Closed dwinter closed 8 years ago
FYI on another package I am seeing new NOTEs that I never saw before:
Undefined global functions or variables:
abline axTicks axis barplot box col2rgb colorRampPalette cor.test
dbinom density dev.off dgeom gray grconvertX grconvertY hasArg image
kruskal.test layout legend lines locator loess lsf.str median mtext
optim par pdf plot plot.new plot.window points polygon quantile
read.csv read.table rect reorder rgb runif sd segments setNames tail
terrain.colors text wilcox.test write.csv write.table
Consider adding
importFrom("grDevices", "col2rgb", "colorRampPalette", "dev.off",
"gray", "pdf", "rgb", "terrain.colors")
importFrom("graphics", "abline", "axTicks", "axis", "barplot", "box",
"grconvertX", "grconvertY", "image", "layout", "legend",
"lines", "locator", "mtext", "par", "plot", "plot.new",
"plot.window", "points", "polygon", "rect", "segments",
"text")
importFrom("methods", "hasArg")
importFrom("stats", "cor.test", "dbinom", "density", "dgeom",
"kruskal.test", "loess", "median", "optim", "quantile",
"reorder", "runif", "sd", "setNames", "wilcox.test")
importFrom("utils", "lsf.str", "read.csv", "read.table", "tail",
"write.csv", "write.table")
to your NAMESPACE file (and ensure that your DESCRIPTION Imports field
contains 'methods').
This proved simple to "fix", but probably worth running check()
on all of our packages.
I'm not sure why you didn't get this NOTE earlier. It's something that was changed in R 3.2.0. You need to use importFrom even for base packages (which was not required in the past).
On Thu, May 5, 2016 at 4:48 PM, Joseph W. Brown notifications@github.com wrote:
FYI on another package I am seeing new NOTEs that I never saw before:
Undefined global functions or variables: abline axTicks axis barplot box col2rgb colorRampPalette cor.test dbinom density dev.off dgeom gray grconvertX grconvertY hasArg image kruskal.test layout legend lines locator loess lsf.str median mtext optim par pdf plot plot.new plot.window points polygon quantile read.csv read.table rect reorder rgb runif sd segments setNames tail terrain.colors text wilcox.test write.csv write.table Consider adding importFrom("grDevices", "col2rgb", "colorRampPalette", "dev.off", "gray", "pdf", "rgb", "terrain.colors") importFrom("graphics", "abline", "axTicks", "axis", "barplot", "box", "grconvertX", "grconvertY", "image", "layout", "legend", "lines", "locator", "mtext", "par", "plot", "plot.new", "plot.window", "points", "polygon", "rect", "segments", "text") importFrom("methods", "hasArg") importFrom("stats", "cor.test", "dbinom", "density", "dgeom", "kruskal.test", "loess", "median", "optim", "quantile", "reorder", "runif", "sd", "setNames", "wilcox.test") importFrom("utils", "lsf.str", "read.csv", "read.table", "tail", "write.csv", "write.table") to your NAMESPACE file (and ensure that your DESCRIPTION Imports field contains 'methods').
This proved simple to "fix", but probably worth running check() on all of our packages.
— You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub https://github.com/ropensci/rotl/issues/81#issuecomment-217273896
@fmichonneau This is what I was asking you about before: these imports were hand written before, but R CMD check
did not complain, document()
didn't generate them (obviously).
I still see the NOTE, both locally and on travis.
As pointed out by @josephwb, the newest version of R turns up a new NOTE about the meta-analysis vignette.
Since this vignette is mostly my doing the bug if my fault and I'll sort ito ut.