Movatterモバイル変換

1. All queries in theWITH list are computed. These effectively serve as temporary tables that can be referenced in theFROM list. AWITH query that is referenced more than once inFROM is computed only once, unless specified otherwise withNOT MATERIALIZED. (SeeWITH Clause below.)

2. All elements in theFROM list are computed. (Each element in theFROM list is a real or virtual table.) If more than one element is specified in theFROM list, they are cross-joined together. (SeeFROM Clause below.)

3. If theWHERE clause is specified, all rows that do not satisfy the condition are eliminated from the output. (SeeWHERE Clause below.)

4. If theGROUP BY clause is specified, or if there are aggregate function calls, the output is combined into groups of rows that match on one or more values, and the results of aggregate functions are computed. If theHAVING clause is present, it eliminates groups that do not satisfy the given condition. (SeeGROUP BY Clause andHAVING Clause below.) Although query output columns are nominally computed in the next step, they can also be referenced (by name or ordinal number) in theGROUP BY clause.

5. The actual output rows are computed using theSELECT output expressions for each selected row or row group. (SeeSELECT List below.)

6. SELECT DISTINCT eliminates duplicate rows from the result.SELECT DISTINCT ON eliminates rows that match on all the specified expressions.SELECT ALL (the default) will return all candidate rows, including duplicates. (SeeDISTINCT Clause below.)

7. Using the operatorsUNION,INTERSECT, andEXCEPT, the output of more than oneSELECT statement can be combined to form a single result set. TheUNION operator returns all rows that are in one or both of the result sets. TheINTERSECT operator returns all rows that are strictly in both result sets. TheEXCEPT operator returns the rows that are in the first result set but not in the second. In all three cases, duplicate rows are eliminated unlessALL is specified. The noise wordDISTINCT can be added to explicitly specify eliminating duplicate rows. Notice thatDISTINCT is the default behavior here, even thoughALL is the default forSELECT itself. (SeeUNION Clause,INTERSECT Clause, andEXCEPT Clause below.)

8. If theORDER BY clause is specified, the returned rows are sorted in the specified order. IfORDER BY is not given, the rows are returned in whatever order the system finds fastest to produce. (SeeORDER BY Clause below.)

9. If theLIMIT (orFETCH FIRST) orOFFSET clause is specified, theSELECT statement only returns a subset of the result rows. (SeeLIMIT Clause below.)

10. IfFOR UPDATE,FOR NO KEY UPDATE,FOR SHARE orFOR KEY SHARE is specified, theSELECT statement locks the selected rows against concurrent updates. (SeeThe Locking Clause below.)

WITH Clause#

TheWITH clause allows you to specify one or more subqueries that can be referenced by name in the primary query. The subqueries effectively act as temporary tables or views for the duration of the primary query. Each subquery can be aSELECT,TABLE,VALUES,INSERT,UPDATE,DELETE, orMERGE statement. When writing a data-modifying statement (INSERT,UPDATE,DELETE, orMERGE) inWITH, it is usual to include aRETURNING clause. It is the output ofRETURNING,not the underlying table that the statement modifies, that forms the temporary table that is read by the primary query. IfRETURNING is omitted, the statement is still executed, but it produces no output so it cannot be referenced as a table by the primary query.

A name (without schema qualification) must be specified for eachWITH query. Optionally, a list of column names can be specified; if this is omitted, the column names are inferred from the subquery.

IfRECURSIVE is specified, it allows aSELECT subquery to reference itself by name. Such a subquery must have the form

non_recursive_term UNION [ ALL | DISTINCT ]recursive_term

where the recursive self-reference must appear on the right-hand side of theUNION. Only one recursive self-reference is permitted per query. Recursive data-modifying statements are not supported, but you can use the results of a recursiveSELECT query in a data-modifying statement. SeeSection 7.8 for an example.

Another effect ofRECURSIVE is thatWITH queries need not be ordered: a query can reference another one that is later in the list. (However, circular references, or mutual recursion, are not implemented.) WithoutRECURSIVE,WITH queries can only reference siblingWITH queries that are earlier in theWITH list.

When there are multiple queries in theWITH clause,RECURSIVE should be written only once, immediately afterWITH. It applies to all queries in theWITH clause, though it has no effect on queries that do not use recursion or forward references.

The optionalSEARCH clause computes asearch sequence column that can be used for ordering the results of a recursive query in either breadth-first or depth-first order. The supplied column name list specifies the row key that is to be used for keeping track of visited rows. A column namedsearch_seq_col_name will be added to the result column list of theWITH query. This column can be ordered by in the outer query to achieve the respective ordering. SeeSection 7.8.2.1 for examples.

