<?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: Brand New Compiler Mode in IntelliJ IDEA 12 Leda</title>
	<atom:link href="http://blogs.jetbrains.com/idea/2012/06/brand-new-compiler-mode-in-intellij-idea-12-leda/feed/" rel="self" type="application/rss+xml" />
	<link>http://blogs.jetbrains.com/idea/2012/06/brand-new-compiler-mode-in-intellij-idea-12-leda/</link>
	<description>tips &#38; tricks, news, how-to's</description>
	<pubDate>Thu, 23 May 2013 00:28:05 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.6.5</generator>
		<item>
		<title>By: Jim W</title>
		<link>http://blogs.jetbrains.com/idea/2012/06/brand-new-compiler-mode-in-intellij-idea-12-leda/#comment-377072</link>
		<dc:creator>Jim W</dc:creator>
		<pubDate>Tue, 18 Dec 2012 19:56:20 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.jetbrains.com/idea/?p=7296#comment-377072</guid>
		<description>For projects with multiple modules containing source, the display of errors in the "Problems" tab populated by automake, and those files outlined in red in the project pane, only appear to take into account errors in the module where the error was originated.

For instance, if module A contains an interface that contains implementors in modules A, B, and C, and I change the interface in A to force an error with automake options turned on, the errors reported by the automake and those files outlined in red in the project pane only include the implementation in A.  However, if I open the implementors in B or C, then those errors are immediately shown in the project pane.  It's as if the automake did compile the code in modules B and C, but fails to render them to the project pane or to the errors reported by the automake.

This of course makes determining quickly all the errors produced by a non-automated change (one not created by a refactoring) more difficult to determine.

Is there some view option I'm missing, or is this a feature that is coming soon?</description>
		<content:encoded><![CDATA[<p>For projects with multiple modules containing source, the display of errors in the &#8220;Problems&#8221; tab populated by automake, and those files outlined in red in the project pane, only appear to take into account errors in the module where the error was originated.</p>
<p>For instance, if module A contains an interface that contains implementors in modules A, B, and C, and I change the interface in A to force an error with automake options turned on, the errors reported by the automake and those files outlined in red in the project pane only include the implementation in A.  However, if I open the implementors in B or C, then those errors are immediately shown in the project pane.  It&#8217;s as if the automake did compile the code in modules B and C, but fails to render them to the project pane or to the errors reported by the automake.</p>
<p>This of course makes determining quickly all the errors produced by a non-automated change (one not created by a refactoring) more difficult to determine.</p>
<p>Is there some view option I&#8217;m missing, or is this a feature that is coming soon?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: stevo</title>
		<link>http://blogs.jetbrains.com/idea/2012/06/brand-new-compiler-mode-in-intellij-idea-12-leda/#comment-314179</link>
		<dc:creator>stevo</dc:creator>
		<pubDate>Mon, 05 Nov 2012 22:34:12 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.jetbrains.com/idea/?p=7296#comment-314179</guid>
		<description>hello, i have the same issue as Lukasz.
when I remove setter that is used in other classes, nothing happens. no red underscore on the affected classes / tree. but when I’ve added the setter back, the underscore disappears in few seconds.. 

(IDEA: 122.639)

cheers, stevo</description>
		<content:encoded><![CDATA[<p>hello, i have the same issue as Lukasz.<br />
when I remove setter that is used in other classes, nothing happens. no red underscore on the affected classes / tree. but when I’ve added the setter back, the underscore disappears in few seconds.. </p>
<p>(IDEA: 122.639)</p>
<p>cheers, stevo</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Sebastiano Pilla</title>
		<link>http://blogs.jetbrains.com/idea/2012/06/brand-new-compiler-mode-in-intellij-idea-12-leda/#comment-312403</link>
		<dc:creator>Sebastiano Pilla</dc:creator>
		<pubDate>Fri, 02 Nov 2012 10:26:53 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.jetbrains.com/idea/?p=7296#comment-312403</guid>
		<description>@batto: I second Nick's point, if some code is broken then the only sensible thing to do is to remove it or to fix it. The compiler/IDE should not make this decision for the developer.</description>
		<content:encoded><![CDATA[<p>@batto: I second Nick&#8217;s point, if some code is broken then the only sensible thing to do is to remove it or to fix it. The compiler/IDE should not make this decision for the developer.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Wesslan</title>
		<link>http://blogs.jetbrains.com/idea/2012/06/brand-new-compiler-mode-in-intellij-idea-12-leda/#comment-312246</link>
		<dc:creator>Wesslan</dc:creator>
		<pubDate>Fri, 02 Nov 2012 04:41:47 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.jetbrains.com/idea/?p=7296#comment-312246</guid>
		<description>I have version 122.694 but when I go to preferences-&#62;Compiler I can't see \Use out-of-process build for project compilation\, just \Use external build\.</description>
		<content:encoded><![CDATA[<p>I have version 122.694 but when I go to preferences-&gt;Compiler I can&#8217;t see \Use out-of-process build for project compilation\, just \Use external build\.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Joakim</title>
		<link>http://blogs.jetbrains.com/idea/2012/06/brand-new-compiler-mode-in-intellij-idea-12-leda/#comment-304845</link>
		<dc:creator>Joakim</dc:creator>
		<pubDate>Wed, 24 Oct 2012 13:07:02 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.jetbrains.com/idea/?p=7296#comment-304845</guid>
		<description>Wow, I'm late to the party as I just found this. Sounds like great news, though!

