cloudquery-plugin-sdk 0.1.33__tar.gz → 0.1.34__tar.gz

{cloudquery_plugin_sdk-0.1.33/cloudquery_plugin_sdk.egg-info → cloudquery_plugin_sdk-0.1.34}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: cloudquery-plugin-sdk
-Version: 0.1.33
+Version: 0.1.34
 Summary: CloudQuery Plugin SDK for Python
 Home-page: https://github.com/cloudquery/plugin-sdk-python
 Author: CloudQuery LTD

{cloudquery_plugin_sdk-0.1.33 → cloudquery_plugin_sdk-0.1.34}/cloudquery/sdk/internal/memdb/memdb.py

@@ -5,6 +5,7 @@ from cloudquery.sdk import message
 from cloudquery.sdk import schema
 from typing import List, Generator, Dict
 import pyarrow as pa
+from cloudquery.sdk.schema.table import Table
 from cloudquery.sdk.types import JSONType
 from dataclasses import dataclass, field
 
@@ -109,5 +110,9 @@ class MemDB(plugin.Plugin):
             else:
                 raise NotImplementedError(f"Unknown message type {type(msg)}")
 
+    def read(self, table: Table) -> Generator[message.ReadMessage, None, None]:
+        for table, record in self._db.items():
+            yield message.ReadMessage(record)
+
     def close(self) -> None:
         self._db = {}

{cloudquery_plugin_sdk-0.1.33 → cloudquery_plugin_sdk-0.1.34}/cloudquery/sdk/internal/servers/plugin_v3/plugin.py

@@ -81,8 +81,14 @@ class PluginServicer(plugin_pb2_grpc.PluginServicer):
                 # unknown sync message type
                 raise NotImplementedError()
 
-    def Read(self, request, context):
-        raise NotImplementedError()
+    def Read(
+        self, request: plugin_pb2.Read.Request, context
+    ) -> Generator[plugin_pb2.Read.Response, None, None]:
+        schema = arrow.new_schema_from_bytes(request.table)
+        table = Table.from_arrow_schema(schema)
+        for msg in self._plugin.read(table):
+            buf = arrow.record_to_bytes(msg.record)
+            yield plugin_pb2.Read.Response(record=buf)
 
     def Write(
         self, request_iterator: Generator[plugin_pb2.Write.Request, None, None], context
@@ -93,7 +99,9 @@ class PluginServicer(plugin_pb2_grpc.PluginServicer):
                 if field == "migrate_table":
                     sc = arrow.new_schema_from_bytes(msg.migrate_table.table)
                     table = Table.from_arrow_schema(sc)
-                    yield WriteMigrateTableMessage(table=table)
+                    yield WriteMigrateTableMessage(
+                        table=table, migrate_force=msg.migrate_table.migrate_force
+                    )
                 elif field == "insert":
                     yield WriteInsertMessage(
                         record=arrow.new_record_from_bytes(msg.insert.record)

{cloudquery_plugin_sdk-0.1.33 → cloudquery_plugin_sdk-0.1.34}/cloudquery/sdk/message/__init__.py

@@ -5,3 +5,4 @@ from .write import (
     WriteMigrateTableMessage,
     WriteDeleteStale,
 )
+from .read import ReadMessage

cloudquery_plugin_sdk-0.1.34/cloudquery/sdk/message/read.py

@@ -0,0 +1,6 @@
+import pyarrow as pa
+
+
+class ReadMessage:
+    def __init__(self, record: pa.RecordBatch):
+        self.record = record

{cloudquery_plugin_sdk-0.1.33 → cloudquery_plugin_sdk-0.1.34}/cloudquery/sdk/message/write.py

@@ -12,8 +12,9 @@ class WriteInsertMessage(WriteMessage):
 
 
 class WriteMigrateTableMessage(WriteMessage):
-    def __init__(self, table: Table):
+    def __init__(self, table: Table, migrate_force: bool):
         self.table = table
+        self.migrate_force = migrate_force
 
 
 class WriteDeleteStale(WriteMessage):

{cloudquery_plugin_sdk-0.1.33 → cloudquery_plugin_sdk-0.1.34}/cloudquery/sdk/plugin/plugin.py

@@ -93,5 +93,8 @@ class Plugin:
     def write(self, writer: Generator[message.WriteMessage, None, None]) -> None:
         raise NotImplementedError()
 