The optionalCYCLE clause is used to detect cycles in recursive queries. The supplied column name list specifies the row key that is to be used for keeping track of visited rows. A column namedcycle_mark_col_name will be added to the result column list of theWITH query. This column will be set tocycle_mark_value when a cycle has been detected, else tocycle_mark_default. Furthermore, processing of the recursive union will stop when a cycle has been detected.cycle_mark_value andcycle_mark_default must be constants and they must be coercible to a common data type, and the data type must have an inequality operator. (The SQL standard requires that they be Boolean constants or character strings, but Postgres Pro does not require that.) By default,TRUE andFALSE (of typeboolean) are used. Furthermore, a column namedcycle_path_col_name will be added to the result column list of theWITH query. This column is used internally for tracking visited rows. SeeSection 7.8.2.2 for examples.

Both theSEARCH and theCYCLE clause are only valid for recursiveWITH queries. Thewith_query must be aUNION (orUNION ALL) of twoSELECT (or equivalent) commands (no nestedUNIONs). If both clauses are used, the column added by theSEARCH clause appears before the columns added by theCYCLE clause.

The primary query and theWITH queries are all (notionally) executed at the same time. This implies that the effects of a data-modifying statement inWITH cannot be seen from other parts of the query, other than by reading itsRETURNING output. If two such data-modifying statements attempt to modify the same row, the results are unspecified.

A key property ofWITH queries is that they are normally evaluated only once per execution of the primary query, even if the primary query refers to them more than once. In particular, data-modifying statements are guaranteed to be executed once and only once, regardless of whether the primary query reads all or any of their output.

However, aWITH query can be markedNOT MATERIALIZED to remove this guarantee. In that case, theWITH query can be folded into the primary query much as though it were a simple sub-SELECT in the primary query'sFROM clause. This results in duplicate computations if the primary query refers to thatWITH query more than once; but if each such use requires only a few rows of theWITH query's total output,NOT MATERIALIZED can provide a net savings by allowing the queries to be optimized jointly.NOT MATERIALIZED is ignored if it is attached to aWITH query that is recursive or is not side-effect-free (i.e., is not a plainSELECT containing no volatile functions).

By default, a side-effect-freeWITH query is folded into the primary query if it is used exactly once in the primary query'sFROM clause. This allows joint optimization of the two query levels in situations where that should be semantically invisible. However, such folding can be prevented by marking theWITH query asMATERIALIZED. That might be useful, for example, if theWITH query is being used as an optimization fence to prevent the planner from choosing a bad plan.Postgres Pro versions before v12 never did such folding, so queries written for older versions might rely onWITH to act as an optimization fence.

SeeSection 7.8 for additional information.

FROM Clause#

TheFROM clause specifies one or more source tables for theSELECT. If multiple sources are specified, the result is the Cartesian product (cross join) of all the sources. But usually qualification conditions are added (viaWHERE) to restrict the returned rows to a small subset of the Cartesian product.

TheFROM clause can contain the following elements:

WHERE Clause#

The optionalWHERE clause has the general form

WHEREcondition

wherecondition is any expression that evaluates to a result of typeboolean. Any row that does not satisfy this condition will be eliminated from the output. A row satisfies the condition if it returns true when the actual row values are substituted for any variable references.

GROUP BY Clause#

The optionalGROUP BY clause has the general form

GROUP BY [ ALL | DISTINCT ]grouping_element [, ...]

GROUP BY will condense into a single row all selected rows that share the same values for the grouped expressions. Anexpression used inside agrouping_element can be an input column name, or the name or ordinal number of an output column (SELECT list item), or an arbitrary expression formed from input-column values. In case of ambiguity, aGROUP BY name will be interpreted as an input-column name rather than an output column name.

If any ofGROUPING SETS,ROLLUP orCUBE are present as grouping elements, then theGROUP BY clause as a whole defines some number of independentgrouping sets. The effect of this is equivalent to constructing aUNION ALL between subqueries with the individual grouping sets as theirGROUP BY clauses. The optionalDISTINCT clause removes duplicate sets before processing; it doesnot transform theUNION ALL into aUNION DISTINCT. For further details on the handling of grouping sets seeSection 7.2.4.

Aggregate functions, if any are used, are computed across all rows making up each group, producing a separate value for each group. (If there are aggregate functions but noGROUP BY clause, the query is treated as having a single group comprising all the selected rows.) The set of rows fed to each aggregate function can be further filtered by attaching aFILTER clause to the aggregate function call; seeSection 4.2.7 for more information. When aFILTER clause is present, only those rows matching it are included in the input to that aggregate function.

