crocodilestick
be6cb19cb3
Fix archived book count mismatch by deleting archived_book rows when a book is deleted. Add TaskCleanArchivedBooks to purge stale archived references safely in batches. Schedule cleanup via CWA settings (default 03:00 local) and expose schedule controls in CWA Settings UI. Add new cwa_settings defaults/schema fields for archived cleanup timing. [bug] Deleting An Archived Book Doesn't Remove Archived Book Entry From app.db's archived_book table Fixes #8243
380 lines
17 KiB
Python
Executable File
380 lines
17 KiB
Python
Executable File
# -*- coding: utf-8 -*-
|
||
# Calibre-Web Automated – fork of Calibre-Web
|
||
# Copyright (C) 2018-2025 Calibre-Web contributors
|
||
# Copyright (C) 2024-2025 Calibre-Web Automated contributors
|
||
# SPDX-License-Identifier: GPL-3.0-or-later
|
||
# See CONTRIBUTORS for full list of authors.
|
||
|
||
import datetime
|
||
|
||
from . import config, constants
|
||
from .services.background_scheduler import BackgroundScheduler, CronTrigger, IntervalTrigger, use_APScheduler, DateTrigger
|
||
from .tasks.database import TaskReconnectDatabase, TaskCleanArchivedBooks
|
||
from .tasks.clean import TaskClean
|
||
from .tasks.thumbnail import TaskGenerateCoverThumbnails, TaskGenerateSeriesThumbnails, TaskClearCoverThumbnailCache
|
||
from .tasks.thumbnail_migration import check_and_migrate_thumbnails
|
||
from .services.worker import WorkerThread
|
||
from .tasks.metadata_backup import TaskBackupMetadata
|
||
from .tasks.auto_hardcover_id import TaskAutoHardcoverID
|
||
|
||
def get_scheduled_tasks(reconnect=True):
|
||
tasks = list()
|
||
# Reconnect Calibre database (metadata.db) based on config.schedule_reconnect
|
||
if reconnect:
|
||
tasks.append([lambda: TaskReconnectDatabase(), 'reconnect', False])
|
||
|
||
# Delete temp folder
|
||
tasks.append([lambda: TaskClean(), 'delete temp', True])
|
||
|
||
# Generate metadata.opf file for each changed book
|
||
if config.schedule_metadata_backup:
|
||
tasks.append([lambda: TaskBackupMetadata("en"), 'backup metadata', False])
|
||
|
||
# Generate all missing book cover thumbnails
|
||
if config.schedule_generate_book_covers:
|
||
tasks.append([lambda: TaskClearCoverThumbnailCache(0), 'delete superfluous book covers', True])
|
||
tasks.append([lambda: TaskGenerateCoverThumbnails(), 'generate book covers', False])
|
||
|
||
# Generate all missing series thumbnails
|
||
if config.schedule_generate_series_covers:
|
||
tasks.append([lambda: TaskGenerateSeriesThumbnails(), 'generate book covers', False])
|
||
|
||
return tasks
|
||
|
||
|
||
def end_scheduled_tasks():
|
||
worker = WorkerThread.get_instance()
|
||
for __, __, __, task, __ in worker.tasks:
|
||
if task.scheduled and task.is_cancellable:
|
||
worker.end_task(task.id)
|
||
|
||
|
||
def register_scheduled_tasks(reconnect=True):
|
||
scheduler = BackgroundScheduler()
|
||
|
||
if scheduler:
|
||
# Remove all existing jobs
|
||
scheduler.remove_all_jobs()
|
||
|
||
start = config.schedule_start_time
|
||
duration = config.schedule_duration
|
||
|
||
# Register scheduled tasks
|
||
timezone_info = datetime.datetime.now(datetime.timezone.utc).astimezone().tzinfo
|
||
scheduler.schedule_tasks(tasks=get_scheduled_tasks(reconnect), trigger=CronTrigger(hour=start,
|
||
timezone=timezone_info))
|
||
_schedule_duplicate_scan(scheduler, timezone_info)
|
||
end_time = calclulate_end_time(start, duration)
