refscan 0.2.3__tar.gz → 0.2.4__tar.gz

{refscan-0.2.3 → refscan-0.2.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: refscan
-Version: 0.2.3
+Version: 0.2.4
 Summary: Command-line program that scans the NMDC MongoDB database for referential integrity violations
 Keywords: mongodb,mongo,relationships,reference,database,data,referential integrity,scan
 Requires-Python: >=3.9,<4.0

{refscan-0.2.3 → refscan-0.2.4}/pyproject.toml

@@ -2,7 +2,7 @@
 
 [tool.poetry]
 name = "refscan"
-version = "v0.2.3"
+version = "v0.2.4"
 description = "Command-line program that scans the NMDC MongoDB database for referential integrity violations"
 authors = []
 readme = "README.md"

{refscan-0.2.3 → refscan-0.2.4}/refscan/lib/Finder.py

@@ -1,6 +1,7 @@
 from typing import List, Optional, Dict
 
 from pymongo.database import Database
+from pymongo.client_session import ClientSession
 
 
 class Finder:
@@ -10,6 +11,9 @@ class Finder:
     """
 
     def __init__(self, database: Database):
+        r"""
+        Initializes an instance of the `Finder` class. The instance is bound to the specified MongoDB database.
+        """
         self.db = database
 
         # Initialize our cache of names of collections we most recently found referenced document `id`s in.
@@ -114,15 +118,20 @@ class Finder:
             return self.cached_id_presence_by_collection[collection_name][document_id]
 
     def check_whether_document_having_id_exists_among_collections(
-        self, document_id: str, collection_names: List[str]
+        self, document_id: str, collection_names: List[str], client_session: ClientSession = None
     ) -> Optional[str]:
         r"""
         Checks whether any document in any of the specified collections has the specified value in its `id` field.
         Returns the name of the first collection, if any, containing such a document. If none of the collections
         contain such a document, the function returns `None`.
 
+        If a `pymongo.client_session.ClientSession` is specified, this function will access the database within
+        the context of that session. If a transaction is pending on that session, this feature can be used
+        to check whether the specified document _would_ exist if that transaction were to be committed.
+
         References:
         - https://pymongo.readthedocs.io/en/stable/api/pymongo/collection.html#pymongo.collection.Collection.find_one
+        - https://pymongo.readthedocs.io/en/stable/api/pymongo/client_session.html#transactions
         """
 
         # Initialize the name of the containing collection to `None`.
@@ -155,7 +164,8 @@ class Finder:
             # If we find it in the collection in the database, record the collection name, update both (a) the cache
             # of which documents we have found in which collections, and (b) the cache of names of collections where
             # a referenced document was recently found, and skip searching additional collections.
-            if self.db.get_collection(collection_name).find_one(query_filter, projection=["_id"]) is not None:
+            collection = self.db.get_collection(collection_name)
+            if collection.find_one(query_filter, projection=["_id"], session=client_session) is not None:
                 name_of_collection_containing_target_document = collection_name
 
                 # Record that a document having this `id` is _present_ in this collection.

{refscan-0.2.3 → refscan-0.2.4}/refscan/scanner.py

@@ -1,6 +1,7 @@
 from typing import List, Optional, Dict
 
 from pymongo.database import Database
+from pymongo.client_session import ClientSession
 from rich.console import Console
 from linkml_runtime import SchemaView
 
@@ -24,6 +25,7 @@ def scan_outgoing_references(
     finder: Finder,
     collection_names: List[str],
     source_collection_name: str,
+    client_session: ClientSession = None,
     user_wants_to_locate_misplaced_documents: bool = False,
 ) -> ViolationList:
     r"""
@@ -40,6 +42,10 @@ def scan_outgoing_references(
                                                      all the legal collections
     :param source_collection_name: Name of collection in which source document resides (this is included in the
                                    violation report in an attempt to facilitate investigation of violations)
+    :param client_session: A `pymongo.client_session.ClientSession` instance that, if specified, will be used when
+                           searching for referenced documents. If a transaction happens to be pending on that session,
+                           the scan will effectively happen on the database as it _would_ exist if that transaction
+                           were to be committed.
     :param collection_names: List of collection names that are described by the schema
     """
 
@@ -77,7 +83,7 @@ def scan_outgoing_references(
             for target_id in target_ids:
                 name_of_collection_containing_target_document = (
                     finder.check_whether_document_having_id_exists_among_collections(
-                        collection_names=target_collection_names, document_id=target_id
+                        collection_names=target_collection_names, document_id=target_id, client_session=client_session
                     )
                 )
                 if name_of_collection_containing_target_document is None:
@@ -88,7 +94,9 @@ def scan_outgoing_references(
                         names_of_ineligible_collections = list(set(collection_names) - set(target_collection_names))
                         name_of_collection_containing_target_document = (
                             finder.check_whether_document_having_id_exists_among_collections(
-                                collection_names=names_of_ineligible_collections, document_id=target_id
+                                collection_names=names_of_ineligible_collections,
+                                document_id=target_id,
+                                client_session=client_session,
                             )
                         )