WhenGROUP BY is present, or any aggregate functions are present, it is not valid for theSELECT list expressions to refer to ungrouped columns except within aggregate functions or when the ungrouped column is functionally dependent on the grouped columns, since there would otherwise be more than one possible value to return for an ungrouped column. A functional dependency exists if the grouped columns (or a subset thereof) are the primary key of the table containing the ungrouped column.

Keep in mind that all aggregate functions are evaluated before evaluating any“scalar” expressions in theHAVING clause orSELECT list. This means that, for example, aCASE expression cannot be used to skip evaluation of an aggregate function; seeSection 4.2.14.

Currently,FOR NO KEY UPDATE,FOR UPDATE,FOR SHARE andFOR KEY SHARE cannot be specified withGROUP BY.

HAVING Clause#

The optionalHAVING clause has the general form

HAVINGcondition

wherecondition is the same as specified for theWHERE clause.

HAVING eliminates group rows that do not satisfy the condition.HAVING is different fromWHERE:WHERE filters individual rows before the application ofGROUP BY, whileHAVING filters group rows created byGROUP BY. Each column referenced incondition must unambiguously reference a grouping column, unless the reference appears within an aggregate function or the ungrouped column is functionally dependent on the grouping columns.

The presence ofHAVING turns a query into a grouped query even if there is noGROUP BY clause. This is the same as what happens when the query contains aggregate functions but noGROUP BY clause. All the selected rows are considered to form a single group, and theSELECT list andHAVING clause can only reference table columns from within aggregate functions. Such a query will emit a single row if theHAVING condition is true, zero rows if it is not true.

Currently,FOR NO KEY UPDATE,FOR UPDATE,FOR SHARE andFOR KEY SHARE cannot be specified withHAVING.

WINDOW Clause#

The optionalWINDOW clause has the general form

WINDOWwindow_name AS (window_definition ) [, ...]

wherewindow_name is a name that can be referenced fromOVER clauses or subsequent window definitions, andwindow_definition is

[existing_window_name ][ PARTITION BYexpression [, ...] ][ ORDER BYexpression [ ASC | DESC | USINGoperator ] [ NULLS { FIRST | LAST } ] [, ...] ][frame_clause ]

If anexisting_window_name is specified it must refer to an earlier entry in theWINDOW list; the new window copies its partitioning clause from that entry, as well as its ordering clause if any. In this case the new window cannot specify its ownPARTITION BY clause, and it can specifyORDER BY only if the copied window does not have one. The new window always uses its own frame clause; the copied window must not specify a frame clause.

The elements of thePARTITION BY list are interpreted in much the same fashion as elements of aGROUP BY clause, except that they are always simple expressions and never the name or number of an output column. Another difference is that these expressions can contain aggregate function calls, which are not allowed in a regularGROUP BY clause. They are allowed here because windowing occurs after grouping and aggregation.

Similarly, the elements of theORDER BY list are interpreted in much the same fashion as elements of a statement-levelORDER BY clause, except that the expressions are always taken as simple expressions and never the name or number of an output column.

The optionalframe_clause defines thewindow frame for window functions that depend on the frame (not all do). The window frame is a set of related rows for each row of the query (called thecurrent row). Theframe_clause can be one of

{ RANGE | ROWS | GROUPS }frame_start [frame_exclusion ]{ RANGE | ROWS | GROUPS } BETWEENframe_start ANDframe_end [frame_exclusion ]

whereframe_start andframe_end can be one of

UNBOUNDED PRECEDINGoffset PRECEDINGCURRENT ROWoffset FOLLOWINGUNBOUNDED FOLLOWING

andframe_exclusion can be one of

EXCLUDE CURRENT ROWEXCLUDE GROUPEXCLUDE TIESEXCLUDE NO OTHERS

Ifframe_end is omitted it defaults toCURRENT ROW. Restrictions are thatframe_start cannot beUNBOUNDED FOLLOWING,frame_end cannot beUNBOUNDED PRECEDING, and theframe_end choice cannot appear earlier in the above list offrame_start andframe_end options than theframe_start choice does — for exampleRANGE BETWEEN CURRENT ROW ANDoffset PRECEDING is not allowed.

