I am processing a number of potentially large nested archive files (such
as zips/jars etc). I appear to have hit  as one of the zips contains
a nested path. Simplifying the test down to a trivial zip as per that
jira issue does show the problem - it simply throws an exception and
stops processing. However, I would like to continue processing the zip
file, perhaps logging the offending archive entry.
Using version 2.2, is there any way of overriding the default
ZipFileSystem so I can provide a patched implementation of init() so I
can avoid this issue?