[m-dev.] compiling minizinc tools with java grade

Julien Fischer juliensf at csse.unimelb.edu.au
Tue Jun 2 22:53:37 AEST 2009

On Tue, 2 Jun 2009, Peter Wang wrote:

> On 2009-06-02, Julien Fischer <juliensf at csse.unimelb.edu.au> wrote:
>> minizinc_to_flatzinc.java is 66,833 lines of Java generated from
>> about 9400 lines of Mercury.  Searching for the above error message
>> on the web indicates that this is a result of the class file format
>> limiting each method to 64Kb of bytecode.
> I guess we'll have to break up large methods somehow.  But 9400 lines of
> Mercury is.. I assume it's generated, so the generator can be changed.

Now that the Erlang backend is working again I've managed to also
compile** the MiniZinc tools in the erlang grade (minizinc_to_flatzinc.m
is ~27,000 lines of Erlang if anyone is wondering.)

Furthermore, the tools also work in the Erlang grade (at based on the
rather limited testing I've done so far.)


** actually things went a bit wrong when compiling one of the Zinc
frontend modules, but since the code in question isn't actually useful
aside from for debugging the tool itself I just commented it out - I'll
post a bug report concerning it later.
mercury-developers mailing list
Post messages to:       mercury-developers at csse.unimelb.edu.au
Administrative Queries: owner-mercury-developers at csse.unimelb.edu.au
Subscriptions:          mercury-developers-request at csse.unimelb.edu.au

More information about the developers mailing list