The default framing option isRANGE UNBOUNDED PRECEDING, which is the same asRANGE BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW; it sets the frame to be all rows from the partition start up through the current row's lastpeer (a row that the window'sORDER BY clause considers equivalent to the current row; all rows are peers if there is noORDER BY). In general,UNBOUNDED PRECEDING means that the frame starts with the first row of the partition, and similarlyUNBOUNDED FOLLOWING means that the frame ends with the last row of the partition, regardless ofRANGE,ROWS orGROUPS mode. InROWS mode,CURRENT ROW means that the frame starts or ends with the current row; but inRANGE orGROUPS mode it means that the frame starts or ends with the current row's first or last peer in theORDER BY ordering. TheoffsetPRECEDING andoffsetFOLLOWING options vary in meaning depending on the frame mode. InROWS mode, theoffset is an integer indicating that the frame starts or ends that many rows before or after the current row. InGROUPS mode, theoffset is an integer indicating that the frame starts or ends that many peer groups before or after the current row's peer group, where apeer group is a group of rows that are equivalent according to the window'sORDER BY clause. InRANGE mode, use of anoffset option requires that there be exactly oneORDER BY column in the window definition. Then the frame contains those rows whose ordering column value is no more thanoffset less than (forPRECEDING) or more than (forFOLLOWING) the current row's ordering column value. In these cases the data type of theoffset expression depends on the data type of the ordering column. For numeric ordering columns it is typically of the same type as the ordering column, but for datetime ordering columns it is aninterval. In all these cases, the value of theoffset must be non-null and non-negative. Also, while theoffset does not have to be a simple constant, it cannot contain variables, aggregate functions, or window functions.

Theframe_exclusion option allows rows around the current row to be excluded from the frame, even if they would be included according to the frame start and frame end options.EXCLUDE CURRENT ROW excludes the current row from the frame.EXCLUDE GROUP excludes the current row and its ordering peers from the frame.EXCLUDE TIES excludes any peers of the current row from the frame, but not the current row itself.EXCLUDE NO OTHERS simply specifies explicitly the default behavior of not excluding the current row or its peers.

Beware that theROWS mode can produce unpredictable results if theORDER BY ordering does not order the rows uniquely. TheRANGE andGROUPS modes are designed to ensure that rows that are peers in theORDER BY ordering are treated alike: all rows of a given peer group will be in the frame or excluded from it.

The purpose of aWINDOW clause is to specify the behavior ofwindow functions appearing in the query'sSELECT list orORDER BY clause. These functions can reference theWINDOW clause entries by name in theirOVER clauses. AWINDOW clause entry does not have to be referenced anywhere, however; if it is not used in the query it is simply ignored. It is possible to use window functions without anyWINDOW clause at all, since a window function call can specify its window definition directly in itsOVER clause. However, theWINDOW clause saves typing when the same window definition is needed for more than one window function.

Currently,FOR NO KEY UPDATE,FOR UPDATE,FOR SHARE andFOR KEY SHARE cannot be specified withWINDOW.

Window functions are described in detail inSection 3.5,Section 4.2.8, andSection 7.2.5.

SELECT List#

TheSELECT list (between the key wordsSELECT andFROM) specifies expressions that form the output rows of theSELECT statement. The expressions can (and usually do) refer to columns computed in theFROM clause.

Just as in a table, every output column of aSELECT has a name. In a simpleSELECT this name is just used to label the column for display, but when theSELECT is a sub-query of a larger query, the name is seen by the larger query as the column name of the virtual table produced by the sub-query. To specify the name to use for an output column, writeASoutput_name after the column's expression. (You can omitAS, but only if the desired output name does not match anyPostgres Pro keyword (seeAppendix C). For protection against possible future keyword additions, it is recommended that you always either writeAS or double-quote the output name.) If you do not specify a column name, a name is chosen automatically byPostgres Pro. If the column's expression is a simple column reference then the chosen name is the same as that column's name. In more complex cases a function or type name may be used, or the system may fall back on a generated name such as?column?.

An output column's name can be used to refer to the column's value inORDER BY andGROUP BY clauses, but not in theWHERE orHAVING clauses; there you must write out the expression instead.

Instead of an expression,* can be written in the output list as a shorthand for all the columns of the selected rows. Also, you can writetable_name.* as a shorthand for the columns coming from just that table. In these cases it is not possible to specify new names withAS; the output column names will be the same as the table columns' names.

According to the SQL standard, the expressions in the output list should be computed before applyingDISTINCT,ORDER BY, orLIMIT. This is obviously necessary when usingDISTINCT, since otherwise it's not clear what values are being made distinct. However, in many cases it is convenient if output expressions are computed afterORDER BY andLIMIT; particularly if the output list contains any volatile or expensive functions. With that behavior, the order of function evaluations is more intuitive and there will not be evaluations corresponding to rows that never appear in the output.Postgres Pro will effectively evaluate output expressions after sorting and limiting, so long as those expressions are not referenced inDISTINCT,ORDER BY orGROUP BY. (As a counterexample,SELECT f(x) FROM tab ORDER BY 1 clearly must evaluatef(x) before sorting.) Output expressions that contain set-returning functions are effectively evaluated after sorting and before limiting, so thatLIMIT will act to cut off the output from a set-returning function.

