<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments for An Oak In The Fall</title>
	<atom:link href="http://blog.xzion.net/comments/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.xzion.net</link>
	<description>Struggling against the almighty tides of eventuality</description>
	<lastBuildDate>Tue, 12 Apr 2011 00:57:43 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.1</generator>
	<item>
		<title>Comment on Postgres GUC as Session/Transaction Variables by sycobuny</title>
		<link>http://blog.xzion.net/2011/02/04/postgres-guc-as-session-transaction-variables/comment-page-1/#comment-4209</link>
		<dc:creator>sycobuny</dc:creator>
		<pubDate>Tue, 12 Apr 2011 00:57:43 +0000</pubDate>
		<guid isPermaLink="false">http://blog.xzion.net/?p=330#comment-4209</guid>
		<description>&lt;p&gt;Thanks for the comment, and for the tip - I had no idea about the set_config() or current_setting() functions and they look pretty nice. I&#039;ll definitely replace my existing function to return the values with the current_setting() function. I&#039;ve actually left it up to my clients to directly use SET on the variables; I&#039;m unclear if there&#039;s any way to restrict users from changing variables directly, and without that there&#039;s no way for me to enforce constraints like &quot;must be formatted like user@host&quot; except by including them on the column into which the variables will eventually be inserted. Any inserts would then fail with a (hopefully helpful) error message.&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>Thanks for the comment, and for the tip &#8211; I had no idea about the set_config() or current_setting() functions and they look pretty nice. I&#8217;ll definitely replace my existing function to return the values with the current_setting() function. I&#8217;ve actually left it up to my clients to directly use SET on the variables; I&#8217;m unclear if there&#8217;s any way to restrict users from changing variables directly, and without that there&#8217;s no way for me to enforce constraints like &#8220;must be formatted like user@host&#8221; except by including them on the column into which the variables will eventually be inserted. Any inserts would then fail with a (hopefully helpful) error message.</p>]]></content:encoded>
	</item>
	<item>
		<title>Comment on Postgres GUC as Session/Transaction Variables by oscar</title>
		<link>http://blog.xzion.net/2011/02/04/postgres-guc-as-session-transaction-variables/comment-page-1/#comment-3961</link>
		<dc:creator>oscar</dc:creator>
		<pubDate>Mon, 28 Mar 2011 14:48:35 +0000</pubDate>
		<guid isPermaLink="false">http://blog.xzion.net/?p=330#comment-3961</guid>
		<description>&lt;p&gt;Hey man thanks a lot for this post!&lt;/p&gt;

&lt;p&gt;Basically I find it easier to just use the &quot;admin&quot; procedure set_config rather than SET. It can be used anywhere, both with the interactive psql client and within a stored proc. Here is the doc:&lt;/p&gt;

&lt;p&gt;http://www.postgresql.org/docs/9.0/interactive/functions-admin.html#FUNCTIONS-ADMIN-SET-TABLE&lt;/p&gt;

&lt;p&gt;The nice bonus with this method is that it is working with variable substitution whereas SET won&#039;t let you as you highlighted in your original post. Here is a code sample I have tried:&lt;/p&gt;

&lt;p&gt;...&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;IF v_username IS NOT NULL THEN

    PERFORM set_config(&#039;customaudit.username&#039;, v_username, false);
    PERFORM set_config(&#039;customaudit.user_id&#039;, &#039;&#039;&#124;&#124;p_user_id, false);
    PERFORM set_config(&#039;customaudit.action&#039;, p_procedure_name, false);

    -- Returning
    RETURN p_user_id;

THEN
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;...&lt;/p&gt;

&lt;p&gt;and I must say I am very happy since it will allow me to add an admin user responsible for any modification in the audit tables. Note that the PERFORM is here since 9.0 won&#039;t let you do SELECT yourverycleverfunction(); anymore.&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>Hey man thanks a lot for this post!</p>

<p>Basically I find it easier to just use the &#8220;admin&#8221; procedure set_config rather than SET. It can be used anywhere, both with the interactive psql client and within a stored proc. Here is the doc:</p>

<p><a href="http://www.postgresql.org/docs/9.0/interactive/functions-admin.html#FUNCTIONS-ADMIN-SET-TABLE" rel="nofollow">http://www.postgresql.org/docs/9.0/interactive/functions-admin.html#FUNCTIONS-ADMIN-SET-TABLE</a></p>

<p>The nice bonus with this method is that it is working with variable substitution whereas SET won&#8217;t let you as you highlighted in your original post. Here is a code sample I have tried:</p>

