feat: Update collection sorting, metadata, stats

backend/btrixcloud/basecrawls.py

@@ -49,7 +49,7 @@
 
 
 # ============================================================================
-# pylint: disable=too-many-instance-attributes, too-many-public-methods, too-many-lines
+# pylint: disable=too-many-instance-attributes, too-many-public-methods, too-many-lines, too-many-branches
 class BaseCrawlOps:
     """operations that apply to all crawls"""
 
@@ -300,6 +300,7 @@ async def delete_crawls(
     ) -> tuple[int, dict[UUID, dict[str, int]], bool]:
         """Delete a list of crawls by id for given org"""
         cids_to_update: dict[UUID, dict[str, int]] = {}
+        collection_ids_to_update = set()
 
         size = 0
 
@@ -325,6 +326,10 @@ async def delete_crawls(
 
             await self.page_ops.delete_crawl_pages(crawl_id, org.id)
 
+            if crawl.collectionIds:
+                for coll_id in crawl.collectionIds:
+                    collection_ids_to_update.add(coll_id)
+
             if type_ == "crawl":
                 await self.delete_all_crawl_qa_files(crawl_id, org)
 
@@ -361,6 +366,10 @@ async def delete_crawls(
 
         await self.orgs.set_last_crawl_finished(org.id)
 
+        if collection_ids_to_update:
+            for coll_id in collection_ids_to_update:
+                await self.colls.update_collection_counts_and_tags(coll_id)
+
         quota_reached = self.orgs.storage_quota_reached(org)
 
         return res.deleted_count, cids_to_update, quota_reached

backend/btrixcloud/colls.py

@@ -396,7 +396,7 @@ async def list_collections(
         page = page - 1
         skip = page * page_size
 
-        match_query: dict[str, object] = {"oid": org.id}
+        match_query: Dict[str, Union[str, UUID, int, object]] = {"oid": org.id}
 
         if name:
             match_query["name"] = name
@@ -409,15 +409,33 @@ async def list_collections(
         elif access:
             match_query["access"] = access
 
-        aggregate = [{"$match": match_query}]
+        aggregate: List[Dict[str, Union[str, UUID, int, object]]] = [
+            {"$match": match_query}
+        ]
 
         if sort_by:
-            if sort_by not in ("modified", "name", "description", "totalSize"):
+            if sort_by not in (
+                "created",
+                "modified",
+                "dateLatest",
+                "name",
+                "crawlCount",
+                "pageCount",
+                "totalSize",
+                "description",
+                "caption",
+            ):
                 raise HTTPException(status_code=400, detail="invalid_sort_by")
             if sort_direction not in (1, -1):
                 raise HTTPException(status_code=400, detail="invalid_sort_direction")
 
-            aggregate.extend([{"$sort": {sort_by: sort_direction}}])
+            sort_query = {sort_by: sort_direction}
+
+            # add secondary sort keys:
+            if sort_by == "dateLatest":
+                sort_query["dateEarliest"] = sort_direction
+
+            aggregate.extend([{"$sort": sort_query}])
 
         aggregate.extend(
             [
@@ -564,11 +582,14 @@ async def recalculate_org_collection_counts_tags(self, org: Organization):
 
     async def update_collection_counts_and_tags(self, collection_id: UUID):
         """Set current crawl info in config when crawl begins"""
+        # pylint: disable=too-many-locals
         crawl_count = 0
         page_count = 0
         total_size = 0
         tags = []
 
+        crawl_ids = []
+
         coll = await self.get_collection(collection_id)
         org = await self.orgs.get_org_by_id(coll.oid)
 
@@ -582,25 +603,30 @@ async def update_collection_counts_and_tags(self, collection_id: UUID):
                 total_size += file.size
 
             try:
-                _, crawl_pages = await self.page_ops.list_pages(
+                _, crawl_page_count = await self.page_ops.list_pages(
                     crawl.id, org, page_size=1_000_000
                 )
-                page_count += crawl_pages
+                page_count += crawl_page_count
             # pylint: disable=broad-exception-caught
             except Exception:
                 pass
 
             if crawl.tags:
                 tags.extend(crawl.tags)
 
