mirror/postgresql
2
0
Fork 0
mirror of https://git.postgresql.org/git/postgresql.git synced 2025-02-23 19:39:53 +08:00
Code Issues Packages Projects Releases Wiki Activity

doc: expand description of how non-SELECT queries are processed

Browse Source 
The previous description of how the executor processes non-SELECT
queries was very dense, causing lack of clarity.  This expanded text
spells it out more simply.

Reported-by: fotis.koutoupas@gmail.com

Discussion: https://postgr.es/m/160912275508.676.17469511338925622905@wrigleys.postgresql.org

Backpatch-through: 9.5
This commit is contained in:
Bruce Momjian 2021-01-09 12:11:15 -05:00
parent 0c23f30fad
commit 7bf325c1fe
1 changed files with 30 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

50
doc/src/sgml/arch-dev.sgml
View File

@ -528,26 +528,36 @@
</para>
<para>
The executor mechanism is used to evaluate all four basic SQL query types:
<command>SELECT</>, <command>INSERT</>, <command>UPDATE</>, and
<command>DELETE</>. For <command>SELECT</>, the top-level executor
code only needs to send each row returned by the query plan tree off
to the client. For <command>INSERT</>, each returned row is inserted
into the target table specified for the <command>INSERT</>. This is
done in a special top-level plan node called <literal>ModifyTable</>.
(A simple
<command>INSERT ... VALUES</> command creates a trivial plan tree
consisting of a single <literal>Result</> node, which computes just one
result row, and <literal>ModifyTable</> above it to perform the insertion.
But <command>INSERT ... SELECT</> can demand the full power
of the executor mechanism.) For <command>UPDATE</>, the planner arranges
that each computed row includes all the updated column values, plus
the <firstterm>TID</> (tuple ID, or row ID) of the original target row;
this data is fed into a <literal>ModifyTable</> node, which uses the
information to create a new updated row and mark the old row deleted.
For <command>DELETE</>, the only column that is actually returned by the
plan is the TID, and the <literal>ModifyTable</> node simply uses the TID
to visit each target row and mark it deleted.
The executor mechanism is used to evaluate all four basic SQL query
types: <command>SELECT</command>, <command>INSERT</command>,
<command>UPDATE</command>, and <command>DELETE</command>.
For <command>SELECT</command>, the top-level executor code
only needs to send each row returned by the query plan tree
off to the client. <command>INSERT ... SELECT</command>,
<command>UPDATE</command>, and <command>DELETE</command>
are effectively <command>SELECT</command>s under a special
top-level plan node called <literal>ModifyTable</literal>.
</para>
<para>
<command>INSERT ... SELECT</command> feeds the rows up
to <literal>ModifyTable</literal> for insertion. For
<command>UPDATE</command>, the planner arranges that each
computed row includes all the updated column values, plus the
<firstterm>TID</firstterm> (tuple ID, or row ID) of the original
target row; this data is fed up to the <literal>ModifyTable</literal>
node, which uses the information to create a new updated row and
mark the old row deleted. For <command>DELETE</command>, the only
column that is actually returned by the plan is the TID, and the
<literal>ModifyTable</literal> node simply uses the TID to visit each
target row and mark it deleted.
</para>
<para>
A simple <command>INSERT ... VALUES</command> command creates a
trivial plan tree consisting of a single <literal>Result</literal>
node, which computes just one result row, feeding that up
to<literal>ModifyTable</literal> to perform the insertion.
</para>
</sect1>