<p>&#8230;</p>

<pre><code>IF v_username IS NOT NULL THEN

    PERFORM set_config('customaudit.username', v_username, false);
    PERFORM set_config('customaudit.user_id', ''||p_user_id, false);
    PERFORM set_config('customaudit.action', p_procedure_name, false);

    -- Returning
    RETURN p_user_id;

THEN
</code></pre>

<p>&#8230;</p>

<p>and I must say I am very happy since it will allow me to add an admin user responsible for any modification in the audit tables. Note that the PERFORM is here since 9.0 won&#8217;t let you do SELECT yourverycleverfunction(); anymore.</p>]]></content:encoded>
	</item>
	<item>
		<title>Comment on Running by sycobuny</title>
		<link>http://blog.xzion.net/2011/03/04/running/comment-page-1/#comment-3845</link>
		<dc:creator>sycobuny</dc:creator>
		<pubDate>Mon, 07 Mar 2011 17:21:16 +0000</pubDate>
		<guid isPermaLink="false">http://blog.xzion.net/?p=365#comment-3845</guid>
		<description>&lt;p&gt;Thanks :-)&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>Thanks <img src='http://blog.xzion.net/wp-includes/images/smilies/icon_smile.gif' alt=':-)' class='wp-smiley' /> </p>]]></content:encoded>
	</item>
	<item>
		<title>Comment on Running by Molly</title>
		<link>http://blog.xzion.net/2011/03/04/running/comment-page-1/#comment-3841</link>
		<dc:creator>Molly</dc:creator>
		<pubDate>Sat, 05 Mar 2011 17:29:11 +0000</pubDate>
		<guid isPermaLink="false">http://blog.xzion.net/?p=365#comment-3841</guid>
		<description>&lt;p&gt;This is beautiful.  &lt;3&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>This is beautiful.  &lt;3</p>]]></content:encoded>
	</item>
	<item>
		<title>Comment on On Sexual Identity by Eric Will</title>
		<link>http://blog.xzion.net/2011/02/07/on-sexual-identity/comment-page-1/#comment-3828</link>
		<dc:creator>Eric Will</dc:creator>
		<pubDate>Thu, 03 Mar 2011 18:22:17 +0000</pubDate>
		<guid isPermaLink="false">http://blog.xzion.net/?p=346#comment-3828</guid>
		<description>&lt;p&gt;I think this is mostly right. Labels are fine for those that like them, but what it all really comes down to is who you&#039;re attracted to. I think the labels simply try to take the general group of people you&#039;re attracted to the most, and group you in with other people that are attracted to that group the most.&lt;/p&gt;

&lt;p&gt;I find myself having a range as well. I find petite girls with dark hair and dark eyes the most attractive; but, by &quot;petite&quot; I mean &quot;built modestly&quot; rather than &quot;skinny.&quot; I actually find myself preferring girls with a little bit of meat to them as opposed to girls that society terms &quot;supermodels&quot; and the psychologists term &quot;eating disorder.&quot; I think we&#039;re all just born with certain attractions programmed into us, and they can refine themselves as our lives go on. It just so happens that the vast majority of us born have this programming attuned to &quot;opposite sex of some variety,&quot; but we&#039;ve obviously seen massive &quot;mutations&quot; in that programming.&lt;/p&gt;

&lt;p&gt;At the end of the day, between two consenting adults, whomever they happen to find attractive is no one&#039;s concern but their own.&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>I think this is mostly right. Labels are fine for those that like them, but what it all really comes down to is who you&#8217;re attracted to. I think the labels simply try to take the general group of people you&#8217;re attracted to the most, and group you in with other people that are attracted to that group the most.</p>

<p>I find myself having a range as well. I find petite girls with dark hair and dark eyes the most attractive; but, by &#8220;petite&#8221; I mean &#8220;built modestly&#8221; rather than &#8220;skinny.&#8221; I actually find myself preferring girls with a little bit of meat to them as opposed to girls that society terms &#8220;supermodels&#8221; and the psychologists term &#8220;eating disorder.&#8221; I think we&#8217;re all just born with certain attractions programmed into us, and they can refine themselves as our lives go on. It just so happens that the vast majority of us born have this programming attuned to &#8220;opposite sex of some variety,&#8221; but we&#8217;ve obviously seen massive &#8220;mutations&#8221; in that programming.</p>

