[
https://issues.apache.org/jira/browse/CARBONDATA-2626?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
xuchuanyin updated CARBONDATA-2626:
-----------------------------------
Description:
To reproduce:
Create table and load data to it. The table schema is:
```
sql(
s"""
|CREATE TABLE $bloomDMSampleTable(id INT, name STRING, city STRING, age INT,|
|s1 STRING, s2 STRING, s3 STRING, s4 STRING, s5 STRING, s6 STRING, s7 STRING, s8 STRING)|
|STORED BY 'carbondata' TBLPROPERTIES('table_blocksize'='128', 'dictionary_include'='id', 'sort_columns'='name')|
|""".stripMargin)|
```
In tablespec, the first dimension spec is 'name' which is sort_columns but not dictionary. But Carbondata now treat it as dictionary page. (See TablePage Line94)
Then for 'id' which is dictionary, carbondata treat it as no-dictionary page (see TablePage Line101)
was:
To reproduce:
Create table and load data to it. The table schema is:
```
sql(
s"""
| CREATE TABLE $bloomDMSampleTable(id INT, name STRING, city STRING, age INT,
| s1 STRING, s2 STRING, s3 STRING, s4 STRING, s5 STRING, s6 STRING, s7 STRING, s8 STRING)
| STORED BY 'carbondata' TBLPROPERTIES('table_blocksize'='128', 'dictionary_include'='id', 'sort_columns'='name')
| """.stripMargin)
```
In tablespec, the first dimension spec is 'name' which is sort_columns but not dictionary. But Carbondata now treat it as dictionary page. (See TablePage Line94)
> Logic for dictionary/nodictionary column pages in TablePage is wrong
> --------------------------------------------------------------------
>
> Key: CARBONDATA-2626
> URL:
https://issues.apache.org/jira/browse/CARBONDATA-2626> Project: CarbonData
> Issue Type: Bug
> Reporter: xuchuanyin
> Priority: Major
>
> To reproduce:
> Create table and load data to it. The table schema is:
> ```
> sql(
> s"""
> |CREATE TABLE $bloomDMSampleTable(id INT, name STRING, city STRING, age INT,|
> |s1 STRING, s2 STRING, s3 STRING, s4 STRING, s5 STRING, s6 STRING, s7 STRING, s8 STRING)|
> |STORED BY 'carbondata' TBLPROPERTIES('table_blocksize'='128', 'dictionary_include'='id', 'sort_columns'='name')|
> |""".stripMargin)|
> ```
> In tablespec, the first dimension spec is 'name' which is sort_columns but not dictionary. But Carbondata now treat it as dictionary page. (See TablePage Line94)
> Then for 'id' which is dictionary, carbondata treat it as no-dictionary page (see TablePage Line101)
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)