+    def read(self, table: Table) -> Generator[message.ReadMessage, None, None]:
+        raise NotImplementedError()
+
     def close(self) -> None:
         raise NotImplementedError()

cloudquery_plugin_sdk-0.1.34/cloudquery/sdk/schema/__init__.py

@@ -0,0 +1,16 @@
+from .column import Column
+from .table import (
+    Table,
+    tables_to_arrow_schemas,
+    filter_dfs,
+    TableColumnChangeType,
+    TableColumnChange,
+    TableColumnChangeType,
+    get_table_changes,
+    get_table_column,
+    flatten_tables_recursive,
+    flatten_tables,
+)
+from .resource import Resource
+
+# from .table_resolver import TableReso

{cloudquery_plugin_sdk-0.1.33 → cloudquery_plugin_sdk-0.1.34}/cloudquery/sdk/schema/column.py

@@ -55,7 +55,9 @@ class Column:
                 arrow.METADATA_TRUE if self.incremental_key else arrow.METADATA_FALSE
             ),
         }
-        return pa.field(self.name, self.type, metadata=metadata)
+        return pa.field(
+            self.name, self.type, metadata=metadata, nullable=not self.not_null
+        )
 
     @staticmethod
     def from_arrow_field(field: pa.Field) -> Column:

{cloudquery_plugin_sdk-0.1.33 → cloudquery_plugin_sdk-0.1.34}/cloudquery/sdk/schema/table.py

@@ -1,8 +1,9 @@
 from __future__ import annotations
 
 import copy
+from enum import IntEnum
 import fnmatch
-from typing import List
+from typing import List, Optional
 
 import pyarrow as pa
 
@@ -10,6 +11,10 @@ from cloudquery.sdk.schema import arrow
 from .column import Column
 
 
+CQ_SYNC_TIME_COLUMN = "cq_sync_time"
+CQ_SOURCE_NAME_COLUMN = "cq_source_name"
+
+
 class Client:
     pass
 
@@ -192,9 +197,137 @@ def _filter_dfs_impl(t, parent_matched, include, exclude, skip_dependent_tables)
     return None
 
 
-def flatten_tables(tables: List[Table]) -> List[Table]:
-    flattened: List[Table] = []
+class TableColumnChangeType:
+    ADD = 1
+    REMOVE = 2
+    REMOVE_UNIQUE_CONSTRAINT = 3
+
+
+class TableColumnChange:
+    def __init__(
+        self,
+        type: TableColumnChangeType,
+        column_name: str,
+        current: Optional[Column],
+        previous: Optional[Column],
+    ):
+        self.type = type
+        self.column_name = column_name
+        self.current = current
+        self.previous = previous
+
+
+class TableColumnChangeType(IntEnum):
+    UNKNOWN = 0
+    ADD = 1
+    UPDATE = 2
+    REMOVE = 3
+    REMOVE_UNIQUE_CONSTRAINT = 4
+    MOVE_TO_CQ_ONLY = 5
+
+
+def get_table_changes(new: Table, old: Table) -> List[TableColumnChange]:
+    changes = []
+
+    # Special case: Moving from individual PKs to singular PK on _cq_id
+    new_pks = new.primary_keys
+    if (
+        len(new_pks) == 1
+        and new_pks[0] == "CqIDColumn"
+        and get_table_column(old, "CqIDColumn") is None
+        and len(old.primary_keys) > 0
+    ):
+        changes.append(
+            TableColumnChange(
+                type=TableColumnChangeType.MOVE_TO_CQ_ONLY,
+            )
+        )
+
+    for c in new.columns:
+        other_column = get_table_column(old, c.name)
+        # A column was added to the table definition
+        if other_column is None:
+            changes.append(
+                TableColumnChange(
+                    type=TableColumnChangeType.ADD,
+                    column_name=c.name,
+                    current=c,
+                    previous=None,
+                )
+            )
+            continue
+
+        # Column type or options (e.g. PK, Not Null) changed in the new table definition
+        if (
+            c.type != other_column.type
+            or c.not_null != other_column.not_null
+            or c.primary_key != other_column.primary_key
+        ):
+            changes.append(
+                TableColumnChange(
+                    type=TableColumnChangeType.UPDATE,
+                    column_name=c.name,
+                    current=c,
+                    previous=other_column,
+                )
+            )
+
+        # Unique constraint was removed
+        if not c.unique and other_column.unique:
+            changes.append(
+                TableColumnChange(
+                    type=TableColumnChangeType.REMOVE_UNIQUE_CONSTRAINT,
+                    column_name=c.name,
+                    current=c,
+                    previous=other_column,
+                )
+            )
+
+    # A column was removed from the table definition
+    for c in old.columns:
+        if get_table_column(new, c.name) is None:
+            changes.append(
+                TableColumnChange(
+                    type=TableColumnChangeType.REMOVE,
+                    column_name=c.name,
+                    current=None,
+                    previous=c,
+                )
+            )
+
+    return changes
+
+
+def get_table_column(table: Table, column_name: str) -> Optional[Column]:
+    for c in table.columns:
+        if c.name == column_name:
+            return c
+    return None
+
+
+def flatten_tables_recursive(original_tables: List[Table]) -> List[Table]:
+    tables = []
+    for table in original_tables:
+        table_copy = Table(
+            name=table.name,
+            columns=table.columns,
+            relations=table.relations,
+            title=table.title,
+            description=table.description,
+            is_incremental=table.is_incremental,
+            parent=table.parent,
+        )
+        tables.append(table_copy)
+        tables.extend(flatten_tables_recursive(table.relations))
+    return tables
+
+
+def flatten_tables(original_tables: List[Table]) -> List[Table]:
+    tables = flatten_tables_recursive(original_tables)
+    seen = set()
+    deduped = []
     for table in tables:
-        flattened.append(table)
-        flattened.extend(flatten_tables(table.relations))
-    return flattened
+        if table.name not in seen:
+            deduped.append(table)
+            seen.add(table.name)
+    return deduped

