New Command for 3.1.0.3: udbsql

Overview:

udbsql is a method of quering the data in any of your UDB databases. It allows you to quickly and efficiently explore data that has been loaded into and distributed in a UDB database, and is written in a user friendly, SQL-like language.

The udbsql command structure consists of a select statement to define which columns are desired and what operations you want to perform on them, a from statement telling udbsql which UDB database and table|vector|variable to pull the data from, and a series of filtering and ordering options to modify how the data is output.

udbsql supports the following types of sql statements:

Select Statement Specifications: column_name(s), *, count(column_name), count(*), count(distinct column_name)
From Statement Specifications: database_name:table_name|vector_name|variable_name
Output Modification Options: where column_name operator value, limit number, group by column_name, order by column_name

Syntax:

Command	Arguments	Description
udbsql	`[--help \|\| -h]` `[--verbose \|\| -v]` `[--list \|\| -l]`	Query data in a UDB database using simple sql-like statements

Options:

-v, --verbose  verbosely list aq_udb command to execute
-l, --list     list aq_udb command without execution

Restrictions:

Query Format:

select [column_name] | [*] from [database_name]:[table_name | vector_name | variable_name] where ... order by ... limit ...

select count(distinct [column_name] | [*]) from [database_name]:[table_name | vector_name | variable_name] where ...

select [column_name], count(*) from [database_name]:[table_name | vector_name | variable_name] where ... group by [column_name]

Rules:

The first query format above is a "select" query.
The second and third query formats above are "count" queries.

1. Group By is NOT supported for SELECT queries.
2. Order By is NOT supported for COUNT queries.
3. Limit is NOT supported for COUNT queries.
4. Order By, Limit, or Group By can only be used when there is no DISTINCT in COUNT queries.

Examples:

SELECT:

select * from udb_databasename:tablename
select count(*) from udb_databasename:tablename
SELECT ... FROM database:tableA WHERE pkey_A IN (SELECT pkey_B FROM [database:]tableB WHERE ...) ...

Columns:

select col2, col4 from udb_databasename:tablename

LIMIT:

select * from udb_databasename:tablename limit 10

WHERE:

select * from udb_databasename:tablename where filterspec

ORDER BY:

select * from udb_databasename:tablename order by col

DISTINCT:

select count(distinct col1), count(distinct col2), count(*) from udb_databasename:tablename

GROUP BY:

select count(*) from udb_databasename:tablename group by col1 col2