mirrors/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2025-05-28 05:34:14 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions

rename actors to workers

Browse source
This commit is contained in:
Nick Sweeting 2024-11-18 02:25:00 -08:00
parent 9b8cf7b4f0
commit f5727c7da2
No known key found for this signature in database
5 changed files with 115 additions and 120 deletions
Download patch file Download diff file Expand all files Collapse all files

24
archivebox/crawls/statemachines.py
View file

@ -1,14 +1,13 @@
__package__ = 'archivebox.crawls'
from typing import ClassVar
from django.utils import timezone
from statemachine import State, StateMachine
from actors.actor import ActorType
from crawls.models import Crawl
# State Machine Definitions
#################################################
class CrawlMachine(StateMachine, strict_states=True):
"""State machine for managing Crawl lifecycle."""
@ -22,9 +21,9 @@ class CrawlMachine(StateMachine, strict_states=True):
# Tick Event
tick = (
queued.to.itself(unless='can_start') |
queued.to.itself(unless='can_start', internal=True) |
queued.to(started, cond='can_start') |
started.to.itself(unless='is_finished') |
started.to.itself(unless='is_finished', internal=True) |
started.to(sealed, cond='is_finished')
)
@ -63,3 +62,18 @@ class CrawlMachine(StateMachine, strict_states=True):
self.crawl.retry_at = None
self.crawl.save()
class CrawlWorker(ActorType[Crawl]):
"""The Actor that manages the lifecycle of all Crawl objects"""
Model = Crawl
StateMachineClass = CrawlMachine
ACTIVE_STATE: ClassVar[State] = CrawlMachine.started
FINAL_STATES: ClassVar[list[State]] = CrawlMachine.final_states
STATE_FIELD_NAME: ClassVar[str] = Crawl.state_field_name
MAX_CONCURRENT_ACTORS: ClassVar[int] = 3
MAX_TICK_TIME: ClassVar[int] = 10
CLAIM_FROM_TOP_N: ClassVar[int] = MAX_CONCURRENT_ACTORS * 10