Fabian Greimel

I've been using this branch for reading many shapefiles for many countries without problems. So I think it's mergeable. > * port the MultiPolygon tests from https://github.com/GeospatialPython/pyshp/blob/master/test_shapefile.py > * this...

As an aside, we should add a map of https://en.wikipedia.org/wiki/Baarle-Hertog to the tests. If we can handle this, what else could go wrong.

All tests pass locally. I have not yet integrated the new tests into the test suite. You will see that the old polygon tests yield a warning because they do...

Something that would be nice to have is relative paths. Right now you cannot really use them, because Pluto stores the tomls in some temporary folder. e.g. ``` dev ../MyPkg...

Did you try this on Makie#master? https://github.com/Kolaru/MathTeXEngine.jl/blob/50f8b4cbe2b5b477f1ff59b70f119ae97c49041f/prototype/prototype.jl#L162 It seems that `mathbb` etc should work on [email protected], which isn't compatible with the latest release of Makie. cc @Kolaru EDIT: I've checked...

Does plotting without latex work for you? You generally have to checkout `MakieCore#master` as well. This works for me ```julia begin using Pkg Pkg.activate(temp = true) pkg"add MakieCore#master Makie#master CairoMakie#master"...

Is that in a fresh julia session?

I just encountered the same issue. This problem occurs when reading data from Stata's `.dta` files using `StatFiles.jl`. Stata seems to work with lower precision.

I've gotten rid of the excess allocations.

The math for gridded cubic splines can be found in the Numerical Recipes for Fortran book. The relevant pages are [here](http://www.aip.de/groups/soe/local/numres/bookfpdf/f3-3.pdf). @wupeifan I am afraid that I have other priorities...