DISTINCT Clause#

IfSELECT DISTINCT is specified, all duplicate rows are removed from the result set (one row is kept from each group of duplicates).SELECT ALL specifies the opposite: all rows are kept; that is the default.

SELECT DISTINCT ON (expression [, ...] ) keeps only the first row of each set of rows where the given expressions evaluate to equal. TheDISTINCT ON expressions are interpreted using the same rules as forORDER BY (see above). Note that the“first row” of each set is unpredictable unlessORDER BY is used to ensure that the desired row appears first. For example:

SELECT DISTINCT ON (location) location, time, report    FROM weather_reports    ORDER BY location, time DESC;

retrieves the most recent weather report for each location. But if we had not usedORDER BY to force descending order of time values for each location, we'd have gotten a report from an unpredictable time for each location.

TheDISTINCT ON expression(s) must match the leftmostORDER BY expression(s). TheORDER BY clause will normally contain additional expression(s) that determine the desired precedence of rows within eachDISTINCT ON group.

Currently,FOR NO KEY UPDATE,FOR UPDATE,FOR SHARE andFOR KEY SHARE cannot be specified withDISTINCT.

UNION Clause#

TheUNION clause has this general form:

select_statement UNION [ ALL | DISTINCT ]select_statement

select_statement is anySELECT statement without anORDER BY,LIMIT,FOR NO KEY UPDATE,FOR UPDATE,FOR SHARE, orFOR KEY SHARE clause. (ORDER BY andLIMIT can be attached to a subexpression if it is enclosed in parentheses. Without parentheses, these clauses will be taken to apply to the result of theUNION, not to its right-hand input expression.)

TheUNION operator computes the set union of the rows returned by the involvedSELECT statements. A row is in the set union of two result sets if it appears in at least one of the result sets. The twoSELECT statements that represent the direct operands of theUNION must produce the same number of columns, and corresponding columns must be of compatible data types.

The result ofUNION does not contain any duplicate rows unless theALL option is specified.ALL prevents elimination of duplicates. (Therefore,UNION ALL is usually significantly quicker thanUNION; useALL when you can.)DISTINCT can be written to explicitly specify the default behavior of eliminating duplicate rows.

MultipleUNION operators in the sameSELECT statement are evaluated left to right, unless otherwise indicated by parentheses.

Currently,FOR NO KEY UPDATE,FOR UPDATE,FOR SHARE andFOR KEY SHARE cannot be specified either for aUNION result or for any input of aUNION.

INTERSECT Clause#

TheINTERSECT clause has this general form:

select_statement INTERSECT [ ALL | DISTINCT ]select_statement

select_statement is anySELECT statement without anORDER BY,LIMIT,FOR NO KEY UPDATE,FOR UPDATE,FOR SHARE, orFOR KEY SHARE clause.

TheINTERSECT operator computes the set intersection of the rows returned by the involvedSELECT statements. A row is in the intersection of two result sets if it appears in both result sets.

The result ofINTERSECT does not contain any duplicate rows unless theALL option is specified. WithALL, a row that hasm duplicates in the left table andn duplicates in the right table will appear min(m,n) times in the result set.DISTINCT can be written to explicitly specify the default behavior of eliminating duplicate rows.

MultipleINTERSECT operators in the sameSELECT statement are evaluated left to right, unless parentheses dictate otherwise.INTERSECT binds more tightly thanUNION. That is,A UNION B INTERSECT C will be read asA UNION (B INTERSECT C).

Currently,FOR NO KEY UPDATE,FOR UPDATE,FOR SHARE andFOR KEY SHARE cannot be specified either for anINTERSECT result or for any input of anINTERSECT.

EXCEPT Clause#

TheEXCEPT clause has this general form:

select_statement EXCEPT [ ALL | DISTINCT ]select_statement

select_statement is anySELECT statement without anORDER BY,LIMIT,FOR NO KEY UPDATE,FOR UPDATE,FOR SHARE, orFOR KEY SHARE clause.

TheEXCEPT operator computes the set of rows that are in the result of the leftSELECT statement but not in the result of the right one.

The result ofEXCEPT does not contain any duplicate rows unless theALL option is specified. WithALL, a row that hasm duplicates in the left table andn duplicates in the right table will appear max(m-n,0) times in the result set.DISTINCT can be written to explicitly specify the default behavior of eliminating duplicate rows.

MultipleEXCEPT operators in the sameSELECT statement are evaluated left to right, unless parentheses dictate otherwise.EXCEPT binds at the same level asUNION.

