<?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: NSXReturnThrowError</title>
	<atom:link href="http://mjtsai.com/blog/2007/01/07/nsxreturnthrowerror/feed/" rel="self" type="application/rss+xml" />
	<link>http://mjtsai.com/blog/2007/01/07/nsxreturnthrowerror/</link>
	<description></description>
	<pubDate>Sun, 07 Sep 2008 02:02:32 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.5.1</generator>
		<item>
		<title>By: rentzsch</title>
		<link>http://mjtsai.com/blog/2007/01/07/nsxreturnthrowerror/#comment-30490</link>
		<dc:creator>rentzsch</dc:creator>
		<pubDate>Mon, 08 Jan 2007 00:05:23 +0000</pubDate>
		<guid isPermaLink="false">http://mjtsai.com/blog/2007/01/07/nsxreturnthrowerror/#comment-30490</guid>
		<description>The forerunner to these macros had an "NSXSetError" macro, which I thought was OK at the time. Now going back and looking at the code, I dislike it intensely. Strangely, I'm not sure why.

While writing this new version I had "NSXMakeError" hanging around, doing pretty much what you want. However I cut it thinking nobody would care. I put it back in :-)</description>
		<content:encoded><![CDATA[<p>The forerunner to these macros had an "NSXSetError" macro, which I thought was OK at the time. Now going back and looking at the code, I dislike it intensely. Strangely, I'm not sure why.</p>
<p>While writing this new version I had "NSXMakeError" hanging around, doing pretty much what you want. However I cut it thinking nobody would care. I put it back in :-)</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Michael</title>
		<link>http://mjtsai.com/blog/2007/01/07/nsxreturnthrowerror/#comment-30468</link>
		<dc:creator>Michael</dc:creator>
		<pubDate>Sun, 07 Jan 2007 20:44:57 +0000</pubDate>
		<guid isPermaLink="false">http://mjtsai.com/blog/2007/01/07/nsxreturnthrowerror/#comment-30468</guid>
		<description>Yeah, I agree that it's a tradeoff. It might be nice to have a version that takes a name and define NSXReturnError in terms of that. I think part of the reason it bugs me is that I tend to follow the frameworks and use "error" as the name of the method's output parameter.

Also, incidentally, when I first saw "Return" in the name I thought it meant "return from the calling function" or something, since that's what the "Throw" variant does. Maybe "Store" would be clearer.</description>
		<content:encoded><![CDATA[<p>Yeah, I agree that it's a tradeoff. It might be nice to have a version that takes a name and define NSXReturnError in terms of that. I think part of the reason it bugs me is that I tend to follow the frameworks and use "error" as the name of the method's output parameter.</p>
<p>Also, incidentally, when I first saw "Return" in the name I thought it meant "return from the calling function" or something, since that's what the "Throw" variant does. Maybe "Store" would be clearer.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: rentzsch</title>
		<link>http://mjtsai.com/blog/2007/01/07/nsxreturnthrowerror/#comment-30460</link>
		<dc:creator>rentzsch</dc:creator>
		<pubDate>Sun, 07 Jan 2007 19:57:34 +0000</pubDate>
		<guid isPermaLink="false">http://mjtsai.com/blog/2007/01/07/nsxreturnthrowerror/#comment-30460</guid>
		<description>Thanks for the write-up! I'm glad you appreciated the @encode/typeof trickery.

Ooo, you're right about naming the NSException "NSError". Probably should go with NSXError or some such.

I agree implicit-assigning-to-error isn't great, but I think the tradeoff of always having to put "error" in the macro param list is worse.

Oh yeah, they should be do-while protected. I tend to forget that. I don't get burned because I run religious about always using braces with my if-style statements. The fix is in svn trunk, scheduled for 1.0.</description>
		<content:encoded><![CDATA[<p>Thanks for the write-up! I'm glad you appreciated the @encode/typeof trickery.</p>
<p>Ooo, you're right about naming the NSException "NSError". Probably should go with NSXError or some such.</p>
<p>I agree implicit-assigning-to-error isn't great, but I think the tradeoff of always having to put "error" in the macro param list is worse.</p>
<p>Oh yeah, they should be do-while protected. I tend to forget that. I don't get burned because I run religious about always using braces with my if-style statements. The fix is in svn trunk, scheduled for 1.0.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
