<?xml version="1.0" encoding="utf-8"?>

<!-- name="generator" content="blosxom/2.0.djm" -->

<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Daily News:  R-2-14-branch</title>
    <link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch</link>
    <atom:link href="https://developer.r-project.org/blosxom.cgi/R-2-14-branch/index.rss" rel="self" type="application/rss+xml" />
    <description>Daily updates to R-2-14-branch</description>
    <language>en</language>

<item>
<title>2012-03-01</title>
<pubDate>Wed, 29 Feb 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/29#n2012-03-01</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/29#n2012-03-01</guid>
<description><![CDATA[<H4>2.14.2 UTILITIES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘R CMD check’ has a new option ‘--as-cran’ to turn on most of the customizations that CRAN uses for its incoming checks.</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2012-02-24</title>
<pubDate>Thu, 23 Feb 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/23#n2012-02-24</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/23#n2012-02-24</guid>
<description><![CDATA[<H4>2.14.2 NEW FEATURES </H4>
<UL>
<LI><P>
For consistency with the ‘logLik()’ method, ‘nobs()’ for ‘&quot;nls&quot;’ files
now excludes observations with zero weight.  (Reported by Berwin
Turlach.)
</P></LI>

</UL>
<H4>2.14.2 UTILITIES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘R CMD check’ has a new option ‘-as-cran’ to turn on most of the customizations that CRAN uses for its incoming checks.</INS>
</P></LI>
</UL>
<H4>2.14.2 BUG FIXES </H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
For consistency with the ‘logLik()’ method, ‘nobs()’ for ‘&quot;nls&quot;’ files
now excludes observations with zero weight.  (Reported by Berwin
Turlach.)
</DEL>
</P></LI>

<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">The ‘getS3method()’ implementation had an old computation to find an S4 default method.</INS>
</P></LI>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘R CMD check’ could miss undocumented S4 objects in packages which used S4 classes but did not ‘Depends: methods’ in their ‘DESCRIPTION’ file.</INS>
</P></LI>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘svg()’ could segfault if called with a non-existent file path. (PR#14790)</INS>
</P></LI>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘.Primitive()’ could be mis-used to call ‘.Internal()’ entry points.</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2012-02-23</title>
<pubDate>Wed, 22 Feb 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/22#n2012-02-23</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/22#n2012-02-23</guid>
<description><![CDATA[<H4>2.14.2 UTILITIES </H4>
<UL>
<LI><P>
‘R CMD check’ now reports by default on licenses not according to the
description in ‘Writing R Extensions’.
</P></LI>

<LI><P>
‘R CMD check’ has a new option ‘-as-cran’ to turn most of the
customizations that CRAN uses of their incoming checks.
</P></LI>

</UL>
<H4>2.14.2 NEW FEATURES </H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD check’ now reports by default on licenses not according to the
description in ‘Writing R Extensions’.
</DEL>
</P></LI>

</UL>
<H4>2.14.2 BUG FIXES </H4>
<UL>
<LI><P>
‘.Primitive()’ could be misused to call ‘.Internal()’ entry points.
</P></LI>

<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘format.POSIXlt(x)’ raised an arithmetic exception when ‘x’ was an invalid object of class ‘"POSIXlt"’ and parts were empty.</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2012-02-21</title>
<pubDate>Mon, 20 Feb 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/20#n2012-02-21</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/20#n2012-02-21</guid>
<description><![CDATA[<H4>2.14.2 BUG FIXES </H4>
<UL>
<LI><P>
‘installed.packages()’ has some more protection against package
installs going on in parallel.
</P></LI>

<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘format.POSIXlt(x)’ raised an arithmetic exception when ‘x’ was an in valid object of class ‘"POSIXlt"’ and parts were empty.</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2012-02-20</title>
<pubDate>Sun, 19 Feb 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/19#n2012-02-20</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/19#n2012-02-20</guid>
<description><![CDATA[<H4>2.14.1 patched NEW FEATURES </H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
The internal ‘untar()’ (as used by default by ‘R CMD INSTALL’) now
knows about some ‘pax’ headers which ‘bsdtar’ (e.g., the default ‘tar’
for Mac OS &gt;= 10.6) can incorrectly include in ‘tar’ files, and will
skip them with a warning.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
PCRE has been upgraded to version 8.21: as well as bug fixes and
greater Perl compatibility, this adds a JIT pattern compiler, about
which PCRE's news says ‘large performance benefits can be had in many
situations’.  This is supported on most but not all R platforms.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Function ‘compactPDF()’ in package ‘tools’ now takes the default for
argument ‘gs_quality’ from environment variable ‘GS_QUALITY’: there is
a new value ‘&quot;none&quot;’, the ultimate default, which prevents GhostScript
being used in preference to ‘qpdf’ just because environment variable
‘R_GSCMD’ is set.  If ‘R_GSCMD’ is unset or set to ‘&quot;&quot;’, the function
will try to find a suitable GhostScript executable.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The included version of zlib has been updated to 1.2.6.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD check’ now reports by default on licenses not according to the
description in ‘Writing R Extensions’.
</DEL>
</P></LI>

</UL>
<H4>2.14.1 patched PACKAGE INSTALLATION </H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD INSTALL’ will now no longer install certain file types from
‘inst/doc’: these are almost certainly mistakes and for some packages
are wasting a lot of space.  These are ‘Makefile’, files generated by
running LaTeX, and unless the package uses a ‘vignettes’ directory,
PostScript and image bitmap files.
</P>
<P>
Note that only PDF vignettes have ever been supported: some of these
files come from DVI/PS output from the Sweave defaults prior to R
2.13.0.
</DEL>
</P></LI>

</UL>
<H4>2.14.1 patched BUG FIXES </H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
R configured with ‘--disable-openmp’ would mistakenly set ‘HAVE_OPENMP’
(internal) and ‘SUPPORT_OPENMP’ (in ‘Rconfig.h’) even though no OpenMP
flags were populated.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘getS3method()’ implementation had a very old and incorrect
computation to find an S4 default method.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘readLines()’ could overflow a buffer if the last line of the file was
not terminated.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14766">PR#14766</a>)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD check’ could miss undocumented S4 objects in packages which used
S4 classes but did not ‘Depends: methods’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The HTML Help Search page had malformed links.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14769">PR#14769</a>)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
A couple of instances of lack of protection of ‘SEXP’s have been
squashed. (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14772">PR#14772</a>, <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14773">PR#14773</a>)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘image(x, useRaster=TRUE)’ misbehaved on single-column ‘x’. (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14774">PR#14774</a>)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Negative values for ‘options(&quot;max.print&quot;)’ or the ‘max’ argument to
‘print.default()’ caused crashes.  Now the former are ignored and the
latter trigger an error.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14779">PR#14779</a>)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The text of a function body containing more than 4096 bytes was not
properly saved by the parser when entered at the console.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Forgetting the ‘#endif’ tag in an Rd file could cause the parser to go
into a loop.  (Reported by Hans-Jorg Bibiko.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
For consistency with the ‘logLik()’ method, ‘nobs()’ for ‘&quot;nls&quot;’ files
now excludes observations with zero weight.  (Reported by Berwin
Turlach.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘str(*, ....., strict.width=&quot;cut&quot;)’ now also obeys ‘list.len = n’.
(Reported by Sören Vogel.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Printing of arrays did not have enough protection (C level), e.g., in
the context of ‘capture.output()’.  (Reported by Hervé Pagès and Martin
Morgan.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘pdf(file = NULL)’ would produce a spurious file named ‘NA’.
(<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14808">PR#14808</a>)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘list2env()’ did not check the type of its ‘envir’ argument. (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14807">PR#14807</a>)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘svg()’ could segfault if called with a non-existent filename.
(<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14790">PR#14790</a>)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘make install’ can install to a path containing ‘+’ characters.
(<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14798">PR#14798</a>)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘edit()’ function did not respect the ‘options(&quot;keep.source&quot;)’
setting. (Reported by Cleridy Lennert.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘predict.lm(*, type=&quot;terms&quot;, terms=*, se.fit=TRUE)’ did not work.
(<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14817">PR#14817</a>)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
There is a partial workaround for errors in the TRE regular-expressions
engine with named classes and repeat counts of at least 2 in a MBCS
locale (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14408">PR#14408</a>): these are avoided when TRE is in 8-bit mode (e.g.
for ‘useBytes = TRUE’ and when all the data are ASCII).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R_ReplDLLdo1()’ did not call top-level handlers.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The Quartz device was unable to detect window sessions on Mac OS X 10.7
(Lion) and higher and thus it was not used as the default device on the
console. Since Lion any application can use window sessions, so Quartz
will now be the default device if user's window session is active and R
is not run via ssh which is at least close to the behavior in prior OS
X versions.
</DEL>
</P></LI>

</UL>
<H4>2.14.2 NEW FEATURES </H4>
<UL>
<LI><P>
The internal ‘untar()’ (as used by default by ‘R CMD INSTALL’) now
knows about some ‘pax’ headers which ‘bsdtar’ (e.g., the default ‘tar’
for Mac OS &gt;= 10.6) can incorrectly include in ‘tar’ files, and will
skip them with a warning.
</P></LI>

<LI><P>
PCRE has been upgraded to version 8.21: as well as bug fixes and
greater Perl compatibility, this adds a JIT pattern compiler, about
which PCRE's news says ‘large performance benefits can be had in many
situations’.  This is supported on most but not all R platforms.
</P></LI>

<LI><P>
Function ‘compactPDF()’ in package ‘tools’ now takes the default for
argument ‘gs_quality’ from environment variable ‘GS_QUALITY’: there is
a new value ‘&quot;none&quot;’, the ultimate default, which prevents GhostScript
being used in preference to ‘qpdf’ just because environment variable
‘R_GSCMD’ is set.  If ‘R_GSCMD’ is unset or set to ‘&quot;&quot;’, the function
will try to find a suitable GhostScript executable.
</P></LI>

<LI><P>
The included version of zlib has been updated to 1.2.6.
</P></LI>

<LI><P>
‘R CMD check’ now reports by default on licenses not according to the
description in ‘Writing R Extensions’.
</P></LI>

</UL>
<H4>2.14.2 PACKAGE INSTALLATION </H4>
<UL>
<LI><P>
‘R CMD INSTALL’ will now no longer install certain file types from
‘inst/doc’: these are almost certainly mistakes and for some packages
are wasting a lot of space.  These are ‘Makefile’, files generated by
running LaTeX, and unless the package uses a ‘vignettes’ directory,
PostScript and image bitmap files.
</P>
<P>
Note that only PDF vignettes have ever been supported: some of these
files come from DVI/PS output from the Sweave defaults prior to R
2.13.0.
</P></LI>

</UL>
<H4>2.14.2 BUG FIXES </H4>
<UL>
<LI><P>
R configured with ‘--disable-openmp’ would mistakenly set ‘HAVE_OPENMP’
(internal) and ‘SUPPORT_OPENMP’ (in ‘Rconfig.h’) even though no OpenMP
flags were populated.
</P></LI>

<LI><P>
The ‘getS3method()’ implementation had a very old and incorrect
computation to find an S4 default method.
</P></LI>

<LI><P>
‘readLines()’ could overflow a buffer if the last line of the file was
not terminated.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14766">PR#14766</a>)
</P></LI>

<LI><P>
‘R CMD check’ could miss undocumented S4 objects in packages which used
S4 classes but did not ‘Depends: methods’.
</P></LI>

<LI><P>
The HTML Help Search page had malformed links.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14769">PR#14769</a>)
</P></LI>

<LI><P>
A couple of instances of lack of protection of ‘SEXP’s have been
squashed. (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14772">PR#14772</a>, <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14773">PR#14773</a>)
</P></LI>

<LI><P>
‘image(x, useRaster=TRUE)’ misbehaved on single-column ‘x’. (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14774">PR#14774</a>)
</P></LI>

<LI><P>
Negative values for ‘options(&quot;max.print&quot;)’ or the ‘max’ argument to
‘print.default()’ caused crashes.  Now the former are ignored and the
latter trigger an error.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14779">PR#14779</a>)
</P></LI>

<LI><P>
The text of a function body containing more than 4096 bytes was not
properly saved by the parser when entered at the console.
</P></LI>

<LI><P>
Forgetting the ‘#endif’ tag in an Rd file could cause the parser to go
into a loop.  (Reported by Hans-Jorg Bibiko.)
</P></LI>

<LI><P>
For consistency with the ‘logLik()’ method, ‘nobs()’ for ‘&quot;nls&quot;’ files
now excludes observations with zero weight.  (Reported by Berwin
Turlach.)
</P></LI>

<LI><P>
‘str(*, ....., strict.width=&quot;cut&quot;)’ now also obeys ‘list.len = n’.
(Reported by Sören Vogel.)
</P></LI>

<LI><P>
Printing of arrays did not have enough protection (C level), e.g., in
the context of ‘capture.output()’.  (Reported by Hervé Pagès and Martin
Morgan.)
</P></LI>

<LI><P>
‘pdf(file = NULL)’ would produce a spurious file named ‘NA’.
(<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14808">PR#14808</a>)
</P></LI>

<LI><P>
‘list2env()’ did not check the type of its ‘envir’ argument. (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14807">PR#14807</a>)
</P></LI>

<LI><P>
‘svg()’ could segfault if called with a non-existent filename.
(<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14790">PR#14790</a>)
</P></LI>

<LI><P>
‘make install’ can install to a path containing ‘+’ characters.
(<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14798">PR#14798</a>)
</P></LI>

<LI><P>
The ‘edit()’ function did not respect the ‘options(&quot;keep.source&quot;)’
setting. (Reported by Cleridy Lennert.)
</P></LI>

<LI><P>
‘predict.lm(*, type=&quot;terms&quot;, terms=*, se.fit=TRUE)’ did not work.
(<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14817">PR#14817</a>)
</P></LI>

<LI><P>
There is a partial workaround for errors in the TRE regular-expressions
engine with named classes and repeat counts of at least 2 in a MBCS
locale (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14408">PR#14408</a>): these are avoided when TRE is in 8-bit mode (e.g.
for ‘useBytes = TRUE’ and when all the data are ASCII).
</P></LI>

<LI><P>
The C function ‘R_ReplDLLdo1()’ did not call top-level handlers.
</P></LI>

<LI><P>
The Quartz device was unable to detect window sessions on Mac OS X 10.7
(Lion) and higher and thus it was not used as the default device on the
console.  Since Lion any application can use window sessions, so Quartz
will now be the default device if the user's window session is active
and R is not run via ‘ssh’ which is at least close to the behavior in
prior OS X versions.
</P></LI>

<LI><P>
‘mclapply()’ would fail in code assembling the translated error message
if some (but not all) cores encountered an error.
</P></LI>

<LI><P>
‘format.POSIXlt(x)’ raised an arithmetic exception when parts of ‘x’
were empty.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2012-02-18</title>
<pubDate>Fri, 17 Feb 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/17#c2012-02-18</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/17#c2012-02-18</guid>
<description><![CDATA[<H4>2.14.1 patched NEW FEATURES (Windows)</H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘bitmap()’ and ‘dev2bitmap()’ look for ‘gswin64c.exe’ (as used by
64-bit GhostScript), in preference to ‘gswin32c.exe’.
</DEL>
</P></LI>

</UL>
<H4>2.14.1 patched INSTALLATION (Windows)</H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
The sources (and packages) can now be compiled using the multilib
toolchain developed for R 2.15.x: see the ‘MULTI’ macro in
‘MkRules.dist’.  This toolchain is used for the CRAN binary
distribution.
</DEL>
</P></LI>

</UL>
<H4>2.14.1 patched PACKAGE INSTALLATION (Windows)</H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
The preferred toolchain has changed since the one used for R 2.12.0 to
2.14.1: see the‘R Installation and Administration manual’.  Compiled
code (except DLLs) may be incompatible with previous toolchains (and
compiled C++ code almost certainly will be: users of ‘Rcpp’ take care).
</P>
<P>
Ensure that the settings in ‘MkRules.local’ are appropriate to the
toolchain you use.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
There is a new macro (aka make variable) ‘SHLIB_PTHREAD_FLAGS’.  This
is set to ‘-pthread’ on builds using toolchains which support it, and
should be included in both ‘PKG_CPPFLAGS’ (or the Fortran or F9x
equivalents) and ‘PKG_LIBS’.
</DEL>
</P></LI>

</UL>
<H4>2.14.1 patched BUG FIXES (Windows)</H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
Using a prompt of more than 80 characters in ‘readline()’ could cause a
buffer overflow in Rterm.  (Reported by Henrik Bengtsson.)
</DEL>
</P></LI>

</UL>
<H4>2.14.2 NEW FEATURES (Windows)</H4>
<UL>
<LI><P>
‘bitmap()’ and ‘dev2bitmap()’ look for ‘gswin64c.exe’ (as used by
64-bit GhostScript), in preference to ‘gswin32c.exe’.
</P></LI>

</UL>
<H4>2.14.2 INSTALLATION (Windows)</H4>
<UL>
<LI><P>
The sources (and packages) can now be compiled using the multilib
toolchain developed for R 2.15.x: see the ‘MULTI’ macro in
‘MkRules.dist’.  This toolchain is used for the CRAN binary
distribution.
</P></LI>

</UL>
<H4>2.14.2 PACKAGE INSTALLATION (Windows)</H4>
<UL>
<LI><P>
The preferred toolchain has changed since the one used for R 2.12.0 to
2.14.1: see the‘R Installation and Administration manual’.  Compiled
code (except DLLs) may be incompatible with previous toolchains (and
compiled C++ code almost certainly will be: users of ‘Rcpp’ take care).
</P>
<P>
Ensure that the settings in ‘MkRules.local’ are appropriate to the
toolchain you use.
</P></LI>

<LI><P>
There is a new macro (aka make variable) ‘SHLIB_PTHREAD_FLAGS’.  This
is set to ‘-pthread’ on builds using toolchains which support it, and
should be included in both ‘PKG_CPPFLAGS’ (or the Fortran or F9x
equivalents) and ‘PKG_LIBS’.
</P></LI>