{cloudquery_plugin_sdk-0.1.33 → cloudquery_plugin_sdk-0.1.34}/cloudquery/sdk/types/json.py

@@ -21,3 +21,6 @@ class JSONType(pa.ExtensionType):
         # return an instance of this subclass given the serialized
         # metadata.
         return JSONType()
+
+
+pa.register_extension_type(JSONType())

{cloudquery_plugin_sdk-0.1.33 → cloudquery_plugin_sdk-0.1.34}/cloudquery/sdk/types/uuid.py

@@ -23,3 +23,6 @@ class UUIDType(pa.ExtensionType):
         # return an instance of this subclass given the serialized
         # metadata.
         return UUIDType()
+
+
+pa.register_extension_type(UUIDType())

{cloudquery_plugin_sdk-0.1.33 → cloudquery_plugin_sdk-0.1.34/cloudquery_plugin_sdk.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: cloudquery-plugin-sdk
-Version: 0.1.33
+Version: 0.1.34
 Summary: CloudQuery Plugin SDK for Python
 Home-page: https://github.com/cloudquery/plugin-sdk-python
 Author: CloudQuery LTD

{cloudquery_plugin_sdk-0.1.33 → cloudquery_plugin_sdk-0.1.34}/cloudquery_plugin_sdk.egg-info/SOURCES.txt

@@ -13,6 +13,7 @@ cloudquery/sdk/internal/servers/discovery_v1/discovery.py
 cloudquery/sdk/internal/servers/plugin_v3/__init__.py
 cloudquery/sdk/internal/servers/plugin_v3/plugin.py
 cloudquery/sdk/message/__init__.py
+cloudquery/sdk/message/read.py
 cloudquery/sdk/message/sync.py
 cloudquery/sdk/message/write.py
 cloudquery/sdk/plugin/__init__.py

{cloudquery_plugin_sdk-0.1.33 → cloudquery_plugin_sdk-0.1.34}/setup.py

@@ -51,7 +51,7 @@ packages = [
 ]
 setuptools.setup(
     name=name,
-    version="0.1.33",
+    version="0.1.34",
     description=description,
     long_description=long_description,
     author="CloudQuery LTD",

cloudquery_plugin_sdk-0.1.33/cloudquery/sdk/schema/__init__.py

@@ -1,5 +0,0 @@
-from .column import Column
-from .table import Table, tables_to_arrow_schemas, filter_dfs
-from .resource import Resource
-
-# from .table_resolver import TableReso