Mikael Jagan

Not forgotten ... maybe I'll work on this during the **glmmTMB** hack.

Thanks, I will look today ...

I'm sorry if I am completely misunderstanding, but shouldn't those methods be calling `callGeneric` and not `callNextMethod` ... ?? Edit: Well, be that as it may, I see that there...

Minimal example: ``` > setClass("zzz", contains = "character") > setMethod("Compare", c(e1 = "zzz", e2 = "zzz"), function(e1, e2) callNextMethod(e1, getDataPart(e2))) > setMethod("Compare", c(e1 = "zzz", e2 = "ANY"), function(e1, e2)...

Indeed, I've converged on a simple patch for R-devel that unbreaks **nanotime**, so you should be in the clear pending review/adjustment by Martin.

Note that you may not see a resolution for every one of the check flavours because @mmaechler hasn't yet ported the patch (r88055) to R 4.5.0 beta. And, for whatever...

Thanks. I'll take a closer look tomorrow.

But at first glance, the PR must be incomplete, because `flint.pc.in` is not modified to ensure that `pkg-config --static --libs flint` contains options necessary and sufficient for statically linking FLINT....

A few comments for now: * You should read [this](https://people.freedesktop.org/~dbn/pkg-config-guide.html) very short guide to writing correct `.pc` files. In principle, you should be using all of the fields `Cflags`, `Requires`,...

Your `flint.pc` could contain something like: ``` Cflags: -I${includedir} @FLINT_PC_CFLAGS@ Libs: -L${libdir} -lflint Libs.private: @FLINT_PC_LIBS_PRIVATE@ Requires: Requires.private: @FLINT_PC_REQUIRES_PRIVATE@ ``` Then the basic idea for each dependency `xyz` would be: *...