<?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"
	>
<channel>
	<title>Comments on: That Fine Code</title>
	<atom:link href="http://awads.net/wp/2006/05/17/that-fine-code/feed/" rel="self" type="application/rss+xml" />
	<link>http://awads.net/wp/2006/05/17/that-fine-code/</link>
	<description>News, views, tips and tricks on Oracle and other fun stuff</description>
	<pubDate>Fri, 08 Aug 2008 19:50:13 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.5</generator>
		<item>
		<title>By: Eddie Awad</title>
		<link>http://awads.net/wp/2006/05/17/that-fine-code/#comment-50105</link>
		<dc:creator>Eddie Awad</dc:creator>
		<pubDate>Tue, 24 Oct 2006 07:49:07 +0000</pubDate>
		<guid isPermaLink="false">http://awads.net/wp/2006/05/17/that-fine-code/#comment-50105</guid>
		<description>&lt;p&gt;Does this mean that code written in any non OO language is not good? I do not think so. However, I do see your point.&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>Does this mean that code written in any non OO language is not good? I do not think so. However, I do see your point.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: suraj</title>
		<link>http://awads.net/wp/2006/05/17/that-fine-code/#comment-50095</link>
		<dc:creator>suraj</dc:creator>
		<pubDate>Fri, 20 Oct 2006 18:31:02 +0000</pubDate>
		<guid isPermaLink="false">http://awads.net/wp/2006/05/17/that-fine-code/#comment-50095</guid>
		<description>&lt;p&gt;It is best to use coldfusion if you have to develop application very quickly. The code above is fine if it does the job.&lt;/p&gt;

&lt;p&gt;If you want to write a good code you should move to object oriented programming like java&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>It is best to use coldfusion if you have to develop application very quickly. The code above is fine if it does the job.</p>
<p>If you want to write a good code you should move to object oriented programming like java</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: J Mott</title>
		<link>http://awads.net/wp/2006/05/17/that-fine-code/#comment-20461</link>
		<dc:creator>J Mott</dc:creator>
		<pubDate>Fri, 23 Jun 2006 20:58:21 +0000</pubDate>
		<guid isPermaLink="false">http://awads.net/wp/2006/05/17/that-fine-code/#comment-20461</guid>
		<description>&lt;p&gt;Specifying column names instead of 'select *' also prevents you from having to change your data structure within the program if the table is altered (columns [re/]moved or added, etc.).  Since you're specifically naming the columns/attributes, you don't care if they're moved around.  I also agree that you shouldn't request more cols/atts than necessary.&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>Specifying column names instead of &#8217;select *&#8217; also prevents you from having to change your data structure within the program if the table is altered (columns [re/]moved or added, etc.).  Since you&#8217;re specifically naming the columns/attributes, you don&#8217;t care if they&#8217;re moved around.  I also agree that you shouldn&#8217;t request more cols/atts than necessary.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Tom Fox</title>
		<link>http://awads.net/wp/2006/05/17/that-fine-code/#comment-9013</link>
		<dc:creator>Tom Fox</dc:creator>
		<pubDate>Fri, 19 May 2006 14:14:08 +0000</pubDate>
		<guid isPermaLink="false">http://awads.net/wp/2006/05/17/that-fine-code/#comment-9013</guid>
		<description>&lt;p&gt;Eddie, I would question the reason to drag data from disk to the client if you're not going to use it.  It's a waste of disk I/O and network bandwidth for no reason at all.  Additionally, I would venture to say that the application would use less memory since it only has to store 3 columns versus all of them.  This appears to be a lightweight application, so it may not impact it as much as a larger app.&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>Eddie, I would question the reason to drag data from disk to the client if you&#8217;re not going to use it.  It&#8217;s a waste of disk I/O and network bandwidth for no reason at all.  Additionally, I would venture to say that the application would use less memory since it only has to store 3 columns versus all of them.  This appears to be a lightweight application, so it may not impact it as much as a larger app.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Eddie Awad</title>
		<link>http://awads.net/wp/2006/05/17/that-fine-code/#comment-8665</link>
		<dc:creator>Eddie Awad</dc:creator>
		<pubDate>Thu, 18 May 2006 19:51:46 +0000</pubDate>
		<guid isPermaLink="false">http://awads.net/wp/2006/05/17/that-fine-code/#comment-8665</guid>
		<description>&lt;p&gt;Herod, thanks for your example. I cannot imagine myself going back to the dialup days, I would give up using the Internet if I had to use my 56K modem.&lt;/p&gt;

