bellingcat/auto-archiver-api
1
0
Fork 0
mirror of https://github.com/bellingcat/auto-archiver-api.git synced 2026-06-11 13:08:34 +03:00
Code Issues Packages Projects Releases Wiki Activity

API updates

Browse Source
This commit is contained in:
msramalho
2023-02-26 20:22:20 +01:00
parent a00f3ff8d7
commit 3ef4787313
4 changed files with 37 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
src/worker.py
View File

@@ -4,6 +4,7 @@ import os
from celery import Celery
from dataclasses import asdict
from auto_archiver import Config, ArchivingOrchestrator, Metadata
from auto_archiver.enrichers import ScreenshotEnricher
from loguru import logger
from db import crud, models, schemas
@@ -28,11 +29,12 @@ config.parse(use_cli=False, yaml_config_filename="secrets/orchestration.yaml")
orchestrator = None
@celery.task(name="create_archive_task", bind=True)
def create_archive_task(self, url: str , user_email:str=""):
def create_archive_task(self, url: str, email:str=""):
assert type(url)==str and len(url)>5, f"Invalid URL received: {url}"
global orchestrator
if not orchestrator: orchestrator = ArchivingOrchestrator(config)
result = orchestrator.feed_item(Metadata().set_url(url)).to_json()
with get_db() as session:
db_task = crud.create_task(session, task=schemas.TaskCreate(id=self.request.id, url=url, author=user_email, result=json.loads(result)))
db_task = crud.create_task(session, task=schemas.TaskCreate(id=self.request.id, url=url, author=email, result=json.loads(result)))
logger.debug(f"Added {db_task.id=} to database on {db_task.created_at}")
return result