From 6874d123ebe852e2a41dbf3634c4efa35f1f6b6f Mon Sep 17 00:00:00 2001
From: msramalho <19508417+msramalho@users.noreply.github.com>
Date: Tue, 12 Dec 2023 19:14:10 +0000
Subject: [PATCH] adds logic to test if archive is needed, if specified by the
 user

---
 src/db/crud.py    |  7 +++++--
 src/db/schemas.py |  1 +
 src/worker.py     | 13 ++++++++++++-
 3 files changed, 18 insertions(+), 3 deletions(-)

diff --git a/src/db/crud.py b/src/db/crud.py
index b3e444a..b202a5b 100644
--- a/src/db/crud.py
+++ b/src/db/crud.py
@@ -23,14 +23,17 @@ def get_task(db: Session, task_id: str, email: str):
     return query.first()
 
 
-def search_tasks_by_url(db: Session, url: str, email: str, skip: int = 0, limit: int = 100, archived_after: datetime = None, archived_before: datetime = None):
+def search_tasks_by_url(db: Session, url: str, email: str, skip: int = 0, limit: int = 100, archived_after: datetime = None, archived_before: datetime = None, absolute_search: bool = False):
     # searches for partial URLs, if email is * no ownership filtering happens
     query = base_query(db)
     if email != ALLOW_ANY_EMAIL:
         email = email.lower()
         groups = get_user_groups(db, email)
         query = query.filter(or_(models.Archive.public == True, models.Archive.author_id == email, models.Archive.group_id.in_(groups)))
-    query = query.filter(models.Archive.url.like(f'%{url}%'))
+    if absolute_search:
+        query = query.filter(models.Archive.url == url)
+    else:
+        query = query.filter(models.Archive.url.like(f'%{url}%'))
     if archived_after:
         query = query.filter(models.Archive.created_at >= archived_after)
     if archived_before:
diff --git a/src/db/schemas.py b/src/db/schemas.py
index 14b5e81..28823cc 100644
--- a/src/db/schemas.py
+++ b/src/db/schemas.py
@@ -10,6 +10,7 @@ class ArchiveCreate(BaseModel):
     author_id: str | None = None
     group_id: str | None = None
     tags: set = set()
+    rearchive: bool = True
     # urls: list = []
 
 
diff --git a/src/worker.py b/src/worker.py
index b68deb6..457bc9c 100644
--- a/src/worker.py
+++ b/src/worker.py
@@ -13,6 +13,8 @@ from db.database import SessionLocal
 from contextlib import contextmanager
 import json
 
+from security import ALLOW_ANY_EMAIL
+
 celery = Celery(__name__)
 celery.conf.broker_url = os.environ.get("CELERY_BROKER_URL", "redis://localhost:6379")
 celery.conf.result_backend = os.environ.get("CELERY_RESULT_BACKEND", "redis://localhost:6379")
@@ -26,7 +28,7 @@ def get_db():
     finally: session.close()
 
 
-@celery.task(name="create_archive_task", bind=True, autoretry_for=(Exception,), retry_backoff=True, retry_kwargs={'max_retries': 5})
+@celery.task(name="create_archive_task", bind=True, autoretry_for=(Exception,), retry_backoff=True, retry_kwargs={'max_retries': 3})
 def create_archive_task(self, archive_json: str):
     archive = schemas.ArchiveCreate.parse_raw(archive_json)
 
@@ -36,6 +38,15 @@ def create_archive_task(self, archive_json: str):
 
     url = archive.url
     logger.info(f"{url=} {archive=}")
+
+    if not archive.rearchive:
+        with get_db() as session:
+            archives = crud.search_tasks_by_url(session, url, ALLOW_ANY_EMAIL, absolute_search=True)
+            if len(archives):
+                logger.info(f"Skipping {url=} as it was already archived")
+                # TODO: can we achieve something better than the last result?
+                return archives[0].result
+
     orchestrator = choose_orchestrator(archive.group_id, archive.author_id)
     result = orchestrator.feed_item(Metadata().set_url(url))