&lt;p&gt;Gurdip, there is nothing good about being &lt;i&gt;crap&lt;/i&gt; anything. But I know what you mean.&lt;/p&gt;

&lt;p&gt;So, here is what I understand so far: select the columns that you &lt;i&gt;need&lt;/i&gt;, no more, no less, performance should be a secondary issue. If you need &lt;i&gt;all&lt;/i&gt; columns, it is OK to use &lt;i&gt;select *&lt;/i&gt;. For better readability, try to list the column names instead of using *.&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>Herod, thanks for your example. I cannot imagine myself going back to the dialup days, I would give up using the Internet if I had to use my 56K modem.</p>
<p>Gurdip, there is nothing good about being <i>crap</i> anything. But I know what you mean.</p>
<p>So, here is what I understand so far: select the columns that you <i>need</i>, no more, no less, performance should be a secondary issue. If you need <i>all</i> columns, it is OK to use <i>select *</i>. For better readability, try to list the column names instead of using *.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Gurdip Virdi</title>
		<link>http://awads.net/wp/2006/05/17/that-fine-code/#comment-8491</link>
		<dc:creator>Gurdip Virdi</dc:creator>
		<pubDate>Thu, 18 May 2006 11:21:52 +0000</pubDate>
		<guid isPermaLink="false">http://awads.net/wp/2006/05/17/that-fine-code/#comment-8491</guid>
		<description>&lt;p&gt;The good thing about being a crap coder is there is always a geek around to fix it!&lt;/p&gt;

&lt;p&gt;
SACK DEVELOPER
&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>The good thing about being a crap coder is there is always a geek around to fix it!</p>
<p>
SACK DEVELOPER</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Herod</title>
		<link>http://awads.net/wp/2006/05/17/that-fine-code/#comment-8381</link>
		<dc:creator>Herod</dc:creator>
		<pubDate>Thu, 18 May 2006 03:59:37 +0000</pubDate>
		<guid isPermaLink="false">http://awads.net/wp/2006/05/17/that-fine-code/#comment-8381</guid>
		<description>&lt;p&gt;I have never thought of select * or select column name being a database performance issue.&lt;/p&gt;

&lt;p&gt;But more of an all around performance issue by reducing the amount of information coming back to the client.  Network bandwidth is always a concern in our part of a world as we have entire 100 person divisions connecting over a point to point VPN to our database servers.&lt;/p&gt;

&lt;p&gt;Every little bit helps.
Quick example of what I mean is at
&lt;a href="http://herodt.googlepages.com/ed_test.txt" rel="nofollow"&gt;Spool output from 10gXE &lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Others may argue that the small number of bytes does not make a difference. Especially on a log on screen.  But what if you are processing 300-400 logons an hour? Or a query that the application only processes 1 or 2 columns on a 30 or 40 column table and runs the query every time a use opens a new page... to say see if the user has any customizations to the screen? And your server processes 3000-4000 screen openings an hour.&lt;/p&gt;

&lt;p&gt;Those bytes add up in a hurry.  But most of you who have the pleasure of gigabyte to the desktop probably will just scoff and laugh at us poor no-highspeed folks.  I say this over my TOP speed dial up of 26.4KPS. proof here &lt;a href="http://herodt.googlepages.com/dialup.jpg" rel="nofollow"&gt; Dialup &lt;/a&gt;&lt;/p&gt;

