erioon 0.1.5__py3-none-any.whl → 0.1.6__py3-none-any.whl

erioon/database.py

@@ -0,0 +1,82 @@
+# Copyright 2025-present Erioon, Inc.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# Visit www.erioon.com/dev-docs for more information about the python SDK
+
+import json
+from erioon.collection import Collection
+
+class Database:
+    def __init__(self, user_id, metadata, database=None, cluster=None, sas_url=None):
+        """
+        Initialize a Database instance.
+
+        Args:
+            user_id (str): The ID of the authenticated user.
+            metadata (dict): Metadata containing information about the database and its collections.
+            database (str, optional): The name or identifier of the database.
+            cluster (str, optional): The cluster where the database is hosted.
+            sas_url (str, optional): SAS URL for accessing Azure Blob Storage container.
+        """
+        self.user_id = user_id
+        self.metadata = metadata
+        self.db_id = metadata.get("database_info", {}).get("_id")
+        self.database = database
+        self.cluster = cluster
+        self.sas_url = sas_url
+
+    def __getitem__(self, collection_id):
+        """
+        Enables dictionary-like access to collections within the database.
+
+        Args:
+            collection_id (str): Identifier of the collection to retrieve.
+
+        Returns:
+            Collection: An instance of the Collection class initialized with metadata.
+            str: Error message if the collection is not found.
+        """
+        collections = self.metadata.get("database_info", {}).get("collections", {})
+        coll_meta = collections.get(collection_id)
+
+        if not coll_meta:
+            return "No collection found"
+
+        return Collection(
+            user_id=self.user_id,
+            db_id=self.db_id,
+            coll_id=collection_id,
+            metadata=coll_meta,
+            database=self.database,
+            cluster=self.cluster,
+            sas_url=self.sas_url
+        )
+
+    def __str__(self):
+        """
+        Returns a nicely formatted JSON string of the database metadata.
+        Useful for debugging and inspecting the database info.
+
+        Returns:
+            str: Pretty-printed JSON metadata.
+        """
+        return json.dumps(self.metadata, indent=4)
+
+    def __repr__(self):
+        """
+        Returns a concise, informative string representation of the Database instance.
+
+        Returns:
+            str: Formatted string showing the database ID, cluster, and database name.
+        """
+        return f"<Database db_id={self.db_id}, cluster={self.cluster}, database={self.database}>"

erioon/delete.py

@@ -0,0 +1,342 @@
+# Copyright 2025-present Erioon, Inc.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# Visit www.erioon.com/dev-docs for more information about the python SDK
+
+import json
+import io
+import msgpack
+from azure.storage.blob import ContainerClient
+from erioon.functions import update_index_file_delete, check_nested_key, async_log
+
+# DELETE ONE RECORD
+def handle_delete_one(user_id, db_id, coll_id, data_to_delete, container_url):
+    """
+    Delete a single record from a collection.
+
+    The record can be identified either by the unique '_id' field or by a nested key-value pair.
+
+    Args:
+        user_id: Identifier of the user performing the operation.
+        db_id: Database ID containing the collection.
+        coll_id: Collection ID.
+        data_to_delete: Dictionary containing either '_id' or key-value pair to match.
+        container_url: SAS URL pointing to the storage container.
+
+    Returns:
+        A tuple (response dict, status code) indicating success or failure.
+    """
+    if "_id" in data_to_delete:
+        record_id = data_to_delete["_id"]
+        return handle_delete_with_id(user_id, db_id, coll_id, record_id, container_url)
+    else:
+        return handle_delete_without_id(user_id, db_id, coll_id, data_to_delete, container_url)
+
+# DELETE RECORD USING ID FILTER
+def handle_delete_with_id(user_id, db_id, coll_id, record_id, container_url):
+    """
+    Delete a record exactly matching the given '_id'.
+
+    Steps:
+    - Parse container URL and create a ContainerClient.
+    - Load the index.json file which maps shards to record IDs.
+    - Locate the shard containing the target record_id.
+    - Download and unpack the shard blob.
+    - Remove the record from the shard data.
+    - Repack and upload the updated shard if record found.
+    - Update index.json to reflect deletion.
+    - Log success or errors asynchronously.
+
+    Args:
+        user_id, db_id, coll_id: Identifiers for user, database, and collection.
+        record_id: The unique '_id' of the record to delete.
+        container_url: Azure Blob Storage container SAS URL.
+
+    Returns:
+        Tuple (response dict, status code) indicating operation result.
