Increase urn length to 500 in DDL (#804)

This commit is contained in:
Yi (Alan) Wang 2017-10-18 14:57:58 -07:00 committed by GitHub
parent e2616beddb
commit ae6937af41
3 changed files with 14 additions and 10 deletions

View File

@ -107,7 +107,7 @@ CREATE TABLE dataset_partition (
CREATE TABLE `dataset_compliance` (
`dataset_id` INT(10) UNSIGNED NOT NULL,
`dataset_urn` VARCHAR(200) NOT NULL,
`dataset_urn` VARCHAR(500) NOT NULL,
`compliance_purge_type` VARCHAR(30) DEFAULT NULL
COMMENT 'AUTO_PURGE,CUSTOM_PURGE,LIMITED_RETENTION,PURGE_NOT_APPLICABLE',
`compliance_entities` MEDIUMTEXT DEFAULT NULL
@ -132,7 +132,7 @@ CREATE TABLE `dataset_compliance` (
UNIQUE KEY `dataset_urn` (`dataset_urn`)
)
ENGINE = InnoDB
DEFAULT CHARSET = utf8;
DEFAULT CHARSET = latin1;
CREATE TABLE dataset_constraint (
`dataset_id` INT UNSIGNED NOT NULL,

View File

@ -23,10 +23,10 @@ CREATE TABLE `stg_dict_dataset` (
`properties` TEXT CHARACTER SET utf8,
`fields` MEDIUMTEXT CHARACTER SET utf8,
`db_id` SMALLINT UNSIGNED,
`urn` VARCHAR(200) NOT NULL,
`urn` VARCHAR(500) NOT NULL,
`source` VARCHAR(50) NULL,
`location_prefix` VARCHAR(200) NULL,
`parent_name` VARCHAR(200) NULL COMMENT 'Schema Name for RDBMS, Group Name for Jobs/Projects/Tracking Datasets on HDFS',
`parent_name` VARCHAR(500) NULL COMMENT 'Schema Name for RDBMS, Group Name for Jobs/Projects/Tracking Datasets on HDFS',
`storage_type` ENUM('Table', 'View', 'Avro', 'ORC', 'RC', 'Sequence', 'Flat File', 'JSON', 'BINARY_JSON', 'XML', 'Thrift', 'Parquet', 'Protobuff') NULL,
`ref_dataset_name` VARCHAR(200) NULL,
`ref_dataset_id` INT(11) UNSIGNED NULL COMMENT 'Refer to Master/Main dataset for Views/ExternalTables',
@ -61,11 +61,11 @@ CREATE TABLE `dict_dataset` (
COMMENT 'JSON, Hive, DDL, XML, CSV',
`properties` TEXT CHARACTER SET utf8,
`fields` MEDIUMTEXT CHARACTER SET utf8,
`urn` VARCHAR(200) CHARACTER SET utf8 NOT NULL,
`urn` VARCHAR(500) CHARACTER SET utf8 NOT NULL,
`source` VARCHAR(50) NULL
COMMENT 'The original data source type (for dataset in data warehouse). Oracle, Kafka ...',
`location_prefix` VARCHAR(200) NULL,
`parent_name` VARCHAR(200) NULL
`parent_name` VARCHAR(500) NULL
COMMENT 'Schema Name for RDBMS, Group Name for Jobs/Projects/Tracking Datasets on HDFS ',
`storage_type` ENUM('Table', 'View', 'Avro', 'ORC', 'RC', 'Sequence', 'Flat File', 'JSON', 'BINARY_JSON', 'XML', 'Thrift', 'Parquet', 'Protobuff') NULL,
`ref_dataset_id` INT(11) UNSIGNED NULL

View File

@ -15,7 +15,7 @@
CREATE TABLE dataset_owner (
`dataset_id` INT UNSIGNED NOT NULL,
`dataset_urn` VARCHAR(200) NOT NULL,
`dataset_urn` VARCHAR(500) NOT NULL,
`owner_id` VARCHAR(127) NOT NULL,
`app_id` SMALLINT NOT NULL COMMENT 'application id of the namespace',
`namespace` VARCHAR(127) COMMENT 'the namespace of the user',
@ -36,11 +36,13 @@ CREATE TABLE dataset_owner (
wh_etl_exec_id BIGINT COMMENT 'wherehows etl execution id that modified this record',
PRIMARY KEY (`dataset_id`, `owner_id`, `app_id`, `owner_source`),
UNIQUE KEY `with_urn` (`dataset_urn`, `owner_id`, `app_id`, `owner_source`)
);
)
ENGINE = InnoDB
DEFAULT CHARSET = latin1;
CREATE TABLE stg_dataset_owner (
`dataset_id` INT COMMENT 'dataset_id',
`dataset_urn` VARCHAR(200) NOT NULL,
`dataset_urn` VARCHAR(500) NOT NULL,
`owner_id` VARCHAR(127) NOT NULL,
`sort_id` SMALLINT COMMENT '0 = primary owner, order by priority/importance',
`app_id` INT COMMENT 'application id of the namesapce',
@ -58,7 +60,9 @@ CREATE TABLE stg_dataset_owner (
KEY (dataset_urn, owner_id, namespace, db_name),
KEY dataset_index (dataset_urn),
KEY db_name_index (db_name)
);
)
ENGINE = InnoDB
DEFAULT CHARSET = latin1;
CREATE TABLE stg_dataset_owner_unmatched (
`dataset_urn` VARCHAR(200) NOT NULL,