<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<HTML
><HEAD
><TITLE
>Writing Trigger Functions in C</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 8.3.6 Documentation"
HREF="index.html"><LINK
REL="UP"
TITLE="Triggers"
HREF="triggers.html"><LINK
REL="PREVIOUS"
TITLE="Visibility of Data Changes"
HREF="trigger-datachanges.html"><LINK
REL="NEXT"
TITLE="A Complete Example"
HREF="trigger-example.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="2009-02-03T04:34:16"></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"
>PostgreSQL 8.3.6 Documentation</TH
></TR
><TR
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="top"
><A
HREF="trigger-datachanges.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="top"
><A
HREF="triggers.html"
>Fast Backward</A
></TD
><TD
WIDTH="60%"
ALIGN="center"
VALIGN="bottom"
>Chapter 35. Triggers</TD
><TD
WIDTH="10%"
ALIGN="right"
VALIGN="top"
><A
HREF="triggers.html"
>Fast Forward</A
></TD
><TD
WIDTH="10%"
ALIGN="right"
VALIGN="top"
><A
HREF="trigger-example.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
></TABLE
><HR
ALIGN="LEFT"
WIDTH="100%"></DIV
><DIV
CLASS="SECT1"
><H1
CLASS="SECT1"
><A
NAME="TRIGGER-INTERFACE"
>35.3. Writing Trigger Functions in C</A
></H1
><A
NAME="AEN42327"
></A
><P
> This section describes the low-level details of the interface to a
trigger function. This information is only needed when writing
trigger functions in C. If you are using a higher-level language then
these details are handled for you. In most cases you should consider
using a procedural language before writing your triggers in C. The
documentation of each procedural language explains how to write a
trigger in that language.
</P
><P
> Trigger functions must use the <SPAN
CLASS="QUOTE"
>"version 1"</SPAN
> function manager
interface.
</P
><P
> When a function is called by the trigger manager, it is not passed
any normal arguments, but it is passed a <SPAN
CLASS="QUOTE"
>"context"</SPAN
>
pointer pointing to a <TT
CLASS="STRUCTNAME"
>TriggerData</TT
> structure. C
functions can check whether they were called from the trigger
manager or not by executing the macro:
</P><PRE
CLASS="PROGRAMLISTING"
>CALLED_AS_TRIGGER(fcinfo)</PRE
><P>
which expands to:
</P><PRE
CLASS="PROGRAMLISTING"
>((fcinfo)->context != NULL && IsA((fcinfo)->context, TriggerData))</PRE
><P>
If this returns true, then it is safe to cast
<TT
CLASS="LITERAL"
>fcinfo->context</TT
> to type <TT
CLASS="LITERAL"
>TriggerData
*</TT
> and make use of the pointed-to
<TT
CLASS="STRUCTNAME"
>TriggerData</TT
> structure. The function must
<SPAN
CLASS="emphasis"
><I
CLASS="EMPHASIS"
>not</I
></SPAN
> alter the <TT
CLASS="STRUCTNAME"
>TriggerData</TT
>
structure or any of the data it points to.
</P
><P
> <TT
CLASS="STRUCTNAME"
>struct TriggerData</TT
> is defined in
<TT
CLASS="FILENAME"
>commands/trigger.h</TT
>:
</P><PRE
CLASS="PROGRAMLISTING"
>typedef struct TriggerData
{
NodeTag type;
TriggerEvent tg_event;
Relation tg_relation;
HeapTuple tg_trigtuple;
HeapTuple tg_newtuple;
Trigger *tg_trigger;
Buffer tg_trigtuplebuf;
Buffer tg_newtuplebuf;
} TriggerData;</PRE
><P>
where the members are defined as follows:
<P
></P
></P><DIV
CLASS="VARIABLELIST"
><DL
><DT
><TT
CLASS="STRUCTFIELD"
>type</TT
></DT
><DD
><P
> Always <TT
CLASS="LITERAL"
>T_TriggerData</TT
>.
</P
></DD
><DT
><TT
CLASS="STRUCTFIELD"
>tg_event</TT
></DT
><DD
><P
> Describes the event for which the function is called. You can use the
following macros to examine <TT
CLASS="LITERAL"
>tg_event</TT
>:
<P
></P
></P><DIV
CLASS="VARIABLELIST"
><DL
><DT
><TT
CLASS="LITERAL"
>TRIGGER_FIRED_BEFORE(tg_event)</TT
></DT
><DD
><P
> Returns true if the trigger fired before the operation.
</P
></DD
><DT
><TT
CLASS="LITERAL"
>TRIGGER_FIRED_AFTER(tg_event)</TT
></DT
><DD
><P
> Returns true if the trigger fired after the operation.
</P
></DD
><DT
><TT
CLASS="LITERAL"
>TRIGGER_FIRED_FOR_ROW(tg_event)</TT
></DT
><DD
><P
> Returns true if the trigger fired for a row-level event.
</P
></DD
><DT
><TT
CLASS="LITERAL"
>TRIGGER_FIRED_FOR_STATEMENT(tg_event)</TT
></DT
><DD
><P
> Returns true if the trigger fired for a statement-level event.
</P
></DD
><DT
><TT
CLASS="LITERAL"
>TRIGGER_FIRED_BY_INSERT(tg_event)</TT
></DT
><DD
><P
> Returns true if the trigger was fired by an <TT
CLASS="COMMAND"
>INSERT</TT
> command.
</P
></DD
><DT
><TT
CLASS="LITERAL"
>TRIGGER_FIRED_BY_UPDATE(tg_event)</TT
></DT
><DD
><P
> Returns true if the trigger was fired by an <TT
CLASS="COMMAND"
>UPDATE</TT
> command.