<p>At the end of the day, between two consenting adults, whomever they happen to find attractive is no one&#8217;s concern but their own.</p>]]></content:encoded>
	</item>
	<item>
		<title>Comment on On Sexual Identity by Tweets that mention On Sexual Identity &#124; An Oak In The Fall -- Topsy.com</title>
		<link>http://blog.xzion.net/2011/02/07/on-sexual-identity/comment-page-1/#comment-3675</link>
		<dc:creator>Tweets that mention On Sexual Identity &#124; An Oak In The Fall -- Topsy.com</dc:creator>
		<pubDate>Tue, 08 Feb 2011 14:01:38 +0000</pubDate>
		<guid isPermaLink="false">http://blog.xzion.net/?p=346#comment-3675</guid>
		<description>&lt;p&gt;[...] This post was mentioned on Twitter by Stephen Belcher, Stephen Belcher. Stephen Belcher said: Bloggage: On Sexual Identity http://bit.ly/efpiav [...]&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>[...] This post was mentioned on Twitter by Stephen Belcher, Stephen Belcher. Stephen Belcher said: Bloggage: On Sexual Identity <a href="http://bit.ly/efpiav" rel="nofollow">http://bit.ly/efpiav</a> [...]</p>]]></content:encoded>
	</item>
	<item>
		<title>Comment on On Sexual Identity by Meredith</title>
		<link>http://blog.xzion.net/2011/02/07/on-sexual-identity/comment-page-1/#comment-3672</link>
		<dc:creator>Meredith</dc:creator>
		<pubDate>Tue, 08 Feb 2011 03:17:12 +0000</pubDate>
		<guid isPermaLink="false">http://blog.xzion.net/?p=346#comment-3672</guid>
		<description>&lt;p&gt;&lt;em&gt;like&lt;/em&gt;&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p><em>like</em></p>]]></content:encoded>
	</item>
	<item>
		<title>Comment on Stupid Relationship Terminology by On Sexual Identity &#124; An Oak In The Fall</title>
		<link>http://blog.xzion.net/2009/10/11/stupid-relationship-terminology/comment-page-1/#comment-3670</link>
		<dc:creator>On Sexual Identity &#124; An Oak In The Fall</dc:creator>
		<pubDate>Tue, 08 Feb 2011 02:31:18 +0000</pubDate>
		<guid isPermaLink="false">http://blog.xzion.net/?p=27#comment-3670</guid>
		<description>&lt;p&gt;[...] once again, we are presented with a problem of nomenclature. The world of the last 20 years has had so many [...]&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>[...] once again, we are presented with a problem of nomenclature. The world of the last 20 years has had so many [...]</p>]]></content:encoded>
	</item>
	<item>
		<title>Comment on Postgres GUC as Session/Transaction Variables by rakaur</title>
		<link>http://blog.xzion.net/2011/02/04/postgres-guc-as-session-transaction-variables/comment-page-1/#comment-3668</link>
		<dc:creator>rakaur</dc:creator>
		<pubDate>Mon, 07 Feb 2011 21:16:50 +0000</pubDate>
		<guid isPermaLink="false">http://blog.xzion.net/?p=330#comment-3668</guid>
		<description>&lt;p&gt;I work here, and I don&#039;t know what any of this means.&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>I work here, and I don&#8217;t know what any of this means.</p>]]></content:encoded>
	</item>
	<item>
		<title>Comment on Postgres Session Variables &#8211; Neat. by #Postgres GUC as Session/Transaction Variables &#124; An Oak In The Fall</title>
		<link>http://blog.xzion.net/2010/02/10/postgres-session-variables-neat/comment-page-1/#comment-3642</link>
		<dc:creator>#Postgres GUC as Session/Transaction Variables &#124; An Oak In The Fall</dc:creator>
		<pubDate>Fri, 04 Feb 2011 16:24:44 +0000</pubDate>
		<guid isPermaLink="false">http://blog.xzion.net/?p=177#comment-3642</guid>
		<description>&lt;p&gt;[...] Some time ago, I wrote about session variables in PostgreSQL. I&#8217;ve been using the solution for some time to address the problem of performing a more-or-less-automatic audit trail for certain important tables when using accounts defined by the system and not the database, and it&#8217;s been working pretty well so far. However, I&#8217;ve always been concerned about the idea of potentially creating a new table for every transaction, even if it&#8217;s temporary. The database in question is a very small low-throughput system just used internally, but being inefficient just cause nobody will notice doesn&#8217;t seem like a good enough excuse. [...]&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>[...] Some time ago, I wrote about session variables in PostgreSQL. I&#8217;ve been using the solution for some time to address the problem of performing a more-or-less-automatic audit trail for certain important tables when using accounts defined by the system and not the database, and it&#8217;s been working pretty well so far. However, I&#8217;ve always been concerned about the idea of potentially creating a new table for every transaction, even if it&#8217;s temporary. The database in question is a very small low-throughput system just used internally, but being inefficient just cause nobody will notice doesn&#8217;t seem like a good enough excuse. [...]</p>]]></content:encoded>
	</item>
</channel>
</rss>

