
|
If you were logged in you would be able to see more operations.
|
|
|
|
Environment:
|
Windows XP, JDK 1.5
|
|
Situation: Sun's jaxrpc has a QName class but also put out a fix jar with an update to make that class compatible with JDK 1.5. In mule we missed out on the patch jar but since the .classpath generated by maven-eclipse-plugin-1.9 references the JRE_CONTAINER ahead of other jars, people who used JDK 1.5 for development missed this since a working QName was found when it really should not have been (correctly raising an error caused by the 'bad' class and the lack of the patch jar instead).
Solution and proposed fix: move the JRE_CONTAINER to the end of the dependency list, after the j:forEach block that follows in the current version.
IMHO the JRE_CONTAINER should always come last. This is unlikely to cause problems because you cannot accidentally override essential classes (e.g. java.lang), but it is essential when you do want to handle optional packages correctly (say a working QName or JMX).
|
|
Description
|
Situation: Sun's jaxrpc has a QName class but also put out a fix jar with an update to make that class compatible with JDK 1.5. In mule we missed out on the patch jar but since the .classpath generated by maven-eclipse-plugin-1.9 references the JRE_CONTAINER ahead of other jars, people who used JDK 1.5 for development missed this since a working QName was found when it really should not have been (correctly raising an error caused by the 'bad' class and the lack of the patch jar instead).
Solution and proposed fix: move the JRE_CONTAINER to the end of the dependency list, after the j:forEach block that follows in the current version.
IMHO the JRE_CONTAINER should always come last. This is unlikely to cause problems because you cannot accidentally override essential classes (e.g. java.lang), but it is essential when you do want to handle optional packages correctly (say a working QName or JMX). |
Show » |
Sort Order:
made changes - 26/Nov/05 09:28 AM
| Field |
Original Value |
New Value |
|
Fix Version/s
|
1.10
[ 11350
]
|
|
|
BTW I suspect generated files for other IDEs might have the same problem (IDEA?).