</UL>
<H4>2.14.2 BUG FIXES (Windows)</H4>
<UL>
<LI><P>
Using a prompt of more than 80 characters in ‘readline()’ could cause a
buffer overflow in Rterm.  (Reported by Henrik Bengtsson.)
</P></LI>

<LI><P>
Some of the custom messages in the installer were corrupted: add a BOM
mark to the file as now required by Unicode Inno Setup.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14816">PR#14816</a>)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2012-02-18</title>
<pubDate>Fri, 17 Feb 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/17#n2012-02-18</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/17#n2012-02-18</guid>
<description><![CDATA[<H4>2.14.1 patched BUG FIXES </H4>
<UL>
<LI><P>
‘R_ReplDLLdo1()’ did not call top-level handlers.
</P></LI>

<LI><P>
The Quartz device was unable to detect window sessions on Mac OS X 10.7
(Lion) and higher and thus it was not used as the default device on the
console. Since Lion any application can use window sessions, so Quartz
will now be the default device if user's window session is active and R
is not run via ssh which is at least close to the behavior in prior OS
X versions.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2012-02-17</title>
<pubDate>Thu, 16 Feb 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/16#n2012-02-17</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/16#n2012-02-17</guid>
<description><![CDATA[<H4>2.14.1 patched BUG FIXES </H4>
<UL>
<LI><P>
‘predict.lm(*, type=&quot;terms&quot;, terms=*, se.fit=TRUE)’ did not work.
(<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14817">PR#14817</a>)
</P></LI>

<LI><P>
There is a partial workaround for errors in the TRE regular-expressions
engine with named classes and repeat counts of at least 2 in a MBCS
locale (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14408">PR#14408</a>): these are avoided when TRE is in 8-bit mode (e.g.
for ‘useBytes = TRUE’ and when all the data are ASCII).
</P></LI>

<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘str(*, ....., strict.width="cut")’ now also obeys ‘list.len = n’. (Reported by Sören Vogel.)</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2012-02-15</title>
<pubDate>Tue, 14 Feb 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/14#n2012-02-15</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/14#n2012-02-15</guid>
<description><![CDATA[<H4>2.14.1 patched BUG FIXES </H4>
<UL>
<LI><P>
The ‘edit()’ function did not respect the ‘options(&quot;keep.source&quot;)’
setting. (Reported by Cleridy Lennert.)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2012-02-11</title>
<pubDate>Fri, 10 Feb 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/10#n2012-02-11</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/10#n2012-02-11</guid>
<description><![CDATA[<H4>2.14.1 patched BUG FIXES </H4>
<UL>
<LI><P>
‘pdf(file = NULL)’ would produce a spurious file named ‘NA’.
(<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14808">PR#14808</a>)
</P></LI>

<LI><P>
‘list2env()’ did not check the type of its ‘envir’ argument. (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14807">PR#14807</a>)
</P></LI>

<LI><P>
‘svg()’ could segfault if called with a non-existent filename.
(<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14790">PR#14790</a>)
</P></LI>

<LI><P>
‘make install’ can install to a path containing ‘+’ characters.
(<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14798">PR#14798</a>)
</P></LI>

<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘readLines()’ could overflow a buffer if the last line of the file was not terminated.  (PR#14766)</INS>
</P></LI>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">The text of a function body containing more than 4096 bytes was not properly saved by the parser when entered at the console.</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2012-02-09</title>
<pubDate>Wed, 08 Feb 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/08#n2012-02-09</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/08#n2012-02-09</guid>
<description><![CDATA[<H4>2.14.1 patched NEW FEATURES </H4>
<UL>
<LI><P>
‘R CMD check’ now reports by default on licenses not according to the
description in ‘Writing R Extensions’.
</P></LI>

</UL>
<H4>2.14.1 patched BUG FIXES </H4>
<UL>
<LI><P>
Printing of arrays did not have enough protection (C level), e.g., in
the context of ‘capture.output()’.  (Reported by Hervé Pagès and Martin
Morgan.)
</P></LI>

<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">str(*, ....., strict.width="cut") now also obeys ‘list.len = n’. (Reported by Sören Vogel.)</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2012-02-06</title>
<pubDate>Sun, 05 Feb 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/05#n2012-02-06</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/05#n2012-02-06</guid>
<description><![CDATA[<H4>2.14.1 patched NEW FEATURES </H4>
<UL>
<LI><P>
The included version of zlib has been updated to 1.2.6.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2012-02-04</title>
<pubDate>Fri, 03 Feb 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/03#n2012-02-04</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/02/03#n2012-02-04</guid>
<description><![CDATA[<H4>2.14.1 patched BUG FIXES </H4>
<UL>
<LI><P>
str(*, ....., strict.width=&quot;cut&quot;) now also obeys ‘list.len = n’
(Reported by Sören Vogel.)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2012-01-26</title>
<pubDate>Wed, 25 Jan 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/25#n2012-01-26</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/25#n2012-01-26</guid>
<description><![CDATA[<H4>2.14.1 patched NEW FEATURES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">PCRE has been upgraded to version 8.21: as well as bug fixes and greater Perl compatibility, this adds a JIT pattern compiler, about which PCRE's news says ‘large performance benefits can be had in many situations’.  This is supported on most but not all R platforms.</INS>
</P></LI>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">Function ‘compactPDF()’ in package ‘tools’ now takes the default for argument ‘gs_quality’ from environment variable ‘GS_QUALITY’: there is a new value ‘"none"’, the ultimate default, which prevents GhostScript being used in preference to ‘qpdf’ just because environment variable ‘R_GSCMD’ is set.  If ‘R_GSCMD’ is unset or set to ‘""’, the function will try to find a suitable GhostScript executable.</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2012-01-25</title>
<pubDate>Tue, 24 Jan 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/24#n2012-01-25</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/24#n2012-01-25</guid>
<description><![CDATA[<H4>2.14.1 patched NEW FEATURES </H4>
<UL>
<LI><P>
The internal ‘untar()’ (as used by default by ‘R CMD INSTALL’) now
knows about some ‘pax’ headers which ‘bsdtar’ (e.g., the default ‘tar’
for Mac OS &gt;= 10.6) can incorrectly include in ‘tar’ files, and will
skip them with a warning.
</P></LI>

<LI><P>
PCRE has been upgraded to version 8.21: as well as bug fixes and
greater Perl compatibility, this adds a JIT pattern compiler, about
which PCRE's news says ‘large performance benefits can be had in many
situations’.
</P></LI>

<LI><P>
Function ‘compactPDF()’ in package ‘tools’ now takes the default for
argument ‘gs_quality’ from environment variable ‘GS_QUALITY’: there is
a new value ‘&quot;none&quot;’, the ultimate default, which prevents GhostScript
being used in preference to ‘qpdf’ just because environment variable
‘R_GSCMD’ is set.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2012-01-25</title>
<pubDate>Tue, 24 Jan 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/24#c2012-01-25</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/24#c2012-01-25</guid>
<description><![CDATA[<H4>2.14.1 patched BUG FIXES (Windows)</H4>
<UL>
<LI><P>
Using a prompt of more than 80 characters in ‘readline()’ could cause a
buffer overflow in Rterm.  (Reported by Henrik Bengtsson.)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2012-01-24</title>
<pubDate>Mon, 23 Jan 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/23#c2012-01-24</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/23#c2012-01-24</guid>
<description><![CDATA[<H4>2.14.1 patched BUG FIXES (Windows)</H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
Using a prompt of more than 80 characters in ‘readline()’ could cause a
buffer overflow in Rterm.  (Reported by Henrik Bengtsson.)
</DEL>
</P></LI>

</UL>
<H4>2.14.1 patched PACKAGE INSTALLATION (Windows)</H4>
<UL>
<LI><P>
The preferred toolchain has changed since the one used for R 2.12.0 to
2.14.1: see the‘R Installation and Administration manual’.  Compiled
code (except DLLs) may be incompatible with previous toolchains (and
compiled C++ code almost certainly will be: users of ‘Rcpp’ take care).
</P>
<P>
Ensure that the settings in ‘MkRules.local’ are appropriate to the
toolchain you use.
</P></LI>

<LI><P>
There is a new macro (aka make variable) ‘SHLIB_PTHREAD_FLAGS’.  This
is set to ‘-pthread’ on builds using toolchains which support it, and
should be included in both ‘PKG_CPPFLAGS’ (or the Fortran or F9x
equivalents) and ‘PKG_LIBS’.
</P></LI>

</UL>
<H4>2.14.1 patched INSTALLATION (Windows)</H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">The sources (and packages) can now be compiled using the multilib toolchain developed for R 2.15.x: see the ‘MULTI’ macro in ‘MkRules.dist’.  This toolchain is used for the CRAN binary distribution.</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2012-01-23</title>
<pubDate>Sun, 22 Jan 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/22#n2012-01-23</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/22#n2012-01-23</guid>
<description><![CDATA[<H4>2.14.1 patched NEW FAEATURES </H4>
<UL>
<LI><P>
The internal ‘untar()’ (as used by default by ‘R CMD INSTALL’) now
knows about some ‘pax’ headers which ‘bsdtar’ (e.g., the default ‘tar’
for Mac OS &gt;= 10.6) can incorrectly include in ‘tar’ files, and will
skip them with a warning.
</P></LI>

<LI><P>
PCRE has been upgraded to version 8.21: as well as bug fixes and
greater Perl compatibility, this adds a JIT pattern compiler, about
which PCRE's news says ‘large performance benefits can be had in many
situations’.
</P></LI>

</UL>
<H4>2.14.0 NEW FEATURES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘vector(len)’ and ‘length(x) &lt;- len’ no longer accept ‘TRUE’/‘FALSE’ for ‘len’ (not that they were ever documented to, but there was special-casing in the C code).</INS>
</P></LI>
</UL>
<H4>2.14.0 GRAPHICS DEVICES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">There is a new type ‘X11(type = "dbcairo")’ which updates the screen less frequently: see its help page.</INS>
</P></LI>
</UL>
<H4>2.13.0 NEW FEATURES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">The internal (C) function ‘scientific()’, at the heart of R's ‘format.info(x)’, ‘format(x)’, ‘print(x)’, etc, for numeric ‘x’, has been re-written in order to provide slightly more correct results, fixing PR#14491, notably in border cases including when ‘digits &gt;= 16’, thanks to substantial contributions (code and experiments) from Petr Savicky.  This affects a noticeable amount of numeric output from R.</INS>
</P></LI>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">There is an additional marked encoding ‘"bytes"’ for character strings. This is intended to be used for non-ASCII strings which should be treated as a set of bytes, and never re-encoded as if they were in the encoding of the current locale: ‘useBytes = TRUE’ is automatically selected in functions such as ‘writeBin()’, ‘writeLines()’, ‘grep()’ and ‘strsplit()’.  </P>
<P>  Only a few character operations are supported (such as ‘substr()’).  </P>
<P>  Printing, ‘format()’ and ‘cat()’ will represent non-ASCII bytes in such strings by a ‘\xab’ escape.</INS>
</P></LI>
</UL>
<H4>2.12.2 SIGNIFICANT USER-VISIBLE CHANGES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">Complex arithmetic (notably ‘z^n’ for complex ‘z’ and integer ‘n’) gave incorrect results since R 2.10.0 on platforms without C99 complex support.  This and some lesser issues in trigonometric functions have been corrected.  </P>
<P>  Such platforms were rare (we know of Cygwin and FreeBSD).  However, because of new compiler optimizations in the way complex arguments are handled, the same code was selected on x86_64 Linux with ‘gcc 4.5.x’ at the default ‘-O2’ optimization (but not at ‘-O’).</INS>
</P></LI>
</UL>
<H4>2.11.1 BUG FIXES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘do.call()’ing ‘NextMethod’ in erroneous ways no longer segfaults. (PR#13487)</INS>
</P></LI>
</UL>
<H4>2.11.0 SIGNIFICANT USER-VISIBLE CHANGES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">A port to 64-bit Windows is now available as well as binary package repositories: see the ‘R Administration and Installation Manual’.</INS>
</P></LI>
</UL>
<H4>2.11.0 BUG FIXES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">Internal function ‘matchArgs()’ no longer modifies the general purpose bits of the ‘SEXP’s that make up the formals list of R functions.  This fixes an invalid error message that would occur when a garbage collection triggered a second call to ‘matchArgs’ for the same function _via_ a finalizer.</INS>
</P></LI>
</UL>
<H4>2.10.1 BUG FIXES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘qchisq(p, df, ncp, lower.tail = FALSE)’ with ‘ncp &gt;= 80’ was inaccurate for small ‘p’ (as the help page said): it is now less inaccurate.  (In part, PR#13999.)  </P>
<P>  For ‘ncp’ less than but close to 80, ‘pchisq()’ and ‘qchisq()’ are more accurate for probabilities very close to 1 (a series expansion was truncated slightly too early).  </P>
<P>  ‘pchisq(x, df, ncp)’ can no longer return values just larger than one for large values of ‘ncp’.</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2012-01-22</title>
<pubDate>Sat, 21 Jan 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/21#n2012-01-22</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/21#n2012-01-22</guid>
<description><![CDATA[<H4>2.14.1 patched BUG FIXES </H4>
<UL>
<LI><P>
Forgetting the ‘#endif’ tag in an Rd file could cause the parser to go
into a loop.  (Reported by Hans-Jorg Bibiko.)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2012-01-21</title>
<pubDate>Fri, 20 Jan 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/20#n2012-01-21</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/20#n2012-01-21</guid>
<description><![CDATA[<H4>2.14.1 patched BUG FIXES </H4>
<UL>
<LI><P>
For consistency with the ‘logLik()’ method, ‘nobs()’ for ‘&quot;nls&quot;’ files
now excludes observations with zero weight.  (Reported by Berwin
Turlach.)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2012-01-21</title>
<pubDate>Fri, 20 Jan 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/20#c2012-01-21</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/20#c2012-01-21</guid>
<description><![CDATA[<H4>2.14.1 patched INSTALLATION (Windows)</H4>
<UL>
<LI><P>
The sources (and packages) can now be compiled using the multilib
toolchain recommended for R 2.15.x: see the ‘MULTI’ macro in
‘MkRules.dist’.
</P></LI>

</UL>
<H4>2.14.1 BUG FIXES (Windows)</H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">The fix for PR#14543 caused stack problems with outputting large R objects (e.g. data frames of 25,000 items).  (PR#14698)</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2012-01-20</title>
<pubDate>Thu, 19 Jan 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/19#n2012-01-20</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/19#n2012-01-20</guid>
<description><![CDATA[<H4>2.14.1 patched PACKAGE INSTALLATION </H4>
<UL>
<LI><P>
‘R CMD INSTALL’ will now no longer install certain file types from
‘inst/doc’: these are almost certainly mistakes and for some packages
are wasting a lot of space.  These are ‘Makefile’, files generated by
running LaTeX, and unless the package uses a ‘vignettes’ directory,
PostScript and image bitmap files.
</P>
<P>
Note that only PDF vignettes have ever been supported: some of these
files come from DVI/PS output from the Sweave defaults prior to R
2.13.0.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2012-01-12</title>
<pubDate>Wed, 11 Jan 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/11#n2012-01-12</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/11#n2012-01-12</guid>
<description><![CDATA[<H4>2.14.1 patched BUG FIXES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">The text of a function body containing more than 4096 characters was not properly saved by the parser when entered at the console.</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2012-01-11</title>
<pubDate>Tue, 10 Jan 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/10#n2012-01-11</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/10#n2012-01-11</guid>
<description><![CDATA[<H4>2.14.1 patched BUG FIXES </H4>
<UL>
<LI><P>
Negative values for ‘options(&quot;max.print&quot;)’ or the ‘max’ argument to
‘print.default()’ caused crashes.  Now the former are ignored and the
latter trigger an error.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14779">PR#14779</a>)
</P></LI>

<LI><P>
The text of function bodies containing more than 4096 characters were
not properly saved by the parser when entered at the console.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2012-01-06</title>
<pubDate>Thu, 05 Jan 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/05#n2012-01-06</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/05#n2012-01-06</guid>
<description><![CDATA[<H4>2.14.1 patched BUG FIXES </H4>
<UL>
<LI><P>
A couple of instances of lack of protection of ‘SEXP’s have been
squashed. (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14772">PR#14772</a>, <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14773">PR#14773</a>)
</P></LI>

<LI><P>
‘image(x, useRaster=TRUE)’ misbehaved on single-column ‘x’. (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14774">PR#14774</a>)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2012-01-03</title>
<pubDate>Mon, 02 Jan 2012 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/02#n2012-01-03</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2012/01/02#n2012-01-03</guid>
<description><![CDATA[<H4>2.14.1 patched BUG FIXES </H4>
<UL>
<LI><P>
‘R CMD check’ could miss undocumented S4 objects in packages which used
S4 classes but did not ‘Depends: methods’.
</P></LI>

<LI><P>
The HTML Help Search page had malformed links.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14769">PR#14769</a>)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-12-30</title>
<pubDate>Thu, 29 Dec 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/29#n2011-12-30</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/29#n2011-12-30</guid>
<description><![CDATA[<H4>2.14.1 patched BUG FIXES </H4>
<UL>
<LI><P>
‘readLines()’ could overflow a buffer if the last line was not
terminated.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14766">PR#14766</a>)
</P></LI>

