<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/2.3.3" -->
<rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	>
<channel>
	<title>Comments on: Cycling</title>
	<link>http://www.86db1.com/cycling</link>
	<description>IBM RPG and Databases</description>
	<pubDate>Thu, 09 Sep 2010 20:51:10 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.3.3</generator>
		<item>
		<title>By: Aketi Jyuuzou</title>
		<link>http://www.86db1.com/cycling#comment-91</link>
		<dc:creator>Aketi Jyuuzou</dc:creator>
		<pubDate>Wed, 11 Feb 2009 03:02:24 +0000</pubDate>
		<guid>http://www.86db1.com/cycling#comment-91</guid>
		<description>&lt;p&gt;Wow this is a great solution.
I have seen this DBMS_RANDOM usage in US-OTN-SQL-forum.
However I did not realize that “Connect by nocycle” can be emulated.&lt;/p&gt;
&lt;p&gt;I have researched and tested that on Oracle9i.
Then I derived version like below  &lt;/p&gt;
&lt;p&gt;http://oraclesqlpuzzle.hp.infoseek.co.jp/8-43.html
Point is “and prior dbms_random.value != 1;” ,
Because dbms_random.value returns 0 or greater then 0 and lower then 1
&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>Wow this is a great solution.<br />
I have seen this DBMS_RANDOM usage in US-OTN-SQL-forum.<br />
However I did not realize that “Connect by nocycle” can be emulated.</p>
<p>I have researched and tested that on Oracle9i.<br />
Then I derived version like below  </p>
<p><a href="http://oraclesqlpuzzle.hp.infoseek.co.jp/8-43.html" rel="nofollow">http://oraclesqlpuzzle.hp.infoseek.co.jp/8-43.html</a><br />
Point is “and prior dbms_random.value != 1;” ,<br />
Because dbms_random.value returns 0 or greater then 0 and lower then 1</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Laurent Schneider</title>
		<link>http://www.86db1.com/cycling#comment-92</link>
		<dc:creator>Laurent Schneider</dc:creator>
		<pubDate>Wed, 11 Feb 2009 01:02:25 +0000</pubDate>
		<guid>http://www.86db1.com/cycling#comment-92</guid>
		<description>&lt;p&gt;Frank Zhou gave me the DBMS_RANDOM idea a long time ago, now I find a case to use it 
&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>Frank Zhou gave me the DBMS_RANDOM idea a long time ago, now I find a case to use it</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Laurent Schneider</title>
		<link>http://www.86db1.com/cycling#comment-90</link>
		<dc:creator>Laurent Schneider</dc:creator>
		<pubDate>Tue, 10 Feb 2009 06:02:51 +0000</pubDate>
		<guid>http://www.86db1.com/cycling#comment-90</guid>
		<description>&lt;p&gt;:) good catch with != 1, it makes it 100% accurate instead of 99.999999999999999999999999999999999999%
&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p> <img src='http://www.86db1.com/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /> good catch with != 1, it makes it 100% accurate instead of 99.999999999999999999999999999999999999%</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Karl Reitschuster</title>
		<link>http://www.86db1.com/cycling#comment-89</link>
		<dc:creator>Karl Reitschuster</dc:creator>
		<pubDate>Tue, 10 Feb 2009 05:02:42 +0000</pubDate>
		<guid>http://www.86db1.com/cycling#comment-89</guid>
		<description>&lt;p&gt;with SYS_GUID() you even have no PL/SQL context switch in your SQL - Karl&lt;/p&gt;
&lt;p&gt;

SELECT PARENT,
       Child
  FROM Lsc_t
 WHERE LEVEL = 50
CONNECT BY PARENT = PRIOR Child
       AND LEVEL &lt;= 50
       AND PRIOR sys_guid() != sys_guid();


&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>with SYS_GUID() you even have no PL/SQL context switch in your SQL - Karl</p>
<p>SELECT PARENT,<br />
       Child<br />
  FROM Lsc_t<br />
 WHERE LEVEL = 50<br />
CONNECT BY PARENT = PRIOR Child<br />
       AND LEVEL < = 50<br />
       AND PRIOR sys_guid() != sys_guid();</p></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Laurent Schneider</title>
		<link>http://www.86db1.com/cycling#comment-88</link>
		<dc:creator>Laurent Schneider</dc:creator>
		<pubDate>Tue, 10 Feb 2009 05:02:07 +0000</pubDate>
		<guid>http://www.86db1.com/cycling#comment-88</guid>
		<description>&lt;p&gt;Yes, excellent&lt;/p&gt;
&lt;p&gt;also possible with &lt;/p&gt;
&lt;p&gt;and prior sys_guid() is not null
&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>Yes, excellent</p>
<p>also possible with </p>
<p>and prior sys_guid() is not null</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Brian Tkatch</title>
		<link>http://www.86db1.com/cycling#comment-87</link>
		<dc:creator>Brian Tkatch</dc:creator>
		<pubDate>Tue, 10 Feb 2009 04:02:04 +0000</pubDate>
		<guid>http://www.86db1.com/cycling#comment-87</guid>
		<description>&lt;p&gt;Why does this work?
&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>Why does this work?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Laurent Schneider</title>
		<link>http://www.86db1.com/cycling#comment-86</link>
		<dc:creator>Laurent Schneider</dc:creator>
		<pubDate>Mon, 09 Feb 2009 10:02:44 +0000</pubDate>
		<guid>http://www.86db1.com/cycling#comment-86</guid>
		<description>&lt;p&gt;well, in my humble opinion it is not supposed to work… It probably does not in oracle7 or 8 (anyone can check?)&lt;/p&gt;
&lt;p&gt;it is very similar to the connect by without prior row generator
select * from dual connect by level</description>
		<content:encoded><![CDATA[<p>well, in my humble opinion it is not supposed to work… It probably does not in oracle7 or 8 (anyone can check?)</p>
<p>it is very similar to the connect by without prior row generator<br />
select * from dual connect by level</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Brian Tkatch</title>
		<link>http://www.86db1.com/cycling#comment-85</link>
		<dc:creator>Brian Tkatch</dc:creator>
		<pubDate>Mon, 09 Feb 2009 06:02:26 +0000</pubDate>
		<guid>http://www.86db1.com/cycling#comment-85</guid>
		<description>&lt;p&gt;Dores PRIOR remeber the earlier calculated value instead of doing it a second time?
&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>Dores PRIOR remeber the earlier calculated value instead of doing it a second time?</p>
]]></content:encoded>
	</item>
</channel>
</rss>
