Jacob Williams

Look into the [eiscor](https://github.com/eiscor/eiscor) algorithm. It's a somewhat large library, so not sure if it makes sense to absorb all of that code into this library. Or maybe the code...

Should refactor all the routines so that: * dummy variable names are the same * all accept the coefficients in the same order (currently, some use increasing powers, other decreasing).

Add some timing tests so we can see which methods are the fastest.

json-fortran : An updated JSON API

1

Just wanted to mention that I have forked the FSON code and significantly rewritten/improved it. It can be found at: https://github.com/jacobwilliams/json-fortran. It fixes many of the bugs found in FSON...

0.30102000

1

Is this supposed to be `log10(2.0_wp)`? That would be `0.3010299957`.

See [quadrature-fortran](https://github.com/jacobwilliams/quadrature-fortran). Can we merge that library into this one? All the routines would need to be made recursive. Could allow the method to be independently-selected for each dimension.

tanh-sinh quadrature

1

Consider adding fixed-precision versions of David H. Bailey's routines from [tquad.f90](https://github.com/jacobwilliams/MPFUN2020/blob/master/fortran-var1/tquad.f90) in MPFUN2020. * **quadts**: tanh-sinh quadrature scheme of Takahashi and Mori, for functions on a finite interval such as...

Replace linpack DGTSL with lapack DGTSV

4

Consider replacing linpack `DGTSL` with lapack `DGTSV`. This was done in the SciPy version. See: https://github.com/scipy/scipy/commit/361468a6b08e731936d761f5130799fca962acaf#diff-3bc2ad4ff04dcd0fd46bde7b0a273a6b453177a8494cbc77f08009db5cb8ad93

Need to figure out how to regenerate the remaining coefficients in `dqng`.

Real precision

6

Currently, we are using double precision. Consider having the option of: 1. exporting multiple precisions 2. specify the precision via preprocessor flag For 2, I usually would do something like...