| Current File : //usr/share/doc/postgresql-9.2.24/html/rules-views.html |
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<HTML
><HEAD
><TITLE
>Views and the Rule System</TITLE
><META
NAME="GENERATOR"
CONTENT="Modular DocBook HTML Stylesheet Version 1.79"><LINK
REV="MADE"
HREF="mailto:pgsql-docs@postgresql.org"><LINK
REL="HOME"
TITLE="PostgreSQL 9.2.24 Documentation"
HREF="index.html"><LINK
REL="UP"
TITLE="The Rule System"
HREF="rules.html"><LINK
REL="PREVIOUS"
TITLE="The Query Tree"
HREF="querytree.html"><LINK
REL="NEXT"
TITLE="Rules on INSERT, UPDATE, and DELETE"
HREF="rules-update.html"><LINK
REL="STYLESHEET"
TYPE="text/css"
HREF="stylesheet.css"><META
HTTP-EQUIV="Content-Type"
CONTENT="text/html; charset=ISO-8859-1"><META
NAME="creation"
CONTENT="2017-11-06T22:43:11"></HEAD
><BODY
CLASS="SECT1"
><DIV
CLASS="NAVHEADER"
><TABLE
SUMMARY="Header navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TH
COLSPAN="5"
ALIGN="center"
VALIGN="bottom"
><A
HREF="index.html"
>PostgreSQL 9.2.24 Documentation</A
></TH
></TR
><TR
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="top"
><A
TITLE="The Query Tree"
HREF="querytree.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="top"
><A
HREF="rules.html"
ACCESSKEY="U"
>Up</A
></TD
><TD
WIDTH="60%"
ALIGN="center"
VALIGN="bottom"
>Chapter 37. The Rule System</TD
><TD
WIDTH="20%"
ALIGN="right"
VALIGN="top"
><A
TITLE="Rules on INSERT, UPDATE, and DELETE"
HREF="rules-update.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
></TABLE
><HR
ALIGN="LEFT"
WIDTH="100%"></DIV
><DIV
CLASS="SECT1"
><H1
CLASS="SECT1"
><A
NAME="RULES-VIEWS"
>37.2. Views and the Rule System</A
></H1
><P
> Views in <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> are implemented
using the rule system. In fact, there is essentially no difference
between:
</P><PRE
CLASS="PROGRAMLISTING"
>CREATE VIEW myview AS SELECT * FROM mytab;</PRE
><P>
compared against the two commands:
</P><PRE
CLASS="PROGRAMLISTING"
>CREATE TABLE myview (<TT
CLASS="REPLACEABLE"
><I
>same column list as mytab</I
></TT
>);
CREATE RULE "_RETURN" AS ON SELECT TO myview DO INSTEAD
SELECT * FROM mytab;</PRE
><P>
because this is exactly what the <TT
CLASS="COMMAND"
>CREATE VIEW</TT
>
command does internally. This has some side effects. One of them
is that the information about a view in the
<SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> system catalogs is exactly
the same as it is for a table. So for the parser, there is
absolutely no difference between a table and a view. They are the
same thing: relations.</P
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="RULES-SELECT"
>37.2.1. How <TT
CLASS="COMMAND"
>SELECT</TT
> Rules Work</A
></H2
><P
> Rules <TT
CLASS="LITERAL"
>ON SELECT</TT
> are applied to all queries as the last step, even
if the command given is an <TT
CLASS="COMMAND"
>INSERT</TT
>,
<TT
CLASS="COMMAND"
>UPDATE</TT
> or <TT
CLASS="COMMAND"
>DELETE</TT
>. And they
have different semantics from rules on the other command types in that they modify the
query tree in place instead of creating a new one. So
<TT
CLASS="COMMAND"
>SELECT</TT
> rules are described first.</P
><P
> Currently, there can be only one action in an <TT
CLASS="LITERAL"
>ON SELECT</TT
> rule, and it must
be an unconditional <TT
CLASS="COMMAND"
>SELECT</TT
> action that is <TT
CLASS="LITERAL"
>INSTEAD</TT