+            crawl_ids.append(crawl.id)
+
         sorted_tags = [tag for tag, count in Counter(tags).most_common()]
 
+        unique_page_count = await self.page_ops.get_unique_page_count(crawl_ids)
+
         await self.collections.find_one_and_update(
             {"_id": collection_id},
             {
                 "$set": {
                     "crawlCount": crawl_count,
                     "pageCount": page_count,
+                    "uniquePageCount": unique_page_count,
                     "totalSize": total_size,
                     "tags": sorted_tags,
                 }
@@ -618,6 +644,7 @@ async def recalculate_org_collection_dates(self, org: Organization):
 
     async def update_collection_dates(self, coll_id: UUID):
         """Update collection earliest and latest dates from page timestamps"""
+        # pylint: disable=too-many-locals
         coll = await self.get_collection(coll_id)
         crawl_ids = await self.get_collection_crawl_ids(coll_id)
 

backend/btrixcloud/db.py

@@ -17,7 +17,7 @@
 from .migrations import BaseMigration
 
 
-CURR_DB_VERSION = "0040"
+CURR_DB_VERSION = "0041"
 
 
 # ============================================================================
@@ -96,7 +96,7 @@ async def update_and_prepare_db(
     await ping_db(mdb)
     print("Database setup started", flush=True)
     if await run_db_migrations(
-        mdb, user_manager, page_ops, org_ops, background_job_ops
+        mdb, user_manager, page_ops, org_ops, background_job_ops, coll_ops
     ):
         await drop_indexes(mdb)
 
@@ -117,8 +117,10 @@ async def update_and_prepare_db(
 
 
 # ============================================================================
-# pylint: disable=too-many-locals
-async def run_db_migrations(mdb, user_manager, page_ops, org_ops, background_job_ops):
+# pylint: disable=too-many-locals, too-many-arguments
+async def run_db_migrations(
+    mdb, user_manager, page_ops, org_ops, background_job_ops, coll_ops
+):
     """Run database migrations."""
 
     # if first run, just set version and exit
@@ -155,6 +157,7 @@ async def run_db_migrations(mdb, user_manager, page_ops, org_ops, background_job
                 page_ops=page_ops,
                 org_ops=org_ops,
                 background_job_ops=background_job_ops,
+                coll_ops=coll_ops,
             )
             if await migration.run():
                 migrations_run = True

backend/btrixcloud/main.py

@@ -255,7 +255,7 @@ def main() -> None:
     crawls.set_page_ops(page_ops)
     upload_ops.set_page_ops(page_ops)
 
-    org_ops.set_ops(base_crawl_ops, profiles, coll_ops, background_job_ops)
+    org_ops.set_ops(base_crawl_ops, profiles, coll_ops, background_job_ops, page_ops)
 
     user_manager.set_ops(org_ops, crawl_config_ops, base_crawl_ops)
 

backend/btrixcloud/migrations/migration_0040_archived_item_page_count.py

@@ -31,13 +31,13 @@ async def migrate_up(self):
             )
             return
 
-        async for crawl_raw in crawls_mdb.find({"pageCount": None}):
+        async for crawl_raw in crawls_mdb.find({}):
             crawl_id = crawl_raw["_id"]
             try:
-                await self.page_ops.set_archived_item_page_count(crawl_id)
+                await self.page_ops.set_archived_item_page_counts(crawl_id)
             # pylint: disable=broad-exception-caught
             except Exception as err:
                 print(
-                    f"Error saving pageCount for archived item {crawl_id}: {err}",
+                    f"Error saving page counts for archived item {crawl_id}: {err}",
                     flush=True,
                 )

