In test bugs.R only show differences (if any) rather
than making the test fail.
Improve rep() calls in pixmap() to
avoid replicating NULL vectors and to avoid partial
argument matching.
NAMESPACE problems.Added dummy NAMESPACE file.
Move class definitions to separate file, no Collate
in DESCRIPTION.
Explicitly mention in the help file that read.pnm
only works for files, not other conntection.
Fixed a bug in write.pnm that tried to open the same
connection twice.
write.pnm.LazyLoad instead of SaveImage.terrain.colors() now return transparency information ->
do not use it in example.DESCRIPTION file."pixmap".help(pixmap)
submitted by Stephan Matthiesen.Adjust for R 2.0.0.
Fixed a bug in coercion from pixmapIndexed to
pixmapRGB.
There was a bug in the methods package of R 1.9.x
which was triggered by functions in pixmap, hence this
version of the package depends on R >= 2.0.0.
Fixed a bug in write.pnm() that wrote grey images in
PPM format.
The channel information was not changed when converting between RGB and grey pixmaps.
The maxval in PNM headers must be less than
65536, not less than 256.
read.pnm(): Vectorized (and renamed)
as.integer.bytes() which provides a huge performance gain
for reading “PBM” (b/w bitmaps).
New function addlogo().
read.pnm (file name
extensions are ignored).read.pnm(): Changes made to function reading and
parsing PNM file headers to permit comments of arbitrary length.The whole package has moved to S4 classes and methods, hence all classes have a new representation. This also means that the code and the API are not fully backwards compatible with earlier versions of the package.
Added support for subsetting, see
example(pixmap).
New: addChannels() and
getChannels().
bbcent and cellres.nrow and ncol default to the respective
dimensions of the data argument (if present). Hence, pixmap does the
expected when given a matrix or an array.data is rescaled to [0,1] for RGB and grey, and coerced
to positive integers for indexed.col can also be a function like
rainbow()read.pnm and
write.pnm which confused dimensions (rows versus columns),
but together let plots look OK.