>. This restriction was
required to make rules safe enough to open them for ordinary users, and
it restricts <TT
CLASS="LITERAL"
>ON SELECT</TT
> rules to act like views.</P
><P
> The examples for this chapter are two join views that do some
calculations and some more views using them in turn. One of the
two first views is customized later by adding rules for
<TT
CLASS="COMMAND"
>INSERT</TT
>, <TT
CLASS="COMMAND"
>UPDATE</TT
>, and
<TT
CLASS="COMMAND"
>DELETE</TT
> operations so that the final result will
be a view that behaves like a real table with some magic
functionality. This is not such a simple example to start from and
this makes things harder to get into. But it's better to have one
example that covers all the points discussed step by step rather
than having many different ones that might mix up in mind.</P
><P
>For the example, we need a little <TT
CLASS="LITERAL"
>min</TT
> function that
returns the lower of 2 integer values. We create that as:
</P><PRE
CLASS="PROGRAMLISTING"
>CREATE FUNCTION min(integer, integer) RETURNS integer AS $$
SELECT CASE WHEN $1 < $2 THEN $1 ELSE $2 END
$$ LANGUAGE SQL STRICT;</PRE
><P></P
><P
> The real tables we need in the first two rule system descriptions
are these:
</P><PRE
CLASS="PROGRAMLISTING"
>CREATE TABLE shoe_data (
shoename text, -- primary key
sh_avail integer, -- available number of pairs
slcolor text, -- preferred shoelace color
slminlen real, -- minimum shoelace length
slmaxlen real, -- maximum shoelace length
slunit text -- length unit
);
CREATE TABLE shoelace_data (
sl_name text, -- primary key
sl_avail integer, -- available number of pairs
sl_color text, -- shoelace color
sl_len real, -- shoelace length
sl_unit text -- length unit
);
CREATE TABLE unit (
un_name text, -- primary key
un_fact real -- factor to transform to cm
);</PRE
><P>
As you can see, they represent shoe-store data.</P
><P
> The views are created as:
</P><PRE
CLASS="PROGRAMLISTING"
>CREATE VIEW shoe AS
SELECT sh.shoename,
sh.sh_avail,
sh.slcolor,
sh.slminlen,
sh.slminlen * un.un_fact AS slminlen_cm,
sh.slmaxlen,
sh.slmaxlen * un.un_fact AS slmaxlen_cm,
sh.slunit
FROM shoe_data sh, unit un
WHERE sh.slunit = un.un_name;
CREATE VIEW shoelace AS
SELECT s.sl_name,
s.sl_avail,
s.sl_color,
s.sl_len,
s.sl_unit,
s.sl_len * u.un_fact AS sl_len_cm
FROM shoelace_data s, unit u
WHERE s.sl_unit = u.un_name;
CREATE VIEW shoe_ready AS
SELECT rsh.shoename,
rsh.sh_avail,
rsl.sl_name,
rsl.sl_avail,
min(rsh.sh_avail, rsl.sl_avail) AS total_avail
FROM shoe rsh, shoelace rsl
WHERE rsl.sl_color = rsh.slcolor
AND rsl.sl_len_cm >= rsh.slminlen_cm
AND rsl.sl_len_cm <= rsh.slmaxlen_cm;</PRE
><P>
The <TT
CLASS="COMMAND"
>CREATE VIEW</TT
> command for the
<TT
CLASS="LITERAL"
>shoelace</TT
> view (which is the simplest one we
have) will create a relation <TT
CLASS="LITERAL"
>shoelace</TT
> and an entry in
<TT
CLASS="STRUCTNAME"
>pg_rewrite</TT
> that tells that there is a
rewrite rule that must be applied whenever the relation <TT
CLASS="LITERAL"
>shoelace</TT
>
is referenced in a query's range table. The rule has no rule
qualification (discussed later, with the non-<TT
CLASS="COMMAND"
>SELECT</TT
> rules, since
<TT
CLASS="COMMAND"
>SELECT</TT
> rules currently cannot have them) and it is <TT
CLASS="LITERAL"
>INSTEAD</TT
>. Note
that rule qualifications are not the same as query qualifications.