</P
></DD
><DT
><TT
CLASS="LITERAL"
>TRIGGER_FIRED_BY_DELETE(tg_event)</TT
></DT
><DD
><P
> Returns true if the trigger was fired by a <TT
CLASS="COMMAND"
>DELETE</TT
> command.
</P
></DD
></DL
></DIV
><P>
</P
></DD
><DT
><TT
CLASS="STRUCTFIELD"
>tg_relation</TT
></DT
><DD
><P
> A pointer to a structure describing the relation that the trigger fired for.
Look at <TT
CLASS="FILENAME"
>utils/rel.h</TT
> for details about
this structure. The most interesting things are
<TT
CLASS="LITERAL"
>tg_relation->rd_att</TT
> (descriptor of the relation
tuples) and <TT
CLASS="LITERAL"
>tg_relation->rd_rel->relname</TT
>
(relation name; the type is not <TT
CLASS="TYPE"
>char*</TT
> but
<TT
CLASS="TYPE"
>NameData</TT
>; use
<TT
CLASS="LITERAL"
>SPI_getrelname(tg_relation)</TT
> to get a <TT
CLASS="TYPE"
>char*</TT
> if you
need a copy of the name).
</P
></DD
><DT
><TT
CLASS="STRUCTFIELD"
>tg_trigtuple</TT
></DT
><DD
><P
> A pointer to the row for which the trigger was fired. This is
the row being inserted, updated, or deleted. If this trigger
was fired for an <TT
CLASS="COMMAND"
>INSERT</TT
> or
<TT
CLASS="COMMAND"
>DELETE</TT
> then this is what you should return
from the function if you don't want to replace the row with
a different one (in the case of <TT
CLASS="COMMAND"
>INSERT</TT
>) or
skip the operation.
</P
></DD
><DT
><TT
CLASS="STRUCTFIELD"
>tg_newtuple</TT
></DT
><DD
><P
> A pointer to the new version of the row, if the trigger was
fired for an <TT
CLASS="COMMAND"
>UPDATE</TT
>, and <TT
CLASS="SYMBOL"
>NULL</TT
> if
it is for an <TT
CLASS="COMMAND"
>INSERT</TT
> or a
<TT
CLASS="COMMAND"
>DELETE</TT
>. This is what you have to return
from the function if the event is an <TT
CLASS="COMMAND"
>UPDATE</TT
>
and you don't want to replace this row by a different one or
skip the operation.
</P
></DD
><DT
><TT
CLASS="STRUCTFIELD"
>tg_trigger</TT
></DT
><DD
><P
> A pointer to a structure of type <TT
CLASS="STRUCTNAME"
>Trigger</TT
>,
defined in <TT
CLASS="FILENAME"
>utils/rel.h</TT
>:
</P><PRE
CLASS="PROGRAMLISTING"
>typedef struct Trigger
{
Oid tgoid;
char *tgname;
Oid tgfoid;
int16 tgtype;
bool tgenabled;
bool tgisconstraint;
Oid tgconstrrelid;
Oid tgconstraint;
bool tgdeferrable;
bool tginitdeferred;
int16 tgnargs;
int16 tgnattr;
int16 *tgattr;
char **tgargs;
} Trigger;</PRE
><P>
where <TT
CLASS="STRUCTFIELD"
>tgname</TT
> is the trigger's name,
<TT
CLASS="STRUCTFIELD"
>tgnargs</TT
> is number of arguments in
<TT
CLASS="STRUCTFIELD"
>tgargs</TT
>, and <TT
CLASS="STRUCTFIELD"
>tgargs</TT
> is an array of
pointers to the arguments specified in the <TT
CLASS="COMMAND"
>CREATE
TRIGGER</TT
> statement. The other members are for internal use
only.
</P
></DD
><DT
><TT
CLASS="STRUCTFIELD"
>tg_trigtuplebuf</TT
></DT
><DD
><P
> The buffer containing <TT
CLASS="STRUCTFIELD"
>tg_trigtuple</TT
>, or <TT
CLASS="SYMBOL"
>InvalidBuffer</TT
> if there
is no such tuple or it is not stored in a disk buffer.
</P
></DD
><DT
><TT
CLASS="STRUCTFIELD"
>tg_newtuplebuf</TT
></DT
><DD
><P
> The buffer containing <TT
CLASS="STRUCTFIELD"
>tg_newtuple</TT
>, or <TT
CLASS="SYMBOL"
>InvalidBuffer</TT
> if there
is no such tuple or it is not stored in a disk buffer.
</P
></DD
></DL
></DIV
><P>
</P
><P
> A trigger function must return either a
<TT
CLASS="STRUCTNAME"
>HeapTuple</TT
> pointer or a <TT
CLASS="SYMBOL"
>NULL</TT
> pointer
(<SPAN
CLASS="emphasis"
><I
CLASS="EMPHASIS"
>not</I
></SPAN
> an SQL null value, that is, do not set <TT
CLASS="PARAMETER"
>isNull</TT
> true).
Be careful to return either
<TT
CLASS="STRUCTFIELD"
>tg_trigtuple</TT
> or <TT
CLASS="STRUCTFIELD"
>tg_newtuple</TT
>,
as appropriate, if you don't want to modify the row being operated on.
</P
></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="trigger-datachanges.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="trigger-example.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
>Visibility of Data Changes</TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="triggers.html"
ACCESSKEY="U"
>Up</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
>A Complete Example</TD
></TR
></TABLE
></DIV
></BODY
></HTML
>
distrib
>
Mandriva
>
2008.1
>
x86_64
>
media
>
main-testing
>
by-pkgid
>
bab02a23fa9f3df8d66a9a3231b50245
>
files
>
857