Currently,FOR NO KEY UPDATE,FOR UPDATE,FOR SHARE andFOR KEY SHARE cannot be specified either for anEXCEPT result or for any input of anEXCEPT.

ORDER BY Clause#

The optionalORDER BY clause has this general form:

ORDER BYexpression [ ASC | DESC | USINGoperator ] [ NULLS { FIRST | LAST } ] [, ...]

TheORDER BY clause causes the result rows to be sorted according to the specified expression(s). If two rows are equal according to the leftmost expression, they are compared according to the next expression and so on. If they are equal according to all specified expressions, they are returned in an implementation-dependent order.

Eachexpression can be the name or ordinal number of an output column (SELECT list item), or it can be an arbitrary expression formed from input-column values.

The ordinal number refers to the ordinal (left-to-right) position of the output column. This feature makes it possible to define an ordering on the basis of a column that does not have a unique name. This is never absolutely necessary because it is always possible to assign a name to an output column using theAS clause.

It is also possible to use arbitrary expressions in theORDER BY clause, including columns that do not appear in theSELECT output list. Thus the following statement is valid:

SELECT name FROM distributors ORDER BY code;

A limitation of this feature is that anORDER BY clause applying to the result of aUNION,INTERSECT, orEXCEPT clause can only specify an output column name or number, not an expression.

If anORDER BY expression is a simple name that matches both an output column name and an input column name,ORDER BY will interpret it as the output column name. This is the opposite of the choice thatGROUP BY will make in the same situation. This inconsistency is made to be compatible with the SQL standard.

Optionally one can add the key wordASC (ascending) orDESC (descending) after any expression in theORDER BY clause. If not specified,ASC is assumed by default. Alternatively, a specific ordering operator name can be specified in theUSING clause. An ordering operator must be a less-than or greater-than member of some B-tree operator family.ASC is usually equivalent toUSING < andDESC is usually equivalent toUSING >. (But the creator of a user-defined data type can define exactly what the default sort ordering is, and it might correspond to operators with other names.)

IfNULLS LAST is specified, null values sort after all non-null values; ifNULLS FIRST is specified, null values sort before all non-null values. If neither is specified, the default behavior isNULLS LAST whenASC is specified or implied, andNULLS FIRST whenDESC is specified (thus, the default is to act as though nulls are larger than non-nulls). WhenUSING is specified, the default nulls ordering depends on whether the operator is a less-than or greater-than operator.

Note that ordering options apply only to the expression they follow; for exampleORDER BY x, y DESC does not mean the same thing asORDER BY x DESC, y DESC.

Character-string data is sorted according to the collation that applies to the column being sorted. That can be overridden at need by including aCOLLATE clause in theexpression, for exampleORDER BY mycolumn COLLATE "en_US". For more information seeSection 4.2.10 andSection 22.2.

LIMIT Clause#

TheLIMIT clause consists of two independent sub-clauses:

LIMIT {count | ALL }OFFSETstart

The parametercount specifies the maximum number of rows to return, whilestart specifies the number of rows to skip before starting to return rows. When both are specified,start rows are skipped before starting to count thecount rows to be returned.

If thecount expression evaluates to NULL, it is treated asLIMIT ALL, i.e., no limit. Ifstart evaluates to NULL, it is treated the same asOFFSET 0.

SQL:2008 introduced a different syntax to achieve the same result, whichPostgres Pro also supports. It is:

OFFSETstart { ROW | ROWS }FETCH { FIRST | NEXT } [count ] { ROW | ROWS } { ONLY | WITH TIES }

In this syntax, thestart orcount value is required by the standard to be a literal constant, a parameter, or a variable name; as aPostgres Pro extension, other expressions are allowed, but will generally need to be enclosed in parentheses to avoid ambiguity. Ifcount is omitted in aFETCH clause, it defaults to 1. TheWITH TIES option is used to return any additional rows that tie for the last place in the result set according to theORDER BY clause;ORDER BY is mandatory in this case, andSKIP LOCKED is not allowed.ROW andROWS as well asFIRST andNEXT are noise words that don't influence the effects of these clauses. According to the standard, theOFFSET clause must come before theFETCH clause if both are present; butPostgres Pro is laxer and allows either order.

When usingLIMIT, it is a good idea to use anORDER BY clause that constrains the result rows into a unique order. Otherwise you will get an unpredictable subset of the query's rows — you might be asking for the tenth through twentieth rows, but tenth through twentieth in what ordering? You don't know what ordering unless you specifyORDER BY.

