readNEWS {tools} | R Documentation |
Read/check R's ‘NEWS’ file or a similarly formatted one. This was an experimental feature added in R 2.4.0: as from R 2.12.0 the preferred format is ‘NEWS.Rd’.
readNEWS(file = file.path(R.home(), "NEWS"), trace = FALSE, chop = c("first", "1", "par1", "keepAll")) checkNEWS(file = file.path(R.home(), "NEWS"))
file |
the name of the file which the data are to be read from.
Alternatively, |
trace |
logical indicating if the recursive reading should be traced, i.e., print what it is doing. |
chop |
a character string specifying how the news entries should
be chopped; |
readNEWS()
reads a pre-R-2.12.0-style ‘NEWS’ file;
checkNEWS()
checks for common errors in formatting. Currently it
detects an incorrect number of spaces before the "o"
item marker.
If non-ASCII characters are needed, the ‘NEWS’ file may be encoded in
UTF-8 with a byte-order mark (BOM) at the beginning, which
readNEWS()
will recognize. Other encodings will display
incorrectly on some systems. However, BOMs are discouraged on many
systems and not all editors recognize them, so ‘NEWS’ files should
normally be written in ASCII.
readNEWS()
returns an (S3) object of class "newsTree"
;
effectively a list
of lists which is a tree of NEWS
entries.
checkNEWS()
returns TRUE
if no suspected errors are found,
or prints a message for each suspected error and returns FALSE
.
Note that this was only ever experimental and may be removed soon, since the NEWS-file format is not longer supported (but there are examples ‘ONEWS’, ‘OONEWS’ and perhaps in packages).
NEWStr <- readNEWS(trace = TRUE)# chop = "first" ( = "first non-empty") ## keep the full NEWS entry text i.e. "no chopping": NEWStrA <- readNEWS(chop = "keepAll") object.size(NEWStr) object.size(NEWStrA) ## (no chopping) ==> about double the size str(NEWStr, max.level = 3) str(NEWStr[[c("2.3", "2.3.1")]], max.level=2, vec.len=1) NEWStr [[c("2.3", "2.3.1", "NEW FEATURES")]] NEWStrA[[c("2.4", "2.4.0", "NEW FEATURES")]] # Check the current NEWS file stopifnot(checkNEWS())