[Gretl-devel] testcases and pet projects
cottrell at wfu.edu
Sun Jul 5 17:47:05 EDT 2009
On Sun, 5 Jul 2009, Klein, Christoph wrote:
> I've attached a badly specified mle model which provokes the
> error. I don't know whether one can find a well specified model
> which has these problems, too. The script is named
Thanks very much. Will take a look shortly.
> I've attached two more patches. The first one adds a function
> debugger to gretl... The second patch speeds up the execution of
> R functions by calling directly into libr...
These are very substantial contributions. This is excellent.
It'll take us a while to assess them properly, and gretl 1.8.2 is
pretty much ready to roll out after I leave a little more time for
the translators, but I expect we'll get both of these things into
The libR addition is something I've thought about, but never got
around to coding. It looks good, but obviously we'll have to
merge into it into the "configure" process. For the purposes of
the Windows and OS X packages, maybe we'd want to dlopen libR on
demand rather than linking to it directly? (That way we might be
able to support use of libR as a runtime option, depending on
whether or not the library is found on the host system. Although
the headers might be a problem, depending on how stable the libR
More information about the Gretl-devel