|
||
scheduler.schedule(func=end_scheduled_tasks, trigger=CronTrigger(hour=end_time.hour, minute=end_time.minute,
|
||
timezone=timezone_info),
|
||
name="end scheduled task")
|
||
|
||
_schedule_hardcover_auto_fetch(scheduler, timezone_info)
|
||
_schedule_archived_book_cleanup(scheduler, timezone_info)
|
||
|
||
# Kick-off tasks, if they should currently be running
|
||
if should_task_be_running(start, duration):
|
||
scheduler.schedule_tasks_immediately(tasks=get_scheduled_tasks(reconnect))
|
||
|
||
|
||
def register_startup_tasks():
|
||
scheduler = BackgroundScheduler()
|
||
|
||
if scheduler:
|
||
start = config.schedule_start_time
|
||
duration = config.schedule_duration
|
||
|
||
# Run thumbnail migration on startup (one-time operation)
|
||
try:
|
||
check_and_migrate_thumbnails()
|
||
except Exception as ex:
|
||
# Don't let migration failures stop the application
|
||
pass
|
||
|
||
# Rehydrate scheduled auto-send jobs from cwa.db (if any)
|
||
try:
|
||
import sys as _sys
|
||
if '/app/calibre-web-automated/scripts/' not in _sys.path:
|
||
_sys.path.insert(1, '/app/calibre-web-automated/scripts/')
|
||
from cwa_db import CWA_DB
|
||
from .tasks.auto_send import TaskAutoSend
|
||
from .services.worker import WorkerThread
|
||
from datetime import datetime, timezone
|
||
|
||
db = CWA_DB()
|
||
delay_minutes = int(db.cwa_settings.get('auto_send_delay_minutes', 0) or 0)
|
||
pending = db.scheduled_get_pending_autosend()
|
||
for row in pending:
|
||
try:
|
||
# Parse UTC run time, convert to local naive for DateTrigger
|
||
run_at_utc = datetime.fromisoformat(row['run_at_utc'].replace('Z', '+00:00'))
|
||
run_at_local = run_at_utc.astimezone().replace(tzinfo=None)
|
||
username = row.get('username') or 'System'
|
||
title = row.get('title') or 'Book'
|
||
book_id = int(row['book_id']) if row.get('book_id') is not None else None
|
||
user_id = int(row['user_id']) if row.get('user_id') is not None else None
|
||
schedule_id = int(row['id'])
|
||
|
||
def _rehydrate_enqueue(uid=user_id, bid=book_id, u=username, t=title, sid=schedule_id):
|
||
# Mark dispatched and enqueue the task only if state moved from scheduled
|
||
should_enqueue = False
|
||
try:
|
||
should_enqueue = bool(CWA_DB().scheduled_mark_dispatched(int(sid)))
|
||
except Exception:
|
||
pass
|
||
if should_enqueue and bid is not None and uid is not None:
|
||
WorkerThread.add(u, TaskAutoSend(f"Auto-sending '{t}' to user's eReader(s)", bid, uid, delay_minutes), hidden=False)
|
||
|
||
job = scheduler.schedule(func=_rehydrate_enqueue, trigger=DateTrigger(run_date=run_at_local), name=f"rehydrated auto-send {schedule_id}")
|
||
try:
|
||
if job is not None:
|
||
db.scheduled_update_job_id(schedule_id, str(job.id))
|
||
except Exception:
|
||
pass
|
||
except Exception:
|
||
# Never break startup on rehydration issues
|
||
pass
|
||
except Exception:
|
||
# If scripts not available or table missing, skip
|
||
pass
|
||
|
||
# Rehydrate other scheduled ops (convert_library, epub_fixer)
|
||
try:
|
||
import sys as _sys
|
||
if '/app/calibre-web-automated/scripts/' not in _sys.path:
|
||
_sys.path.insert(1, '/app/calibre-web-automated/scripts/')
|
||
from cwa_db import CWA_DB
|
||
from datetime import datetime
|
||
# wrappers will trigger internal routes themselves
|
||
from .tasks.ops import TaskConvertLibraryRun, TaskEpubFixerRun
|
||
|
||
db = CWA_DB()
|
||