The query planner takesLIMIT into account when generating a query plan, so you are very likely to get different plans (yielding different row orders) depending on what you use forLIMIT andOFFSET. Thus, using differentLIMIT/OFFSET values to select different subsets of a query resultwill give inconsistent results unless you enforce a predictable result ordering withORDER BY. This is not a bug; it is an inherent consequence of the fact that SQL does not promise to deliver the results of a query in any particular order unlessORDER BY is used to constrain the order.

It is even possible for repeated executions of the sameLIMIT query to return different subsets of the rows of a table, if there is not anORDER BY to enforce selection of a deterministic subset. Again, this is not a bug; determinism of the results is simply not guaranteed in such a case.

The Locking Clause#

FOR UPDATE,FOR NO KEY UPDATE,FOR SHARE andFOR KEY SHARE arelocking clauses; they affect howSELECT locks rows as they are obtained from the table.

The locking clause has the general form

FORlock_strength [ OFfrom_reference [, ...] ] [ NOWAIT | SKIP LOCKED ]

wherelock_strength can be one of

UPDATENO KEY UPDATESHAREKEY SHARE

from_reference must be a tablealias or non-hiddentable_name referenced in theFROM clause. For more information on each row-level lock mode, refer toSection 13.3.2.

To prevent the operation from waiting for other transactions to commit, use either theNOWAIT orSKIP LOCKED option. WithNOWAIT, the statement reports an error, rather than waiting, if a selected row cannot be locked immediately. WithSKIP LOCKED, any selected rows that cannot be immediately locked are skipped. Skipping locked rows provides an inconsistent view of the data, so this is not suitable for general purpose work, but can be used to avoid lock contention with multiple consumers accessing a queue-like table. Note thatNOWAIT andSKIP LOCKED apply only to the row-level lock(s) — the requiredROW SHARE table-level lock is still taken in the ordinary way (seeChapter 13). You can useLOCK with theNOWAIT option first, if you need to acquire the table-level lock without waiting.

If specific tables are named in a locking clause, then only rows coming from those tables are locked; any other tables used in theSELECT are simply read as usual. A locking clause without a table list affects all tables used in the statement. If a locking clause is applied to a view or sub-query, it affects all tables used in the view or sub-query. However, these clauses do not apply toWITH queries referenced by the primary query. If you want row locking to occur within aWITH query, specify a locking clause within theWITH query.

Multiple locking clauses can be written if it is necessary to specify different locking behavior for different tables. If the same table is mentioned (or implicitly affected) by more than one locking clause, then it is processed as if it was only specified by the strongest one. Similarly, a table is processed asNOWAIT if that is specified in any of the clauses affecting it. Otherwise, it is processed asSKIP LOCKED if that is specified in any of the clauses affecting it.

The locking clauses cannot be used in contexts where returned rows cannot be clearly identified with individual table rows; for example they cannot be used with aggregation.

When a locking clause appears at the top level of aSELECT query, the rows that are locked are exactly those that are returned by the query; in the case of a join query, the rows locked are those that contribute to returned join rows. In addition, rows that satisfied the query conditions as of the query snapshot will be locked, although they will not be returned if they were updated after the snapshot and no longer satisfy the query conditions. If aLIMIT is used, locking stops once enough rows have been returned to satisfy the limit (but note that rows skipped over byOFFSET will get locked). Similarly, if a locking clause is used in a cursor's query, only rows actually fetched or stepped past by the cursor will be locked.

When a locking clause appears in a sub-SELECT, the rows locked are those returned to the outer query by the sub-query. This might involve fewer rows than inspection of the sub-query alone would suggest, since conditions from the outer query might be used to optimize execution of the sub-query. For example,

SELECT * FROM (SELECT * FROM mytable FOR UPDATE) ss WHERE col1 = 5;

will lock only rows havingcol1 = 5, even though that condition is not textually within the sub-query.

Previous releases failed to preserve a lock which is upgraded by a later savepoint. For example, this code:

BEGIN;SELECT * FROM mytable WHERE key = 1 FOR UPDATE;SAVEPOINT s;UPDATE mytable SET ... WHERE key = 1;ROLLBACK TO s;

would fail to preserve theFOR UPDATE lock after theROLLBACK TO. This has been fixed in release 9.3.

SELECT * FROM (SELECT * FROM mytable FOR UPDATE) ss ORDER BY column1;

TABLE Command#

The command

TABLEname

is equivalent to

SELECT * FROMname

It can be used as a top-level command or as a space-saving syntax variant in parts of complex queries. Only theWITH,UNION,INTERSECT,EXCEPT,ORDER BY,LIMIT,OFFSET,FETCH andFOR locking clauses can be used withTABLE; theWHERE clause and any form of aggregation cannot be used.

