-
Notifications
You must be signed in to change notification settings - Fork 236
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Misc fixes found when running package checks on newer releases of R #3237
Conversation
@@ -30,7 +30,7 @@ mcmc.list2init <- function(dat) { | |||
|
|||
## define variables | |||
ic <- list() | |||
n <- nrow(dat[[1]]) | |||
nr <- nrow(dat[[1]]) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
surprised this was not found earlier.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Me too! It looks like the typo ought to make this function return all samples instead of just the last ones, which seems like it ought to be pretty obvious. And yet...!
Needed to avoid being bitten by tidyverse/dbplyr#1240 on R 4.2 image
This comment was marked as outdated.
This comment was marked as outdated.
@infotroph there are merge conflicts that need to be resolved in order to pull this one in |
This comment was marked as outdated.
This comment was marked as outdated.
not sure why it just now started complaining about these...
64c8877
to
5647ae8
Compare
R CMD check complains that data directory is large and that json is a type not allowed in data/. Moved ecoregion JSONs to inst/extdata; resaved BADM.csv (2 MB) as BADM.rda (74 KB!)
It is not used in this package or elsewhere, so does not need to be installed with the package (especially since it weighs >2 MB and triggers check warnings about installed package size.)
8472eda
to
2111b1e
Compare
ef3b66f
to
e9eea4a
Compare
e16174c
to
b25b6eb
Compare
@@ -12,6 +12,6 @@ generalized_plate_model <- function(k, refractive, N) { | |||
length(N) == 1 | |||
) | |||
RT <- matrix(0, 2101, 2) | |||
outlist <- .Fortran("gpm", k, refractive, N, RT) | |||
outlist <- .Fortran("gpm", k, refractive, N, RT, PACKAGE="PEcAnRTM") |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@ashiklom are you still lurking in the PEcAn world? Would it be possible to get your eyes on this commit?
TL;DR: Does this patch look OK to you?
This is a proposed fix for intermittent test failures from RTM that have me puzzled for three reasons:
- Tests are only failing on R 4.2, not on older or newer R versions.
- Failure is traceable to .Fortran calls to
foursail
sometimes, but not always, returning NaNs. Repeated calls with the same input fail maybe 2/3 of the time but succeed often enough to make me nervous. - Adding the PACKAGE argument to each .Fortran call seems to fix it, which would seem to imply some kind of namespace collision... but I don't know what other package would contain a
foursail
symbol or why symbol lookup would behave nondeterministically from one invocation to another.
All checks are passing, so I'll merge this today unless anyone speaks up before then. Substantive changes since the last review:
|
Description
Motivation and Context
Review Time Estimate
Types of changes
Checklist: