There's this JIRA issue https://issues.apache.org/jira/browse/IVY-804 where the user reports a genuine bug in our typedef.properties file
where we list the "cache" property twice with different values. Once to
consider the <cache...> element as a resolver and the other to consider
the <cache...> element as the cache configurations.
This is a genuine bug and I suspect what this implies is that the "cache
resolver" has never worked in the past decade (that's when this entry
was added). Furthermore, although there's a CacheResolver which
represents this element, I don't see any references to that class in our
code and no tests either. Also, our documentation doesn't mention it
either. There's a JIRA asking for the docs to be updated to explain
Is this resolver being used and should this be supported? If it has to
be supported, from what I see, it will involve fixing this
typedef.properties and adding some kind of enhancement to our Ivy
settings parser so that it can uniquely understand what this duplicated
"cache" name means in different context. Probably won't be easy, but if
we do want to support this, I think that's what we will have to do.
On the other hand, if we don't have to support this, then I can just
remove it from that typedef.properties file.