The action of our rule has a query qualification.
The action of the rule is one query tree that is a copy of the
<TT
CLASS="COMMAND"
>SELECT</TT
> statement in the view creation command.</P
><DIV
CLASS="NOTE"
><BLOCKQUOTE
CLASS="NOTE"
><P
><B
>Note: </B
> The two extra range
table entries for <TT
CLASS="LITERAL"
>NEW</TT
> and <TT
CLASS="LITERAL"
>OLD</TT
> that you can see in
the <TT
CLASS="STRUCTNAME"
>pg_rewrite</TT
> entry aren't of interest
for <TT
CLASS="COMMAND"
>SELECT</TT
> rules.
</P
></BLOCKQUOTE
></DIV
><P
> Now we populate <TT
CLASS="LITERAL"
>unit</TT
>, <TT
CLASS="LITERAL"
>shoe_data</TT
>
and <TT
CLASS="LITERAL"
>shoelace_data</TT
> and run a simple query on a view:
</P><PRE
CLASS="PROGRAMLISTING"
>INSERT INTO unit VALUES ('cm', 1.0);
INSERT INTO unit VALUES ('m', 100.0);
INSERT INTO unit VALUES ('inch', 2.54);
INSERT INTO shoe_data VALUES ('sh1', 2, 'black', 70.0, 90.0, 'cm');
INSERT INTO shoe_data VALUES ('sh2', 0, 'black', 30.0, 40.0, 'inch');
INSERT INTO shoe_data VALUES ('sh3', 4, 'brown', 50.0, 65.0, 'cm');
INSERT INTO shoe_data VALUES ('sh4', 3, 'brown', 40.0, 50.0, 'inch');
INSERT INTO shoelace_data VALUES ('sl1', 5, 'black', 80.0, 'cm');
INSERT INTO shoelace_data VALUES ('sl2', 6, 'black', 100.0, 'cm');
INSERT INTO shoelace_data VALUES ('sl3', 0, 'black', 35.0 , 'inch');
INSERT INTO shoelace_data VALUES ('sl4', 8, 'black', 40.0 , 'inch');
INSERT INTO shoelace_data VALUES ('sl5', 4, 'brown', 1.0 , 'm');
INSERT INTO shoelace_data VALUES ('sl6', 0, 'brown', 0.9 , 'm');
INSERT INTO shoelace_data VALUES ('sl7', 7, 'brown', 60 , 'cm');
INSERT INTO shoelace_data VALUES ('sl8', 1, 'brown', 40 , 'inch');
SELECT * FROM shoelace;
sl_name | sl_avail | sl_color | sl_len | sl_unit | sl_len_cm
-----------+----------+----------+--------+---------+-----------
sl1 | 5 | black | 80 | cm | 80
sl2 | 6 | black | 100 | cm | 100
sl7 | 7 | brown | 60 | cm | 60
sl3 | 0 | black | 35 | inch | 88.9
sl4 | 8 | black | 40 | inch | 101.6
sl8 | 1 | brown | 40 | inch | 101.6
sl5 | 4 | brown | 1 | m | 100
sl6 | 0 | brown | 0.9 | m | 90
(8 rows)</PRE
><P>
</P
><P
> This is the simplest <TT
CLASS="COMMAND"
>SELECT</TT
> you can do on our
views, so we take this opportunity to explain the basics of view
rules. The <TT
CLASS="LITERAL"
>SELECT * FROM shoelace</TT
> was
interpreted by the parser and produced the query tree:
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT shoelace.sl_name, shoelace.sl_avail,
shoelace.sl_color, shoelace.sl_len,
shoelace.sl_unit, shoelace.sl_len_cm
FROM shoelace shoelace;</PRE
><P>
and this is given to the rule system. The rule system walks through the
range table and checks if there are rules
for any relation. When processing the range table entry for
<TT
CLASS="LITERAL"
>shoelace</TT
> (the only one up to now) it finds the
<TT
CLASS="LITERAL"
>_RETURN</TT
> rule with the query tree:
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT s.sl_name, s.sl_avail,
s.sl_color, s.sl_len, s.sl_unit,
s.sl_len * u.un_fact AS sl_len_cm
FROM shoelace old, shoelace new,
shoelace_data s, unit u
WHERE s.sl_unit = u.un_name;</PRE
><P></P
><P
> To expand the view, the rewriter simply creates a subquery range-table
entry containing the rule's action query tree, and substitutes this
range table entry for the original one that referenced the view. The