OmittedFROM Clauses

Postgres Pro allows one to omit theFROM clause. It has a straightforward use to compute the results of simple expressions:

SELECT 2+2; ?column?----------        4

Some otherSQL databases cannot do this except by introducing a dummy one-row table from which to do theSELECT.

EmptySELECT Lists

The list of output expressions afterSELECT can be empty, producing a zero-column result table. This is not valid syntax according to the SQL standard.Postgres Pro allows it to be consistent with allowing zero-column tables. However, an empty list is not allowed whenDISTINCT is used.

Omitting theAS Key Word

In the SQL standard, the optional key wordAS can be omitted before an output column name whenever the new column name is a valid column name (that is, not the same as any reserved keyword).Postgres Pro is slightly more restrictive:AS is required if the new column name matches any keyword at all, reserved or not. Recommended practice is to useAS or double-quote output column names, to prevent any possible conflict against future keyword additions.

InFROM items, both the standard andPostgres Pro allowAS to be omitted before an alias that is an unreserved keyword. But this is impractical for output column names, because of syntactic ambiguities.

Omitting Sub-SELECT Aliases inFROM

According to the SQL standard, a sub-SELECT in theFROM list must have an alias. InPostgres Pro, this alias may be omitted.

ONLY and Inheritance

The SQL standard requires parentheses around the table name when writingONLY, for exampleSELECT * FROM ONLY (tab1), ONLY (tab2) WHERE ....Postgres Pro considers these parentheses to be optional.

Postgres Pro allows a trailing* to be written to explicitly specify the non-ONLY behavior of including child tables. The standard does not allow this.

(These points apply equally to all SQL commands supporting theONLY option.)

TABLESAMPLE Clause Restrictions

TheTABLESAMPLE clause is currently accepted only on regular tables and materialized views. According to the SQL standard it should be possible to apply it to anyFROM item.

Function Calls inFROM

Postgres Pro allows a function call to be written directly as a member of theFROM list. In the SQL standard it would be necessary to wrap such a function call in a sub-SELECT; that is, the syntaxFROMfunc(...)alias is approximately equivalent toFROM LATERAL (SELECTfunc(...))alias. Note thatLATERAL is considered to be implicit; this is because the standard requiresLATERAL semantics for anUNNEST() item inFROM.Postgres Pro treatsUNNEST() the same as other set-returning functions.

Namespace Available toGROUP BY andORDER BY

In the SQL-92 standard, anORDER BY clause can only use output column names or numbers, while aGROUP BY clause can only use expressions based on input column names.Postgres Pro extends each of these clauses to allow the other choice as well (but it uses the standard's interpretation if there is ambiguity).Postgres Pro also allows both clauses to specify arbitrary expressions. Note that names appearing in an expression will always be taken as input-column names, not as output-column names.

SQL:1999 and later use a slightly different definition which is not entirely upward compatible with SQL-92. In most cases, however,Postgres Pro will interpret anORDER BY orGROUP BY expression the same way SQL:1999 does.

Functional Dependencies

Postgres Pro recognizes functional dependency (allowing columns to be omitted fromGROUP BY) only when a table's primary key is included in theGROUP BY list. The SQL standard specifies additional conditions that should be recognized.

LIMIT andOFFSET

The clausesLIMIT andOFFSET arePostgres Pro-specific syntax, also used byMySQL. The SQL:2008 standard has introduced the clausesOFFSET ... FETCH {FIRST|NEXT} ... for the same functionality, as shown above inLIMIT Clause. This syntax is also used byIBM DB2. (Applications written forOracle frequently use a workaround involving the automatically generatedrownum column, which is not available in Postgres Pro, to implement the effects of these clauses.)

FOR NO KEY UPDATE,FOR UPDATE,FOR SHARE,FOR KEY SHARE

AlthoughFOR UPDATE appears in the SQL standard, the standard allows it only as an option ofDECLARE CURSOR.Postgres Pro allows it in anySELECT query as well as in sub-SELECTs, but this is an extension. TheFOR NO KEY UPDATE,FOR SHARE andFOR KEY SHARE variants, as well as theNOWAIT andSKIP LOCKED options, do not appear in the standard.

Data-Modifying Statements inWITH

Postgres Pro allowsINSERT,UPDATE,DELETE, andMERGE to be used asWITH queries. This is not found in the SQL standard.

Nonstandard Clauses

DISTINCT ON ( ... ) is an extension of the SQL standard.

ROWS FROM( ... ) is an extension of the SQL standard.

TheMATERIALIZED andNOT MATERIALIZED options ofWITH are extensions of the SQL standard.