Errata

It says:
hive> set env:HOME;
env:HOME=/home/yourusername

Which is correct, and again on the next line it is saying the same thing but different output (which is not possible):
hive> set env:HOME;
env:* variables can not be set.

I think what the author intent to do here is to demonstrate that setting environment variable (env.HOME) results in error i.e.:
hive> set env:HOME=/abcd;
env:* variables can not be set.

Note from the Author or Editor:
Ah yes. Correct. The suggested change is fine.

"If you specify the OVERWRITE keyword, any data already present in the target directory will be deleted first. Without the keyword, the new files are simply added to the target directory. However, if files already exist in the target directory that match filenames being loaded, the old files are overwritten.

Versions of Hive before v0.9.0 had the following bug: when the OVER WRITE keyword was not used, an existing data file in the target directory would be overwritten if its name matched the name of a data file being written to the directory. Hence, data would be lost. This bug was fixed in the v0.9.0 release."

There is a discrepancy as both imply the same - before and after the v0.9.0/bug-fix.

Note from the Author or Editor:
I think what's confusing is the fact that the callout says the same thing, in more detail, as the last sentence of the previous paragraph. So, I suggest deleting the sentence at the end of the paragraph. I.e., delete "However, if files already exist.. overwritten."

In 'CREATE TABLE ca_employees' code snippet, should the

WHERE se.state = 'CA'

clause be

WHERE state= 'CA'

instead (i.e., without the 'se' prefix, which is not defined in the code snippet)?

Note from the Author or Editor:
Correct. Remove the "se." from the clause as described

Towards the top of page 103 it says, "The partition filters are ignored for OUTER JOINTS" but the last word should instead be "JOINS"

Note from the Author or Editor:
Unless we're smoking something ;)

Correct. should be "JOINS".

In the below example:
CREATE VIEW shipments(time, part) AS
SELECT cols["time"], cols["parts"]
FROM dynamictable
WHERE cols["type"] = "response";

The cols["parts"] should cols["part"] i.e. it should be part and not parts.

Note from the Author or Editor:
Yes, should be 'cols["parts"]'.

The statement:
ALTER INDEX employees_index ON TABLE employees PARTITION (country = 'US') REBUILD;

should not have the word TABLE. It should be

ALTER INDEX employees_index ON employees PARTITION (country = 'US') REBUILD;

Also, later on the page, the same error occurs in a different statement:

DROP INDEX IF EXISTS employees_index ON TABLE employees;

should be

DROP INDEX IF EXISTS employees_index ON employees;

The book suggests the following command for dropping an index:

DROP INDEX IF EXISTS employees_index ON TABLE employees;

However, Hive 0.10 gives a ParseException ("extraneous input 'TABLE' expecting Identifier near '<EOF>'") in response.

It seems that the TABLE keyword is not allowed here, so the command should be:

DROP INDEX IF EXISTS employees_index ON employees;

Note from the Author or Editor:
Correct. This might be a SQL language change since the book was written, or it's just a dumb mistake ;)

The local execution mode example shows setting a property named 'mapred.tmp.dir' but I don't believe Hadoop (at least widely-used versions) use a property with this name.

There's a property named 'mapred.temp.dir', though the original JIRA (https://issues.apache.org/jira/browse/HIVE-1408) describes setting the 'mapred.local.dir' property instead (which is what I do).

Note from the Author or Editor:
It should be hadoop.tmp.dir, not mapred.tmp.dir

This code uses the @Description annotation, but the class isn't imported. The code does not compile as written, but this problem can be fixed by adding a line at the top of page 167:

import org.apache.hadoop.hive.ql.exec.Description;

Note from the Author or Editor:
Yes, the code should have the line:

import org.apache.hadoop.hive.ql.exec.Description;

Hi,

I understand that the intention of the code is to identify the external tables whose data reside inside the warehouse directory, which is /user/hive/warehouse.

The if clause code is there as:


if (t.getTableType().equals("MANAGED_TABLE") &&
! u.getPath()contains("/user/hive/warehouse") ) {
System.out.println (t.getTableName()
+ " is a non external table mounted inside /user/hive/warehouse" );
bad.add (t.getTableName());
}

There are two issues here:
1. The check inside the if clause.
2. The message in the println statement.

The code should be

if (! t.getTableType().equals("MANAGED_TABLE") &&
u.getPath()contains("/user/hive/warehouse") ) {
System.out.println (t.getTableName()
+ " is an external table mounted inside /user/hive/warehouse" );
bad.add (t.getTableName());
}

or

if (t.getTableType().equals("EXTERNAL") &&
u.getPath()contains("/user/hive/warehouse") ) {
System.out.println (t.getTableName()
+ " is an external table mounted inside /user/hive/warehouse" );
bad.add (t.getTableName());
}


Regards,
Ramki.

Note from the Author or Editor:
UPDATE: Missing "." in one expression:

I think the second proposed alternative is better:

if (t.getTableType().equals("EXTERNAL") &&

u.getPath().contains("/user/hive/warehouse") ) {

System.out.println (t.getTableName()

+ " is an external table mounted inside /user/hive/warehouse" );

bad.add (t.getTableName());

}

The "CREATE TABLE" statements that are listed for HBase are incorrect. Both the internal and external DDL that is listed contain only a single non-key column mapping in the hbase.columns.mapping SERDEPROPERTIES, while they use two non-key columns in the actual table creation statement. Both of these create table statements are invalid.

The DDL statements can be corrected by either adding an additional column mapping, or reducing the number of columns defined in table definition.

Note from the Author or Editor:
Correct. Should be:
CREATE TABLE hbase_stocks(key INT, name STRING, price FLOAT) STORED BY 'org.apache.hadoop.hive.hbase.HBaseStorageHandler'
WITH SERDEPROPERTIES ("hbase.columns.mapping" = ":key,stock:val,price:val") TBLPROPERTIES ("hbase.table.name" = "stocks");
(I added ,"price:val" to the SERDEPROPERTIES clause.