resulting rewritten query tree is almost the same as if you had typed:
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT shoelace.sl_name, shoelace.sl_avail,
shoelace.sl_color, shoelace.sl_len,
shoelace.sl_unit, shoelace.sl_len_cm
FROM (SELECT s.sl_name,
s.sl_avail,
s.sl_color,
s.sl_len,
s.sl_unit,
s.sl_len * u.un_fact AS sl_len_cm
FROM shoelace_data s, unit u
WHERE s.sl_unit = u.un_name) shoelace;</PRE
><P>
There is one difference however: the subquery's range table has two
extra entries <TT
CLASS="LITERAL"
>shoelace old</TT
> and <TT
CLASS="LITERAL"
>shoelace new</TT
>. These entries don't
participate directly in the query, since they aren't referenced by
the subquery's join tree or target list. The rewriter uses them
to store the access privilege check information that was originally present
in the range-table entry that referenced the view. In this way, the
executor will still check that the user has proper privileges to access
the view, even though there's no direct use of the view in the rewritten
query.</P
><P
> That was the first rule applied. The rule system will continue checking
the remaining range-table entries in the top query (in this example there
are no more), and it will recursively check the range-table entries in
the added subquery to see if any of them reference views. (But it
won't expand <TT
CLASS="LITERAL"
>old</TT
> or <TT
CLASS="LITERAL"
>new</TT
> — otherwise we'd have infinite recursion!)
In this example, there are no rewrite rules for <TT
CLASS="LITERAL"
>shoelace_data</TT
> or <TT
CLASS="LITERAL"
>unit</TT
>,
so rewriting is complete and the above is the final result given to
the planner.</P
><P
> Now we want to write a query that finds out for which shoes currently in the store
we have the matching shoelaces (color and length) and where the
total number of exactly matching pairs is greater or equal to two.
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT * FROM shoe_ready WHERE total_avail >= 2;
shoename | sh_avail | sl_name | sl_avail | total_avail
----------+----------+---------+----------+-------------
sh1 | 2 | sl1 | 5 | 2
sh3 | 4 | sl7 | 7 | 4
(2 rows)</PRE
><P></P
><P
> The output of the parser this time is the query tree:
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT shoe_ready.shoename, shoe_ready.sh_avail,
shoe_ready.sl_name, shoe_ready.sl_avail,
shoe_ready.total_avail
FROM shoe_ready shoe_ready
WHERE shoe_ready.total_avail >= 2;</PRE
><P>
The first rule applied will be the one for the
<TT
CLASS="LITERAL"
>shoe_ready</TT
> view and it results in the
query tree:
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT shoe_ready.shoename, shoe_ready.sh_avail,
shoe_ready.sl_name, shoe_ready.sl_avail,
shoe_ready.total_avail
FROM (SELECT rsh.shoename,
rsh.sh_avail,
rsl.sl_name,
rsl.sl_avail,
min(rsh.sh_avail, rsl.sl_avail) AS total_avail
FROM shoe rsh, shoelace rsl
WHERE rsl.sl_color = rsh.slcolor
AND rsl.sl_len_cm >= rsh.slminlen_cm
AND rsl.sl_len_cm <= rsh.slmaxlen_cm) shoe_ready
WHERE shoe_ready.total_avail >= 2;</PRE
><P>
Similarly, the rules for <TT
CLASS="LITERAL"
>shoe</TT
> and
<TT
CLASS="LITERAL"
>shoelace</TT
> are substituted into the range table of
the subquery, leading to a three-level final query tree:
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT shoe_ready.shoename, shoe_ready.sh_avail,
shoe_ready.sl_name, shoe_ready.sl_avail,
shoe_ready.total_avail
FROM (SELECT rsh.shoename,
rsh.sh_avail,
rsl.sl_name,
rsl.sl_avail,
min(rsh.sh_avail, rsl.sl_avail) AS total_avail
FROM (SELECT sh.shoename,
sh.sh_avail,
sh.slcolor,
sh.slminlen,
sh.slminlen * un.un_fact AS slminlen_cm,
sh.slmaxlen,
sh.slmaxlen * un.un_fact AS slmaxlen_cm,
sh.slunit
FROM shoe_data sh, unit un
WHERE sh.slunit = un.un_name) rsh,
(SELECT s.sl_name,
s.sl_avail,
s.sl_color,
s.sl_len,
s.sl_unit,
s.sl_len * u.un_fact AS sl_len_cm
FROM shoelace_data s, unit u
WHERE s.sl_unit = u.un_name) rsl
WHERE rsl.sl_color = rsh.slcolor
AND rsl.sl_len_cm >= rsh.slminlen_cm
AND rsl.sl_len_cm <= rsh.slmaxlen_cm) shoe_ready
WHERE shoe_ready.total_avail > 2;</PRE
><P>
</P
><P
> It turns out that the planner will collapse this tree into a
two-level query tree: the bottommost <TT
CLASS="COMMAND"
>SELECT</TT
>
commands will be <SPAN
CLASS="QUOTE"
>"pulled up"</SPAN
> into the middle
<TT
CLASS="COMMAND"
>SELECT</TT
> since there's no need to process them
separately. But the middle <TT
CLASS="COMMAND"
>SELECT</TT
> will remain
separate from the top, because it contains aggregate functions.
If we pulled those up it would change the behavior of the topmost
<TT
CLASS="COMMAND"
>SELECT</TT
>, which we don't want. However,
collapsing the query tree is an optimization that the rewrite
system doesn't have to concern itself with.
</P
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="AEN55506"
>37.2.2. View Rules in Non-<TT
CLASS="COMMAND"
>SELECT</TT
> Statements</A
></H2
><P
> Two details of the query tree aren't touched in the description of
view rules above. These are the command type and the result relation.
In fact, the command type is not needed by view rules, but the result
relation may affect the way in which the query rewriter works, because
special care needs to be taken if the result relation is a view.</P
><P
> There are only a few differences between a query tree for a
<TT
CLASS="COMMAND"
>SELECT</TT
> and one for any other
command. Obviously, they have a different command type and for a
command other than a <TT
CLASS="COMMAND"
>SELECT</TT
>, the result
relation points to the range-table entry where the result should
go. Everything else is absolutely the same. So having two tables
<TT
CLASS="LITERAL"
>t1</TT
> and <TT
CLASS="LITERAL"
>t2</TT
> with columns <TT
CLASS="LITERAL"
>a</TT
> and
<TT
CLASS="LITERAL"
>b</TT
>, the query trees for the two statements:
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT t2.b FROM t1, t2 WHERE t1.a = t2.a;
UPDATE t1 SET b = t2.b FROM t2 WHERE t1.a = t2.a;</PRE
><P>
are nearly identical. In particular:
<P
></P
></P><UL
><LI
><P
> The range tables contain entries for the tables <TT
CLASS="LITERAL"
>t1</TT
> and <TT
CLASS="LITERAL"
>t2</TT
>.
</P
></LI
><LI
><P
> The target lists contain one variable that points to column
<TT
CLASS="LITERAL"
>b</TT
> of the range table entry for table <TT
CLASS="LITERAL"
>t2</TT
>.
</P
></LI
><LI
><P
> The qualification expressions compare the columns <TT
CLASS="LITERAL"
>a</TT
> of both
range-table entries for equality.
</P
></LI
><LI
><P
> The join trees show a simple join between <TT
CLASS="LITERAL"
>t1</TT
> and <TT
CLASS="LITERAL"
>t2</TT
>.
