[ 
http://jira.codehaus.org/browse/MJAVADOC-286?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=240200#action_240200
 ] 

Herve Boutemy commented on MJAVADOC-286:
----------------------------------------

I have a question: why is a module trying to reference javadoc section of a 
module which it doesn't depend on?
since the reactor should build modules in an order that preserves dependencies 
compilation order, IMHO, when a module hasn't been previously built (with its 
javadoc), it shows that there is no dependency on it, then it can safely be 
ignored
did I miss something?

> Version 2.7 in Maven Reactor may cause infinite recursion, symptoms similar 
> to MJAVADOC-268
> -------------------------------------------------------------------------------------------
>
>                 Key: MJAVADOC-286
>                 URL: http://jira.codehaus.org/browse/MJAVADOC-286
>             Project: Maven 2.x Javadoc Plugin
>          Issue Type: Bug
>    Affects Versions: 2.6
>         Environment: Apache Maven 2.2.0 (r788681; 2009-06-26 15:04:01+0200)
> Java version: 1.6.0_13
> Java home: /home/mkleint/javatools/jdk1.6.0_13/jre
> Default locale: en_US, platform encoding: UTF-8
> OS name: "linux" version: "2.6.29.6-desktop-2mnb" arch: "i386" Family: "unix"
>            Reporter: Parag Mehta
>            Assignee: Vincent Siveton
>            Priority: Critical
>             Fix For: 2.6.1
>
>
> The getModulesLinks() method is unacceptably slow under certain conditions:
> 1. project's url is defined
> 2. one or more projects in reactor do not have any java sources and are not 
> of "pom" packaging.
> For such projects the apidocs/ output folder is never created resulting in 
> repeated invokation of a forked javadoc goal. It's more severe with high 
> number of modules in reactor and high number of modules without any java 
> sources.
> as an example checkout "hg clone https://hg.kenai.com/hg/forceten~src";
> The immediate problem is in the apidocsFile.exists() condition that 
> re-triggers the forked invokation. The attached patch fixes that. However it 
> looks suspicitions that the method is being called repeatedly for each module 
> at all. Maybe the aborting condition at the start of the method body is wrong 
> (I was not able to decypher that)
> workaround is to use 2.5 or not to specify the url in pom.xml or set the 
> detectOfflineLinks parameter to "false".

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: 
http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Reply via email to