Branko Čibej wrote: > There were two options for a fix: > > dispose of the object from the native code when it goes out of scope; or, > close the stream (causing an implicit disposal) from Java code. > > Picking the former would mean that the Java code must not close the stream, > since doing so would result in a crash. Picking the latter means that > forgetting to close the stream results in a memory leak. Faced with this > dilemma, I decided that a potential memory leak with documented workaround > is the less horrible option. > > Of course it's clear that memory management in JavaHL needs a serious > overhaul, but this is not the time. Also, at some later date we may be able > to remove the requirement to close the stream whilst making closing it > always safe. Given the current JavaHL RA API, the only (Java) editor > implementation likely to exist for a while is the one within JavaHL itself > that's hidden under the hood of the status receiver.
Thanks, Brane. While there is clearly some interesting longer term work to do, and while I don't completely follow all the details, that helps understand the current situation. - Julian
