Version 1.1 beta

[AngusJohnson]

I've just now updated the Beta branch. And with these latest changes I think the code is pretty solid.

This beta (ver 1.1) contains a major rewrite of code that merges touching polygons in clipping solutions.
The old code was really ugly just as it was in Clipper1, and that incidentally was the main reason I embarked on Clipper2.
Anyhow finally with ver 1.1, this merging code is miuch simpler (while still not simple) and will be much easier to maintain.
And it does a better job of merging polygons too. (Note: I don't think this could ever be perfect without massively compromising performance.) WRT Performance, there's little or no difference between Clipper2 1.0 and ver 1.1.

So far I haven't had any feedback on this beta version. So now I'm asking users to try it out, and please let me know if you do (or don't) encounter issues/bugs. Thanks A.

ps: While C++ CI testing passes on Windows machines, it's currently failing in Linux. I'm pretty sure this is just an issue with a CI test and not a problem with the Clipper2 code. Unfortunately I don't currently have a Linux machine to locate exactly where (and why) it's failing.

[reunanen]

ps: While C++ CI testing passes on Windows machines, it's currently failing in Linux.

I added printing the output on test failure even for Linux (and MacOS) builds, and it suggests that the problem is here:

[ RUN      ] Clipper2Tests.TestMultiplePolygons
/home/runner/work/Clipper2/Clipper2/CPP/Tests/TestPolygons.cpp:81: Failure
Expected equality of these values:
  count_diff
    Which is: 1
  0
/home/runner/work/Clipper2/Clipper2/CPP/Tests/TestPolygons.cpp:81: Failure
Expected equality of these values:
  count_diff
    Which is: 1
  0
/home/runner/work/Clipper2/Clipper2/CPP/Tests/TestPolygons.cpp:81: Failure
Expected equality of these values:
  count_diff
    Which is: 2
  0
/home/runner/work/Clipper2/Clipper2/CPP/Tests/TestPolygons.cpp:81: Failure
Expected equality of these values:
  count_diff
    Which is: 1
  0
[  FAILED  ] Clipper2Tests.TestMultiplePolygons (270 ms)

That still doesn't quite tell us what exact test that is coming from, but let me have a quick look into that.

[reunanen]

Added some more output, which (when cleaned up just a bit) looks like this:

/home/runner/work/Clipper2/Clipper2/CPP/Tests/TestPolygons.cpp:81: Failure
---------------------
Test number    : 128
Stored count   : 419
Measured count : 418
Stored area    : 386221
Measured area  : 386173
---------------------
Test number    : 156
Stored count   : 312
Measured count : 311
Stored area    : 441900
Measured area  : 441217
---------------------
Test number    : 163
Stored count   : 120
Measured count : 122
Stored area    : 536602
Measured area  : 536247
---------------------
Test number    : 168
Stored count   : 106
Measured count : 105
Stored area    : 491019
Measured area  : 491897
---------------------

I'll leave it to you to consider if the test criteria pertaining to those specific test numbers should just be relaxed a bit (similar to how many other tests are already treated here).

In the meanwhile, I'm running my entire application-side test suite, on both Windows and Linux, against your v1.1 beta.`

[AngusJohnson]

And I've just spotted and fixed a bug but it's bedtime so I'll upload it in the morning. 💤

[AngusJohnson]

I've just uploaded the promised bugfix (to address Issue #308).
I also addressed the C++ testing issues and it's now behaving in Windows & Linux but still not in MacOS.

[reunanen]

At the risk of stating the obvious, these are the tests that are currently erroring out on MacOS:

/Users/runner/work/Clipper2/Clipper2/CPP/Tests/TestPolygons.cpp:79: Failure
---------------------
Test number    : 168
Stored count   : 106
Measured count : 102
Stored area    : 491019
Measured area  : 491888
---------------------
Test number    : 181
Stored count   : 830
Measured count : 828
Stored area    : 54685
Measured area  : 54566
---------------------