</UL>
<H4>2.14.0 BUG FIXES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">Several stack trampling and overflow issues have been fixed in TRE, triggered by ‘agrep’ and friends with long patterns.  (PR#14627)</INS>
</P></LI>
</UL>
<H4>2.13.1 NEW FEATURES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘persp(box = TRUE)’ now warns if the surface extends outside the box (since occlusion for the box and axes is computed assuming the box is a bounding box). (PR#202)</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2011-12-30</title>
<pubDate>Thu, 29 Dec 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/29#c2011-12-30</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/29#c2011-12-30</guid>
<description><![CDATA[<H4>2.14.1 patched BUG FIXES (Windows)</H4>
<UL>
<LI><P>
Using a prompt of more than 80 characters in ‘readline()’ could cause a
buffer overflow in Rterm.  (Reported by Henrik Bengtsson.)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-12-24</title>
<pubDate>Fri, 23 Dec 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/23#n2011-12-24</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/23#n2011-12-24</guid>
<description><![CDATA[<H4>2.14.1 NEW FEATURES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘tempfile()’ on a Unix-alike now takes the process ID into account. This is needed with ‘multicore’ (and as part of ‘parallel’) because the parent and all the children share a session temporary directory, and they can share the C random number stream used to produce the unique part.  Further, two children can call ‘tempfile()’ simultaneously.</INS>
</P></LI>
</UL>
<H4>2.14.1 BUG FIXES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘options(warn = 0)’ failed to end a (C-level) context with more than 50 accumulated warnings.  (Spotted by Jeffrey Horner.)</INS>
</P></LI>
</UL>
<H4>2.14.0 NEW FEATURES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">Automatic printing for reference classes is now done by the ‘$show()’ method.  A method is defined for class ‘envRefClass’ and may be overridden for user classes (see the ‘?ReferenceClasses’ example).  S4 ‘show()’ methods should no longer be needed for reference classes.</INS>
</P></LI>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">The body of a closure can be one of further types of R objects, including environments and external pointers.</INS>
</P></LI>
</UL>
<H4>2.14.0 PACKAGE INSTALLATION </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">A dependency on SVN revision is allowed for ‘R’, e.g. ‘R (&gt;= r56550)’. This should be used in conjunction with a version number, e.g. ‘R (&gt;= 2.14.0), R (&gt;= r56550)’ to distinguish between R-patched and R-devel versions with the same SVN revision.</INS>
</P></LI>
</UL>
<H4>2.13.1 BUG FIXES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘pbirthday()’ and ‘qbirthday()’ did not implement the algorithm exactly as given in their reference and so were unnecessarily inaccurate.  </P>
<P>  ‘pbirthday()’ now solves the approximate formula analytically rather than using ‘uniroot()’ on a discontinuous function.  </P>
<P>  The description of the problem was inaccurate: the probability is a tail probability (‘2 _or more_ people share a birthday’)</INS>
</P></LI>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">Rd conversion to latex mishandled multi-line titles (including cases where there was a blank line in the ‘\title’ section).  (It seems this happened only in 2.13.0 patched.)</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2011-12-23</title>
<pubDate>Thu, 22 Dec 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/22#n2011-12-23</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/22#n2011-12-23</guid>
<description><![CDATA[<H4>2.14.1 NEW FEATURES </H4>
<UL>
<LI><P>
‘parallel::detectCores()’ is now able to find the number of physical
cores (rather than CPUs) on Sparc Solaris.
</P>
<P>
It can also do so on most versions of Windows; however the default
remains ‘detectCores(logical = TRUE)’ on that platform.
</P></LI>

<LI><P>
Reference classes now keep a record of which fields are locked.
‘$lock()’ with no arguments returns the names of the locked fields.
</P></LI>

<LI><P>
‘HoltWinters()’ reports a warning rather than an error for some
optimization failures (where the answer might be a reasonable one).
</P></LI>

<LI><P>
‘tools::dependsOnPkg()’ now accepts the shorthand ‘dependencies =
&quot;all&quot;’.
</P></LI>

<LI><P>
‘parallel::clusterExport()’ now allows specification of an environment
from which to export.
</P></LI>

<LI><P>
The ‘quartz()’ device now does tilde expansion on its ‘file’ argument.
</P></LI>

<LI><P>
‘tempfile()’ on a Unix-alike now takes the process ID into account.
This is needed with ‘multicore’ (and as part of ‘parallel’) because the
parent and all the children share a session temporary directory, and
they can share the C random number stream used to produce the uniaue
part.  Further, two children can call ‘tempfile()’ simultaneously.
</P></LI>

<LI><P>
Option ‘print’ in Sweave's ‘RweaveLatex()’ driver now emulates
auto-printing rather than printing (which can differ for an S4 object
by calling ‘show()’ rather than ‘print()’).
</P></LI>

<LI><P>
‘filled.contour()’ now accepts infinite values: previously it might
have generated invalid graphics files (e.g. containing NaN values).
</P></LI>

</UL>
<H4>2.14.1 INSTALLATION </H4>
<UL>
<LI><P>
On 64-bit Linux systems, ‘configure’ now only sets ‘LIBnn’ to ‘lib64’
if ‘/usr/lib64’ exists.  This may obviate setting ‘LIBnn’ explicitly on
Debian-derived systems.
</P>
<P>
It is still necessary to set ‘LIBnn = lib’ (or ‘lib32’) for 32-bit
builds of R on a 64-bit OS on those Linux distributions capable for
supporting that concept.
</P></LI>

<LI><P>
‘configure’ looks for ‘inconsolata.sty’, and if not found adjusts the
default ‘R_RD4PDF’ to not use it (with a warning, since it is needed
for high-quality rendering of manuals).
</P></LI>

</UL>
<H4>2.14.1 PACKAGE INSTALLATION </H4>
<UL>
<LI><P>
‘R CMD INSTALL’ will now do a test load for all sub-architectures for
which code was compiled (rather than just the primary
sub-architecture).
</P></LI>

</UL>
<H4>2.14.1 UTILITIES </H4>
<UL>
<LI><P>
When checking examples under more than one sub-architecture, ‘R CMD
check’ now uses a separate directory ‘examples_arch’ for each
sub-architecture, and leaves the output in file ‘pkgname-Ex_arch.Rout’.
Some packages expect their examples to be run in a clean directory ....
</P></LI>

</UL>
<H4>2.14.1 BUG FIXES </H4>
<UL>
<LI><P>
‘stack()’ now gives an error if no vector column is selected, rather
than returning a 1-column data frame (contrary to its documentation).
</P></LI>

<LI><P>
‘summary.mlm()’ did not handle objects where the formula had been
specified by an expression.  (Reported by Helios de Rosario Martinez).
</P></LI>

<LI><P>
‘tools::deparseLatex(dropBraces=TRUE)’ could drop text as well as
braces.
</P></LI>

<LI><P>
‘colormodel = &quot;grey&quot;’ (new in R 2.14.0)) did not always work in
‘postscript()’ and ‘pdf()’.
</P></LI>

<LI><P>
‘file.append()’ could return ‘TRUE’ for failures.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14727">PR#14727</a>)
</P></LI>

<LI><P>
‘gzcon()’ connections are no longer subject to garbage collection: it
was possible for this to happen when unintended (e.g. when calling
‘load()’).
</P></LI>

<LI><P>
‘nobs()’ does not count zero-weight observations for ‘glm()’ fits, for
consistency with ‘lm()’.  This affects the ‘BIC()’ values reported for
such ‘glm()’ fits.  (Spotted by Bill Dunlap.)
</P></LI>

<LI><P>
‘options(warn = 0)’ failed to end a (C-level) context with more than 50
accumulated warnings.  (Spotted by Jeffery Horner.)
</P></LI>

<LI><P>
The internal ‘plot.default()’ code did not do sanity checks on a ‘cex’
argument, so invalid input could cause problems.  (Reported by Ben
Bolker.)
</P></LI>

<LI><P>
‘anyDuplicated(&lt;array&gt;, MARGIN=0)’ no longer fails.  (Reported by Hervé
Pagès.)
</P></LI>

<LI><P>
‘read.dcf()’ removes trailing blanks: unfortunately on some platforms
this included ‘\xa0’ (non-breaking space) which is the trailing byte of
a UTF-8 character.  It now only considers ASCII space and tab to be
‘blank’.
</P></LI>

<LI><P>
There was a sign error in part of the calculations for the variance
returned by ‘KalmanSmooth()’.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14738">PR#14738</a>)
</P></LI>

<LI><P>
‘pbinom(10, 1e6, 0.01, log.p = TRUE)’ was ‘NaN’ thanks to the buggy fix
to <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14320">PR#14320</a> in R 2.11.0.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14739">PR#14739</a>)
</P></LI>

<LI><P>
‘RweaveLatex()’ now emulates auto-printing rather than printing, by
calling ‘methods::show()’ when auto-printing would.
</P></LI>

<LI><P>
‘duplicated()’ ignored ‘fromLast’ for a one-column data frame.
(<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14742">PR#14742</a>)
</P></LI>

<LI><P>
‘source()’ and related functions did not put the correct timestamp on
the source references; ‘srcfilecopy()’ has gained a new argument
‘timestamp’ to support this fix. (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14750">PR#14750</a>)
</P></LI>

<LI><P>
LaTeX conversion of Rd files did not correctly handle preformatted
backslashes.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14751">PR#14751</a>)
</P></LI>

<LI><P>
HTML conversion of Rd files did not handle markup within tabular cells
properly.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14708">PR#14708</a>)
</P></LI>

<LI><P>
‘source()’ on an empty file with ‘keep.source = TRUE’ tried to read
from ‘stdin()’, in R 2.14.0 only.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14753">PR#14753</a>)
</P></LI>

<LI><P>
The code to check Rd files in packages would abort if duplicate
description sections were present.
</P></LI>

</UL>
<H4>2.14.1 patched BUG FIXES </H4>
<UL>
<LI><P>
R configured with ‘--disable-openmp’ would mistakenly set ‘HAVE_OPENMP’
(internal) and ‘SUPPORT_OPENMP’ (in ‘Rconfig.h’) even though no OpenMP
flags were populated.
</P></LI>

<LI><P>
The ‘getS3method()’ implementation had a very old and incorrect
computation to find an S4 default method.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-12-23</title>
<pubDate>Thu, 22 Dec 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/22#c2011-12-23</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/22#c2011-12-23</guid>
<description><![CDATA[<H4>2.14.1 NEW FEATURES (Windows)</H4>
<UL>
<LI><P>
The ‘Save as’ menu item on the script editor adds extension ‘.R’ to a
file name without an extension.
</P></LI>

<LI><P>
In package ‘parallel’, ‘detectCores(logical = FALSE)’ makes an
OS-dependent attempt to find the number of physical cores.  It usually
succeeds, even on XP.
</P></LI>

</UL>
<H4>2.14.1 PACKAGE INSTALLATION (Windows)</H4>
<UL>
<LI><P>
The directory pointed to by ‘USER_LOCAL’ can now have
architecture-specific sub-directories ‘lib/i386’ and ‘lib/x64’.
</P></LI>

</UL>
<H4>2.14.1 BUG FIXES (Windows)</H4>
<UL>
<LI><P>
The fix for <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14543">PR#14543</a> caused stack problems with outputing large R
objects (e.g. data frames of 25,000 items).  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14698">PR#14698</a>)
</P></LI>

<LI><P>
In a double-byte locale (Chinese, Japanese, Korean), ‘grep()’ and
friends might have used byte-wise matching of strings in the native
encoding.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14622">PR#14622</a>)
</P></LI>

</UL>
<H4>2.14.1 patched NEW FEATURES (Windows)</H4>
<UL>
<LI><P>
‘bitmap()’ and ‘dev2bitmap()’ look for ‘gswin64c.exe’ (as used by
64-bit GhostScript), in preference to ‘gswin32c.exe’.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-12-18</title>
<pubDate>Sat, 17 Dec 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/17#n2011-12-18</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/17#n2011-12-18</guid>
<description><![CDATA[<H4>2.14.0 patched BUG FIXES </H4>
<UL>
<LI><P>
The code to check Rd files in packages would abort if duplicate
description sections were present.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-12-12</title>
<pubDate>Sun, 11 Dec 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/11#n2011-12-12</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/11#n2011-12-12</guid>
<description><![CDATA[<H4>2.14.0 patched BUG FIXES </H4>
<UL>
<LI><P>
‘source()’ on an empty file with ‘keep.source = TRUE’ tried to read
from ‘stdin()’, in R 2.14.0 only.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14753">PR#14753</a>)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-12-10</title>
<pubDate>Fri, 09 Dec 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/09#n2011-12-10</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/09#n2011-12-10</guid>
<description><![CDATA[<H4>2.14.0 patched BUG FIXES </H4>
<UL>
<LI><P>
‘source()’ and related functions did not put the correct timestamp on
the source references; ‘srcfilecopy()’ has gained a new argument
‘timestamp’ to support this fix. (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14750">PR#14750</a>)
</P></LI>

<LI><P>
LaTeX conversion of Rd files did not correctly handle preformatted
backslashes.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14751">PR#14751</a>)
</P></LI>

<LI><P>
HTML conversion of Rd files did not handle markup within tabular cells
properly.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14708">PR#14708</a>)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-12-05</title>
<pubDate>Sun, 04 Dec 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/04#n2011-12-05</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/12/04#n2011-12-05</guid>
<description><![CDATA[<H4>2.14.0 patched BUG FIXES </H4>
<UL>
<LI><P>
‘duplicated()’ ignored ‘fromLast’ for a one-column data frame.
(<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14742">PR#14742</a>)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-12-01</title>
<pubDate>Wed, 30 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/30#n2011-12-01</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/30#n2011-12-01</guid>
<description><![CDATA[<H4>2.14.0 patched NEW FEATURES </H4>
<UL>
<LI><P>
‘filled.contour()’ now accepts infinite values: previously it might
have generated invalid graphics files (e.g. containing NaN values).
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-11-30</title>
<pubDate>Tue, 29 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/29#n2011-11-30</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/29#n2011-11-30</guid>
<description><![CDATA[<H4>2.14.0 patched NEW FEATURES </H4>
<UL>
<LI><P>
Option ‘print’ in Sweave's ‘RweaveLatex()’ driver now emulates
auto-printing rather than printing (which can differ for an S4 object
by calling ‘show()’ rather than ‘print()’).
</P></LI>

</UL>
<H4>2.14.0 patched BUG FIXES </H4>
<UL>
<LI><P>
‘RweaveLatex()’ now emulates auto-printing rather than printing, by
calling ‘methods::show()’ when auto-printing would.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-11-27</title>
<pubDate>Sat, 26 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/26#n2011-11-27</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/26#n2011-11-27</guid>
<description><![CDATA[<H4>2.14.0 patched BUG FIXES </H4>
<UL>
<LI><P>
‘pbinom(10, 1e6, 0.01, log.p = TRUE)’ was ‘NaN’ thanks to the buggy fix
to <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14320">PR#14320</a> in R 2.11.0.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14739">PR#14739</a>)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-11-26</title>
<pubDate>Fri, 25 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/25#n2011-11-26</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/25#n2011-11-26</guid>
<description><![CDATA[<H4>2.14.0 patched BUG FIXES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">There was a sign error in part of the calculations for the variance returned by ‘KalmanSmooth()’.  (PR#14738)</INS>
</P></LI>
</UL>
<H4>2.14.0 NEW FEATURES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">The ‘"source"’ attribute on functions created with ‘keep.source=TRUE’ has been replaced with a ‘"srcref"’ attribute.  The ‘"srcref"’ attribute references an in-memory copy of the source file using the ‘"srcfilecopy"’ class or the new ‘"srcfilealias"’ class.  </P>
<P>  *NB:* This means that functions sourced with ‘keep.source = TRUE’ and saved (e.g., by ‘save()’ or ‘readRDS()’) in earlier versions of R will no longer show the original sources (including comments).</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2011-11-25</title>
<pubDate>Thu, 24 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/24#n2011-11-25</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/24#n2011-11-25</guid>
<description><![CDATA[<H4>2.14.0 patched NEW FEATURES </H4>
<UL>
<LI><P>
‘tempfile()’ on a Unix-alike now takes the process ID into account.
This is needed with ‘multicore’ (and as part of ‘parallel’) because the
parent and all the children share a session temporary directory, and
they can share the C random number stream used to produce the uniaue
part.  Further, two children can call ‘tempfile()’ simultaneously.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-11-24</title>
<pubDate>Wed, 23 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/23#n2011-11-24</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/23#n2011-11-24</guid>
<description><![CDATA[<H4>2.14.0 patched BUG FIXES </H4>
<UL>
<LI><P>
There was sign error in part of the calculations for the variance
returned by ‘KalmanSmooth()’.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14738">PR#14738</a>)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-11-21</title>
<pubDate>Sun, 20 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/20#n2011-11-21</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/20#n2011-11-21</guid>
<description><![CDATA[<H4>2.14.0 patched NEW FEATURES </H4>
<UL>
<LI><P>
The ‘quartz()’ device now does tilde expansion on its ‘file’ argument.
</P></LI>

</UL>
<H4>2.14.0 patched BUG FIXES </H4>
<UL>
<LI><P>
‘read.dcf()’ removes trailing blanks: unfortunately on some platforms
this included ‘\xa0’ (non-breaking space) which is the trailing byte of
a UTF-8 character.  It now only considers ASCII space and tab to be
‘blank’.
</P></LI>

<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">The internal ‘plot.default()’ code did not do sanity checks on a ‘cex’ argument, so invalid input could cause problems.  (Reported by Ben Bolker.)</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2011-11-20</title>
<pubDate>Sat, 19 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/19#n2011-11-20</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/19#n2011-11-20</guid>
<description><![CDATA[<H4>2.14.0 patched NEW FEATURES </H4>
<UL>
<LI><P>
‘tools::dependsOnPkg()’ now accepts the shorthand ‘dependencies =
&quot;all&quot;’.
</P></LI>

