huxreg issues with felm object from lfe::felm

[roboton]

Great package! I'd like to report an issue using huxreg on felm objects. felm objects have no problems with broom::tidy:

tidy(mdlGroup)
term estimate std.error statistic p.value
1 groupG1 0.006795107 0.002494202 2.724361 0.0064433668
2 groupG2 0.009430680 0.002622499 3.596067 0.0003231721
3 groupG3 0.006620490 0.002064475 3.206864 0.0013421950
huxreg(mdlGroup)
Error in huxreg(mdlGroup) : Unrecognized statistics: logLik, AIC
Try setting "statistics" explicitly in the call to huxreg()
huxreg(mdlGroup, statistics=c())
Error: all(sapply(value, function(x) is.numeric(x) || is.character(x) || .... is not TRUE

But when calling huxreg on the object I get the following stack trace:

Error: all(sapply(value, function(x) is.numeric(x) || is.character(x) || .... is not TRUE
5. stop(sprintf(ngettext(length(r), "%s is not TRUE", "%s are not all TRUE"), ch), call. = FALSE, domain = NA)
4. stopifnot(all(sapply(value, function(x) is.numeric(x) || is.character(x) || is.function(x))))
3. number_format<-.huxtable(*tmp*, value = list(0, NULL, NULL, NULL, NULL, NULL, NULL, NULL))
2. number_format<-(*tmp*, value = list(0, NULL, NULL, NULL, NULL, NULL, NULL, NULL))

1. huxreg(mdlGroup, statistics = c())

[hughjonesd]

Should work now. If you want to have no statistics, do statistics = character(0), which is different from statistics = c() as c() is NULL.

[roboton]

You are a fantastic man, thanks for your help, speedy response, and great work on this! #1 is true and I have a tremendous amount of respect for you (not me) taking them on.