How do I hook up the AspectJ bytecode weaver to this new compiler mode?</description>
		<content:encoded><![CDATA[<p>Wow, I&#8217;m late to the party as I just found this. Sounds like great news, though!</p>
<p>How do I hook up the AspectJ bytecode weaver to this new compiler mode?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Dave</title>
		<link>http://blogs.jetbrains.com/idea/2012/06/brand-new-compiler-mode-in-intellij-idea-12-leda/#comment-295799</link>
		<dc:creator>Dave</dc:creator>
		<pubDate>Wed, 10 Oct 2012 20:17:11 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.jetbrains.com/idea/?p=7296#comment-295799</guid>
		<description>I'd like to second the request for a remote compiler - being able to use the resources of a fast machine for compilation is a big benefit for us road warrior/telecommute types.

In addition, parallelizing the build (maven style or better) would be fantastic. Parallelizing the build across arbitrary remote backends would be bliss! (far quicker compile times, much better utilization of hardware in a dev team).</description>
		<content:encoded><![CDATA[<p>I&#8217;d like to second the request for a remote compiler - being able to use the resources of a fast machine for compilation is a big benefit for us road warrior/telecommute types.</p>
<p>In addition, parallelizing the build (maven style or better) would be fantastic. Parallelizing the build across arbitrary remote backends would be bliss! (far quicker compile times, much better utilization of hardware in a dev team).</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Christian</title>
		<link>http://blogs.jetbrains.com/idea/2012/06/brand-new-compiler-mode-in-intellij-idea-12-leda/#comment-293393</link>
		<dc:creator>Christian</dc:creator>
		<pubDate>Thu, 04 Oct 2012 11:19:24 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.jetbrains.com/idea/?p=7296#comment-293393</guid>
		<description>Hey guys,
I'm running 122.395 and I'm getting OOMEs from the new compilation process because it is always using the following (hardcoded?) settings

-XX:MaxPermSize=150m -XX:ReservedCodeCacheSize=64m -Xms350m -Xmx700m

The memory settings for javac at
Settings &#62; Compiler &#62; Java Compiler &#62; javac &#62; Maximum Heap Size (MB)
seem to be ignored by the new compilation process as I've set this to 2048.

is there any other way to overwrite the default memory settings?

Kind regards,
Christian</description>
		<content:encoded><![CDATA[<p>Hey guys,<br />
I&#8217;m running 122.395 and I&#8217;m getting OOMEs from the new compilation process because it is always using the following (hardcoded?) settings</p>
<p>-XX:MaxPermSize=150m -XX:ReservedCodeCacheSize=64m -Xms350m -Xmx700m</p>
<p>The memory settings for javac at<br />
Settings &gt; Compiler &gt; Java Compiler &gt; javac &gt; Maximum Heap Size (MB)<br />
seem to be ignored by the new compilation process as I&#8217;ve set this to 2048.</p>
<p>is there any other way to overwrite the default memory settings?</p>
<p>Kind regards,<br />
Christian</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Stephane Nicoll</title>
		<link>http://blogs.jetbrains.com/idea/2012/06/brand-new-compiler-mode-in-intellij-idea-12-leda/#comment-279044</link>
		<dc:creator>Stephane Nicoll</dc:creator>
		<pubDate>Sat, 04 Aug 2012 10:17:44 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.jetbrains.com/idea/?p=7296#comment-279044</guid>
		<description>The rewrite of the annotation processor management in Leda still misses a very important point: you may want to trigger annotation processor in production code and in test code.

Right now, the UI allows you to specify ONE directory where the generated source files will be stored. That's obviously not going to work in case you are in this case with a build tool, like Maven who generates this by default in two separate locations. Worse, the user experience is pretty bad since you easily get an error because you have a duplicated source file (idea generating in its own directory, maven in another).