<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘parallel::clusterExport()’ now allows specification of an environment from which to export.</INS>
</P></LI>
</UL>
<H4>2.14.0 patched UTILITIES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">When checking examples under more than one sub-architecture, ‘R CMD check’ now uses a separate directory ‘examples_arch’ for each sub-architecture, and leaves the output in file ‘pkgname-Ex_arch.Rout’. Some packages expect their examples to be run in a clean directory ....</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2011-11-20</title>
<pubDate>Sat, 19 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/19#c2011-11-20</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/19#c2011-11-20</guid>
<description><![CDATA[<H4>2.14.0 patched PACKAGE INSTALLATION (Windows)</H4>
<UL>
<LI><P>
The directory pointed to by ‘USER_LOCAL’ can now have
architecture-specific sub-directories ‘lib/i386’ and ‘lib/x64’.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-11-19</title>
<pubDate>Fri, 18 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/18#n2011-11-19</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/18#n2011-11-19</guid>
<description><![CDATA[<H4>2.14.0 patched UTILITIES </H4>
<UL>
<LI><P>
When checking examples under more than one sub-architecture, ‘R CMD
check’ now uses a separate directory ‘examples-arch’ for each
sub-architecture, and leaves the output in file ‘pkgname-Ex_arch.Rout’.
Some packages expect their examples to be run in a clean directory ....
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-11-18</title>
<pubDate>Thu, 17 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/17#n2011-11-18</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/17#n2011-11-18</guid>
<description><![CDATA[<H4>2.14.0 patched PACKAGE INSTALLATION </H4>
<UL>
<LI><P>
‘R CMD INSTALL’ will now do a test load for all sub-architectures for
which code was compiled (rather than just the primary
sub-architecture).
</P></LI>

</UL>
<H4>2.14.0 patched NEW FEATURES </H4>
<UL>
<LI><P>
‘parallel::clusterExport’ now allows specification of an environment
from which to export.
</P></LI>

</UL>
<H4>2.14.0 patched BUG FIXES </H4>
<UL>
<LI><P>
The internal ‘plot.default()’ code did not do sanity checks on ‘cex’,
so invalid input could cause problems.  (Reported by Ben Bolker.)
</P></LI>

<LI><P>
‘anyDuplicated(&lt;array&gt;, MARGIN=0)’ no longer fails.  (Reported by Hervé
Pagès.)
</P></LI>

<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘options(warn = 0)’ failed to end a (C-level) context with more than 50 accumulated warnings.  (Spotted by Jeffery Horner.)</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2011-11-17</title>
<pubDate>Wed, 16 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/16#c2011-11-17</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/16#c2011-11-17</guid>
<description><![CDATA[<H4>2.14.0 patched NEW FEATURES (Windows)</H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">The ‘Save as’ menu item on the script editor adds extension ‘.R’ to a file name without an extension.</INS>
</P></LI>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">In package ‘parallel’, ‘detectCores(logical = FALSE)’ makes an OS-dependent attempt to find the number of physical cores.  It usually succeeds, even on XP.</INS>
</P></LI>
</UL>
<H4>2.14.0 patched BUG FIXES (Windows)</H4>
<UL>
<LI><P>
The fix for <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14543">PR#14543</a> caused stack problems with outputing large R
objects (e.g. data frames of 25,000 items).  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14698">PR#14698</a>)
</P></LI>

<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">In a double-byte locale (Chinese, Japanese, Korean), ‘grep()’ and friends might have used byte-wise matching of strings in the native encoding.  (PR#14622)</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2011-11-17</title>
<pubDate>Wed, 16 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/16#n2011-11-17</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/16#n2011-11-17</guid>
<description><![CDATA[<H4>2.14.0 patched BUG FIXES </H4>
<UL>
<LI><P>
‘options(warn = 0)’ failed to end a (C-level) contest with more than 50
accumulated warnings.  (Spotted by Jeffery Horner.)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-11-16</title>
<pubDate>Tue, 15 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/15#n2011-11-16</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/15#n2011-11-16</guid>
<description><![CDATA[<H4>2.14.0 patched BUG FIXES </H4>
<UL>
<LI><P>
‘nobs()’ does not count zero-weight observations for ‘glm()’ fits, for
consistency with ‘lm()’.  This affects the ‘BIC()’ values reported for
such ‘glm()’ fits.  (Spotted by Bill Dunlap.)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-11-15</title>
<pubDate>Mon, 14 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/14#n2011-11-15</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/14#n2011-11-15</guid>
<description><![CDATA[<H4>2.14.0 patched NEW FEATURES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘parallel::detectCores()’ is now able to find the number of physical cores (rather than CPUs) on Sparc Solaris.  </P>
<P>  It can also do so on most versions of Windows; however the default remains ‘detectCores(logical = TRUE)’ on that platform.</INS>
</P></LI>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">Reference classes now keep a record of which fields are locked. ‘$lock()’ with no arguments returns the names of the locked fields.</INS>
</P></LI>
</UL>
<H4>2.14.0 BUG FIXES </H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
The creation of ticks on log axes (including ‘axTicks()’ sometimes
incorrectly omitted a tick at one end
</DEL>
</P></LI>

<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘mapply()’ now gives an explicit error message (rather than an obscure one) if inputs of zero and positive length are mixed.</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2011-11-13</title>
<pubDate>Sat, 12 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/12#n2011-11-13</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/12#n2011-11-13</guid>
<description><![CDATA[<H4>2.14.0 patched INSTALLATION </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">On 64-bit Linux systems, ‘configure’ now only sets ‘LIBnn’ to ‘lib64’ if ‘/usr/lib64’ exists.  This may obviate setting ‘LIBnn’ explicitly on Debian-derived systems.  </P>
<P>  It is still necessary to set ‘LIBnn = lib’ (or ‘lib32’) for 32-bit builds of R on a 64-bit OS on those Linux distributions capable for supporting that concept.</INS>
</P></LI>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘configure’ looks for ‘inconsolata.sty’, and if not found adjusts the default ‘R_RD4PDF’ to not use it (with a warning, since it is needed for high-quality rendering of manuals).</INS>
</P></LI>
</UL>
<H4>2.14.0 patched BUG FIXES </H4>
<UL>
<LI><P>
‘file.append()’ could return ‘TRUE’ for failures.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14727">PR#14727</a>)
</P></LI>

<LI><P>
‘gzcon()’ connections are no longer subject to garbage collection: it
was possible for this to happen when unintended (e.g. when calling
‘load()’).
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-11-10</title>
<pubDate>Wed, 09 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/09#n2011-11-10</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/09#n2011-11-10</guid>
<description><![CDATA[<H4>2.14.0 patched INSTALLATION </H4>
<UL>
<LI><P>
‘configure’ looks for ‘inconsolata.sty’, and if not found adjusts the
default ‘R_RD4PDF’ to not use it (with a warning, since it is needed
for high-quality manuals).
</P></LI>

<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">On 64-bit Linux systems, ‘configure’ now only sets ‘LIBnn’ to ‘lib64’ if ‘/usr/lib64’ exists.  This may obviate setting ‘LIBnn’ explicitly on Debian-derived systems.  </P>
<P>  It is still necessary to set ‘LIBnn = lib’ (or ‘lib32’)for 32-bit builds of R on a 64-bit OS on those Linux distributions capable for supporting that concept.</INS>
</P></LI>
</UL>
<H4>2.14.0 patched BUG FIXES </H4>
<UL>
<LI><P>
‘colormodel = &quot;grey&quot;’ (new in R 2.14.0)) did not always work in
‘postscript()’ and ‘pdf()’.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-11-09</title>
<pubDate>Tue, 08 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/08#n2011-11-09</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/08#n2011-11-09</guid>
<description><![CDATA[<H4>2.14.0 patched NEW FEATURES </H4>
<UL>
<LI><P>
‘HoltWinters()’ reports a warning rather than an error for some
optimization failures (where the answer might be a reasonable one).
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-11-07</title>
<pubDate>Sun, 06 Nov 2011 18:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/06#n2011-11-07</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/06#n2011-11-07</guid>
<description><![CDATA[<H4>2.14.0 patched NEW FEATURES </H4>
<UL>
<LI><P>
Reference classes now keep a record of which fields are locked.
‘\$lock()’ with no arguments returns the names of the locked fields.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-11-06</title>
<pubDate>Sat, 05 Nov 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/05#c2011-11-06</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/05#c2011-11-06</guid>
<description><![CDATA[<H4>2.14.0 patched BUG FIXES (Windows)</H4>
<UL>
<LI><P>
The fix for <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14543">PR#14543</a> caused stack problems with outputing large R
objects (e.g. data frames of 25,000 items).  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14698">PR#14698</a>)
</P>
<P>
In a double-byte locale (Chinese, Japanese, Korean), ‘grep()’ and
friends might have used byte-wise matching of strings in the native
encoding.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14622">PR#14622</a>)
</P></LI>

</UL>
<H4>2.14.0 patched NEW FEATURES (Windows)</H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">In package ‘parallel’, ‘detectCores(logical = FALSE)’ makes an OS-dependent attempt to find the number of physical cores. It usually succeeds, even on XP.</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2011-11-06</title>
<pubDate>Sat, 05 Nov 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/05#n2011-11-06</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/05#n2011-11-06</guid>
<description><![CDATA[<H4>2.14.0 patched BUG FIXES </H4>
<UL>
<LI><P>
‘tools::deparseLatex(dropBraces=TRUE)’ could drop text as well as
braces.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-11-02</title>
<pubDate>Tue, 01 Nov 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/01#n2011-11-02</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/11/01#n2011-11-02</guid>
<description><![CDATA[<H4>2.14.0 patched NEW FEATURES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘parallel::detectCores()’ is now able to find the number of physical cores (rather than CPUs) on Sparc Solaris.  </P>
<P>  It can also do so on recent versions of Windows; however the default remains ‘detectCores(logical = TRUE)’ on that platform.</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2011-11-01</title>
<pubDate>Mon, 31 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/31#c2011-11-01</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/31#c2011-11-01</guid>
<description><![CDATA[<H4>2.14.0 patched NEW FEATURES (Windows)</H4>
<UL>
<LI><P>
The ‘Save as’ menu item on the script editor appends ‘.R’ to a file
name without an extension.
</P></LI>

<LI><P>
In package ‘parallel’, ‘detectCores(logical = FALSE)’ makes an
OS-dependent attempt to find the number of physical cores.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-11-01</title>
<pubDate>Mon, 31 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/31#n2011-11-01</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/31#n2011-11-01</guid>
<description><![CDATA[<H4>2.14.0 patched NEW FEATURES </H4>
<UL>
<LI><P>
‘parallel::detectCores()’ is now able to find the number of physical
cores (rather than CPUs) on Sparc Solaris.
</P>
<P>
It can also do so on recent versions of Windows (but not Windows XP);
however the default remains ‘detectCores(logical = TRUE)’ on that
platform.
</P></LI>

</UL>
<H4>2.14.0 patched INSTALLATION </H4>
<UL>
<LI><P>
On 64-bit Linux systems, ‘configure’ now only sets ‘LIBnn’ to ‘lib64’
if ‘/usr/lib64’ exists.  This may obviate setting ‘LIBnn’ explicitly on
Debian-derived systems.
</P>
<P>
It is still necessary to set ‘LIBnn = lib’ for 32-bit builds of R on a
64-bit OS on those Linux distributions capable for supporting that
concept.
</P></LI>

</UL>
<H4>2.14.0 patched BUG FIXES </H4>
<UL>
<LI><P>
‘stack()’ now gives an error if no vector column is selected, rather
than returning a 1-column data frame (contrary to its documentation).
</P></LI>

<LI><P>
‘summary.mlm()’ did not handle objects where the formula had been
specified by an expression.  (Reported by Helios de Rosario Martinez).
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-10-30</title>
<pubDate>Sat, 29 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/29#n2011-10-30</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/29#n2011-10-30</guid>
<description><![CDATA[<H4>2.14.0 NEW FEATURES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">There are two new options for typesetting PDF manuals from Rd files. These are ‘beramono’ and ‘inconsolata’, and used the named font for monospaced output. They are intended to be used in combination with ‘times’, and ‘times,inconsolata,hyper’ is now the default for the reference manual and package manuals.  If you do not have that font installed, you can set ‘R_RD4PF’ to one of the other options: see the ‘R Installation and Administration Manual’.</INS>
</P></LI>
</UL>
<H4>2.14.0 UTILITIES </H4>
<UL>
<LI><P>
‘R CMD check’ now checks line-endings of makefiles and C/C++/Fortran
sources in subdirectories of ‘src’ as well as in ‘src’ itself.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-10-26</title>
<pubDate>Tue, 25 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/25#n2011-10-26</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/25#n2011-10-26</guid>
<description><![CDATA[<H4>2.14.0 DEPRECATED AND DEFUNCT </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">It is deprecated to use ‘mean(x)’ and ‘sd(x)’ directly on data frames (or also matrices, for ‘sd’) ‘x’, instead of simply using ‘sapply’.  </P>
<P>  In the same spirit, ‘median(x)’ now gives an error for a data frame ‘x’ (it often gave nonsensical results).</INS>
</P></LI>
</UL>
<H4>2.14.0 BUG FIXES </H4>
<UL>
<LI><P>
Setting a Hershey font family followed by string height query would
crash R.
</P></LI>

<LI><P>
‘R CMD javareconf -e’ would fail for some shells due to a shift error.
Also the resulting paths will no longer contain ‘$(JAVA_HOME)’ as that
can result in an unintended substitution based on ‘Makeconf’ instead of
the shell setting.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-10-24</title>
<pubDate>Sun, 23 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/23#n2011-10-24</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/23#n2011-10-24</guid>
<description><![CDATA[<H4>2.14.0 SIGNIFICANT USER-VISIBLE CHANGES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">There is a new package ‘parallel’.  </P>
<P>  It incorporates (slightly revised) copies of packages ‘multicore’ and ‘snow’ (excluding MPI, PVM and NWS clusters).  Code written to use the higher-level API functions in those packages should work unchanged (apart from changing any references to their namespaces to a reference to ‘parallel’, and links explicitly to ‘multicore’ or ‘snow’ on help pages).  </P>
<P>  It also contains support for multiple RNG streams following L'Ecuyer _et al_ (2002), with support for both ‘mclapply’ and ‘snow’ clusters. This replaces functions like ‘clusterSetupRNG()’ from ‘snow’ (which are not in ‘parallel’).  </P>
<P>  The version released for R 2.14.0 contains base functionality: higher-level convenience functions are planned (and some are already available in the ‘R-devel’ version of R).</INS>
</P></LI>
</UL>
<H4>2.14.0 DEPRECATED AND DEFUNCT </H4>
<UL>
<LI><P>
Using a false value for the ‘DESCRIPTION’ field ‘LazyLoad’ is
deprecated.
</P></LI>

</UL>
<H4>2.14.0 BUG FIXES </H4>
<UL>
<LI><P>
‘mapply()’ now gives an explicit error message (rather than an obscure
one) is inputs of zero and positive length are mixed.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-10-23</title>
<pubDate>Sat, 22 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/22#n2011-10-23</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/22#n2011-10-23</guid>
<description><![CDATA[<H4>2.14.0 NEW FEATURES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘socketConnection()’ now has a ‘timeout’ argument.  It is now documented that large values (package ‘snow’ used a year) do not work on some OSes.</INS>
</P></LI>
</UL>
<H4>2.14.0 BUG FIXES </H4>
<UL>
<LI><P>
‘qgamma()’ for small ‘shape’ underflows to ‘0’ rather than sometimes
giving ‘NaN’.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=8528">PR#8528</a>, <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14710">PR#14710</a>)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-10-22</title>
<pubDate>Fri, 21 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/21#n2011-10-22</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/21#n2011-10-22</guid>
<description><![CDATA[<H4>2.14.0 SIGNIFICANT USER-VISIBLE CHANGES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">All packages must have a namespace, and one is created on installation if not supplied in the sources.  This means that any package without a namespace must be re-installed under this version of R (but previously-installed data-only packages without R code can still be used).</INS>
</P></LI>
</UL>
<H4>2.14.0 NEW FEATURES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘socketConnection()’ now has a ‘timeout’ argument.  It is now documented that large values (package ‘snow’ uses a year) do not work on some OSes.</INS>
</P></LI>
</UL>
<H4>2.14.0 GRAPHICS DEVICES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">The default ‘colormodel’ for ‘pdf()’ and ‘postscript()’ is now called ‘"srgb"’ to more accurately describe it.  (Instead of ‘"rgb"’, and in the case of ‘postscript()’ it no longer switches to and from the gray colorspace, by default.)  </P>
<P>  The ‘colormodel’ for ‘postscript()’ which does use both gray and sRGB colorspaces is now called ‘"srgb+gray"’.  </P>
<P>  Plots which are known to use only black/white/transparent can advantageously use ‘colormodel = "gray"’ (just as before, but there is now slightly more advantage in doing so).</INS>
</P></LI>
</UL>
<H4>2.14.0 DEPRECATED AND DEFUNCT </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">It is eprecated to use ‘mean(x)’ and ‘sd(x)’ directly on data frames (or also matrices, for ‘sd’) ‘x’, instead of simply using ‘sapply’.  </P>
<P>  In the same spirit, ‘median(x)’ now gives an error for a data frame ‘x’ (it often gave nonsensical results).</INS>
</P></LI>
</UL>
<H4>2.14.0 BUG FIXES </H4>
<UL>
<LI><P>
A late change in R 2.13.2 broke ‘\Sexpr’ expressions in Rd files.
</P></LI>

<LI><P>
The creation of ticks on log axes (including ‘axTicks()’ sometimes
incorrectly omitted a tick at one end
</P></LI>

<LI><P>
The creation of ticks on log axes (including by ‘axTicks()’) sometimes
incorrectly omitted a tick at one end of the range by rounding error in
a platform-dependent way.  This could be seen in the examples for
‘axTicks()’, where with axis limits ‘c(0.2, 88)’ the tick for ‘0.2’ was
sometimes omitted.
</P></LI>

