You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@iceberg.apache.org by GitBox <gi...@apache.org> on 2022/01/22 01:05:58 UTC
[GitHub] [iceberg] jzhuge commented on a change in pull request #3188: Add common view format spec
jzhuge commented on a change in pull request #3188:
URL: https://github.com/apache/iceberg/pull/3188#discussion_r790081040
##########
File path: site/docs/view-spec.md
##########
@@ -0,0 +1,256 @@
+<!--
+ - Licensed to the Apache Software Foundation (ASF) under one or more
+ - contributor license agreements. See the NOTICE file distributed with
+ - this work for additional information regarding copyright ownership.
+ - The ASF licenses this file to You 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.
+ -->
+
+# Iceberg View Spec
+
+## Background and Motivation
+
+Most compute engines (e.g. Trino and Apache Spark) support logical views, commonly known as ‘views’. A view is a logical table that can be referenced by future queries. Views do not contain any data. Instead, the query stored by the view is executed every time the view is referenced by another query. Views and tables occupy the same namespace.
+Each compute engine stores the metadata of the view in its proprietary format in the metastore of choice. Thus, views created from one engine can not be read or altered easily from another engine even when engines share the metastore as well as the storage system. This document standardizes the view metadata for ease of sharing the views across engines.
+
+## Goals
+
+* A common metadata format for view metadata, similar to how Iceberg supports a common table format for tables.
+* The view metadata format specification
+ * Includes storage format as well as APIs to write/read the metadata.
+ * Supports versioning of views to track how a view evolved over time.
+
+## Overview
+
+The view metadata storage and retrieval mirrors how Iceberg table metadata is stored and retrieved. The view metadata is stored in a JSON file on object storage for ease of tracking the evolution of the view. Metastore continues to hold the view object with some properties such as database name, owner, create time, last access time and an indication that the object is a view.
+
+Each ‘CREATE OR REPLACE VIEW’ statement creates a new view version metadata file for that view.
+Each metadata file is self-sufficient. It contains the history of the last few operations performed on the view and can be used to roll back the view to a previous version.
+
+### Metadata Location
+
+The view metadata location is managed exactly like table metadata location.
+
+### Operations
+
+* Create a view
+* Drop the view
+* Load a view to read the metadata
+* Replace the view
+* Change the view definition
+* Add/delete/edit column comments
+
+## Specification
+
+### Terms
+
+* **Schema** -- Names and types of fields in a view.
+* **Version** -- The state of a view at some point in time.
+
+### View Metadata
+
+The view version metadata file has the following fields:
+
+| Required/Optional | Field Name | Description |
+|-------------------|------------|-------------|
+| Required | format-version | Json format version number for the view metadata spec. The view metadata spec and the corresponding format-version is independent of table spec. Starts with 1 and is incremented when there is a breaking change to view metadata. |
+| Required | object-type | Type of object this metadata file is for: "table" or "view". It must be set to "view" for all objects covered in this spec. |
+| Required | location | Location of the view metadata files |
+| Required | current-version-id | Current version of the view. Set to ‘1’ when the view is first created. |
+| Optional | properties | A string to string map of view properties. Contains pre-set properties such as ‘comment’ describing the view, does not contain arbitrary metadata. |
+| Required | versions | An array of structs describing the last few versions of the view. Controlled by the table property: “version.history.num_entries”. See more below. |
+| Required | version-log | An array of structs describing the log of created versions. See more below. |
+| Optional | schemas | A list of schemas, the same as the ‘schemas’ field from Iceberg table spec. |
+| Optional | current-schema-id | ID of the current schema of the view |
+
+
+‘Versions’ is an array of structs with fields as shown below:
+
+| Required/Optional | Field Name | Description |
+|-------------------|------------|-------------|
+| Required | version-id | Monotonically increasing id indicating the version of the view. Starts with “1”. |
+| Required | timestamp-ms | Timestamp expressed in ms since epoch at which the version of the view was created. |
+| Required | summary | A string to string map of view properties to track version metadata. This field can be used by engines to store any necessary properties. Two currently required properties are described below. |
+| Required | representations | A list of "representations" as described below. |
+
+
+Note that each version is stored in a separate AVRO file. This is to ensure that the metadata file stays readable in the case the view definition is huge. As a future extension, an engine-agnostic intermediate representation or a serialized abstract syntax tree of the SQL definition may also be stored in each version, exacerbating the problem.
+
+“summary” is a string-string map with the following required keys. Engines may store additional key-value pairs in this map.