+    """
+    
+    container_client = ContainerClient.from_container_url(container_url)
+
+    index_blob_client = container_client.get_blob_client(f"{db_id}/{coll_id}/index.json")
+
+    if not index_blob_client.exists():
+        return {"error": "Index file does not exist"}, 404
+
+    index_data = json.loads(index_blob_client.download_blob().readall())
+    shard_number = None
+
+    for shard in index_data:
+        for shard_key, ids in shard.items():
+            if record_id in ids:
+                shard_number = int(shard_key.split("_")[-1])
+                break
+        if shard_number:
+            break
+
+    if shard_number is None:
+        async_log(user_id, db_id, coll_id, "DELETE", "ERROR", f"Record with _id {record_id} not found", 1, container_url)
+        return {"error": f"Record with _id {record_id} not found"}, 404
+
+    msgpack_blob_client = container_client.get_blob_client(f"{db_id}/{coll_id}/{coll_id}_{shard_number}.msgpack")
+
+    try:
+        msgpack_data = msgpack_blob_client.download_blob().readall()
+        with io.BytesIO(msgpack_data) as buffer:
+            records = []
+            original_length = 0
+
+            unpacked_data = msgpack.unpackb(buffer.read(), raw=False)
+            if isinstance(unpacked_data, list):
+                for record in unpacked_data:
+                    original_length += 1
+                    if record.get("_id") == record_id:
+                        continue
+                    records.append(record)
+
+            if len(records) < original_length:
+                with io.BytesIO() as out_file:
+                    packed_data = msgpack.packb(records)
+                    out_file.write(packed_data)
+                    out_file.seek(0)
+                    msgpack_blob_client.upload_blob(out_file, overwrite=True)
+
+                update_index_file_delete(user_id, db_id, coll_id, record_id, shard_number, container_url)
+                async_log(user_id, db_id, coll_id, "DELETE", "SUCCESS", f"Record with _id {record_id} deleted successfully", 1, container_url)
+                return {"success": f"Record with _id {record_id} deleted successfully"}, 200
+            else:
+                async_log(user_id, db_id, coll_id, "DELETE", "ERROR", f"Record with _id {record_id} not found in shard", 1, container_url)
+                return {"error": f"Record with _id {record_id} not found in shard"}, 404
+
+    except Exception as e:
+        async_log(user_id, db_id, coll_id, "DELETE", "ERROR", f"Error deleting record {record_id}: {str(e)}", 1, container_url)
+        return {"error": f"Error deleting record {record_id}: {str(e)}"}, 500
+
+# DELETE RECORD USING KEY FILTER
+def handle_delete_without_id(user_id, db_id, coll_id, data_to_delete, container_url):
+    """
+    Delete a single record matching a nested key-value pair when '_id' is not provided.
+    Behaves like MongoDB's delete_one: deletes only the first matched record.
+    """
+    container_client = ContainerClient.from_container_url(container_url)
+
+    nested_key = list(data_to_delete.keys())[0]
+    key, value = nested_key, data_to_delete[nested_key]
+
+    directory_path = f"{db_id}/{coll_id}/"
+    blob_list = container_client.list_blobs(name_starts_with=directory_path)
+
+    for blob in blob_list:
+        if blob.name.endswith(".msgpack"):
+            try:
+                blob_client = container_client.get_blob_client(blob.name)
+                msgpack_data = blob_client.download_blob().readall()
+
+                with io.BytesIO(msgpack_data) as buffer:
+                    unpacked_data = msgpack.unpackb(buffer.read(), raw=False)
+                    if isinstance(unpacked_data, list):
+                        for record in unpacked_data:
+                            if check_nested_key(record, key, value):
+                                delete_response, status = handle_delete_with_id(user_id, db_id, coll_id, record["_id"], container_url)
+                                if status == 200:
+                                    return {"success": f"Record with _id {record['_id']} deleted successfully"}, 200
+                                else:
+                                    return delete_response, status
+            except Exception as e:
+                continue
+
+    async_log(user_id, db_id, coll_id, "DELETE", "ERROR", f"No matching record found for key-value pair {key}:{value}", 1, container_url)
+    return {"error": f"No matching record found for the specified key-value pair {key}:{value}"}, 404
+
+# DELETE MULTIPLE RECORDS
+def handle_delete_many(user_id, db_id, coll_id, data_to_delete_list, container_url, batch_size=10):
+    """
+    Delete multiple records from a collection.
+
+    Supports a mix of deletions by '_id' and by key-value pair filters.
+    Processes deletions in batches for performance and error isolation.
+
+    Args:
+        user_id: Identifier of the user making the request.
+        db_id: The database identifier.
+        coll_id: The collection identifier.
+        data_to_delete_list: List of dictionaries representing deletion filters (must contain either '_id' or a key-value pair).
+        container_url: SAS URL of the blob storage container.
+        batch_size: Number of deletions to process per batch.
+
+    Returns:
+        Tuple (response dict, status code). The response includes a summary of successes and failures.
