David,

Thanks. Unfortunately, after pulling your fix, I get the same assertion failure 
2 files further
on in the ProofPower build. The attached tarball contains files similar to the 
ones I sent
yesterday to exhibit the problem.

Regards,

Rob.

> On 12 Nov 2017, at 15:41, David Matthews <[email protected]> 
> wrote:
> 
> Rob,
> Thanks for doing that.  I've pushed a commit that seems to have fixed it.
> Regards,
> David
> 
> On 11/11/2017 18:47, Rob Arthan wrote:
>> David,
>>> On 8 Nov 2017, at 14:10, David Matthews <[email protected]> 
>>> wrote:
>>> 
>>> We are approaching the point at which the current version of Git master is 
>>> ready for release as Poly/ML 5.7.1.  Version 5.7 introduced a number of 
>>> significant changes and it has taken quite a bit of work since then to fix 
>>> various bugs and sort out performance issues.  I've been working with 
>>> Makarius on dealing with those that affect Isabelle and we now seem to have 
>>> dealt with everything.  I'd like to ask everyone to try out the current 
>>> version and let me know if there is anything that would stand in the way of 
>>> a release.
>> The ProofPower build fails with an assertion failure:
>> Assertion failed: (t->tree[r] == 0), function AddTreeRange, file 
>> savestate.cpp, line 896.
>> This is on Mac OS Sierra 10.12.6 with Poly/ML version v5.7-283-g04d3c95 .
>> I haven't tried any other OSs.  I presume this is happening where my main 
>> program calls
>> PolyML.SaveState.loadState.
>> I've attached a tarball of a cut-down set of source files that exhibits the 
>> problem
>> together with a shell script that simulates what the ProofPower make file 
>> does.
>> Regards,
>> Rob.
> _______________________________________________
> polyml mailing list
> [email protected]
> http://lists.inf.ed.ac.uk/mailman/listinfo/polyml

Attachment: fordavidm20171112.tgz
Description: Binary data


_______________________________________________
polyml mailing list
[email protected]
http://lists.inf.ed.ac.uk/mailman/listinfo/polyml

Reply via email to