+
+| Required/Optional | Key | Value |
+|-------------------|-----|-------|
+| Required | operation |A string value indicating the view operation that caused this metadata to be created. Allowed values are “CREATE” and “REPLACE” |
+| Optional | engine-version | A string value indicating the version of the engine that performed the operation (create / replace) |
+
+“representations” is a list of structs with fields as shown below:
+
+| Required/Optional | Field Name | Description |
+|-------------------|------------|-------------|
+| Required | type | A string indicating the type of representation. The only valid choice is "sql". The rest of the fields are interpreted by the type of representation. |
+
+Here are the fields for "sql" representation type:
+
+| Required/Optional | Field Name | Description |
+|-------------------|------------|-------------|
+| Required | type | It must be set to "sql" |
+| Required | sql | A string representing SQL definition of the view as input |
+| Required | dialect | A string specifying the dialect of the ‘sql’ field above. Used by engines to perform necessary translations to the SQL dialect supported by the engine. |
+| Optional | session-catalog | A string that specifies the catalog of the user session when the view was created / replaced. Used to resolve the tables in the view definition. |
+| Optional | session-namespace | An array of strings indicating namespace at the time view was created / replaced. Used similar to ‘session-catalog’ above. |
+| Optional | field-aliases | A list of strings of field aliases E.g. a list of alias_name info specified in the following create view statement. `CREATE VIEW v (alias_name COMMENT 'docs', alias_name2, ...) AS SELECT ...` |
+| Optional | field-docs | A list of strings of field comments E.g. a list of ‘comment’ info specified in the following create view statement. `CREATE VIEW v (alias_name COMMENT 'docs', alias_name2, ...) AS SELECT ...` |
+
+“version-log” is an array of structs describing the log of the versions created. The struct has the following fields:
+
+| Required/Optional | Field Name | Description |
+|-------------------|------------|-------------|
+| Required | timestamp-ms | Timestamp expressed as ms since epoch for when the version of the view indicated in version-id field below was created |
+| Required | version-id | Version id of the view |
+
+## Appendix A: An Example
+
+The JSON metadata file format is described using an example below.
+
+Imagine the following sequence of operations:
+
+* `CREATE TABLE base_tab(c1 int, c2 varchar);`
+* `INSERT INTO base_tab VALUES (1,’one’), (2,’two’);`
+* `CREATE VIEW common_view AS SELECT * FROM base_tab;`
+* `CREATE OR REPLACE VIEW common_view AS SELECT count(*) AS my_cnt FROM base_tab;`
+
+The metadata json file created at the end of step 3 looks as follows. The file path looks like:
+`s3://my_company/my/warehouse/anorwood.db/common_view`
+
+The path is intentionally similar to the path for iceberg tables and contains a ‘metadata’ directory. (`METASTORE_WAREHOUSE_DIR/<dbname>.db/<viewname>/metadata`)
+
+The metadata directory contains View Version Metadata files. The text after '=>' symbols describes the fields.
+```
+{
+ "format-version" : 1, =>Json format. Will change as format evolves.
+ "location" : "s3n://my_company/my/warehouse/anorwood.db/common_view",
+ "current-version-id" : 1, => current / latest version of the view. ‘1’ here since this metadata was created when the view was created.
+ "properties" : { => shows properties of the view
+ "comment" : "View captures all the data from the table" => View comment
+ },
+ "versions" : [ { => Last few versions of the view. Controlled by the table property: “version.history.num_entries”
+ "version-id" : 1,
+ "parent-version-id" : -1,
+ "timestamp-ms" : 1573518431292,
+ "summary" : {
+ "operation" : "create", => View operation that caused this metadata to be created
+ "engineVersion" : "presto-350",=> Version of the engine that performed the operation (create / replace)
+ },
+ "representations" : [ { => SQL metadata of the view
+ "type" : "sql",
+ "sql" : "SELECT *\nFROM\n base_tab\n", => original view SQL
+ "schema" : { => Schema of the view expressed in Iceberg types
+ "type" : "struct",
+ "fields" : [ {
+ "id" : 0,
+ "name" : "c1",
+ "required" : false,
+ "type" : "int",
+ "doc" : "" => Column comment
+ }, {
+ "id" : 1,
+ "name" : "c2",
+ "required" : false,
+ "type" : "string",
+ "doc" : ""
+ } ]
+ },
+ "sessionCatalog" : "iceberg",=> The catalog of the user session when the view was created / replaced. Used to resolve the tables used in the view definition.
+ "sessionNamespace" : [ "anorwood" ] => Namespace at the time view was created / replaced. Used similar to ‘sessionCatalog’ above.
+ } ],
+ } ],
+ "version-log" : [ { => Log of the created versions
+ "timestamp-ms" : 1573518431292,
+ "version-id" : 1
+ } ]
+}
+```
+
+The Iceberg / view library creates a new metadata json file every time the view undergoes a DDL change. This way the history of how the view evolved can be maintained. Following metadata json file was created at the end of Step 4.
+
+```
+{
+ "format-version" : 1,
+ "location" : "s3n://my_company/my/warehouse/anorwood.db/common_view",
+ "current-version-id" : 2,
+ "properties" : { => shows properties of the view
+ "comment" : "View captures count of the data from the table"
+ },
+ "versions" : [ {
+ "version-id" : 1,
+ "parent-version-id" : -1,
+ "timestamp-ms" : 1573518431292,
+ "summary" : {
+ "operation" : "create",
+ "engineVersion" : "presto-350",
+ },
+ "representations" : [ {
+ "type" : "sql",
+ "sql" : "SELECT *\nFROM\n base_tab\n",
+ "schema" : {
+ "type" : "struct",
+ "fields" : [ {
+ "id" : 0,
+ "name" : "c1",
+ "required" : false,
+ "type" : "int",
+ "doc" : ""
+ }, {
+ "id" : 1,
+ "name" : "c2",
+ "required" : false,
+ "type" : "string",
+ "doc" : ""
+ } ]
+ },
Review comment:
Not planned. What use case do you have in mind?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@iceberg.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@iceberg.apache.org
For additional commands, e-mail: issues-help@iceberg.apache.org