david-cortes

Question about the tests: are they somehow connected to the tests from the GSL library? I see the examples here are exact translations of GSL's examples, but I'm wondering about...

Created pull request: https://github.com/twiecki/CythonGSL/pull/32

A small comment: I'm using the sparse modules from SciPy in one of the tests that I created, so that's a new dependency to add to the tests.

@pdet But after the PR it produces incorrect results. This returns an empty data frame: ```python import pandas as pd import duckdb df = pd.DataFrame({ "col1" : pd.Categorical(pd.Series(["a", "b", "c"]))...

In the parquet file, they are lists of strings, thus reading them should not yield float NaN.

> On further inspection, I don't think there is any bug here. > > NaN is what Pandas uses to represent NULL inside a dataframe, if you don't export this...

Among other reasons, because (a) series and dataframes returned by pandas operations would have `None` or `pd.NA` values for this kind of column, but not `np.nan`, (b) code that is...

@jameslamb : another potential obstacle for getting accepted at CRAN, and which would be good to solve anyway, is that the documentations of the package's functions do not describe the...

@kizill : Something very helpful for anyone working on potential CRAN obstacles would be to provide some build system that would allow compiling it from source within R. Right now...

@kizill : I also notice that the R wrapper doesn't use R' own exporting and registration methods for the C/C++ functions (`R_registerRoutines` + `R_useDynamicSymbols` + `.register=TRUE`), and the windows version...