for job_type in ('convert_library', 'epub_fixer'):
|
||
try:
|
||
rows = db.scheduled_get_pending_by_type(job_type)
|
||
except Exception:
|
||
rows = []
|
||
for row in rows:
|
||
try:
|
||
run_at_utc = datetime.fromisoformat(row['run_at_utc'].replace('Z', '+00:00'))
|
||
run_at_local = run_at_utc.astimezone().replace(tzinfo=None)
|
||
schedule_id = int(row['id'])
|
||
username = row.get('username') or 'System'
|
||
|
||
def _rehydrate_trigger(sid=schedule_id, jt=job_type, u=username):
|
||
should_run = False
|
||
try:
|
||
should_run = bool(CWA_DB().scheduled_mark_dispatched(int(sid)))
|
||
except Exception:
|
||
pass
|
||
if not should_run:
|
||
return
|
||
if jt == 'convert_library':
|
||
WorkerThread.add(u, TaskConvertLibraryRun(), hidden=False)
|
||
elif jt == 'epub_fixer':
|
||
WorkerThread.add(u, TaskEpubFixerRun(), hidden=False)
|
||
|
||
job = scheduler.schedule(func=_rehydrate_trigger, trigger=DateTrigger(run_date=run_at_local), name=f"rehydrated {job_type} {schedule_id}")
|
||
try:
|
||
if job is not None:
|
||
db.scheduled_update_job_id(schedule_id, str(job.id))
|
||
except Exception:
|
||
pass
|
||
except Exception:
|
||
pass
|
||
except Exception:
|
||
pass
|
||
|
||
# Run scheduled tasks immediately for development and testing
|
||
# Ignore tasks that should currently be running, as these will be added when registering scheduled tasks
|
||
if constants.APP_MODE in ['development', 'test'] and not should_task_be_running(start, duration):
|
||
scheduler.schedule_tasks_immediately(tasks=get_scheduled_tasks(False))
|
||
else:
|
||
scheduler.schedule_tasks_immediately(tasks=[[lambda: TaskClean(), 'delete temp', True]])
|
||
|
||
|
||
def should_task_be_running(start, duration):
|
||
now = datetime.datetime.now()
|
||
start_time = datetime.datetime.now().replace(hour=start, minute=0, second=0, microsecond=0)
|
||
end_time = start_time + datetime.timedelta(hours=duration // 60, minutes=duration % 60)
|
||
return start_time < now < end_time
|
||
|
||
|
||
def calclulate_end_time(start, duration):
|
||
start_time = datetime.datetime.now().replace(hour=start, minute=0)
|
||
return start_time + datetime.timedelta(hours=duration // 60, minutes=duration % 60)
|
||
|
||
|
||
def _schedule_duplicate_scan(scheduler, timezone_info):
|
||
"""Schedule background duplicate scan based on CWA settings."""
|
||
try:
|
||
import sys as _sys
|
||
if '/app/calibre-web-automated/scripts/' not in _sys.path:
|
||
_sys.path.insert(1, '/app/calibre-web-automated/scripts/')
|
||
from cwa_db import CWA_DB
|
||
from .tasks.duplicate_scan import TaskDuplicateScan
|
||
from apscheduler.triggers.cron import CronTrigger
|
||
|
||
db = CWA_DB()
|
||
enabled = bool(db.cwa_settings.get('duplicate_scan_enabled', 0))
|
||
cron_expr = (db.cwa_settings.get('duplicate_scan_cron') or '').strip()
|
||
|
||
if not enabled:
|
||
return
|
||
|
||
if cron_expr:
|
||
trigger = CronTrigger.from_crontab(cron_expr, timezone=timezone_info)
|
||
else:
|
||
# manual/after_import handled elsewhere
|
||
return
|
||
|
||
scheduler.schedule_task(lambda: TaskDuplicateScan(full_scan=True, trigger_type='scheduled'),
|
||
user='System', trigger=trigger, name='duplicate scan', hidden=False)
|
||
except Exception:
|
||
# Scheduling is best-effort; never block startup
|
||
pass
|
||
|
||
|
||
def _schedule_hardcover_auto_fetch(scheduler, timezone_info):
|
||
"""Schedule background Hardcover auto-fetch based on CWA settings."""
