Roy,
Obviously the packages you mention are meant to work with what sounds like large
data and would be "slow" if using such data. But a vignette for
illustration may not need to use large data files just for illustration. Is
there any reason you cannot create something small, perhaps a subset of a larger
set of data, that would run fast enough and still illustrate how it is meant to
be used?
If you wrote a package to do something like compute pi to a bazillion places,
could it be illustrated by just calculating to five decimal places? How is your
package different in being able to come up with reasonable results on reasonable
data? R already has other packages that allow you to open databases and do SQL
queries that can theoretically download more data than your machine can hold.
But an example does not need to do that just to show proof of concept.
If you feel the FULL treatment is useful, can you supply optional vignettes a
user can deliberately ask for that will not be reviewed by CRAN, alongside any
that are?
-----Original Message-----
From: R-help <r-help-bounces at r-project.org> On Behalf Of Roy
Mendelssohn - NOAA Federal via R-help
Sent: Tuesday, May 12, 2026 2:36 PM
To: r-help <r-help at r-project.org>
Subject: [R] ERDDAP? and R
ERDDAP? is a web data server/middleware that is used throughout the world (there
are well over 100 ERDDAP? servers) that provides easy access to petabytes of
data that can be subsetted and converted into many different formats and
downloaded, including graphics. At present there are three main CRAN packages
that simplify the access and use of ERDDAP? within R. These are ?rerddap?,
?rerddapXtracto? and ?plotdap?. ?rerddap? is the main package,
?rerddapXtracto? adds the ability to extract along a track or within a polygon,
and ?plotdap? simplifies the plotting of the results of the extracts. These
packages work with any ERDDAP? server. All of these packages have been
updated in the last number of months and if you use these packages it is worth
updating them.
Besides just general information, there are two things worth noting. First,
?plotdap? now can make contour plots using the ?isoband? package (and thanks to
the developers of that package), as well as simplified procedures to make
interactive plots using ?plotly? (again thanks to the developers of that
package).
Also there is a fourth R package under development, ?rerddapUtils?, which is a
collection of utility functions that work with ?rerddap?, These include a
function to make an extract only in a season defined by the user, a function to
split a very large extract into parts, so that the request is not too large, a
function to estimate the size of an ?rerddap? request as well as a function that
estimates the size of each split request, and two functions to make it easier to
work with ?rerddap? requests for projected datasets.
If this package interests you, or if you are interested in testing the package,
at present it is only available on Github (or pre-built binaries are now on
r-universe) - that is because there are a few CRAN requirements I haven?t quite
worked out (not a complaint about CRAN requirements, I just have to figure out
how to adequately reduce the runtime of the examples), and even when it is ready
for CRAN since it is a new package it will take longer to appear, as new
packages are given closer scrutiny before acceptance. To install the package
from Github:
pak::pkg_install('rmendels/rerddapUtils?)
and to install from r-universe:
install.packages('rerddapUtils', repos =
c('https://rmendels.r-universe.dev',
'https://cloud.r-project.org'))
Any comments, recommendations, bug reports etc would be greatly appreciated.
Thanks,
-Roy
Roy Mendelssohn
NOAA/NMFS
110 MCAllister Way
Santa Cruz, CA 95060
Phone: (831)-420-3666
Fax: (831) 420-3980
e-mail: Roy.Mendelssohn at noaa.gov
______________________________________________
R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide https://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.