Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Failed to query Iceberg TIMESTAMP type

See original GitHub issue

名称	版本	描述
flink	1.3.2	开源版本
cdh	6.3.2	开源版本
hive	2.1.1-cdh6.3.2	包含cdh中
hadoop	3.0.0-cdh6.3.2	cdh原生版本
presto	2.591	开源版本
trino	360	开源版本
Iceberg	0.13	master分支编译

Flink create table

CREATE TABLE ic17 (
   tinyint0 INT
  , smallint1 SMALLINT
  , int2 INT
  , bigint3 BIGINT
  , float4 FLOAT
  , double5 DOUBLE  
  , decimal6 DECIMAL(12,3)
  , boolean7 BOOLEAN
  ,char8 CHAR(64) PRIMARY KEY NOT ENFORCED
  ,varchar9 VARCHAR(64)
  , string10 STRING
  , timestamp11 TIMESTAMP(3)
) 
PARTITIONED BY (tinyint0)
WITH (
     'connector'='iceberg'                    -- 主键和分区均可多个，分割
   , 'format-version' = '1'                   -- iceberg表版本，可选1，2
   , 'engine.hive.enabled' = 'true'           -- 启用hive同步
   , 'catalog-name'='hive_catalog'            -- 指定catalog
   , 'catalog-database'='iceberg'             -- 指定hive database
   , 'uri'='thrift://cdh2:9083'               -- hive hms地址，分割 
);

Flink insert

INSERT INTO ic17 VALUES(
      cast(1218 as INT)
    , cast(295 as SMALLINT)
    , cast(-210121792 as INT)
    , cast(-3697946268377828253 as BIGINT)
    , cast(1.123456789111111 as FLOAT)
    , cast(1111111.123411 as DOUBLE)
    , cast(1111.1234111 as DECIMAL(12, 3))
    , cast(123123123123 as BOOLEAN)
 , cast('`[s1tX213ysdasdasdgfq3wqwdqwqd速度速度pGPYl`AggMaHNRJv\[CkIYzcgMlmVvLSjtYmnlBEcwH^kEgDSxGIwGNLDP' as CHAR(64))
    , cast('daQOIE[n_eJsYLBJLttyFHnBXiCoT`RWeCO\G[JZZTdFFnFZFCODoI`X[SbMVAjq' as VARCHAR(64))
    , cast('e1916697-e626-4446-bd18-0142bfb9417b' as STRING)
    , cast('2021-09-13 03:08:50.810' as TIMESTAMP(3))
);

presto query
trino query
hive query

Issue Analytics

State:
Created 2 years ago
Comments:6 (4 by maintainers)

Top GitHub Comments

1reaction

ChunxuTangcommented, Jan 14, 2022

#17190

0reactions

ChunxuTangcommented, Jan 14, 2022

Hi @Sudeepam97, I’m looking at the issue and plan to send an initial PR to support the feature soon. Cc @beinan

Top Results From Across the Web

Querying Iceberg table data and performing time travel

To run a time travel query, use FOR TIMESTAMP AS OF timestamp after the table name in the SELECT statement, as in the...

"Iceberg query cannot be parsed" when trying to create ...

1 Answer 1 ... Neither map (without specifying the key/value types) nor varchar are valid Iceberg types. See the Iceberg documentation for valid ......

Spark Writes - Apache Iceberg

When writing with the v1 DataFrame API in Spark 3, use saveAsTable or insertInto to load tables with a catalog. Using format("iceberg") loads...

15: Iceberg right ahead! - Trino

... Write data from Trino and check data and snapshots */ INSERT INTO iceberg.logging.logs VALUES ( 'ERROR', timestamp '2021-04-01' AT TIME ...

Maintaining Iceberg Tables - Compaction, Expiring Snapshots ...

The Apache Iceberg table format creates many benefits when working ... When it comes to querying the data it would be more efficient...