STL assert while doing union for two polys

[enduguXL]

I suspect that I'm doing something fundamentally wrong.
VS2015. LLVM-vs2014 toolset has the same effect.

`

#define TYPE std::int64_t
mapbox::geometry::wagyu::wagyu<TYPE> Wagyu;

bool bRes;

//A	
mapbox::geometry::linear_ring<TYPE> RingA;

RingA.push_back({ 200986, 661584 });
RingA.push_back({ 100000, 670000 });
RingA.push_back({ 45000,  648310 });
RingA.push_back({ 45715,  125631 });

bRes = Wagyu.add_ring(RingA);

//B
mapbox::geometry::linear_ring<TYPE> RingB;

RingB.push_back({ 45000,   122869 });
RingB.push_back({ 45000,   648310 });
RingB.push_back({ -610000, 390000 });
RingB.push_back({ -650000, 220000 });
RingB.push_back({ -660000, 60000 });
RingB.push_back({ -610000, -320000 });
RingB.push_back({ -409999, -459999 });
RingB.push_back({ -50000,  -550000 });
RingB.push_back({ 150000,  -550000 });
RingB.push_back({ 470000,  -409999 });
RingB.push_back({ 730000,  -50000 });
RingB.push_back({ 750000,  80000 });
RingB.push_back({ 350000,  600000 });
RingB.push_back({ 227060,  656741 });
RingB.push_back({ 62080,   110243 });
RingB.push_back({ 45000,   115399 });
RingB.push_back({ 42933,   116023 });

bRes = Wagyu.add_ring(RingB);

//
mapbox::geometry::multi_polygon<TYPE> OutPolys;
bRes = Wagyu.execute(mapbox::geometry::wagyu::clip_type_union, OutPolys, mapbox::geometry::wagyu::fill_type_non_zero, mapbox::geometry::wagyu::fill_type_non_zero);

`

[flippmoke]

How are you compiling the code above? I am not having any issue with this code - though I may have different build flags.

https://github.com/mapbox/wagyu/blob/master/Makefile#L3

Are you using -std=c++11?

[enduguXL]

No, because this is GCC switch and I'm using both Visual Studio 2015 Compiler and Clang.
I will try to build it with GCC (on Windows) in next couple of days.

[enduguXL]

I've finally found the time to sort it out.
MinGW (GCC 6.1.0) built code doesn't crashes.
But msvs2015/Clang built does.

It produces "list iterator not incrementable" assert for the following code:

        auto bnd = active_bounds.begin();
        auto bnd_next = std::next(bnd);

(\mapbox\geometry\wagyu\intersect_util.hpp, line 82)
And active_bounds.size() == 0

[flippmoke]

@enduguXL are you using the latest from master? What geometry/test is this failing on?

[enduguXL]

Yes, I think so (my SVN path is https://github.com/mapbox/wagyu.git/trunk).
This is piece of code that I've posted in the first comment.

[flippmoke]

@enduguXL I just pushed a new fix that hopefully will resolve that issue, can you confirm?

[enduguXL]

Confirming.
This one is resolved.
But same assert triggered in the other place:

template <typename T>
void next_edge_in_bound(active_bound_list_itr<T>& bnd, scanbeam_list<T>& scanbeam) {
    ++((*bnd)->current_edge);
    ++((*bnd)->next_edge);

mapbox\geometry\wagyu\active_bound_list.hpp, line 165
(*bnd)->next_edge is NULL

Looks like the problem is the same: you are incrementing iterators but not checking them for end() / NULL prior to this.
I think if you build with clang you will get the same results.

[flippmoke]

@enduguXL I use clang, not sure why this isn't triggering on any of my builds...

[flippmoke]

another fix pushed

[enduguXL]

Are you really sure about clang? The makefile you've posted above is for GCC because compiler options there are in GCC format. For example, clang doesn't understand -std=c++11, it says "unknown argument ignored in clang-cl".

Anyway, thanks for the fix, I'll check it later today when I return home.

[flippmoke]

@enduguXL - definitely using clang and clang supports -std=c++11
http://clang.llvm.org/cxx_status.html

[enduguXL]

Looks like it's an STL assert in Microsoft's STL implementation. In Release configuration it works, producing no asserts.
So if you not supporting Microsoft Visual Studio, you can close the issue.

[flippmoke]

@enduguXL I am definitely supporting Microsoft Visual Studio, but I do not have a setup for this currently - so I would love to make sure that everything compiles properly!

Microsoft's compilers are typically stricter, so this is often a very good thing!

[enduguXL]

Sounds good!
Maybe I could fix some of these asserts while you don't have Microsoft environment. When I have time.
What is the best way to contribute? Branch? Patch?

[flippmoke]

Creating a new pull request is the easiest way by far to contribute any changes to the code.

Additionally we should look into getting wagyu to build on a windows test environment, much like we do with travis currently.

/cc @springmeyer @BergWerkGIS