[
https://issues.jboss.org/browse/JBIDE-24881?page=com.atlassian.jira.plugi...
]
Rob Stryker commented on JBIDE-24881:
-------------------------------------
Further info: the -i flag is likely to introduce color codes, and all sorts of other
behavior completely undesirable for what we're using it for. The only evidence we have
that the -i is needed at all is a bunch of old comments from Xavier with no explanation,
just asserting that it's required. However, since it's now causing problems, and
we have no evidence it ever was needed, removing it seems the correct solution.
Strange strings in cdk's PATH environmental variable
----------------------------------------------------
Key: JBIDE-24881
URL:
https://issues.jboss.org/browse/JBIDE-24881
Project: Tools (JBoss Tools)
Issue Type: Bug
Components: cdk
Affects Versions: 4.5.1.AM1
Reporter: Martin Malina
Assignee: Rob Stryker
Priority: Blocker
Fix For: 4.5.1.AM1
Hardy was trying to test cdk in eclipse and got an error when starting cdk:
{code}
org.xml.sax.SAXParseException; lineNumber: 5; columnNumber: 67; Character reference
"" is an invalid XML character. occurred while reading launch
configuration file:
/Volumes/Gram/Hardy/tmp/workspace/.metadata/.plugins/org.eclipse.debug.core/.launches/Container
Development Environment 3.launch.
Character reference "" is an invalid XML character.
{code}
It turns out that his PATH env var for cdk is this (from JBIDE-24222):
{code}
]1337;RemoteHost=hardy@nineveh.lan]1337;CurrentDir=/Volumes/Gram/Hardy/tmp/devstudio/studio/devstudio.app/Contents/MacOS]1337;ShellIntegrationVersion=3;shell=bash/opt/local/bin:/opt/local/sbin:/Users/hardy/.rvm/gems/ruby-2.0.0-p648/bin:/Users/hardy/.rvm/gems/ruby-2.0.0-p648@global/bin:/Users/hardy/.rvm/rubies/ruby-2.0.0-p648/bin:/opt/local/bin:/opt/local/sbin:/Users/hardy/work/developer-platform/openshift/oc/1.3.1:/opt/java/wildfly/bin:/opt/java/tomcat/bin:/opt/java/ant/bin:/opt/java/maven/bin:/opt/java/scala/bin:/opt/java/gradle/bin:/bin:/Applications/Development/Vagrant/bin:/Users/hardy/work/developer-platform/cdk-3/go/bin:/opt/local/bin:/opt/local/sbin:/Users/hardy/bin:/opt/openshift:/opt/local/lib/mariadb/bin:/Applications/Internet/Firefox.app/Contents/MacOS:/Applications/Dev/Editor/Sublime
Text.app/Contents/SharedSupport/bin:/usr/local/bin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/go/bin:/Applications/Wireshark.app/Contents/MacOS:/Library/Java/JavaVirtualMachines/jdk-9.jdk/Contents/Home/bin/:/opt/local/bin:/opt/local/sbin:/Users/hardy/.rvm/gems/ruby-2.0.0-p648/bin:/Users/hardy/.rvm/gems/ruby-2.0.0-p648@global/bin:/Users/hardy/.rvm/rubies/ruby-2.0.0-p648/bin:/Users/hardy/work/developer-platform/openshift/oc/1.3.1:/opt/java/wildfly/bin:/opt/java/tomcat/bin:/opt/java/ant/bin:/opt/java/maven/bin:/opt/java/scala/bin:/opt/java/gradle/bin:/Applications/Development/Vagrant/bin:/Users/hardy/work/developer-platform/cdk-3/go/bin:/Users/hardy/bin:/opt/openshift:/opt/local/lib/mariadb/bin:/Applications/Internet/Firefox.app/Contents/MacOS:/Applications/Dev/Editor/Sublime
Text.app/Contents/SharedSupport/bin:/Users/hardy/.rvm/bin:/Users/hardy/.rvm/bin
{code}
This was in devstudio nightly. So I checked too and I have a similar problem. In my case
I have this in the PATH variable in launch config:
{code}
Restored session: Tue Aug 22 15:34:26 CEST 2017
/Library/Java/JavaVirtualMachines/jdk1.8.0_131.jdk/Contents/Home/bin:/usr/local/bin:/usr/bin:/bin:/usr/sbin:/sbin/bin
Saving session...
...saving history...
...completed.:/Users/rasp/jbossqa/cdk/cdk-3.1.0
{code}
So something in the way you get the PATH is wrong. I would say this is definitely a
regression - I remember seeing correct PATH there in the past.
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)