11R News
22
3+ CHANGES IN R VERSION 2.14.0 patched:
4+
5+ NEW FEATURES:
6+
7+ o parallel::detectCores() is now able to find the number of
8+ physical cores (rather than CPUs) on Sparc Solaris.
9+
10+ It can also do so on most versions of Windows; however the
11+ default remains detectCores(logical = TRUE) on that platform.
12+
13+ o Reference classes now keep a record of which fields are locked.
14+ $lock() with no arguments returns the names of the locked fields.
15+
16+ o HoltWinters() reports a warning rather than an error for some
17+ optimization failures (where the answer might be a reasonable
18+ one).
19+
20+ o tools::dependsOnPkg() now accepts the shorthand dependencies =
21+ "all".
22+
23+ o parallel::clusterExport() now allows specification of an
24+ environment from which to export.
25+
26+ o The quartz() device now does tilde expansion on its file
27+ argument.
28+
29+ o tempfile() on a Unix-alike now takes the process ID into account.
30+ This is needed with multicore (and as part of parallel) because
31+ the parent and all the children share a session temporary
32+ directory, and they can share the C random number stream used to
33+ produce the uniaue part. Further, two children can call
34+ tempfile() simultaneously.
35+
36+ o Option print in Sweave's RweaveLatex() driver now emulates
37+ auto-printing rather than printing (which can differ for an S4
38+ object by calling show() rather than print()).
39+
40+ o filled.contour() now accepts infinite values: previously it might
41+ have generated invalid graphics files (e.g. containing NaN
42+ values).
43+
44+ INSTALLATION:
45+
46+ o On 64-bit Linux systems, configure now only sets LIBnn to lib64
47+ if /usr/lib64 exists. This may obviate setting LIBnn explicitly
48+ on Debian-derived systems.
49+
50+ It is still necessary to set LIBnn = lib (or lib32) for 32-bit
51+ builds of R on a 64-bit OS on those Linux distributions capable
52+ for supporting that concept.
53+
54+ o configure looks for inconsolata.sty, and if not found adjusts the
55+ default R_RD4PDF to not use it (with a warning, since it is
56+ needed for high-quality rendering of manuals).
57+
58+ PACKAGE INSTALLATION:
59+
60+ o R CMD INSTALL will now do a test load for all sub-architectures
61+ for which code was compiled (rather than just the primary
62+ sub-architecture).
63+
64+ UTILITIES:
65+
66+ o When checking examples under more than one sub-architecture, R
67+ CMD check now uses a separate directory examples_arch for each
68+ sub-architecture, and leaves the output in file
69+ pkgname-Ex_arch.Rout. Some packages expect their examples to be
70+ run in a clean directory ....
71+
72+ BUG FIXES:
73+
74+ o stack() now gives an error if no vector column is selected,
75+ rather than returning a 1-column data frame (contrary to its
76+ documentation).
77+
78+ o summary.mlm() did not handle objects where the formula had been
79+ specified by an expression. (Reported by Helios de Rosario
80+ Martinez).
81+
82+ o tools::deparseLatex(dropBraces=TRUE) could drop text as well as
83+ braces.
84+
85+ o colormodel = "grey" (new in R 2.14.0)) did not always work in
86+ postscript() and pdf().
87+
88+ o file.append() could return TRUE for failures. (PR#14727)
89+
90+ o gzcon() connections are no longer subject to garbage collection:
91+ it was possible for this to happen when unintended (e.g. when
92+ calling load()).
93+
94+ o nobs() does not count zero-weight observations for glm() fits,
95+ for consistency with lm(). This affects the BIC() values
96+ reported for such glm() fits. (Spotted by Bill Dunlap.)
97+
98+ o options(warn = 0) failed to end a (C-level) context with more
99+ than 50 accumulated warnings. (Spotted by Jeffery Horner.)
100+
101+ o The internal plot.default() code did not do sanity checks on a
102+ cex argument, so invalid input could cause problems. (Reported
103+ by Ben Bolker.)
104+
105+ o anyDuplicated(<array>, MARGIN=0) no longer fails. (Reported by
106+ Herv'e Pag`es.)
107+
108+ o read.dcf() removes trailing blanks: unfortunately on some
109+ platforms this included \xa0 (non-breaking space) which is the
110+ trailing byte of a UTF-8 character. It now only considers ASCII
111+ space and tab to be 'blank'.
112+
113+ o There was a sign error in part of the calculations for the
114+ variance returned by KalmanSmooth(). (PR#14738)
115+
116+ o pbinom(10, 1e6, 0.01, log.p = TRUE) was NaN thanks to the buggy
117+ fix to PR#14320 in R 2.11.0. (PR#14739)
118+
119+ o RweaveLatex() now emulates auto-printing rather than printing, by
120+ calling methods::show() when auto-printing would.
121+
122+ o duplicated() ignored fromLast for a one-column data frame.
123+ (PR#14742)
124+
125+ o source() and related functions did not put the correct timestamp
126+ on the source references; srcfilecopy() has gained a new argument
127+ timestamp to support this fix. (PR#14750)
128+
129+ o LaTeX conversion of Rd files did not correctly handle
130+ preformatted backslashes. (PR#14751)
131+
132+ o HTML conversion of Rd files did not handle markup within tabular
133+ cells properly. (PR#14708)
134+
135+ o source() on an empty file with keep.source = TRUE tried to read
136+ from stdin(), in R 2.14.0 only. (PR#14753)
137+
138+ o The code to check Rd files in packages would abort if duplicate
139+ description sections were present.
140+
3141CHANGES IN R VERSION 2.14.0:
4142
5143 SIGNIFICANT USER-VISIBLE CHANGES:
@@ -259,6 +397,10 @@ CHANGES IN R VERSION 2.14.0:
259397 attribute references an in-memory copy of the source file using
260398 the "srcfilecopy" class or the new "srcfilealias" class.
261399
400+ *NB:* This means that functions sourced with keep.source = TRUE
401+ and saved (e.g., by save() or readRDS()) in earlier versions of R
402+ will no longer show the original sources (including comments).
403+
262404 o New items User Manuals and Technical Papers have been added to
263405 the HTML help main page. These link to vignettes in the base and
264406 recommended packages and to a collection of papers about R
@@ -833,9 +975,6 @@ CHANGES IN R VERSION 2.14.0:
833975
834976 o A late change in R 2.13.2 broke \Sexpr expressions in Rd files.
835977
836- o The creation of ticks on log axes (including axTicks() sometimes
837- incorrectly omitted a tick at one end
838-
839978 o The creation of ticks on log axes (including by axTicks())
840979 sometimes incorrectly omitted a tick at one end of the range by
841980 rounding error in a platform-dependent way. This could be seen
@@ -846,7 +985,7 @@ CHANGES IN R VERSION 2.14.0:
846985 giving NaN. (PR#8528, PR#14710)
847986
848987 o mapply() now gives an explicit error message (rather than an
849- obscure one) is inputs of zero and positive length are mixed.
988+ obscure one) if inputs of zero and positive length are mixed.
850989
851990 o Setting a Hershey font family followed by string height query
852991 would crash R.
0 commit comments