<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">The ‘X11()’ device's cairo back-end produced incorrect capture snapshot images on big-endian machines.</INS>
</P></LI>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘loglin()’ gave a spurious error when argument ‘margin’ consisted of a single element of length one.  (PR#14690)</INS>
</P></LI>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘stack()’ and ‘unstack()’ converted character columns to factors.  </P>
<P>  ‘unstack()’ sometimes produced incorrect results (a list or a vector) if the factor on which to un-split had only one level.</INS>
</P></LI>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">Non-blocking listening socket connections continued to report ‘isIncomplete()’ as true even when the peer had closed down and all available input had been read.</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2011-10-21</title>
<pubDate>Thu, 20 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/20#n2011-10-21</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/20#n2011-10-21</guid>
<description><![CDATA[<H4>2.14.0 NEW FEATURES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘help.search()’ and ‘??’ can now display vignettes and demos as well as help pages.  The new option ‘"help.search.types"’ controls the types of documentation and the order of their display.  </P>
<P>  This also applies to HTML searches, which now give results in all of help pages, vignettes and demos.</INS>
</P></LI>
</UL>
<H4>2.14.0 BUG FIXES </H4>
<UL>
<LI><P>
The revised HTML search system now generates better hyperlinks to help
topics found: previously it gave problems with help pages with names
containing e.g. spaces and slashes.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-10-20</title>
<pubDate>Wed, 19 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/19#n2011-10-20</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/19#n2011-10-20</guid>
<description><![CDATA[<H4>2.14.0 SIGNIFICANT USER-VISIBLE CHANGES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">There is a new package ‘parallel’.  </P>
<P>  It incorporates (slightly revised) copies of packages ‘multicore’ and ‘snow’ (excluding MPI, PVM and NWS clusters).  Code written to use the higher-level API functions in those packages should work unchanged (apart from changing any references to their namespaces to a reference to ‘parallel’, and links explicitly to ‘multicore’ or ‘snow’ on help pages).  </P>
<P>  It also contains support for multiple RNG streams following L'Ecuyer _et al_ (2002), with support for both ‘mclapply’ and ‘snow’ clusters. This replaces functions like ‘clusterSetupRNG()’ from ‘snow’ (which are not in ‘parallel’).</INS>
</P></LI>
</UL>
<H4>2.14.0 NEW FEATURES </H4>
<UL>
<LI><P>
Using ‘fix()’ or ‘edit()’ on a R object (except perhaps a matrix or
data frame) writes its temporary file with extension ‘.R’ so editors
which select their mode based on the extension will select a suitable
mode.
</P></LI>

</UL>
<H4>2.14.0 INSTALLATION </H4>
<UL>
<LI><P>
Java detection (‘R CMD javareconf’) works around bogus
‘java.library.path’ property in recent Oracle Java binaries.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-10-19</title>
<pubDate>Tue, 18 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/18#n2011-10-19</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/18#n2011-10-19</guid>
<description><![CDATA[<H4>2.14.0 SIGNIFICANT USER-VISIBLE CHANGES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">There is a new package ‘parallel’.  </P>
<P>  It incorporates (slightly revised) copies of packages ‘multicore’ and ‘snow’ (excluding MPI, PVM and NWS clusters).  Code written to use the higher-level API functions in those packages should work unchanged (apart from changing any references to their namespaces to a reference to ‘parallel’).  </P>
<P>  It also contains support for multiple RNG streams following L'Ecuyer _et al_ (2002), with support for both ‘mclapply’ and ‘snow’ clusters. This replaces functions like ‘clusterSetupRNG()’ from ‘snow’ (which are not in ‘parallel’).</INS>
</P></LI>
</UL>
<H4>2.14.0 BUG FIXES </H4>
<UL>
<LI><P>
On some systems ‘help(&quot;.C&quot;, help_type = &quot;pdf&quot;)’ and similar generated
file names that TeX was unable to handle.
</P></LI>

<LI><P>
Non-blocking listening socket connections continued to report
‘isIncomplete’ as true even when the peer had closed down.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-10-18</title>
<pubDate>Mon, 17 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/17#n2011-10-18</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/17#n2011-10-18</guid>
<description><![CDATA[<H4>2.14.0 BUG FIXES </H4>
<UL>
<LI><P>
‘stack()’ and ‘unstack()’ converted character columns to factors.
‘unstack()’ sometimes produced incorrect results (a list or a vector)
if the factor on which to un-split had only one level.
</P></LI>

<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">The ‘"table"’ method for ‘Axis()’ hardcoded ‘side = 1’, hence calls to ‘plot(&lt;vector&gt;, &lt;table&gt;)’ labelled the wrong axis. (PR#14699)</INS>
</P></LI>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">Creating a connection might fail under ‘gctorture(TRUE)’.</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2011-10-16</title>
<pubDate>Sat, 15 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/15#n2011-10-16</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/15#n2011-10-16</guid>
<description><![CDATA[<H4>2.14.0 DEPRECATED AND DEFUNCT </H4>
<UL>
<LI><P>
The ‘keep.source’ argument to ‘library()’ and ‘require()’ is
deprecated: it was only used for packages installed without
lazy-loading, and now all packages are lazy-loaded.
</P></LI>

</UL>
<H4>2.14.0 BUG FIXES </H4>
<UL>
<LI><P>
Creating almost any connection would fail under ‘gctorture()’.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-10-15</title>
<pubDate>Fri, 14 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/14#c2011-10-15</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/14#c2011-10-15</guid>
<description><![CDATA[<H4>2.14.0 INSTALLATION (Windows)</H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">Where both 32- and 64-bit versions of R are installed, the file association for ‘.RData’ files defaults to 64-bit R (it defaulted to 32-bit in R 2.12.x and 2.13.x).</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2011-10-14</title>
<pubDate>Thu, 13 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/13#n2011-10-14</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/13#n2011-10-14</guid>
<description><![CDATA[<H4>2.14.0 UTILITIES </H4>
<UL>
<LI><P>
Reference output for a vignette can be supplied when checking a package
by ‘R CMD check’: see ‘Writing R Extensions’.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-10-14</title>
<pubDate>Thu, 13 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/13#c2011-10-14</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/13#c2011-10-14</guid>
<description><![CDATA[<H4>2.14.0 BUG FIXES (Windows)</H4>
<UL>
<LI><P>
Raster drawing on ‘win.metafile()’ (or copying a plot that includes a
raster image from another device to a Metafile) now does not crash.
(Reported by Stefan Gelissen.)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-10-13</title>
<pubDate>Wed, 12 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/12#n2011-10-13</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/12#n2011-10-13</guid>
<description><![CDATA[<H4>2.14.0 SIGNIFICANT USER-VISIBLE CHANGES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">There is a new package ‘parallel’.  </P>
<P>  It incorporates (slightly revised) copies of packages ‘multicore’ and ‘snow’ (excluding MPI, PVM and NWS clusters).  Code written to use those packages should work unchanged (apart from changing any references to their namespaces to a reference to ‘parallel’).  </P>
<P>  It also contains support for multiple RNG streams following L'Ecuyer _et al_ (2002), with support for both ‘mclapply’ and ‘snow’ clusters. This replaces functions like ‘clusterSetupRNG()’ from ‘snow’ (which are not in ‘parallel’).</INS>
</P></LI>
</UL>
<H4>2.14.0 GRAPHICS DEVICES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">The default ‘colormodel’ for ‘pdf()’ and ‘postscript()’ is now called ‘"srgb"’ to more accurately describe it.  (Instead of ‘"rgb"’, and in the case of ‘postscript()’ it no longer switches to and from the gray colorspace, by default.)  </P>
<P>  The ‘colormodel’ for ‘postscript()’ which does use both gray and sRGB colorspace is now called ‘"srgb+gray"’.  </P>
<P>  Plots which are known to use only black/white/transparent can advantageously use ‘colormodel = "gray"’ (just as before, but there is now slightly more advantage in doing so).</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2011-10-11</title>
<pubDate>Mon, 10 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/10#n2011-10-11</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/10#n2011-10-11</guid>
<description><![CDATA[<H4>2.14.0 DEPRECATED AND DEFUNCT </H4>
<UL>
<LI><P>
It is deprecated now to use ‘mean(x)’ and ‘sd(x)’ directly on data
frames (or also matrices, for ‘sd’) ‘x’, instead of simply using
‘sapply’.  In the same spirit, ‘median(x)’ now gives an error for a
data frame ‘x’ (it often gave nonsensical results).
</P></LI>

<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">Use of ‘library.dynam()’ without specifying all the first three arguments is deprecated.  (It is often called from a namespace, and the defaults are only appropriate to a package.)  </P>
<P>  Use of ‘chname’ in ‘library.dynam()’ with the extension ‘.so’ or ‘.dll’ (which is clearly not allowed according to the help page) is deprecated.  This also applies to ‘library.dynam.unload()’ and ‘useDynLib’ directives in ‘NAMESPACE’ files.</INS>
</P></LI>
</UL>
<H4>2.14.0 BUG FIXES </H4>
<UL>
<LI><P>
The ‘&quot;table&quot;’ method for ‘Axis()’ hardcoded ‘side = 1’, hence calls
‘plot(&lt;vector&gt;, &lt;table&gt;)’ labelled the wrong axis. (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14699">PR#14699</a>)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-10-10</title>
<pubDate>Sun, 09 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/09#n2011-10-10</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/09#n2011-10-10</guid>
<description><![CDATA[<H4>2.14.0 BUG FIXES </H4>
<UL>
<LI><P>
‘round(x, -Inf)’ now does something sensible (return zero rather than
‘NA’).
</P></LI>

<LI><P>
‘signif(x, -Inf)’ now behaves as documented (‘signif(x, 1)’) rather
than giving zero.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-10-10</title>
<pubDate>Sun, 09 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/09#c2011-10-10</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/09#c2011-10-10</guid>
<description><![CDATA[<H4>2.14.0 NEW FEATURES (Windows)</H4>
<UL>
<LI><P>
There is a Danish translation of the RGui menus.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-10-09</title>
<pubDate>Sat, 08 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/08#n2011-10-09</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/08#n2011-10-09</guid>
<description><![CDATA[<H4>2.14.0 DEPRECATED AND DEFUNCT </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">Use of ‘library.dynam()’ without specifying all the first three arguments is deprecated.  (It is often called from a namespace, and the defaults are only appropriate to a package.)  </P>
<P>  Use of ‘chname’ in ‘library.dynam()’ with the extension ‘.so’ or ‘.dll’ (which is clearly not allowed according to the help page) is deprecated.  This also applies to ‘library.dynam.unload()’ and ‘useDynLib’ directives in ‘NAMESPACE’ files.  </P>
<P>  It is deprecated now to use ‘mean(x)’ and ‘sd(x)’ directly on data frames (or also matrices, for ‘sd’) ‘x’, instead of simply using ‘sapply’.  </P>
<P>  In the same spirit, ‘median(x)’ now gives an error for a data frame ‘x’, rather than often non-sense.</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2011-10-08</title>
<pubDate>Fri, 07 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/07#n2011-10-08</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/07#n2011-10-08</guid>
<description><![CDATA[<H4>2.14.0 BUG FIXES </H4>
<UL>
<LI><P>
The (undocumented) inclusion of superclass objects in default
initializing of reference classes overwrote explicit field arguments.
The bug is fixed, the feature documented and a test added.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-10-07</title>
<pubDate>Thu, 06 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/06#n2011-10-07</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/06#n2011-10-07</guid>
<description><![CDATA[<H4>2.14.0 SIGNIFICANT USER-VISIBLE CHANGES </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">There is a new package ‘parallel’.  </P>
<P>  It incorporates (slightly revised) copies of packages ‘multicore’ and ‘snow’ (excluding MPI, PVM and NWS clusters).  Code written to use those packages should work unchanged (apart from changing any references to their namespaces to a reference to ‘parallel’).  </P>
<P>  It also contains support for multiple RNG streams following L'Ecuyer _et al_ (2002), with support for both ‘mclapply’ and ‘snow’ clusters.  </P>
<P>  (It is not yet complete: some enhancements are planned by R 2.14.0.)</INS>
</P></LI>
</UL>]]></description>  
</item>
   <item>
<title>2011-10-05</title>
<pubDate>Tue, 04 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/04#n2011-10-05</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/04#n2011-10-05</guid>
<description><![CDATA[<H4>2.14.0 DEPRECATED AND DEFUNCT </H4>
<UL>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">‘R CMD Rd2dvi’ without the ‘--pdf’ option is now deprecated: only PDF output will be supported in future releases (since this allows the use of fonts only supported for PDF), and only ‘R CMD Rd2pdf’ will be available.</INS>
</P></LI>
<LI><P>
<INS STYLE="background:#E6FFE6;" TITLE="i=0">Use of ‘library.dynam()’ without specifying all the first three arguments is deprecated.  (It is often called from a namespace, and the defaults are only appropriate to a package.)  </P>
<P>  Use of ‘chname’ in ‘library.dynam()’ with the extension ‘.so’ or ‘.dll’ (which is clearly not allowed according to the help page) is deprecated.  This also applies to ‘library.dynam.unload()’ and ‘useDynLib’ directives in ‘NAMESPACE’ files.</INS>
</P></LI>
</UL>
<H4>2.14.0 BUG FIXES </H4>
<UL>
<LI><P>
‘HoltWinters()’ checks that the optimization succeeded. (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14694">PR#14694</a>)
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-10-04</title>
<pubDate>Mon, 03 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/03#c2011-10-04</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/03#c2011-10-04</guid>
<description><![CDATA[<H4>CHANGES IN R-devel NEW FEATURES (Windows)</H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘yLineBias’ of the ‘windows()’ family of devices has been changed
from 0.1 to 0.2: this changes slightly the vertical positioning of text
(including axis annotations).  This can be overridden by setting the
new ‘&quot;ylbias&quot;’ graphical parameter.  This was done for consistency with
other devices such as ‘pdf()’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD build’ once again attempts to preserve file permissions on
Windows.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
There is support for cairographics-based devices using the same code as
on Unix-alikes.  This can be selected by the new ‘type’ argument of the
bitmap devices ‘bmp()’, ‘jpeg()’, ‘png()’ and ‘tiff()’, and devices
‘svg()’, ‘cairo_pdf()’ and ‘cairo_ps()’ are now available on Windows.
</P>
<P>
These are not compiled in by default when building from source: see the
instructions in the ‘R Installation and Administration Manual’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
All the Windows-specific graphics devices now have a ‘family’ argument.
If non-empty this specifies an initial family to be used for fonts 1-4.
If empty the fonts specified in the ‘Rdevga’ configuration file are
used for the Windows GDI devices and ‘&quot;sans&quot;’ for cairographics-based
devices.
</P>
<P>
This will generally be a Windows font name such as ‘&quot;Lucida Bright&quot;’ or
one of the device-independent names (‘&quot;sans&quot;’, ‘&quot;serif&quot;’ and ‘&quot;mono&quot;’).
Outside Western Europe you may need to select a family that better
supports your locale such as ‘&quot;Arial MS Unicode&quot;’ or one specific to
Chinese/Korean/Thai ....
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
There is a new ‘antialias’ argument to ‘windows()’, ‘win.print()’ and
the bitmap devices.  This is an option that can be set in
‘windows.options()’ to set the default for ‘windows()’ (and
‘win.graph()’).
</P>
<P>
This gives a hint to the Windows plotting system.  Whether
anti-aliasing is actually used principally depends on the OS settings:
this argument should at least be able to turn it off.  The default
behaviour (unchanged from before) is that Windows will use
anti-aliasing for screen devices (and bitmap devices, as they plot on a
hidden screen) if ClearType has been enabled.  For those not using
ClearType, ‘windows.options(antialias = &quot;cleartype&quot;)’ will make this
the default, and it will probably give more legible plots.
</P>
<P>
The argument can also be used for the cairographics-based versions of
the bitmap devices.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘Update packages ...’ menu item now runs
‘update.packages(ask=&quot;graphics&quot;, checkBuilt=TRUE)’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD INSTALL’ preserves the package-directory modification time when
it restores an earlier install of the package.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
File extensions ‘.xz’, ‘.rda’ and ‘.RData’ have been added to those
which default to binary transfer for ‘download.file()’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘install.packages()’ and ‘R CMD check’ have a small delay after
removing a directory to counteract some interference from anti-virus
software.
</DEL>
</P></LI>

</UL>
<H4>CHANGES IN R-devel INSTALLATION (Windows)</H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
Compilation of C and Fortran code now uses the optimization flag
‘-mtune=core2’: this will improve performance a few percent on recent
CPUs at the expense of those which are several years old.  Its effect
is particularly evident on 64-bit builds.
</P>
<P>
This can be overridden when building from the sources: see the ‘EOPTS’
macro defined in file ‘MkRules.dist’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Where both 32- and 64-bit versions of R are installed, the file
association for @file.RData files defaults to 64-bit R (it defaulted to
32-bit in R 2.12.x and 2.13.x).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
There is preliminary support for ‘multilib’ toolchains which use
options ‘--m32’ or ‘--m64’ to select the architecture; set the
appropriate macros in ‘MkRules.local’.
</P>
<P>
It is the intention to move to such a toolchain when they become mature
enough.
</DEL>
</P></LI>

</UL>
<H4>CHANGES IN R-devel PACKAGE INSTALLATION (Windows)</H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘Rzlib.dll’ (sometimes used in packages _via_ ‘$(ZLIB_LIBS)’) does not
include the buggy gzio interface from zlib 1.2.5.
</DEL>
</P></LI>

</UL>
<H4>CHANGES IN R-devel DEFUNCT (Windows)</H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
zip.unpack() (deprecated in R 2.13.0) is defunct: use ‘unzip()’
instead.
</DEL>
</P></LI>

</UL>
<H4>2.14.0 NEW FEATURES (Windows)</H4>
<UL>
<LI><P>
The ‘yLineBias’ of the ‘windows()’ family of devices has been changed
from 0.1 to 0.2: this changes slightly the vertical positioning of text
(including axis annotations).  This can be overridden by setting the
new ‘&quot;ylbias&quot;’ graphical parameter.  This was done for consistency with
other devices such as ‘pdf()’.
</P></LI>

<LI><P>
‘R CMD build’ once again attempts to preserve file permissions on
Windows.
</P></LI>

<LI><P>
There is support for cairographics-based devices using the same code as
on Unix-alikes.  This can be selected by the new ‘type’ argument of the
bitmap devices ‘bmp()’, ‘jpeg()’, ‘png()’ and ‘tiff()’, and devices
‘svg()’, ‘cairo_pdf()’ and ‘cairo_ps()’ are now available on Windows.
</P>
<P>
These are not compiled in by default when building from source: see the
instructions in the ‘R Installation and Administration Manual’.
</P></LI>