</P
></LI
></UL
><P>
</P
><P
> The consequence is, that both query trees result in similar
execution plans: They are both joins over the two tables. For the
<TT
CLASS="COMMAND"
>UPDATE</TT
> the missing columns from <TT
CLASS="LITERAL"
>t1</TT
> are added to
the target list by the planner and the final query tree will read
as:
</P><PRE
CLASS="PROGRAMLISTING"
>UPDATE t1 SET a = t1.a, b = t2.b FROM t2 WHERE t1.a = t2.a;</PRE
><P>
and thus the executor run over the join will produce exactly the
same result set as:
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT t1.a, t2.b FROM t1, t2 WHERE t1.a = t2.a;</PRE
><P>
But there is a little problem in
<TT
CLASS="COMMAND"
>UPDATE</TT
>: the part of the executor plan that does
the join does not care what the results from the join are
meant for. It just produces a result set of rows. The fact that
one is a <TT
CLASS="COMMAND"
>SELECT</TT
> command and the other is an
<TT
CLASS="COMMAND"
>UPDATE</TT
> is handled higher up in the executor, where
it knows that this is an <TT
CLASS="COMMAND"
>UPDATE</TT
>, and it knows that
this result should go into table <TT
CLASS="LITERAL"
>t1</TT
>. But which of the rows
that are there has to be replaced by the new row?</P
><P
> To resolve this problem, another entry is added to the target list
in <TT
CLASS="COMMAND"
>UPDATE</TT
> (and also in
<TT
CLASS="COMMAND"
>DELETE</TT
>) statements: the current tuple ID
(<ACRONYM
CLASS="ACRONYM"
>CTID</ACRONYM
>).
This is a system column containing the
file block number and position in the block for the row. Knowing
the table, the <ACRONYM
CLASS="ACRONYM"
>CTID</ACRONYM
> can be used to retrieve the
original row of <TT
CLASS="LITERAL"
>t1</TT
> to be updated. After adding the
<ACRONYM
CLASS="ACRONYM"
>CTID</ACRONYM
> to the target list, the query actually looks like:
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT t1.a, t2.b, t1.ctid FROM t1, t2 WHERE t1.a = t2.a;</PRE
><P>
Now another detail of <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> enters
the stage. Old table rows aren't overwritten, and this
is why <TT
CLASS="COMMAND"
>ROLLBACK</TT
> is fast. In an <TT
CLASS="COMMAND"
>UPDATE</TT
>,
the new result row is inserted into the table (after stripping the
<ACRONYM
CLASS="ACRONYM"
>CTID</ACRONYM
>) and in the row header of the old row, which the
<ACRONYM
CLASS="ACRONYM"
>CTID</ACRONYM
> pointed to, the <TT
CLASS="LITERAL"
>cmax</TT
> and
<TT
CLASS="LITERAL"
>xmax</TT
> entries are set to the current command counter
and current transaction ID. Thus the old row is hidden, and after
the transaction commits the vacuum cleaner can eventually remove
the dead row.</P
><P
> Knowing all that, we can simply apply view rules in absolutely
the same way to any command. There is no difference.</P
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="AEN55562"
>37.2.3. The Power of Views in <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
></A
></H2
><P
> The above demonstrates how the rule system incorporates view
definitions into the original query tree. In the second example, a
simple <TT
CLASS="COMMAND"
>SELECT</TT
> from one view created a final
query tree that is a join of 4 tables (<TT
CLASS="LITERAL"
>unit</TT
> was used twice with
different names).</P
><P
> The benefit of implementing views with the rule system is,
that the planner has all
the information about which tables have to be scanned plus the
relationships between these tables plus the restrictive
qualifications from the views plus the qualifications from
the original query
in one single query tree. And this is still the situation
when the original query is already a join over views.
