I am not sure if that gives you a hint or not but I can't explain why that fixed it. Now we are down to 2 classes left not being shaded and we have no idea why.
The relocation is a chain so order matters so sounds very possible and unrelated to this project technically (could be to maven shade to enable a kind of routing relocation - case vs for)
Hey @rmannibucau its me again.
We had this reported in Primefaces: https://github.com/primefaces/primefaces/issues/9657
It was not transforming
javax.el
tojakarta.el
even though our pom.xml has the correct transforms.https://github.com/primefaces/primefaces/blob/ae87e9e33205f8a0d108780963544fed61f86a34/primefaces/pom.xml#L938-L1168
However if I moved this line to the bottom of the list it then started shading
javax.el
correctly tojakarta.el
Previously that had been declared right above
javax.el
like this..I am not sure if that gives you a hint or not but I can't explain why that fixed it. Now we are down to 2 classes left not being shaded and we have no idea why.
PrimeApplicationContext.java (is not shading javax.validation): https://github.com/primefaces/primefaces/blob/master/primefaces/src/main/java/org/primefaces/context/PrimeApplicationContext.java
JpaLazyDataModel (is not shading javax.persistence): https://github.com/primefaces/primefaces/blob/master/primefaces/src/main/java/org/primefaces/model/JpaLazyDataModel.java
And we have no idea why or where to begin debugging?