<LI><P>
All the Windows-specific graphics devices now have a ‘family’ argument.
If non-empty this specifies an initial family to be used for fonts 1-4.
If empty the fonts specified in the ‘Rdevga’ configuration file are
used for the Windows GDI devices and ‘&quot;sans&quot;’ for cairographics-based
devices.
</P>
<P>
This will generally be a Windows font name such as ‘&quot;Lucida Bright&quot;’ or
one of the device-independent names (‘&quot;sans&quot;’, ‘&quot;serif&quot;’ and ‘&quot;mono&quot;’).
Outside Western Europe you may need to select a family that better
supports your locale such as ‘&quot;Arial MS Unicode&quot;’ or one specific to
Chinese/Korean/Thai ....
</P></LI>

<LI><P>
There is a new ‘antialias’ argument to ‘windows()’, ‘win.print()’ and
the bitmap devices.  This is an option that can be set in
‘windows.options()’ to set the default for ‘windows()’ (and
‘win.graph()’).
</P>
<P>
This gives a hint to the Windows plotting system.  Whether
anti-aliasing is actually used principally depends on the OS settings:
this argument should at least be able to turn it off.  The default
behaviour (unchanged from before) is that Windows will use
anti-aliasing for screen devices (and bitmap devices, as they plot on a
hidden screen) if ClearType has been enabled.  For those not using
ClearType, ‘windows.options(antialias = &quot;cleartype&quot;)’ will make this
the default, and it will probably give more legible plots.
</P>
<P>
The argument can also be used for the cairographics-based versions of
the bitmap devices.
</P></LI>

<LI><P>
The ‘Update packages ...’ menu item now runs
‘update.packages(ask=&quot;graphics&quot;, checkBuilt=TRUE)’.
</P></LI>

<LI><P>
‘R CMD INSTALL’ preserves the package-directory modification time when
it restores an earlier install of the package.
</P></LI>

<LI><P>
File extensions ‘.xz’, ‘.rda’ and ‘.RData’ have been added to those
which default to binary transfer for ‘download.file()’.
</P></LI>

<LI><P>
‘install.packages()’ and ‘R CMD check’ have a small delay after
removing a directory to counteract some interference from anti-virus
software.
</P></LI>

</UL>
<H4>2.14.0 INSTALLATION (Windows)</H4>
<UL>
<LI><P>
Compilation of C and Fortran code now uses the optimization flag
‘-mtune=core2’: this will improve performance a few percent on recent
CPUs at the expense of those which are several years old.  Its effect
is particularly evident on 64-bit builds.
</P>
<P>
This can be overridden when building from the sources: see the ‘EOPTS’
macro defined in file ‘MkRules.dist’.
</P></LI>

<LI><P>
Where both 32- and 64-bit versions of R are installed, the file
association for @file.RData files defaults to 64-bit R (it defaulted to
32-bit in R 2.12.x and 2.13.x).
</P></LI>

<LI><P>
There is preliminary support for ‘multilib’ toolchains which use
options ‘--m32’ or ‘--m64’ to select the architecture; set the
appropriate macros in ‘MkRules.local’.
</P>
<P>
It is the intention to move to such a toolchain when they become mature
enough.
</P></LI>

</UL>
<H4>2.14.0 PACKAGE INSTALLATION (Windows)</H4>
<UL>
<LI><P>
‘Rzlib.dll’ (sometimes used in packages _via_ ‘$(ZLIB_LIBS)’) does not
include the buggy gzio interface from zlib 1.2.5.
</P></LI>

</UL>
<H4>2.14.0 DEFUNCT (Windows)</H4>
<UL>
<LI><P>
zip.unpack() (deprecated in R 2.13.0) is defunct: use ‘unzip()’
instead.
</P></LI>

</UL>]]></description>  
</item>
   <item>
<title>2011-10-04</title>
<pubDate>Mon, 03 Oct 2011 19:00:00 GMT</pubDate>
<link>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/03#n2011-10-04</link>
<guid>https://developer.r-project.org/blosxom.cgi/R-2-14-branch/2011/10/03#n2011-10-04</guid>
<description><![CDATA[<H4>CHANGES IN R-devel SIGNIFICANT USER-VISIBLE CHANGES </H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
All packages must have a namespace, and one is created on installation
if not supplied in the sources.  This means that any package without a
namespace must be re-installed under this version of R (but data-only
packages without R code can still be used).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘yLineBias’ of the ‘X11()’ and ‘windows()’ families of devices has
been changed from 0.1 to 0.2: this changes slightly the vertical
positioning of text in the margins (including axis annotations).  This
is mainly for consistency with other devices such as ‘quartz()’ and
‘pdf()’.  (Wish of <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14538">PR#14538</a>.)
</P>
<P>
There is a new graphics parameter ‘&quot;ylbias&quot;’ which allows the y-line
bias of the graphics device to be tweaked, including to reproduce
output from earlier versions of R.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Labeling of the p-values in various anova tables has been rationalized
to be either ‘&quot;Pr(&gt;F)&quot;’ or ‘&quot;Pr(&gt;Chi)&quot;’ (i.e. the ‘&quot;Pr(F)&quot;’,
‘&quot;Pr(Chi)&quot;’ and ‘&quot;P(&gt;|Chi|)&quot;’ variants have been eliminated).  Code
which extracts the p value _via_ indexing by name may need adjustment.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘::’ can now be used for datasets made available for lazy-loading in
packages with namespaces (which makes it consistent with its use for
data-only packages without namespaces in earlier versions of R).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
There is a new package ‘parallel’.
</P>
<P>
It incorporates (slightly revised) copies of packages ‘multicore’ and
‘snow’ (excluding MPI, PVM and NWS clusters).  Code written to use
those packages should work unchanged (apart from changing any
references to their namespaces to a reference to ‘parallel’).
</P>
<P>
It also contains support for multiple RNG streams following L'Ecuyer
_et al_ (2002), with support for both ‘mclapply’ and ‘snow’ clusters.
</P>
<P>
(It is still incomplete/experimental.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Building PDF manuals (for R itself or packages, e.g. _via_ ‘R CMD
check’) by default requires the LaTeX package ‘inconsolata’: see the
section on ‘Making the manuals’ in the ‘R Installation and
Administration Manual’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘axTicks(*, log=TRUE)’ has changed in some cases to satisfy the
documented behavior and be consistent.
</DEL>
</P></LI>

</UL>
<H4>CHANGES IN R-devel NEW FEATURES </H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘txtProgressBar()’ can write to an open connection instead of the
console.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Non-portable package names ending in ‘.’ are no longer allowed.  Nor
are single-character package names (‘R’ was already disallowed).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘regexpr()’ and ‘gregexpr()’ with ‘perl = TRUE’ allows Python-style
named captures.  (Wish and contribution of <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14518">PR#14518</a>.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The placement of ‘plotmath’ text in the margins of plots done by base
graphics now makes the same vertical adjustment as ordinary text, so
using ordinary and plotmath text on the same margin line will seem
better aligned (but not exactly aligned, since ordinary text has
descenders below the baseline and plotmath places them on the
baseline).  (Related to <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14537">PR#14537</a>.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘sunflowerplot()’ now has a formula interface.  (Wish of <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14541">PR#14541</a>.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘iconv()’ has a new argument ‘toRaw’ to handle encodings such as
‘UTF-16’ with embedded nuls (as was possible before the ‘CHARSXP’ cache
was introduced).
</P>
<P>
It will also accept as input the type of list generated with ‘toRaw =
TRUE’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Garbage-collecting an unused input text connection no longer gives a
warning (since it ‘connects’ to nothing outside R).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘read.table()’ and ‘scan()’ have gained a ‘text’ argument, to allow
reading data from a (possibly literal) character string.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘optim(*, method = .)’ now allows ‘method = &quot;Brent&quot;’ as an interface to
‘optimize()’, for use in cases such as ‘mle()’ where ‘optim()’ is used
internally.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘mosaicplot()’ gains a ‘border’ argument.  (Wish of <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14550">PR#14550</a>.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘smooth.spline()’ gains a ‘tol’ argument which controls how different
‘x’ values need to be to be treated as distinct.  The default has been
changed to be more reliable for inputs whose range is small compared to
their maximum absolute value.  (Wish of <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14452">PR#14452</a>.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘gl()’ runs faster by avoiding calling ‘factor()’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘print()’ method for ‘object.size()’ accepts ‘B’ as well as ‘b’ as
an abbreviation for ‘bytes’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘unlink()’ gains a ‘force’ argument to work like ‘rm -f’ and if
possible override restrictive permissions.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘pbirthday()’ and ‘qbirthday()’ now use exact calculations for
‘coincident = 2’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘unzip()’ and ‘unz()’ connections have been updated with support for
more recent Zip64 features (including large file sizes and ‘bzip2’
compression, but not UTF-8 file names).
</P>
<P>
‘unzip()’ has a new option to restore file times from those recorded
(in an unknown timezone) in the zip file.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘update.packages()’ now accepts a character vector of package names for
the ‘oldPkgs’ argument.  (Suggestion of Tal Galili.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The special reference class fields ‘.self’ and ‘.refClassDef’ are now
read-only to prevent corrupting the object.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘decompose()’ now returns the original series as part of its value, so
it can be used (rather than reconstructed) when plotting.  (Suggestion
of Rob Hyndman.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Rao's efficient score test has been implemented for ‘glm’ objects.
Specifically, the ‘add1’, ‘drop1’, and ‘anova’ methods now allow ‘test
= &quot;Rao&quot;’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
If a saved workspace (e.g. ‘.RData’) contains objects that cannot be
loaded, R will now start with an warning message and an empty
workspace, rather than failing to start.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘strptime()’ now accepts times such as ‘24:00’ for midnight at the end
of the day, for although these are disallowed by POSIX 1003.1-2008, ISO
8601:2004 allows them.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Assignment of ‘names()’ to S4 objects now checks for a corresponding
‘&quot;names&quot;’ slot, and generates a warning or an error if that slot is not
defined.  See the section on slots in ‘?Classes’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The default methods for ‘is.finite()’, ‘is.infinite()’ and ‘is.nan()’
now signal an error if their argument is not an atomic vector.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The formula method for ‘plot()’ no longer places package ‘stats’ on the
search path (it loads the namespace instead).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
There now is a genuine ‘&quot;function&quot;’ method for ‘plot()’ rather than the
generic dispatching internally to ‘graphics::plot.function()’.  It is
now exported, so can be called directly as ‘plot.function()’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The one-sided ‘ks.test()’ allows ‘exact = TRUE’ to be specified in the
presence of ties (but the approximate calculation remains the default:
the ‘exact’ computation makes assumptions known to be invalid in the
presence of ties).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The behaviour of ‘curve(add = FALSE)’ has changed: it now no longer
takes the default x limits from the previous plot (if any): rather they
default to ‘c(0, 1)’ just as the ‘&quot;function&quot;’ method for ‘plot()’.  To
get the previous behaviour use ‘curve(add = NA)’, which also takes the
default for log-scaling of the x-axis from the previous plot.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Both ‘curve()’ and the ‘plot()’ method for functions have a new
argument ‘xname’ to facilitate plots such as ‘sin(t)’ _vs_ ‘t’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘local’ argument to ‘source()’ can specify an environment as well
as ‘TRUE’ (‘parent.env()’) and ‘FALSE’ (‘.GlobalEnv’).  It gives better
error messages for other values, such as ‘NA’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘vcov()’ gains methods for classes ‘&quot;summary.lm&quot;’ and ‘&quot;summary.glm&quot;’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘plot()’ method for class ‘&quot;profile.nls&quot;’ gains ‘ylab’ and ‘lty’
arguments, and passes ‘...’ on to ‘plot.default’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Character-string arguments such as the ‘mode’ argument of ‘vector()’,
‘as.vector()’ and ‘is.vector()’ and the ‘description’ argument of
‘file()’ are required to be of length exactly one, rather than any
further elements being silently discarded.  This helps catch incorrect
usage in programming.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘length’ argument of ‘vector()’ and its wrappers such as
‘numeric()’ is required to be of length exactly one (other values are
now an error rather than giving a warning as previously).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘vector(len)’ and ‘length(x) &lt;- len’ no longer acccept ‘TRUE’/‘FALSE’
for ‘len’ (not that they were ever documented to, but there was
special-casing in the C code).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
There is a new function ‘Sys.setFileTime()’ to set the time of a file
(including a directory).  See its help for exactly which times it sets
on various OSes.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The file times reported by ‘file.info()’ are reported to sub-second
resolution on systems which support it.  (Currently the POSIX 2008 and
FreeBSD/Darwin/NetBSD methods are detected.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
New function ‘getCall(m)’ as an abstraction for ‘m$call’, enabling
‘update()’'s default method to apply more universally.  (NB: this can
be masked by existing functions in packages.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘Sys.info()’ gains a ‘euser’ component to report the ‘effective’ user
on OSes which have that concept.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The result returned by ‘try()’ now contains the original error
condition object as the ‘&quot;condition&quot;’ attribute.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
All packages with R code are lazy-loaded irrespective of the ‘LazyLoad’
field in the ‘DESCRIPTION’ file.  A warning is given if the ‘LazyLoad’
field is overridden.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘Rd’ markup has a new ‘\figure’ tag so that figures can be included in
help pages when converted to HTML or LaTeX.  There are examples on the
help pages for ‘par()’ and ‘points()’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The built-in httpd server now allows access to files in the session
temporary directory ‘tempdir()’, addressed as the ‘/session’ directory
on the httpd server.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Development versions of R are no longer referred to by the number under
which they might be released, e.g. in the startup banner, ‘R --version’
and ‘sessionUtils()’.  The correct way to refer to a development
version of R is ‘R-devel’, preferably with the date and SVN version
number.
E.g. ‘R-devel (2011-07-04 r56266)’
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
There is a new function ‘texi2pdf()’ in package ‘tools’, currently a
convenience wrapper for ‘texi2dvi(pdf = TRUE)’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
There are two new options for typesetting PDF manuals from Rd files.
These are ‘beramono’ and ‘inconsolata’, and used the named font for
mononspaced output. They are intended to be used in combination with
‘times’, and ‘times,inconsolata,hyper’ is now the default for the
reference manual and package manuals.  If you do not have that font
installed, you can set ‘R_RD4PF’ to one of the other options: see the
‘R Installation and Administration Manual’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Automatic printing for reference classes is now done by the ‘$show()’
method.  A method is defined for class ‘envRefClass’ and may be
overriden for user classes (see the ‘?ReferenceClasses’ example).  S4
‘show()’ methods should no longer be needed for reference classes.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘tools::Rdiff’ (by default) and ‘R CMD Rdiff’ now ignore differences in
pointer values when comparing printed environments, compiled byte code,
etc.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘&quot;source&quot;’ attribute on functions created with ‘keep.source=TRUE’
has been replaced with a ‘&quot;srcref&quot;’ attribute.  The ‘&quot;srcref&quot;’
attribute references an in-memory copy of the source file using the
‘&quot;srcfilecopy&quot;’ class or the new ‘&quot;srcfilealias&quot;’ class.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
New items User Manuals and Technical Papers have been added to the HTML
help main page.  These link to vignettes in the base and recommended
packages and to a collection of papers about R issues, respectively.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Documentation and messages have been standardized to use “namespace”
rather than “name space”.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘setGeneric()’ now looks in the default packages for a non-generic
version of a function if called from a package with a namespace.  (It
always did for packages without a namespace.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Setting the environment variable ‘_R_WARN_ON_LOCKED_BINDINGS_’ will
give a warning if an attempt is made to change a locked binding.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘\SweaveInput’ is now supported when generating concordances in
‘Sweave()’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘findLineNum()’ and ‘setBreakpoint()’ now allow the environment to be
specified indirectly; the latter gains a ‘clear’ argument to allow it
to call ‘untrace()’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The body of a closure can be one of further types of R objects,
including enviroments and external pointers.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘Rd2HTML()’ function in package ‘tools’ now has a ‘stylesheet’
argument, allowing pages to be displayed in alternate formats.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
New function ‘requireNamespace()’ analogous to ‘require()’, returning a
logical value after attempting to load a namespace.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
There is a new type of RNG, ‘&quot;L'Ecuyer-CMRG&quot;’, implementing L'Ecuyer
(1999)'s ‘combined multiple-recursive generator’ ‘MRG32k3a’.  See the
comments on ‘?RNG’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘help.search()’ and ‘??’ can now display vignettes and demos as well as
help pages.  The new option ‘&quot;help.search.types&quot;’ controls the types of
documentation and the order of their display.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘socketConnection()’ now has a ‘timeout’ argument.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The initialization of the random-number generator now uses the process
ID as well as the current time, just in case two R processes are
launched very rapidly on a machine with low-resolution wall clock (some
have a resolution of a second; modern systems have microsecond-level
resolution).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
New function ‘pskill()’ in the ‘tools’ package to send a terminate
signal to one or more processes, plus constants such as ‘SIGTERM’ to
provide a portable way to refer to signals (since the numeric values
are OS-dependent).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
New function ‘psnice()’ in the ‘tools’ package to return or change the
‘niceness’ of a process.  (Refers to the ‘priority class’ on Windows.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘list.dirs()’ gains a ‘recursive’ argument.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
An ‘Authors@R’ field in a package ‘DESCRIPTION’ file can now be used to
generate ‘Author’ and ‘Maintainer’ fields if needed, and to
auto-generate package citations.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
New utility ‘getElement()’ for accessing either a list component or a
slot in an S4 object.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘stars()’ gains a ‘col.lines’ argument, thanks to Dustin Sallings.
(Wish of <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14657">PR#14657</a>.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
New function ‘regmatches()’ for extracting or replacing matched or
non-matched substrings from match data obtained by ‘regexpr()’,
‘gregexpr()’ and ‘regexec()’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘help(package = &quot;pkg_name&quot;, help_type = &quot;HTML&quot;)’ now gives HTML help on
the package rather than text help.  (This gives direct access to the
HTML version of the package manual shown _via_ ‘help.start()’'s
‘Packages’ menu.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘agrep()’ gains a ‘fixed’ argument to optionally allow approximate
regular expression matching, and a ‘costs’ argument to specify possibly
different integer match costs for insertions, deletions and
substitutions.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘read.dcf()’ and ‘write.dcf()’ gain a ‘keep.white’ argument to indicate
fields where whitespace should be kept as is.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘available.packages()’ now works around servers that fail to return an
error code when ‘PACKAGES.gz’ does not exist.  (Patch submitted by Seth
Schommer.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘readBin()’ can now read more than 2^31 - 1 bytes in a single call (the
previously documented limitation).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
New function ‘regexec()’ for finding the positions of matches as well
as all substrings corresponding to parenthesized subexpressions of the
given regular expression.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
New function ‘adist()’ in package ‘utils’ for computing ‘edit’
(generalized Levenshtein) distances between strings.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Class ‘&quot;raster&quot;’ gains an ‘is.na’ method to avoid confusion from the
misuse of the matrix method (such as <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14618">PR#14618</a>).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘identical()’ function gains an ‘ignore.bytecode’ argument to
control comparison of compiled functions.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘pmin’ and ‘pmax’ now warn if an argument is partially recycled (wish
of <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14638">PR#14638</a>).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The default for ‘image(useRaster=)’ is now taken from option
‘&quot;preferRaster&quot;’: for the small print see ‘?image’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘str()’ now displays reference class objects and their fields, rather
than treating them as classical S4 classes.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
New function ‘aregexec()’ in package ‘utils’ for finding the positions
of approximate string matches as well as all substrings corresponding
to parenthesized subexpressions of the given regular expression.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘download.file()’ has an ‘extra’ argument to pass additional
command-line options to the non-default methods using command-line
utilities.
</P>
<P>
‘cacheOK = FALSE’ is now supported for ‘method = &quot;curl&quot;’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘interaction.plot(*, type = .)’ now also allows type ‘&quot;o&quot;’ or ‘&quot;c&quot;’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘axTicks(*, log=TRUE)’ did sometimes give more values than the ticks in
the corresponding ‘graphics::axis()’.  By default, it now makes use of
the new (‘graphics’-package independent) ‘axisTicks()’ which can make
use of a new utility ‘.axisPars()’.  Further, it now returns a
decreasing sequence (as for ‘log=FALSE’) when ‘usr’ is decreasing.
</DEL>
</P></LI>