backend/btrixcloud/migrations/migration_0041_pages_snapshots.py

@@ -0,0 +1,43 @@
+"""
+Migration 0041 - Rationalize page counts
+"""
+
+from btrixcloud.migrations import BaseMigration
+
+
+MIGRATION_VERSION = "0041"
+
+
+class Migration(BaseMigration):
+    """Migration class."""
+
+    # pylint: disable=unused-argument
+    def __init__(self, mdb, **kwargs):
+        super().__init__(mdb, migration_version=MIGRATION_VERSION)
+
+        self.coll_ops = kwargs.get("coll_ops")
+
+    async def migrate_up(self):
+        """Perform migration up.
+
+        Recalculate collections to get new page and unique page counts
+        """
+        colls_mdb = self.mdb["collections"]
+
+        if self.coll_ops is None:
+            print(
+                "Unable to set collection page counts, missing coll_ops",
+                flush=True,
+            )
+            return
+
+        async for coll in colls_mdb.find({}):
+            coll_id = coll["_id"]
+            try:
+                await self.coll_ops.update_collection_counts_and_tags(coll_id)
+            # pylint: disable=broad-exception-caught
+            except Exception as err:
+                print(
+                    f"Unable to update page counts for collection {coll_id}: {err}",
+                    flush=True,
+                )

backend/btrixcloud/models.py

@@ -798,6 +798,7 @@ class BaseCrawl(CoreCrawlable, BaseMongoModel):
     reviewStatus: ReviewStatus = None
 
     pageCount: Optional[int] = 0
+    uniquePageCount: Optional[int] = 0
 
     filePageCount: Optional[int] = 0
     errorPageCount: Optional[int] = 0
@@ -875,6 +876,7 @@ class CrawlOut(BaseMongoModel):
     lastQAStarted: Optional[datetime] = None
 
     pageCount: Optional[int] = 0
+    uniquePageCount: Optional[int] = 0
     filePageCount: Optional[int] = 0
     errorPageCount: Optional[int] = 0
 
@@ -1250,6 +1252,7 @@ class Collection(BaseMongoModel):
 
     crawlCount: Optional[int] = 0
     pageCount: Optional[int] = 0
+    uniquePageCount: Optional[int] = 0
     totalSize: Optional[int] = 0
 
     dateEarliest: Optional[datetime] = None
@@ -1303,6 +1306,7 @@ class CollOut(BaseMongoModel):
 
     crawlCount: Optional[int] = 0
     pageCount: Optional[int] = 0
+    uniquePageCount: Optional[int] = 0
     totalSize: Optional[int] = 0
 
     dateEarliest: Optional[datetime] = None
@@ -1339,6 +1343,7 @@ class PublicCollOut(BaseMongoModel):
 
     crawlCount: Optional[int] = 0
     pageCount: Optional[int] = 0
+    uniquePageCount: Optional[int] = 0
     totalSize: Optional[int] = 0
 
     dateEarliest: Optional[datetime] = None
@@ -1919,6 +1924,9 @@ class OrgMetrics(BaseModel):
     pageCount: int
     crawlPageCount: int
     uploadPageCount: int
+    uniquePageCount: int
+    crawlUniquePageCount: int
+    uploadUniquePageCount: int
     profileCount: int
     workflowsRunningCount: int
     maxConcurrentCrawls: int

backend/btrixcloud/operator/crawls.py

@@ -1534,7 +1534,7 @@ async def do_crawl_finished_tasks(
         )
 
         if state in SUCCESSFUL_STATES and crawl.oid:
-            await self.page_ops.set_archived_item_page_count(crawl.id)
+            await self.page_ops.set_archived_item_page_counts(crawl.id)
             await self.org_ops.inc_org_bytes_stored(
                 crawl.oid, status.filesAddedSize, "crawl"
             )

backend/btrixcloud/ops.py

@@ -97,7 +97,7 @@ def init_ops() -> Tuple[
 
     background_job_ops.set_ops(crawl_ops, profile_ops)
 
-    org_ops.set_ops(base_crawl_ops, profile_ops, coll_ops, background_job_ops)
+    org_ops.set_ops(base_crawl_ops, profile_ops, coll_ops, background_job_ops, page_ops)
 
     user_manager.set_ops(org_ops, crawl_config_ops, base_crawl_ops)