Summary: | java-utils-2.eclass should set variables to empty instead of unset | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | radekd2 |
Component: | [OLD] Java | Assignee: | Java team <java> |
Status: | RESOLVED FIXED | ||
Severity: | normal | ||
Priority: | High | ||
Version: | unspecified | ||
Hardware: | x86 | ||
OS: | Linux | ||
URL: | http://forums.gentoo.org/viewtopic-t-577526-highlight-.html | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- |
Description
radekd2
2007-11-25 19:53:46 UTC
(In reply to comment #0) > CLASSPATH="/usr/share/ant-core/lib/ant.jar:/usr/share/cyrus-sasl-2/lib/cyrus-sasl.jar:/usr/share/ant-core/lib/ant-launcher.jar:/usr/share/libidn/lib/libidn-0.5.15.jar:/usr/share/pdflib-5/lib/pdflib.jar:/usr/lib/db-4.2.jar:." Please try again with CLASSPATH unset in the shell you call emerge. Looks like the "unset CLASSPATH" in our eclass doesn't work without phase hooks. [00:03] Caster seems like unsetting var isn't preserved across stages :/ [00:05] zmedico Caster: set it to empty if possible [00:07] Caster zmedico: is that new regression or again something we didn't check enough when removing our phase hooks? :) [00:08] zmedico it's not a regression. it's long standing behavior Looks like another instance of bug 189417 (dupe of 56407) which we somehow lost sight of. The eclass has had "export CLASSPATH=" for a long time now. |