|
||
try:
|
||
import sys as _sys
|
||
if '/app/calibre-web-automated/scripts/' not in _sys.path:
|
||
_sys.path.insert(1, '/app/calibre-web-automated/scripts/')
|
||
from cwa_db import CWA_DB
|
||
from .tasks.auto_hardcover_id import TaskAutoHardcoverID
|
||
from os import getenv
|
||
|
||
db = CWA_DB()
|
||
cwa_settings = db.get_cwa_settings()
|
||
|
||
# Check if enabled and token available
|
||
enabled = bool(cwa_settings.get('hardcover_auto_fetch_enabled', False))
|
||
token_available = bool(
|
||
getattr(config, "config_hardcover_token", None) or
|
||
getenv("HARDCOVER_TOKEN")
|
||
)
|
||
|
||
if not enabled or not token_available:
|
||
return
|
||
|
||
schedule_type = cwa_settings.get('hardcover_auto_fetch_schedule', 'weekly')
|
||
schedule_day = cwa_settings.get('hardcover_auto_fetch_schedule_day', 'sunday')
|
||
schedule_hour = int(cwa_settings.get('hardcover_auto_fetch_schedule_hour', 2))
|
||
min_confidence = float(cwa_settings.get('hardcover_auto_fetch_min_confidence', 0.85))
|
||
batch_size = int(cwa_settings.get('hardcover_auto_fetch_batch_size', 50))
|
||
rate_limit = float(cwa_settings.get('hardcover_auto_fetch_rate_limit', 5.0))
|
||
|
||
# Create lambda that returns task instance with configured settings
|
||
task_lambda = lambda: TaskAutoHardcoverID(
|
||
min_confidence=min_confidence,
|
||
batch_size=batch_size,
|
||
rate_limit_delay=rate_limit
|
||
)
|
||
|
||
# Map day names to APScheduler format
|
||
day_map = {
|
||
'monday': 'mon', 'tuesday': 'tue', 'wednesday': 'wed',
|
||
'thursday': 'thu', 'friday': 'fri', 'saturday': 'sat', 'sunday': 'sun'
|
||
}
|
||
|
||
# Determine trigger based on schedule type
|
||
trigger = None
|
||
name = "hardcover auto-fetch"
|
||
|
||
if schedule_type == '15min':
|
||
trigger = IntervalTrigger(minutes=15, timezone=timezone_info)
|
||
elif schedule_type == '30min':
|
||
trigger = IntervalTrigger(minutes=30, timezone=timezone_info)
|
||
elif schedule_type == '1hour':
|
||
trigger = IntervalTrigger(hours=1, timezone=timezone_info)
|
||
elif schedule_type == '2hours':
|
||
trigger = IntervalTrigger(hours=2, timezone=timezone_info)
|
||
elif schedule_type == '4hours':
|
||
trigger = IntervalTrigger(hours=4, timezone=timezone_info)
|
||
elif schedule_type == '6hours':
|
||
trigger = IntervalTrigger(hours=6, timezone=timezone_info)
|
||
elif schedule_type == '12hours':
|
||
trigger = IntervalTrigger(hours=12, timezone=timezone_info)
|
||
elif schedule_type == 'daily':
|
||
trigger = CronTrigger(hour=schedule_hour, minute=0, timezone=timezone_info)
|
||
elif schedule_type == 'weekly':
|
||
day_abbr = day_map.get(schedule_day.lower(), 'sun')
|
||
trigger = CronTrigger(day_of_week=day_abbr, hour=schedule_hour, minute=0, timezone=timezone_info)
|
||
elif schedule_type == 'monthly':
|
||
# For monthly, schedule_day contains day of month (1-28)
|
||
try:
|
||
day_of_month = int(schedule_day) if str(schedule_day).isdigit() else 1
|
||
day_of_month = max(1, min(28, day_of_month)) # Clamp to 1-28
|
||
except (ValueError, TypeError):
|
||
day_of_month = 1
|
||
trigger = CronTrigger(day=day_of_month, hour=schedule_hour, minute=0, timezone=timezone_info)
|
||
|
||
if trigger:
|
||
scheduler.schedule_task(task_lambda, user='System', trigger=trigger, name=name, hidden=False)
|
||
except Exception:
|
||
# Scheduling is best-effort; never block startup
|
||
pass
|
||
|
||
|
||
def _schedule_archived_book_cleanup(scheduler, timezone_info):
|
||
"""Schedule cleanup for stale archived_book entries (default 03:00 local)."""