&lt;p&gt;My 2 cents, might be worth less than that.&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>I have never thought of select * or select column name being a database performance issue.</p>
<p>But more of an all around performance issue by reducing the amount of information coming back to the client.  Network bandwidth is always a concern in our part of a world as we have entire 100 person divisions connecting over a point to point VPN to our database servers.</p>
<p>Every little bit helps.<br />
Quick example of what I mean is at<br />
<a href="http://herodt.googlepages.com/ed_test.txt" rel="nofollow">Spool output from 10gXE </a></p>
<p>Others may argue that the small number of bytes does not make a difference. Especially on a log on screen.  But what if you are processing 300-400 logons an hour? Or a query that the application only processes 1 or 2 columns on a 30 or 40 column table and runs the query every time a use opens a new page&#8230; to say see if the user has any customizations to the screen? And your server processes 3000-4000 screen openings an hour.</p>
<p>Those bytes add up in a hurry.  But most of you who have the pleasure of gigabyte to the desktop probably will just scoff and laugh at us poor no-highspeed folks.  I say this over my TOP speed dial up of 26.4KPS. proof here <a href="http://herodt.googlepages.com/dialup.jpg" rel="nofollow"> Dialup </a></p>
<p>My 2 cents, might be worth less than that.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Eddie Awad</title>
		<link>http://awads.net/wp/2006/05/17/that-fine-code/#comment-8350</link>
		<dc:creator>Eddie Awad</dc:creator>
		<pubDate>Thu, 18 May 2006 02:02:29 +0000</pubDate>
		<guid isPermaLink="false">http://awads.net/wp/2006/05/17/that-fine-code/#comment-8350</guid>
		<description>&lt;p&gt;I hear you Michael. I'm sure I will end up re-writing most of the pages.&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>I hear you Michael. I&#8217;m sure I will end up re-writing most of the pages.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Michael White</title>
		<link>http://awads.net/wp/2006/05/17/that-fine-code/#comment-8344</link>
		<dc:creator>Michael White</dc:creator>
		<pubDate>Thu, 18 May 2006 01:38:52 +0000</pubDate>
		<guid isPermaLink="false">http://awads.net/wp/2006/05/17/that-fine-code/#comment-8344</guid>
		<description>&lt;p&gt;some times I use SELECT * when the real query is done by a sql view.
as far as your project to "fix" this code... sometimes it's easier to just start over than fix something that bad.&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>some times I use SELECT * when the real query is done by a sql view.<br />
as far as your project to &#8220;fix&#8221; this code&#8230; sometimes it&#8217;s easier to just start over than fix something that bad.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Eddie Awad</title>
		<link>http://awads.net/wp/2006/05/17/that-fine-code/#comment-8295</link>
		<dc:creator>Eddie Awad</dc:creator>
		<pubDate>Wed, 17 May 2006 19:54:24 +0000</pubDate>
		<guid isPermaLink="false">http://awads.net/wp/2006/05/17/that-fine-code/#comment-8295</guid>
		<description>&lt;p&gt;Thanks for the suggestions.&lt;/p&gt;

&lt;p&gt;Regarding the &lt;i&gt;select *&lt;/i&gt;, I do see the point of having the columns explicitly listed instead of just the * as a way to improve readability. But what I do not see clearly is how this impacts the performance of a query.&lt;/p&gt;

&lt;p&gt;When you have the columns explicitly listed after the select, doesn't the DB go to the data dictionary anyway when parsing and validating the query?&lt;/p&gt;

&lt;p&gt;Do you have an example that shows this performance impact when using *?&lt;/p&gt;

&lt;p&gt;By the way, the database I use is Oracle (10gXE as far as this application is concerned). In fact, the reason why I'm even touching this application is because we've decided to use Oracle 10gXE instead of MS Access.&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>Thanks for the suggestions.</p>
<p>Regarding the <i>select *</i>, I do see the point of having the columns explicitly listed instead of just the * as a way to improve readability. But what I do not see clearly is how this impacts the performance of a query.</p>
<p>When you have the columns explicitly listed after the select, doesn&#8217;t the DB go to the data dictionary anyway when parsing and validating the query?</p>
<p>Do you have an example that shows this performance impact when using *?</p>
<p>By the way, the database I use is Oracle (10gXE as far as this application is concerned). In fact, the reason why I&#8217;m even touching this application is because we&#8217;ve decided to use Oracle 10gXE instead of MS Access.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
