.\" This manpage has been automatically generated by docbook2man 
.\" from a DocBook document.  This tool can be found at:
.\" <http://shell.ipoline.com/~elmert/comp/docbook2X/> 
.\" Please send any bug reports, improvements, comments, patches, 
.\" etc. to Steve Cheng <steve@ggi-project.org>.
.TH "SELECT" "" "02 July 1999" "SQL - Language Statements" ""
.SH NAME
SELECT \- Retrieve rows from a table or view.
.SH SYNOPSIS
1998-09-24

.nf
SELECT [ALL|DISTINCT [ON \fIcolumn\fR] ]
    \fIexpression\fR [ AS
   \fIname\fR ] [, ...]
    [ INTO [TEMP] [TABLE] \fInew_table\fR ]
    [ FROM \fItable\fR
   [\fIalias\fR ] [, ...] ]
    [ WHERE \fIcondition\fR ]
    [ GROUP BY \fIcolumn\fR [, ...] ]
    [ HAVING \fIcondition\fR [, ...] ]
    [ { UNION [ALL] | INTERSECT | EXCEPT } \fIselect\fR ]
    [ ORDER BY \fIcolumn\fR [ ASC | DESC ] [, ...] ]
    [ FOR UPDATE [OF class_name...]]
    [ LIMIT count [OFFSET|, count]]
  
.fi
.SS    "INPUTS"
.PP
.TP
\fB       \fIexpression\fB \fR
The name of a table's column or an expression.
.TP
\fB       \fIname\fB \fR
Specifies another name for a column or an expression using
the AS clause. \fIname\fR
cannot be used in the WHERE
condition. It can, however, be referenced in associated
ORDER BY or GROUP BY clauses.
.TP
\fB     TEMP \fR
The table is created unique to this session, and is
automatically dropped on session exit.
.TP
\fB       \fInew_table\fB \fR
If the INTO TABLE clause is specified, the result of the
query will be stored in another table with the indicated
name.
The target table (\fInew_table\fR) will
be created automatically and should not exist before this command.
Refer to \fBSELECT INTO\fR for more information.
.sp
.RS
.B "Note:"
The \fBCREATE TABLE AS\fR statement will also
create a new  table from a select query.
.RE
.TP
\fB       \fItable\fB \fR
The name of an existing table referenced by the FROM clause.
.TP
\fB       \fIalias\fB \fR
An alternate name for the preceding
\fItable\fR.
It is used for brevity or to eliminate ambiguity for joins
within a single table.
.TP
\fB       \fIcondition\fB \fR
A boolean expression giving a result of true or false.
See the WHERE clause.
.TP
\fB       \fIcolumn\fB \fR
The name of a table's column.
.TP
\fB       \fIselect\fB \fR
A select statement with all features except the ORDER BY clause.
.SS    "OUTPUTS"
.PP
.TP
\fB    Rows \fR
The complete set of rows resulting from the query specification.
.TP
\fB    \fIcount\fB \fR
The count of rows returned by the query.
.SH   "DESCRIPTION"
.PP
\fBSELECT\fR will return rows from one or more tables.
Candidates for selection are rows which satisfy the WHERE condition;
if WHERE is omitted, all rows are candidates.
.PP
\fBDISTINCT\fR will eliminate all duplicate rows from the
selection.
\fBDISTINCT ON \fIcolumn\fB\fR will eliminate all duplicates in the specified column; this is
equivalent to using \fBGROUP BY \fIcolumn\fB\fR.  \fBALL\fR will return all candidate rows,
including duplicates.
.PP
The GROUP BY clause allows a user to divide a table
conceptually into groups. (See GROUP BY clause).
.PP
The HAVING clause specifies a grouped table derived by the
elimination of groups from the result of the previously
specified clause. (See HAVING clause).
.PP
The ORDER BY clause allows a user to specify that he/she
wishes the rows sorted according to the ASCending or 
DESCending mode operator. (See ORDER BY clause)
.PP
The UNION clause allows the result to be the collection of rows
returned by the queries involved. (See UNION clause).
.PP
The INTERSECT give you the rows that are common to both queries.
(See INTERSECT clause).
.PP
The EXCEPT give you the rows in the upper query not in the lower query.
(See EXCEPT clause).
.PP
The FOR UPDATE clause allows the SELECT statement to perform 
exclusive locking of selected rows.
(See EXCEPT clause).
.PP
The LIMIT...OFFSET clause allows control over which rows are
returned by the query.
.PP
You must have SELECT privilege to a table to read its values
(See \fBGRANT\fR/\fBREVOKE\fR statements).
.SS    "WHERE CLAUSE"
.PP
The optional WHERE condition has the general form:

.nf
WHERE \fIexpr\fR \fIETER">c\fRe\fI"PAR\fRreplaceable> [ \fIlog_op\fR ... ]
    
.fi
where \fIcond_op\fR can be
one of: =, , =, , = or ,
a conditional operator like ALL, ANY, IN, LIKE, et cetera or a
locally-defined operator, 
and \fIlog_op\fR can be one 
of: AND, OR, NOT.
The comparison returns either TRUE or FALSE and all
instances will be discarded
if the expression evaluates to FALSE.
.SS    "GROUP BY CLAUSE"
.PP
GROUP BY specifies a grouped table derived by the application
of this clause:

.nf
GROUP BY \fIcolumn\fR [, ...]
    