|
||
try:
|
||
import sys as _sys
|
||
if '/app/calibre-web-automated/scripts/' not in _sys.path:
|
||
_sys.path.insert(1, '/app/calibre-web-automated/scripts/')
|
||
from cwa_db import CWA_DB
|
||
|
||
db = CWA_DB()
|
||
cwa_settings = db.get_cwa_settings()
|
||
|
||
enabled = bool(cwa_settings.get('archived_cleanup_enabled', True))
|
||
if not enabled:
|
||
return
|
||
|
||
schedule_type = (cwa_settings.get('archived_cleanup_schedule') or 'daily').strip().lower()
|
||
schedule_day = (cwa_settings.get('archived_cleanup_schedule_day') or 'sunday')
|
||
schedule_hour = int(cwa_settings.get('archived_cleanup_schedule_hour', 3))
|
||
|
||
day_map = {
|
||
'monday': 'mon', 'tuesday': 'tue', 'wednesday': 'wed',
|
||
'thursday': 'thu', 'friday': 'fri', 'saturday': 'sat', 'sunday': 'sun'
|
||
}
|
||
|
||
trigger = None
|
||
name = 'clean archived book references'
|
||
|
||
if schedule_type == '15min':
|
||
trigger = IntervalTrigger(minutes=15, timezone=timezone_info)
|
||
elif schedule_type == '30min':
|
||
trigger = IntervalTrigger(minutes=30, timezone=timezone_info)
|
||
elif schedule_type == '1hour':
|
||
trigger = IntervalTrigger(hours=1, timezone=timezone_info)
|
||
elif schedule_type == '2hours':
|
||
trigger = IntervalTrigger(hours=2, timezone=timezone_info)
|
||
elif schedule_type == '4hours':
|
||
trigger = IntervalTrigger(hours=4, timezone=timezone_info)
|
||
elif schedule_type == '6hours':
|
||
trigger = IntervalTrigger(hours=6, timezone=timezone_info)
|
||
elif schedule_type == '12hours':
|
||
trigger = IntervalTrigger(hours=12, timezone=timezone_info)
|
||
elif schedule_type == 'daily':
|
||
trigger = CronTrigger(hour=schedule_hour, minute=0, timezone=timezone_info)
|
||
elif schedule_type == 'weekly':
|
||
day_abbr = day_map.get(str(schedule_day).lower(), 'sun')
|
||
trigger = CronTrigger(day_of_week=day_abbr, hour=schedule_hour, minute=0, timezone=timezone_info)
|
||
elif schedule_type == 'monthly':
|
||
try:
|
||
day_of_month = int(schedule_day) if str(schedule_day).isdigit() else 1
|
||
day_of_month = max(1, min(28, day_of_month))
|
||
except (ValueError, TypeError):
|
||
day_of_month = 1
|
||
trigger = CronTrigger(day=day_of_month, hour=schedule_hour, minute=0, timezone=timezone_info)
|
||
|
||
if trigger:
|
||
scheduler.schedule_task(lambda: TaskCleanArchivedBooks(), user='System',
|
||
trigger=trigger, name=name, hidden=True)
|
||
except Exception:
|
||
# Scheduling is best-effort; never block startup
|
||
pass
|