The planner has to decide which is
the best path to execute the query, and the more information
the planner has, the better this decision can be. And
the rule system as implemented in <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
>
ensures, that this is all information available about the query
up to that point.</P
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="RULES-VIEWS-UPDATE"
>37.2.4. Updating a View</A
></H2
><P
> What happens if a view is named as the target relation for an
<TT
CLASS="COMMAND"
>INSERT</TT
>, <TT
CLASS="COMMAND"
>UPDATE</TT
>, or
<TT
CLASS="COMMAND"
>DELETE</TT
>? Simply doing the substitutions
described above would give a query tree in which the result
relation points at a subquery range-table entry, which will not
work. Instead, the rewriter assumes that the operation will be
handled by an <TT
CLASS="LITERAL"
>INSTEAD OF</TT
> trigger on the view.
(If there is no such trigger, the executor will throw an error
when execution starts.) Rewriting works slightly differently
in this case. For <TT
CLASS="COMMAND"
>INSERT</TT
>, the rewriter does
nothing at all with the view, leaving it as the result relation
for the query. For <TT
CLASS="COMMAND"
>UPDATE</TT
> and
<TT
CLASS="COMMAND"
>DELETE</TT
>, it's still necessary to expand the
view query to produce the <SPAN
CLASS="QUOTE"
>"old"</SPAN
> rows that the command will
attempt to update or delete. So the view is expanded as normal,
but another unexpanded range-table entry is added to the query
to represent the view in its capacity as the result relation.</P
><P
> The problem that now arises is how to identify the rows to be
updated in the view. Recall that when the result relation
is a table, a special <ACRONYM
CLASS="ACRONYM"
>CTID</ACRONYM
> entry is added to the target
list to identify the physical locations of the rows to be updated.
This does not work if the result relation is a view, because a view
does not have any <ACRONYM
CLASS="ACRONYM"
>CTID</ACRONYM
>, since its rows do not have
actual physical locations. Instead, for an <TT
CLASS="COMMAND"
>UPDATE</TT
>
or <TT
CLASS="COMMAND"
>DELETE</TT
> operation, a special <TT
CLASS="LITERAL"
>wholerow</TT
>
entry is added to the target list, which expands to include all
columns from the view. The executor uses this value to supply the
<SPAN
CLASS="QUOTE"
>"old"</SPAN
> row to the <TT
CLASS="LITERAL"
>INSTEAD OF</TT
> trigger. It is
up to the trigger to work out what to update based on the old and
new row values.</P
><P
> If there are no <TT
CLASS="LITERAL"
>INSTEAD OF</TT
> triggers to update the view,
the executor will throw an error, because it cannot automatically
update a view by itself. To change this, we can define rules that
modify the behavior of <TT
CLASS="COMMAND"
>INSERT</TT
>,
<TT
CLASS="COMMAND"
>UPDATE</TT
>, and <TT
CLASS="COMMAND"
>DELETE</TT
> commands on
a view. These rules will rewrite the command, typically into a command
that updates one or more tables, rather than views. That is the topic
of the next section.</P
><P
> Note that rules are evaluated first, rewriting the original query
before it is planned and executed. Therefore, if a view has
<TT
CLASS="LITERAL"
>INSTEAD OF</TT
> triggers as well as rules on <TT
CLASS="COMMAND"
>INSERT</TT
>,
<TT
CLASS="COMMAND"
>UPDATE</TT
>, or <TT
CLASS="COMMAND"
>DELETE</TT
>, then the rules will be
evaluated first, and depending on the result, the triggers may not be
used at all.</P
></DIV
></DIV
><DIV
CLASS="NAVFOOTER"
><HR
ALIGN="LEFT"
WIDTH="100%"><TABLE
SUMMARY="Footer navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
><A
HREF="querytree.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="index.html"
ACCESSKEY="H"
>Home</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
><A
HREF="rules-update.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
>The Query Tree</TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="rules.html"
ACCESSKEY="U"
>Up</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
>Rules on <TT
CLASS="COMMAND"
>INSERT</TT
>, <TT
CLASS="COMMAND"
>UPDATE</TT
>, and <TT
CLASS="COMMAND"
>DELETE</TT
></TD
></TR
></TABLE
></DIV
></BODY
></HTML
>