.fi
.PP
GROUP BY will condense into a single row all rows that share the same values for the
grouped columns; aggregates return values derived from all rows that make up the group.  The value returned for an ungrouped
and unaggregated column is dependent on the order in which rows happen to be read from the database.
.SS    "HAVING CLAUSE"
.PP
The optional HAVING condition has the general form:

.nf
HAVING \fIcond_expr\fR
    
.fi
where \fIcond_expr\fR is the same
as specified for the WHERE clause.
.PP
HAVING specifies a grouped table derived by the elimination
of groups from the result of the previously specified clause
that do not meet the \fIcond_expr\fR.
.PP
Each column referenced in 
\fIcond_expr\fR shall unambiguously
reference a grouping column.
.SS    "ORDER BY CLAUSE"
.PP

.nf
ORDER BY \fIcolumn\fR [ ASC | DESC ] [, ...]
    
.fi
.PP
\fIcolumn\fR can be either a column
name or an ordinal number.
.PP
The ordinal numbers refers to the ordinal (left-to-right) position
of the column. This feature makes it possible to define an ordering
on the basis of a column that does not have a proper name.
This is never absolutely necessary because it is always possible
assign a name
to a calculated column using the AS clause, e.g.:

.nf
SELECT title, date_prod + 1 AS newlen FROM films ORDER BY newlen;
    
.fi
.PP
>From release 6.4 of PostgreSQL, the columns in the ORDER BY clause do not need to appear in the SELECT clause.
Thus the following statement is now legal:

.nf
SELECT name FROM distributors ORDER BY code;
    
.fi
.PP
Optionally one may add the keyword DESC (descending)
or ASC (ascending) after each column name in the ORDER BY clause.
If not specified, ASC is assumed by default.
.SS    "UNION CLAUSE"
.PP

.nf
\fItable_query\fR UNION [ ALL ]
\fItable_query\fR
     [ ORDER BY \fIcolumn\fR [ ASC | DESC ] [, ...] ]
    
.fi
where
\fItable_query\fR
specifies any select expression without an ORDER BY clause.
.PP
The UNION clause allows the result to be the collection of rows
returned by the queries involved. (See UNION clause).
The two tables that represent the direct operands of the UNION must
have the same number of columns, and corresponding columns must be
of compatible data types.
.PP
By default, the result of UNION does not contain any duplicate rows
unless the ALL clause is specified.
.PP
Multiple UNION operators in the same SELECT statement are
evaluated left to right.
Note that the ALL keyword is not global in nature, being 
applied only for the current pair of table results.
.SS    "INTERSECT CLAUSE"
.PP

.nf
\fItable_query\fR INTERSECT
\fItable_query\fR
     [ ORDER BY \fIcolumn\fR [ ASC | DESC ] [, ...] ]
    
.fi
where
\fItable_query\fR
specifies any select expression without an ORDER BY clause.
.PP
The INTERSECT clause allows the result to be all rows that are 
common to the involved queries.  (See INTERSECT clause).
The two tables that represent the direct operands of the INTERSECT must
have the same number of columns, and corresponding columns must be
of compatible data types.
.PP
Multiple INTERSECT operators in the same SELECT statement are
evaluated left to right.
.SS    "EXCEPT CLAUSE"
.PP

.nf
\fItable_query\fR EXCEPT
     \fItable_query\fR
     [ ORDER BY \fIcolumn\fR [ ASC | DESC ] [, ...] ]
    
.fi
where
\fItable_query\fR
specifies any select expression without an ORDER BY clause.
.PP
The EXCEPT clause allows the result to be rows from the upper query
that are not in the lower query.  (See EXCEPT clause).
The two tables that represent the direct operands of the EXCEPT must
have the same number of columns, and corresponding columns must be
of compatible data types.
.PP
Multiple EXCEPT operators in the same SELECT statement are
evaluated left to right.
.SH   "USAGE"
.PP
To join the table films with the table
distributors:

.nf
SELECT f.title, f.did, d.name, f.date_prod, f.kind
    FROM distributors d, films f
    WHERE f.did = d.did

title                    |did|name            | date_prod|kind
-------------------------+---+----------------+----------+----------
The Third Man            |101|British Lion    |1949-12-23|Drama
The African Queen        |101|British Lion    |1951-08-11|Romantic
Une Femme est une Femme  |102|Jean Luc Godard |1961-03-12|Romantic
Vertigo                  |103|Paramount       |1958-11-14|Action
Becket                   |103|Paramount       |1964-02-03|Drama
48 Hrs                   |103|Paramount       |1982-10-22|Action
War and Peace            |104|Mosfilm         |1967-02-12|Drama
West Side Story          |105|United Artists  |1961-01-03|Musical
Bananas                  |105|United Artists  |1971-07-13|Comedy
Yojimbo                  |106|Toho            |1961-06-16|Drama
There's a Girl in my Soup|107|Columbia        |1970-06-11|Comedy
Taxi Driver              |107|Columbia        |1975-05-15|Action
Absence of Malice        |107|Columbia        |1981-11-15|Action
Storia di una donna      |108|Westward        |1970-08-15|Romantic
The King and I           |109|20th Century Fox|1956-08-11|Musical
Das Boot                 |110|Bavaria Atelier |1981-11-11|Drama
Bed Knobs and Broomsticks|111|Walt Disney     |          |Musical
  
.fi
.PP
To sum the column len of all films and group
the results by kind:

.nf
SELECT kind, SUM(len) AS total FROM films GROUP BY kind;

    kind      |total
    ----------+------
    Action    | 07:34
    Comedy    | 02:58
    Drama     | 14:28
    Musical   | 06:42
    Romantic  | 04:38
  
.fi
.PP
To sum the column len of all films, group
the results by kind and show those group totals
that are less than 5 hours:

.nf
SELECT kind, SUM(len) AS total
    FROM films
    GROUP BY kind
    HAVING SUM(len) < INTERVAL '5 hour';

    kind      |total
    ----------+------
    Comedy    | 02:58
    Romantic  | 04:38
  
.fi
.PP
The following two examples are identical ways of sorting the individual
results according to the contents of the second column
(name):

.nf
SELECT * FROM distributors ORDER BY name;
SELECT * FROM distributors ORDER BY 2;

    did|name
    ---+----------------
    109|20th Century Fox
    110|Bavaria Atelier
    101|British Lion
    107|Columbia
    102|Jean Luc Godard
    113|Luso films
    104|Mosfilm
    103|Paramount
    106|Toho
    105|United Artists
    111|Walt Disney
    112|Warner Bros.
    108|Westward
  
.fi
.PP
This example shows how to obtain the union of the tables
distributors and
actors, restricting the results to those that begin
with letter W in each table.  Only distinct rows are to be used, so the
ALL keyword is omitted:

.nf
    --        distributors:                actors:
    --        did|name                     id|name
    --        ---+------------             --+--------------
    --        108|Westward                  1|Woody Allen
    --        111|Walt Disney               2|Warren Beatty
    --        112|Warner Bros.              3|Walter Matthau
    --        ...                           ...

SELECT distributors.name
    FROM   distributors
    WHERE  distributors.name LIKE 'W%'
UNION
SELECT actors.name
    FROM   actors
    WHERE  actors.name LIKE 'W%'

name
--------------
Walt Disney
Walter Matthau
Warner Bros.
Warren Beatty
Westward
Woody Allen
  
.fi
.SH   "COMPATIBILITY"
.PP
.SS    "EXTENSIONS"
.PP
Postgres allows one to omit 
the \fBFROM\fR clause from a query. This feature
was retained from the original PostQuel query language:

.nf
SELECT distributors.* WHERE name = 'Westwood';

    did|name
    ---+----------------
    108|Westward
  
.fi
.SS    "SQL92"
.PP
.SS     "SELECT CLAUSE"
.PP
In the SQL92 standard, the optional keyword "AS"
is just noise and can be 
omitted without affecting the meaning.
The Postgres parser requires this keyword when
renaming columns because the type extensibility features lead to
parsing ambiguities
in this context.
.PP
In the SQL92 standard, the new column name 
specified in an
"AS" clause may be referenced in GROUP BY and HAVING clauses.
This is not currently
allowed in Postgres.
.PP
The DISTINCT ON phrase is not part of SQL92.
.SS     "UNION CLAUSE"
.PP
The SQL92 syntax for UNION allows an
additional CORRESPONDING BY clause:

.nf
 
\fItable_query\fR UNION [ALL]
    [CORRESPONDING [BY (\fIcolumn\fR [,...])]]
    \fItable_query\fR
     
.fi
.PP
The CORRESPONDING BY clause is not supported by
Postgres.