exceeded the file system's
+255-byte limit and the upload failed with a 500.
+"""
+
+import os
+from datetime import datetime
+
+from src.services.storage import get_storage_service
+from src.utils.filenames import shorten_filename
+from src.utils.titles import placeholder_title, placeholder_titles, resolve_upload_title
+
+
+def test_short_names_are_unchanged():
+ assert shorten_filename("meeting.mp3") == "meeting.mp3"
+
+
+def test_long_names_keep_their_extension():
+ name = shorten_filename("a" * 300 + ".m4a")
+ assert name.endswith(".m4a") and len(name.encode("utf-8")) == 150
+
+
+def test_multibyte_names_are_cut_on_a_character():
+ name = shorten_filename("会议" * 200 + ".mp3")
+ assert name.endswith(".mp3") and len(name.encode("utf-8")) <= 150
+ name.encode("utf-8").decode("utf-8")
+
+
+def test_storage_key_name_fits_the_file_system():
+ key = get_storage_service().build_recording_key("会议" * 200 + ".mp3", 12345, now=datetime(2026, 10, 19))
+ assert len(os.path.basename(key).encode("utf-8")) <= 255
+
+
+def test_titles_fit_the_title_column():
+ long_name = "x" * 400 + ".mp3"
+ assert len(placeholder_title(long_name)) == 200
+ assert all(len(t) == 200 for t in placeholder_titles(long_name))
+ assert placeholder_title(long_name) in placeholder_titles(long_name)
+ assert len(resolve_upload_title("y" * 300, long_name)) == 200