I agree with Enis. I think now that the number of javadoc warnings
is 0, we should not mask that. Also the new patch process message
helps mitigate the problem somewhat. Jim can this be closed?
On Aug 19, 2007, at 11:09 PM, Enis Soztutar (JIRA) wrote:
[ https://issues.apache.org/jira/browse/HADOOP-1738?
page=com.atlassian.jira.plugin.system.issuetabpanels:comment-
tabpanel#action_12521018 ]
Enis Soztutar commented on HADOOP-1738:
---------------------------------------
Before removing ant.jar from the lib, javadoc tool used to generate
no warnings. So we may want to prefer to keep the number at zero at
all times, instead.
Hadoop-Patch build should also compute the number of existing
javadoc warnings in the pre-build stage
---------------------------------------------------------------------
--------------------------------
Key: HADOOP-1738
URL: https://issues.apache.org/jira/browse/
HADOOP-1738
Project: Hadoop
Issue Type: Improvement
Components: build
Affects Versions: 0.15.0
Reporter: Jim Kellerman
Priority: Minor
Fix For: 0.15.0
The Hadoop-Patch build does a "pre-build" phase to determine the
number of javac and findbugs warnings so it can tell if the patch
generates new warnings:
########################################################
Pre-building trunk to determine current
number of javac and Findbugs warnings
########################################################
It would be nice if it also computed the number of javadoc
warnings so that a new test would not get a -1 because javadoc
warnings already exist.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
