hadoop-hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Namit Jain (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HIVE-558) describe extended table/partition output is cryptic
Date Mon, 23 Aug 2010 22:53:30 GMT

    [ https://issues.apache.org/jira/browse/HIVE-558?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12901649#action_12901649
] 

Namit Jain commented on HIVE-558:
---------------------------------

What I was thinking was like this:


Change:

  public String getSchema() {
    return schema;
  }


in DescTableDesc.java to a static funtion:

  public static String getSchema() {
    return schema;
  }





    public static String getAllColumnsInformation(Table tbl) {
        StringBuilder columnInformation = new StringBuilder(1024);
        formatOutput("Column", "Type", "Comments", columnInformation);
        formatOutput("------", "----", "--------", columnInformation);
        getColumnInfoAsString(columnInformation, tbl);

        // Partitions
        if (tbl.isPartitioned()) {
            columnInformation.append(LINE_DELIM);
            formatOutput("Partition", "Type", "Comments", columnInformation);
            formatOutput("---------", "----", "--------", columnInformation);
            getPartitionInfoAsString(columnInformation, tbl);
        }
        return columnInformation.toString();
    }



can be::



    public static String getAllColumnsInformation(Table tbl) {
        StringBuilder columnInformation = new StringBuilder(1024);
        String schema = DescTableDesc.getSchema();
        String[] colsArr = parseTheSchemaToGetListOfCols(); // it should return ["col_name","data_type","comment"]
        formatOutput(colsArr, columnInformation);   // takes in array as an input
        getColumnInfoAsString(columnInformation, tbl);

        // Partitions
        if (tbl.isPartitioned()) {
            columnInformation.append(LINE_DELIM);
// same as above

            formatOutput("Partition", "Type", "Comments", columnInformation);
            formatOutput("---------", "----", "--------", columnInformation);




// If you want to show partitioned columns, have a seperate line: something like:
PARTITIONED COLUMNS

after the initial columns

            getPartitionInfoAsString(columnInformation, tbl);
        }
        return columnInformation.toString();
    }




It is pretty close to what you have - only the schema is defined in one place. 
What do you think ?

> describe extended table/partition output is cryptic
> ---------------------------------------------------
>
>                 Key: HIVE-558
>                 URL: https://issues.apache.org/jira/browse/HIVE-558
>             Project: Hadoop Hive
>          Issue Type: Improvement
>          Components: Metastore
>            Reporter: Prasad Chakka
>         Attachments: HIVE-558_PrelimPatch.patch, SampleOutputDescribe.txt
>
>
> describe extended table prints out the Thrift metadata object directly. The information
from it is not easy to read or parse. Output should be easily read and can be simple parsed
to get table location etc by programs.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message