There are also weird issue related to annotation processor. There is a sample project that reproduces these issues, please have a look at http://youtrack.jetbrains.com/issue/IDEA-60774#comment=27-366379

Thanks!</description>
		<content:encoded><![CDATA[<p>The rewrite of the annotation processor management in Leda still misses a very important point: you may want to trigger annotation processor in production code and in test code.</p>
<p>Right now, the UI allows you to specify ONE directory where the generated source files will be stored. That&#8217;s obviously not going to work in case you are in this case with a build tool, like Maven who generates this by default in two separate locations. Worse, the user experience is pretty bad since you easily get an error because you have a duplicated source file (idea generating in its own directory, maven in another).</p>
<p>There are also weird issue related to annotation processor. There is a sample project that reproduces these issues, please have a look at <a href="http://youtrack.jetbrains.com/issue/IDEA-60774#comment=27-366379" rel="nofollow">http://youtrack.jetbrains.com/issue/IDEA-60774#comment=27-366379</a></p>
<p>Thanks!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: batto</title>
		<link>http://blogs.jetbrains.com/idea/2012/06/brand-new-compiler-mode-in-intellij-idea-12-leda/#comment-276741</link>
		<dc:creator>batto</dc:creator>
		<pubDate>Tue, 24 Jul 2012 12:26:10 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.jetbrains.com/idea/?p=7296#comment-276741</guid>
		<description>@Nick

&#62; “Running with errors” feature is stupid. If the code does not run, it should be fixed; there is no point in being “intelligent” and running in half-broken way, wasting developer time by testing code which, for all practical purposes, runs non-deterministically, because it has been “patched” by the IDE.

Because some unfinished and UNREALTED code in your project doesn not compile doesn't mean that you can't compile &#38; run COMPILABLE code. I like IDEA for being good all-purpose Java IDE that lets me write code how I like. On the other hand I hate doctrine tellers like you.

I'm working on some algorithm that depends on some common code. Friend sends me his version with tests in main() that depends on same common code. So I create a class in IDEA and paste it there. I try to run it but I can't because my algorithm code is unfinished. So I add dumb things like "return null", comment that, comment whole file, ...

You enterprise folks and your doctrines...</description>
		<content:encoded><![CDATA[<p>@Nick</p>
<p>&gt; “Running with errors” feature is stupid. If the code does not run, it should be fixed; there is no point in being “intelligent” and running in half-broken way, wasting developer time by testing code which, for all practical purposes, runs non-deterministically, because it has been “patched” by the IDE.</p>
<p>Because some unfinished and UNREALTED code in your project doesn not compile doesn&#8217;t mean that you can&#8217;t compile &amp; run COMPILABLE code. I like IDEA for being good all-purpose Java IDE that lets me write code how I like. On the other hand I hate doctrine tellers like you.</p>
<p>I&#8217;m working on some algorithm that depends on some common code. Friend sends me his version with tests in main() that depends on same common code. So I create a class in IDEA and paste it there. I try to run it but I can&#8217;t because my algorithm code is unfinished. So I add dumb things like &#8220;return null&#8221;, comment that, comment whole file, &#8230;</p>
<p>You enterprise folks and your doctrines&#8230;</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Lukasz</title>
		<link>http://blogs.jetbrains.com/idea/2012/06/brand-new-compiler-mode-in-intellij-idea-12-leda/#comment-274651</link>
		<dc:creator>Lukasz</dc:creator>
		<pubDate>Wed, 18 Jul 2012 14:26:27 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.jetbrains.com/idea/?p=7296#comment-274651</guid>
		<description>Eugene I don't see my answer I've posted last week (don't know why :-( ), so again:

I've enabled "automake" option and what I expect is to see errors when a class cannot be compiled (like the above example with setter). Right now I have to switch to other application (another window) and switch back to see red underscore on the affected classes. It looks like to invoke autocompile I must take an action (like it was with autosaving sometime ago).</description>
		<content:encoded><![CDATA[<p>Eugene I don&#8217;t see my answer I&#8217;ve posted last week (don&#8217;t know why <img src='http://blogs.jetbrains.com/idea/wp-includes/images/smilies/icon_sad.gif' alt=':-(' class='wp-smiley' /> ), so again:</p>
<p>I&#8217;ve enabled &#8220;automake&#8221; option and what I expect is to see errors when a class cannot be compiled (like the above example with setter). Right now I have to switch to other application (another window) and switch back to see red underscore on the affected classes. It looks like to invoke autocompile I must take an action (like it was with autosaving sometime ago).</p>
]]></content:encoded>
	</item>
</channel>
</rss>