</UL>
<H4>CHANGES IN R-devel GRAPHICS DEVICES </H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘pdf()’ device makes use of Flate compression: this is controlled
by the new logical argument ‘compress’, and is enabled by default.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Devices ‘svg()’, ‘cairo_pdf()’ and ‘cairo_ps()’ gain a ‘family’
argument.
</P>
<P>
On a Unix-alike ‘X11()’ gains a ‘family’ argument.  This is one of the
‘x11.options()’ and so can be passed as an argument to the ‘bmp()’,
‘jpeg()’, ‘png()’ and ‘tiff()’ devices.
</P>
<P>
Analogous changes have been made on Windows, so all built-in R graphics
devices now have a ‘family’ argument except ‘pictex()’ (which has no
means to change fonts).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘bmp()’, ‘jpeg()’, ‘png()’ and ‘tiff()’ devices now make use of the
‘antialias’ argument for ‘type = &quot;quartz&quot;’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
There are several new built-in font mappings for ‘X11(type = &quot;Xlib&quot;)’:
see the help on ‘X11Fonts()’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
There is a new type ‘X11(type = &quot;dbcairo&quot;)’ which updates the screeen
less frequently: see its help page.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘X11()’ device now makes use of cursors to distinguish its states.
The normal cursor is an arrow (rather than a crosshair); the crosshair
is used when the locator is in use, and a watch cursor is shown when
plotting computations are being done.  (These are the standard names
for X11 cursors: how they are actually displayed depends on the window
manager.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
New functions ‘dev.hold()’ and ‘dev.flush()’ for use with graphics
devices with buffering.  These are used for most of the high-level
graphics functions such as ‘boxplot()’, so that the plot is only
displayed when the page is complete.
</P>
<P>
Currently implemented for ‘windows(buffered = TRUE)’, ‘quartz()’ and
the cairographics-based ‘X11()’ types with buffering (which are the
default on-screen devices).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
New function ‘dev.capture()’ for capture of bitmap snapshots of
image-based devices (a superset of the functionality provided by
‘grid.cap()’ in ‘grid’).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The default ‘colormodel’ for ‘pdf()’ and ‘postscript()’ is now called
‘&quot;srgb&quot;’ to more accurately describe it.  (Instead of ‘&quot;rgb&quot;’, and in
the case of ‘postscript()’ it no longer switches to and from the gray
colorspace, by default.)
</P>
<P>
The‘colormodel’ for ‘postscript()’ which does use both gray and sRGB
colorspace is now called ‘&quot;srgb+gray&quot;’.
</P>
<P>
Plots which are known to use only black/white/transparent can
advantageously use ‘colormodel = &quot;gray&quot;’ (just as before, but there is
now slightly more advantage in doing so).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘postscript()’ with values ‘colormodel = &quot;rgb&quot;’ and ‘colormodel =
&quot;rgb-nogray&quot;’ give the behaviour prior to R 2.13.0 of uncalibrated RGB,
which under some circumstances can be rendered much faster by a viewer.
</P>
<P>
‘pdf(colormodel = &quot;rgb&quot;)’ gives the behaviour prior to R 2.13.0 of
uncalibrated RGB, which under some circumstances can be rendered faster
by a viewer, and the files will be smaller (by about 9KB if compression
is not used).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘postscript()’ device only includes the definition of the sRGB
colorspace in the output file for the colormodels which use it.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘postscript()’ and ‘pdf()’ devices now output greyscale raster
images (and not RGB) when ‘colormodel = &quot;gray&quot;’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘postscript(colormodel = &quot;gray&quot;)’ now accepts non-grey colours and uses
their luminance (as ‘pdf()’ long has).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘colormodel = &quot;grey&quot;’ is allowed as an alternative name for
‘postscript()’ and ‘pdf()’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘pdf()’ in the default sRGB colorspace outputs many fewer changes of
colorspace, which may speed up rendering in some viewing applications.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
There is a new function ‘dev.capabilities()’ to query the capabilities
of the current device.  The initial set of capabilities are support for
semi-transparent colours, rendering and capturing raster images, the
locator and for interactive events.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
For ‘pdf()’, ‘maxRasters’ is increased as needed so the argument is no
longer used.
</DEL>
</P></LI>

</UL>
<H4>CHANGES IN R-devel SWEAVE &amp; VIGNETTES </H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
Options ‘keep.source = TRUE, figs.only = FALSE’ are now the default.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The way the type of user-defined options is determined has changed.
Previously they were all regarded as logical: now the type is
determined by the value given at first use.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The allowed values of logical options are now precisely those allowed
for character inputs to ‘as.logical()’: this means that ‘t’ and ‘f’ are
no longer allowed (although ‘T’ and ‘F’ still are).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The preferred location for vignette sources is now the directory
‘vignettes’ and not ‘inst/doc’: ‘R CMD build’ will now re-build
vignettes in directory ‘vignettes’ and copy the ‘.Rnw’ (etc) files and
the corresponding PDFs to ‘inst/doc’.  Further files to be copied to
‘inst/doc’ can be specified _via_ the file ‘vignettes/.install_extras’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD Sweave’ now supports a ‘--driver’ option to select the Sweave
driver: the default is equivalent to ‘--driver=RweaveLatex’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD Sweave’ and ‘R CMD Stangle’ support options ‘--encoding’ and
‘--options’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘Rtangle()’ driver allows ‘output = &quot;stdout&quot;’ or ‘output =
&quot;stderr&quot;’ to select the output or message connection.  This is
convenient for scripting using something like
</P>
<P>
    R CMD Stangle --options='output=&quot;stdout&quot;' foo.Rnw &gt; foo2.R
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
There is a new option ‘pdf.compress’ controlling whether PDF figures
are generated using Flate compression (they are by default).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD Sweave’ now has a ‘--pdf’ option to produce a PDF version of the
processed Sweave document.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
It is no longer allowed to have two vignettes with the same vignette
basename (e.g. ‘vig.Rnw’ and ‘vig.Snw’).  (Previously one vignette hid
the other in the ‘vignette()’ function.)
</DEL>
</P></LI>

</UL>
<H4>CHANGES IN R-devel C-LEVEL FACILITIES </H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
Function ‘R_tmpnam2’ has been added to the API to allow a temporary
filename to include a specified extension.
</DEL>
</P></LI>

</UL>
<H4>CHANGES IN R-devel PACKAGE INSTALLATION </H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
Package ‘DESCRIPTION’ file field ‘KeepSource’ forces the package to be
installed with ‘keep.source = TRUE’ (or ‘FALSE’).  (Suggestion of Greg
Snow.  Note that as all packages are lazy-loaded, this is now only
relevant at installation.)
</P>
<P>
There are corresponding options ‘--with-keep.source’ and
‘--without-keep.source’ for ‘R CMD INSTALL’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD INSTALL’ has a new option ‘--byte-compile’ to byte-compile the
packages during installation (since all packages are now lazy-loaded).
This can be controlled on a per-package basis by the optional field
‘ByteCompile’ in the ‘DESCRIPTION’ file.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
A package R code but without a ‘NAMESPACE’ file will have a default one
created at ‘R CMD build’ or ‘R CMD INSTALL’ time, so all packages will
be installed with namespaces.  A consequence of this is that
‘.First.lib()’ functions need to be copied to ‘.onLoad()’ (usually) or
‘.onAttach()’.  For the time being, if there is an auto-generated
‘NAMESPACE’ file and no ‘.onLoad()’ nor ‘.onAttach()’ function is found
but ‘.First.lib()’ is, it will be run as the attach hook (unless the
package is one of a list of known exceptions, when it will be run as
the load hook).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
A warning is given if test-loading a package changes a locked binding
in a package other than itself.  It is likely that this will be
disallowed in future releases.  (There are _pro tem_ some exceptions to
the warning.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
A dependency on SVN revision is allowed for ‘R’, e.g. ‘R (&gt;= r56550)’.
This should be used in conjunction with a version number, e.g. ‘R (&gt;=
2.14.0), R (&gt;= r56550)’ to distinguish beteen R-patched and R-devel
versions with the same SVN revision.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘installed.packages()’ now hashes the names of its cache files to avoid
very rare problems with excessively long path names.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14669">PR#14669</a>)
</DEL>
</P></LI>

</UL>
<H4>CHANGES IN R-devel UTILITIES </H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD check’ now gives an error if the R code in a vignette fails to
run, unless this is caused by a missing package.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD check’ now unpacks tarballs in the same way as ‘R CMD INSTALL’,
including making use of the environment variable ‘R_INSTALL_TAR’ to
override the default behaviour.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD check’ performs additional code analysis of package startup
functions, and notifies about incorrect argument lists and (incorrect)
calls to functions which modify the search path or inappropriately
generate messages.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD check’ now also checks compiled code for symbols corresponding
to functions which might terminate R or write to ‘stdout’/‘stderr’
instead of the console.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD check’ now uses a ‘pdf()’ device when checking examples (rather
than ‘postscript()’).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD check’ now reports as a NOTE what look like methods documented
with their full names even if there is a namespace and they are
exported.  In almost all cases they are intended to be used only as
methods and should use the ‘\method’ markup.  In the other rare cases
the recommended form is to use a function such as ‘coefHclust’ which
would not get confused with a method, document that and register it in
the ‘NAMESPACE’ file by ‘s3method(coef, hclust, coefHclust)’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The default for the environment variable ‘_R_CHECK_COMPACT_DATA2_’ is
now true: thus if using the newer forms of compression introduced in R
2.10.0 would be beneficial is now checked (by default).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD Rd2dvi’ allows the use of LaTeX package ‘inputenx’ rather than
‘inputenc’: the value of the environment variable ‘RD2DVI_INPUTENC’ is
used.  (LaTeX package ‘inputenx’ is an optional install which provides
greater coverage of the UTF-8 encoding.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘Rscript’ on a Unix-alike now accepts file names containing spaces
(provided these are escaped or quoted in the shell).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD build’ on a Unix-alike (only) now tries to preserve dates on
files it copies from its input directory.  (This was the undocumented
behaviour prior to R 2.13.0.)
</DEL>
</P></LI>

</UL>
<H4>CHANGES IN R-devel DEPRECATED AND DEFUNCT </H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘require()’ no longer has a ‘save’ argument.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘gamma’ argument to ‘hsv()’, ‘rainbow()’, and ‘rgb2hsv()’ has been
removed.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘--no-docs’ option for ‘R CMD build --binary’ is defunct: use
‘--install-args’ instead.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The option ‘--unsafe’ to ‘R CMD INSTALL’ is defunct: use the identical
option ‘--no-lock’ instead.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The entry point ‘pythag’ formerly in ‘Rmath.h’ is defunct: use instead
the C99 function ‘hypot’.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD build --binary’ is formally defunct: ‘R CMD INSTALL --build’ has
long been the preferred alternative.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘zip.file.extract()’ is now defunct: use ‘unzip()’ or ‘unz()’ instead.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
‘R CMD Rd2dvi’ without the ‘--pdf’ option is now deprecated: only PDF
output will be supported in future releases (since this allows the use
of fonts only supported for PDF).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
A top-level ‘COPYING’ file in a package is now deprecated (file names
‘LICENSE’ or ‘LICENCE’ have long been preferred), and will not be
installed as from R 2.15.0.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Options such as ‘--max-nsize’ and the function ‘mem.limits()’ are now
deprecated: these limits are nowadays almost never used, and are
reported by ‘gc()’ when they are in use.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Forms like ‘binomial(link = &quot;link&quot;)’ for GLM families deprecated since
R 2.4.0 are now defunct.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘declarativeOnly’ argument to ‘loadNamespace()’ (not relevant since
R 2.13.0) has been removed.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Use of ‘library.dynam()’ without specifying all the first three
arguments is deprecated.  (It is often called from a namespace, and the
defaults are only appropriate to a package.)
</DEL>
</P></LI>

</UL>
<H4>CHANGES IN R-devel INSTALLATION </H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
The base and recommended packages are now byte-compiled (equivalent to
‘make bytecode’ in R 2.13.x).
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Configure option ‘--with-system-zlib’ now only makes use of the basic
interface of ‘zlib’ and not the C function ‘gzseek’ which has shown
erroneous behaviour in ‘zlib’ 1.2.4 and 1.2.5.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘zlib’ in the R sources is now version 1.2.5.  (This is safe even
on 32-bit Linux systems because only the basic interface is now used.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘.afm’ files in package ‘grDevices’ are now installed as compressed
files (as long done on Windows), saving ca 2MB on the installed size.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The non-screen cairo-based devices are no longer in the X11 module and
so can be installed without X11.  (We have never seen a Unix-alike
system with cairographics installed but not X11, but a user might
select ‘--without-x’.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Configure will try to use ‘-fobjc-exceptions’ for the Objective-C
compiler (if present) to ensure that even compilers that do not enable
exceptions by default (such as vanilla gcc) can be used.  (Objective-C
is currently only used on Mac OS X.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The system call ‘times’ is required.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The C99 functions ‘acosh’, ‘asinh’, ‘atanh’, ‘snprintf’ and ‘vsnprintf’
are now required.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
There is no longer support for making DVI manuals _via_ ‘make dvi’,
‘make install-dvi’ and similar.  Only PDF manuals are supported (to
allow the use of fonts which are only available for PDF.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘configure’ arguments used during configuration of R are included
as a comment in ‘Makeconf’ for informative purposes on Unix-alikes in a
form suitable for shell execution.  Note that those are merely
command-line arguments, they do not include environment variables (one
more reason to use configure variables instead) or site configuration
settings.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Framework installation now supports ‘DESTDIR’ (Mac OS X only).
</DEL>
</P></LI>

</UL>
<H4>CHANGES IN R-devel BUG FIXES </H4>
<UL>
<LI><P>
<DEL STYLE="background:#FFE6E6;">
The locale category ‘LC_MONETARY’ was only being set on startup on
Windows: that is now done on Unix-alikes where supported.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Reference class utilities will detect an attempt to modify methods or
fields in a locked class definition (e.g., in a namespace) and generate
an error.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The formula methods for ‘lines()’, ‘points()’ and ‘text()’ now work
even if package ‘stats’ is not on the search path.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
In principle, S4 classes from different packages could have the same
name.  This has not previously worked.  Changes have now been installed
that should allow such classes and permit methods to use them.  New
functions ‘className()’ and ‘multipleClasses()’ are related tools for
programming.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Work around an issue in Linux (a system ‘select’ call resetting ‘tv’)
which prevented internet operations from timing out properly.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
Several stack trampling and overflow issues have been fixed in TRE,
triggered by ‘agrep’ and friends with long patterns.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14627">PR#14627</a>.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
(“design infelicity”) Field assignments in reference classes are now
consistent with slots in S4 classes: the assigned value must come from
the declared class (if any) for the field or from a subclass.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The methods objects constructed for ‘&quot;coerce&quot;’ and ‘&quot;coerce&lt;-&quot;’ were
lacking some essential information in the ‘generic’, ‘defined’ and
‘target’ slots; ‘as()’ did not handle duplicate class definitions
correctly.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The parser no longer accepts the digit ‘8’ in an octal character code
in a string, nor does it accept unterminated strings in a file.
(Reported by Bill Dunlap.)
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
The ‘print()’ method for class ‘&quot;summary.aov&quot;’ did not pass on argument
‘digits’ when ‘summary()’ was called on a single object, and hence used
more digits than documented.
</DEL>
</P></LI>

<LI><P>
<DEL STYLE="background:#FFE6E6;">
X11 cairo back-end produced incorrect capture snapshot images on
big-endian machines.
</DEL>
</P></LI>

</UL>
<H4>2.14.0 SIGNIFICANT USER-VISIBLE CHANGES </H4>
<UL>
<LI><P>
All packages must have a namespace, and one is created on installation
if not supplied in the sources.  This means that any package without a
namespace must be re-installed under this version of R (but data-only
packages without R code can still be used).
</P></LI>

