throw() from R.oo instead of R.methodsS3._) in Rd files.Added package tests (from examples).
Importing suggested functions using ::.
Declaring all S3 methods.
Bumped package dependencies.
Now celToApd() throws a more informative error message if it
fails to located an APD map file.
Now the aroma.apd Package object is also available when the
package is only loaded (but not attached).
Now writeApd() is atomic, which is achieved by first writing to a
temporary file which is then renamed upon completion.
celToApd() assumed that affxparser was already attached.Removed unnecessary import statements from NAMESPACE.
Some of the examples assumed that the suggested affxparser was installed and attached.
Bumped up package version requirements.
cdfToApdMap() did not validate and assign argument verbose.
Forgot to import trim() from R.oo.
Removed fallback attachments of R.utils and R.huge as these are no longer needed with R.oo (>= 1.15.1).
Package no longer utilizes import(), only importFrom():s.
Now only using Authors@R in DESCRIPTION, which is possible since
R (>= 2.14.0).
Dropped obsolete R (< 2.0.0) code that was never used.
Package no longer attaches/loads R.utils and R.huge (only imports them). However, they will be attached dynamically, when certain methods are called.
Utilizing new startupMessage() of R.oo.
rm() calls with NULL assignments. \usage{} lines are at most 90 characters
long.R CMD check NOTE that
.onAttach() tries to load the affxparser package.packageStartupMessage() instead of cat().{read|update|write}Apd() to coerce argument writeMap to
integer indices. readApd() also coerce argument indices to the
same. Before it used to coerce to doubles, if using R.utils
v1.1.4 or before. These updates should make no difference to
anyone. The main purpose is just to use Arguments$getIndices()
instead of Arguments$getNumerics() where ever possible.autoload("setMethodS3", package="R.oo"), etc.
Package would not install on Linux.Added updateApdUnits().
Now all probe indices are not longer zero-based but one-based.
However, there is still an option to change this to zero-based
indices again by using argument indexOffset = 0. This option
might be remove in the future, because zero-based indices are
really confusing in R.
There is now a quite large set of methods to process APD files.
All of the methods to re-arrange CDF structures and to create cell-index maps have been moved to the affxparser package.