+    """
+    
+    batch_results = []
+
+    for i in range(0, len(data_to_delete_list), batch_size):
+        batch = data_to_delete_list[i : i + batch_size]
+
+        ids_to_delete = [d["_id"] for d in batch if "_id" in d]
+        non_id_queries = [d for d in batch if "_id" not in d]
+
+        batch_success = []
+        batch_errors = []
+
+        if ids_to_delete:
+            results = handle_delete_many_with_id(user_id, db_id, coll_id, ids_to_delete, container_url)
+            for data_to_delete, (response, status_code) in zip([{"_id": rid} for rid in ids_to_delete], results):
+                if 200 <= status_code < 300:
+                    batch_success.append({
+                        "delete_query": data_to_delete,
+                        "message": response.get("success", "Record deleted successfully"),
+                    })
+                else:
+                    batch_errors.append({
+                        "delete_query": data_to_delete,
+                        "error": response.get("error", f"Failed to delete record - Status code {status_code}"),
+                    })
+
+        if non_id_queries:
+            deleted_results, errors = handle_delete_many_without_id(user_id, db_id, coll_id, non_id_queries, container_url)
+
+            for res in deleted_results:
+                batch_success.append({
+                    "delete_query": res["query"],
+                    "message": "Records deleted successfully",
+                })
+
+            for err in errors:
+                batch_errors.append({
+                    "delete_query": err["query"],
+                    "error": err.get("error", "Unknown error"),
+                })
+
+        batch_results.append({
+            "queries": len(batch),
+            "success": batch_success,
+            "errors": batch_errors,
+        })
+
+    total_success = sum(len(batch["success"]) for batch in batch_results)
+    total_errors = sum(len(batch["errors"]) for batch in batch_results)
+
+    if total_errors == 0:
+        return {
+            "success": f"Selected records deleted successfully",
+            "details": batch_results,
+            "total_deleted": total_success,
+        }, 200
+    elif total_success > 0:
+        return {
+            "warning": "Partial success deleting selected records",
+            "details": batch_results,
+            "total_deleted": total_success,
+            "total_errors": total_errors,
+        }, 207
+    else:
+        return {
+            "error": "Error deleting selected records",
+            "details": batch_results,
+        }, 500
+
+# DELETE MULTIPLE RECORDS WITH ID FILTER
+def handle_delete_many_with_id(user_id, db_id, coll_id, record_ids, container_url):
+    """
+    Delete multiple records by their '_id' values.
+
+    Args:
+        user_id: User identifier.
+        db_id: Database identifier.
+        coll_id: Collection identifier.
+        record_ids: List of '_id' values of records to delete.
+        container_url: Azure Blob Storage container SAS URL.
+
+    Returns:
+        List of tuples (response dict, status code) for each deletion attempt.
+    """
+    
+    results = []
+    for record_id in record_ids:
+        resp, status = handle_delete_with_id(user_id, db_id, coll_id, record_id, container_url)
+        results.append((resp, status))
+    return results
+
+# DELETE MULTIPLE RECORDS WITH KEY FILTER
+def handle_delete_many_without_id(user_id, db_id, coll_id, queries, container_url):
+    """
+    Delete multiple records that match key-value queries across all shards.
+
+    For each query in the list, it finds all records matching the key-value condition
+    and deletes them.
+
+    Args:
+        user_id: ID of the user performing the operation.
+        db_id: Database ID containing the collection.
+        coll_id: Collection ID.
+        queries: List of dictionaries containing key-value match conditions.
+        container_url: SAS URL pointing to the blob storage container.
+
+    Returns:
+        Tuple of:
+            - deleted_results (list): List of queries that resulted in deletion.
+            - errors (list): List of dictionaries containing error details for failed deletions.
+    """
+    
+    container_client = ContainerClient.from_container_url(container_url)
+    deleted_results = []
+    errors = []
+
+    directory_path = f"{db_id}/{coll_id}/"
+    blob_list = list(container_client.list_blobs(name_starts_with=directory_path))
+
+    for query in queries:
+        key = list(query.keys())[0]
+        value = query[key]
+        deleted_any = False
+        for blob in blob_list:
+            if blob.name.endswith(".msgpack"):
+                try:
+                    blob_client = container_client.get_blob_client(blob.name)
+                    msgpack_data = blob_client.download_blob().readall()
+
+                    with io.BytesIO(msgpack_data) as buffer:
+                        records = msgpack.unpackb(buffer.read(), raw=False)
+                        if not isinstance(records, list):
+                            continue
+                        new_records = []
+                        deleted_in_blob = False
+                        for record in records:
+                            if check_nested_key(record, key, value):
+                                deleted_in_blob = True
+                                deleted_any = True
+                            else:
+                                new_records.append(record)
+
+                        if deleted_in_blob:
+                            with io.BytesIO() as out_file:
+                                out_file.write(msgpack.packb(new_records))
+                                out_file.seek(0)
+                                blob_client.upload_blob(out_file, overwrite=True)
+
+                            deleted_ids = [r["_id"] for r in records if check_nested_key(r, key, value)]
+                            shard_number = int(blob.name.split("_")[-1].split(".")[0])
+                            for rid in deleted_ids:
+                                update_index_file_delete(user_id, db_id, coll_id, rid, shard_number, container_url)
+                                async_log(user_id, db_id, coll_id, "DELETE", "SUCCESS", f"Record with _id {rid} deleted successfully", 1, container_url)
+
+                except Exception as e:
+                    errors.append({"query": query, "error": str(e)})
+
+        if deleted_any:
+            deleted_results.append({"query": query, "status": "deleted"})
+        else:
+            errors.append({"query": query, "error": "No matching records found"})
+
+    return deleted_results, errors