<LI><P>
The ‘yLineBias’ of the ‘X11()’ and ‘windows()’ families of devices has
been changed from 0.1 to 0.2: this changes slightly the vertical
positioning of text in the margins (including axis annotations).  This
is mainly for consistency with other devices such as ‘quartz()’ and
‘pdf()’.  (Wish of <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14538">PR#14538</a>.)
</P>
<P>
There is a new graphics parameter ‘&quot;ylbias&quot;’ which allows the y-line
bias of the graphics device to be tweaked, including to reproduce
output from earlier versions of R.
</P></LI>

<LI><P>
Labeling of the p-values in various anova tables has been rationalized
to be either ‘&quot;Pr(&gt;F)&quot;’ or ‘&quot;Pr(&gt;Chi)&quot;’ (i.e. the ‘&quot;Pr(F)&quot;’,
‘&quot;Pr(Chi)&quot;’ and ‘&quot;P(&gt;|Chi|)&quot;’ variants have been eliminated).  Code
which extracts the p value _via_ indexing by name may need adjustment.
</P></LI>

<LI><P>
‘::’ can now be used for datasets made available for lazy-loading in
packages with namespaces (which makes it consistent with its use for
data-only packages without namespaces in earlier versions of R).
</P></LI>

<LI><P>
There is a new package ‘parallel’.
</P>
<P>
It incorporates (slightly revised) copies of packages ‘multicore’ and
‘snow’ (excluding MPI, PVM and NWS clusters).  Code written to use
those packages should work unchanged (apart from changing any
references to their namespaces to a reference to ‘parallel’).
</P>
<P>
It also contains support for multiple RNG streams following L'Ecuyer
_et al_ (2002), with support for both ‘mclapply’ and ‘snow’ clusters.
</P>
<P>
(It is still incomplete/experimental.)
</P></LI>

<LI><P>
Building PDF manuals (for R itself or packages, e.g. _via_ ‘R CMD
check’) by default requires the LaTeX package ‘inconsolata’: see the
section on ‘Making the manuals’ in the ‘R Installation and
Administration Manual’.
</P></LI>

<LI><P>
‘axTicks(*, log=TRUE)’ has changed in some cases to satisfy the
documented behavior and be consistent.
</P></LI>

</UL>
<H4>2.14.0 NEW FEATURES </H4>
<UL>
<LI><P>
‘txtProgressBar()’ can write to an open connection instead of the
console.
</P></LI>

<LI><P>
Non-portable package names ending in ‘.’ are no longer allowed.  Nor
are single-character package names (‘R’ was already disallowed).
</P></LI>

<LI><P>
‘regexpr()’ and ‘gregexpr()’ with ‘perl = TRUE’ allows Python-style
named captures.  (Wish and contribution of <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14518">PR#14518</a>.)
</P></LI>

<LI><P>
The placement of ‘plotmath’ text in the margins of plots done by base
graphics now makes the same vertical adjustment as ordinary text, so
using ordinary and plotmath text on the same margin line will seem
better aligned (but not exactly aligned, since ordinary text has
descenders below the baseline and plotmath places them on the
baseline).  (Related to <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14537">PR#14537</a>.)
</P></LI>

<LI><P>
‘sunflowerplot()’ now has a formula interface.  (Wish of <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14541">PR#14541</a>.)
</P></LI>

<LI><P>
‘iconv()’ has a new argument ‘toRaw’ to handle encodings such as
‘UTF-16’ with embedded nuls (as was possible before the ‘CHARSXP’ cache
was introduced).
</P>
<P>
It will also accept as input the type of list generated with ‘toRaw =
TRUE’.
</P></LI>

<LI><P>
Garbage-collecting an unused input text connection no longer gives a
warning (since it ‘connects’ to nothing outside R).
</P></LI>

<LI><P>
‘read.table()’ and ‘scan()’ have gained a ‘text’ argument, to allow
reading data from a (possibly literal) character string.
</P></LI>

<LI><P>
‘optim(*, method = .)’ now allows ‘method = &quot;Brent&quot;’ as an interface to
‘optimize()’, for use in cases such as ‘mle()’ where ‘optim()’ is used
internally.
</P></LI>

<LI><P>
‘mosaicplot()’ gains a ‘border’ argument.  (Wish of <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14550">PR#14550</a>.)
</P></LI>

<LI><P>
‘smooth.spline()’ gains a ‘tol’ argument which controls how different
‘x’ values need to be to be treated as distinct.  The default has been
changed to be more reliable for inputs whose range is small compared to
their maximum absolute value.  (Wish of <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14452">PR#14452</a>.)
</P></LI>

<LI><P>
‘gl()’ runs faster by avoiding calling ‘factor()’.
</P></LI>

<LI><P>
The ‘print()’ method for ‘object.size()’ accepts ‘B’ as well as ‘b’ as
an abbreviation for ‘bytes’.
</P></LI>

<LI><P>
‘unlink()’ gains a ‘force’ argument to work like ‘rm -f’ and if
possible override restrictive permissions.
</P></LI>

<LI><P>
‘pbirthday()’ and ‘qbirthday()’ now use exact calculations for
‘coincident = 2’.
</P></LI>

<LI><P>
‘unzip()’ and ‘unz()’ connections have been updated with support for
more recent Zip64 features (including large file sizes and ‘bzip2’
compression, but not UTF-8 file names).
</P>
<P>
‘unzip()’ has a new option to restore file times from those recorded
(in an unknown timezone) in the zip file.
</P></LI>

<LI><P>
‘update.packages()’ now accepts a character vector of package names for
the ‘oldPkgs’ argument.  (Suggestion of Tal Galili.)
</P></LI>

<LI><P>
The special reference class fields ‘.self’ and ‘.refClassDef’ are now
read-only to prevent corrupting the object.
</P></LI>

<LI><P>
‘decompose()’ now returns the original series as part of its value, so
it can be used (rather than reconstructed) when plotting.  (Suggestion
of Rob Hyndman.)
</P></LI>

<LI><P>
Rao's efficient score test has been implemented for ‘glm’ objects.
Specifically, the ‘add1’, ‘drop1’, and ‘anova’ methods now allow ‘test
= &quot;Rao&quot;’.
</P></LI>

<LI><P>
If a saved workspace (e.g. ‘.RData’) contains objects that cannot be
loaded, R will now start with an warning message and an empty
workspace, rather than failing to start.
</P></LI>

<LI><P>
‘strptime()’ now accepts times such as ‘24:00’ for midnight at the end
of the day, for although these are disallowed by POSIX 1003.1-2008, ISO
8601:2004 allows them.
</P></LI>

<LI><P>
Assignment of ‘names()’ to S4 objects now checks for a corresponding
‘&quot;names&quot;’ slot, and generates a warning or an error if that slot is not
defined.  See the section on slots in ‘?Classes’.
</P></LI>

<LI><P>
The default methods for ‘is.finite()’, ‘is.infinite()’ and ‘is.nan()’
now signal an error if their argument is not an atomic vector.
</P></LI>

<LI><P>
The formula method for ‘plot()’ no longer places package ‘stats’ on the
search path (it loads the namespace instead).
</P></LI>

<LI><P>
There now is a genuine ‘&quot;function&quot;’ method for ‘plot()’ rather than the
generic dispatching internally to ‘graphics::plot.function()’.  It is
now exported, so can be called directly as ‘plot.function()’.
</P></LI>

<LI><P>
The one-sided ‘ks.test()’ allows ‘exact = TRUE’ to be specified in the
presence of ties (but the approximate calculation remains the default:
the ‘exact’ computation makes assumptions known to be invalid in the
presence of ties).
</P></LI>

<LI><P>
The behaviour of ‘curve(add = FALSE)’ has changed: it now no longer
takes the default x limits from the previous plot (if any): rather they
default to ‘c(0, 1)’ just as the ‘&quot;function&quot;’ method for ‘plot()’.  To
get the previous behaviour use ‘curve(add = NA)’, which also takes the
default for log-scaling of the x-axis from the previous plot.
</P></LI>

<LI><P>
Both ‘curve()’ and the ‘plot()’ method for functions have a new
argument ‘xname’ to facilitate plots such as ‘sin(t)’ _vs_ ‘t’.
</P></LI>

<LI><P>
The ‘local’ argument to ‘source()’ can specify an environment as well
as ‘TRUE’ (‘parent.env()’) and ‘FALSE’ (‘.GlobalEnv’).  It gives better
error messages for other values, such as ‘NA’.
</P></LI>

<LI><P>
‘vcov()’ gains methods for classes ‘&quot;summary.lm&quot;’ and ‘&quot;summary.glm&quot;’.
</P></LI>

<LI><P>
The ‘plot()’ method for class ‘&quot;profile.nls&quot;’ gains ‘ylab’ and ‘lty’
arguments, and passes ‘...’ on to ‘plot.default’.
</P></LI>

<LI><P>
Character-string arguments such as the ‘mode’ argument of ‘vector()’,
‘as.vector()’ and ‘is.vector()’ and the ‘description’ argument of
‘file()’ are required to be of length exactly one, rather than any
further elements being silently discarded.  This helps catch incorrect
usage in programming.
</P></LI>

<LI><P>
The ‘length’ argument of ‘vector()’ and its wrappers such as
‘numeric()’ is required to be of length exactly one (other values are
now an error rather than giving a warning as previously).
</P></LI>

<LI><P>
‘vector(len)’ and ‘length(x) &lt;- len’ no longer acccept ‘TRUE’/‘FALSE’
for ‘len’ (not that they were ever documented to, but there was
special-casing in the C code).
</P></LI>

<LI><P>
There is a new function ‘Sys.setFileTime()’ to set the time of a file
(including a directory).  See its help for exactly which times it sets
on various OSes.
</P></LI>

<LI><P>
The file times reported by ‘file.info()’ are reported to sub-second
resolution on systems which support it.  (Currently the POSIX 2008 and
FreeBSD/Darwin/NetBSD methods are detected.)
</P></LI>

<LI><P>
New function ‘getCall(m)’ as an abstraction for ‘m$call’, enabling
‘update()’'s default method to apply more universally.  (NB: this can
be masked by existing functions in packages.)
</P></LI>

<LI><P>
‘Sys.info()’ gains a ‘euser’ component to report the ‘effective’ user
on OSes which have that concept.
</P></LI>

<LI><P>
The result returned by ‘try()’ now contains the original error
condition object as the ‘&quot;condition&quot;’ attribute.
</P></LI>

<LI><P>
All packages with R code are lazy-loaded irrespective of the ‘LazyLoad’
field in the ‘DESCRIPTION’ file.  A warning is given if the ‘LazyLoad’
field is overridden.
</P></LI>

<LI><P>
‘Rd’ markup has a new ‘\figure’ tag so that figures can be included in
help pages when converted to HTML or LaTeX.  There are examples on the
help pages for ‘par()’ and ‘points()’.
</P></LI>

<LI><P>
The built-in httpd server now allows access to files in the session
temporary directory ‘tempdir()’, addressed as the ‘/session’ directory
on the httpd server.
</P></LI>

<LI><P>
Development versions of R are no longer referred to by the number under
which they might be released, e.g. in the startup banner, ‘R --version’
and ‘sessionUtils()’.  The correct way to refer to a development
version of R is ‘R-devel’, preferably with the date and SVN version
number.
E.g. ‘R-devel (2011-07-04 r56266)’
</P></LI>

<LI><P>
There is a new function ‘texi2pdf()’ in package ‘tools’, currently a
convenience wrapper for ‘texi2dvi(pdf = TRUE)’.
</P></LI>

<LI><P>
There are two new options for typesetting PDF manuals from Rd files.
These are ‘beramono’ and ‘inconsolata’, and used the named font for
mononspaced output. They are intended to be used in combination with
‘times’, and ‘times,inconsolata,hyper’ is now the default for the
reference manual and package manuals.  If you do not have that font
installed, you can set ‘R_RD4PF’ to one of the other options: see the
‘R Installation and Administration Manual’.
</P></LI>

<LI><P>
Automatic printing for reference classes is now done by the ‘$show()’
method.  A method is defined for class ‘envRefClass’ and may be
overriden for user classes (see the ‘?ReferenceClasses’ example).  S4
‘show()’ methods should no longer be needed for reference classes.
</P></LI>

<LI><P>
‘tools::Rdiff’ (by default) and ‘R CMD Rdiff’ now ignore differences in
pointer values when comparing printed environments, compiled byte code,
etc.
</P></LI>

<LI><P>
The ‘&quot;source&quot;’ attribute on functions created with ‘keep.source=TRUE’
has been replaced with a ‘&quot;srcref&quot;’ attribute.  The ‘&quot;srcref&quot;’
attribute references an in-memory copy of the source file using the
‘&quot;srcfilecopy&quot;’ class or the new ‘&quot;srcfilealias&quot;’ class.
</P></LI>

<LI><P>
New items User Manuals and Technical Papers have been added to the HTML
help main page.  These link to vignettes in the base and recommended
packages and to a collection of papers about R issues, respectively.
</P></LI>

<LI><P>
Documentation and messages have been standardized to use “namespace”
rather than “name space”.
</P></LI>

<LI><P>
‘setGeneric()’ now looks in the default packages for a non-generic
version of a function if called from a package with a namespace.  (It
always did for packages without a namespace.)
</P></LI>

<LI><P>
Setting the environment variable ‘_R_WARN_ON_LOCKED_BINDINGS_’ will
give a warning if an attempt is made to change a locked binding.
</P></LI>

<LI><P>
‘\SweaveInput’ is now supported when generating concordances in
‘Sweave()’.
</P></LI>

<LI><P>
‘findLineNum()’ and ‘setBreakpoint()’ now allow the environment to be
specified indirectly; the latter gains a ‘clear’ argument to allow it
to call ‘untrace()’.
</P></LI>

<LI><P>
The body of a closure can be one of further types of R objects,
including enviroments and external pointers.
</P></LI>

<LI><P>
The ‘Rd2HTML()’ function in package ‘tools’ now has a ‘stylesheet’
argument, allowing pages to be displayed in alternate formats.
</P></LI>

<LI><P>
New function ‘requireNamespace()’ analogous to ‘require()’, returning a
logical value after attempting to load a namespace.
</P></LI>

<LI><P>
There is a new type of RNG, ‘&quot;L'Ecuyer-CMRG&quot;’, implementing L'Ecuyer
(1999)'s ‘combined multiple-recursive generator’ ‘MRG32k3a’.  See the
comments on ‘?RNG’.
</P></LI>

<LI><P>
‘help.search()’ and ‘??’ can now display vignettes and demos as well as
help pages.  The new option ‘&quot;help.search.types&quot;’ controls the types of
documentation and the order of their display.
</P></LI>

<LI><P>
‘socketConnection()’ now has a ‘timeout’ argument.
</P></LI>

<LI><P>
The initialization of the random-number generator now uses the process
ID as well as the current time, just in case two R processes are
launched very rapidly on a machine with low-resolution wall clock (some
have a resolution of a second; modern systems have microsecond-level
resolution).
</P></LI>

<LI><P>
New function ‘pskill()’ in the ‘tools’ package to send a terminate
signal to one or more processes, plus constants such as ‘SIGTERM’ to
provide a portable way to refer to signals (since the numeric values
are OS-dependent).
</P></LI>

<LI><P>
New function ‘psnice()’ in the ‘tools’ package to return or change the
‘niceness’ of a process.  (Refers to the ‘priority class’ on Windows.)
</P></LI>

<LI><P>
‘list.dirs()’ gains a ‘recursive’ argument.
</P></LI>

<LI><P>
An ‘Authors@R’ field in a package ‘DESCRIPTION’ file can now be used to
generate ‘Author’ and ‘Maintainer’ fields if needed, and to
auto-generate package citations.
</P></LI>

<LI><P>
New utility ‘getElement()’ for accessing either a list component or a
slot in an S4 object.
</P></LI>

<LI><P>
‘stars()’ gains a ‘col.lines’ argument, thanks to Dustin Sallings.
(Wish of <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14657">PR#14657</a>.)
</P></LI>

<LI><P>
New function ‘regmatches()’ for extracting or replacing matched or
non-matched substrings from match data obtained by ‘regexpr()’,
‘gregexpr()’ and ‘regexec()’.
</P></LI>

<LI><P>
‘help(package = &quot;pkg_name&quot;, help_type = &quot;HTML&quot;)’ now gives HTML help on
the package rather than text help.  (This gives direct access to the
HTML version of the package manual shown _via_ ‘help.start()’'s
‘Packages’ menu.)
</P></LI>

<LI><P>
‘agrep()’ gains a ‘fixed’ argument to optionally allow approximate
regular expression matching, and a ‘costs’ argument to specify possibly
different integer match costs for insertions, deletions and
substitutions.
</P></LI>

<LI><P>
‘read.dcf()’ and ‘write.dcf()’ gain a ‘keep.white’ argument to indicate
fields where whitespace should be kept as is.
</P></LI>

<LI><P>
‘available.packages()’ now works around servers that fail to return an
error code when ‘PACKAGES.gz’ does not exist.  (Patch submitted by Seth
Schommer.)
</P></LI>

<LI><P>
‘readBin()’ can now read more than 2^31 - 1 bytes in a single call (the
previously documented limitation).
</P></LI>

<LI><P>
New function ‘regexec()’ for finding the positions of matches as well
as all substrings corresponding to parenthesized subexpressions of the
given regular expression.
</P></LI>

<LI><P>
New function ‘adist()’ in package ‘utils’ for computing ‘edit’
(generalized Levenshtein) distances between strings.
</P></LI>

<LI><P>
Class ‘&quot;raster&quot;’ gains an ‘is.na’ method to avoid confusion from the
misuse of the matrix method (such as <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14618">PR#14618</a>).
</P></LI>

<LI><P>
The ‘identical()’ function gains an ‘ignore.bytecode’ argument to
control comparison of compiled functions.
</P></LI>

<LI><P>
‘pmin’ and ‘pmax’ now warn if an argument is partially recycled (wish
of <a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14638">PR#14638</a>).
</P></LI>

<LI><P>
The default for ‘image(useRaster=)’ is now taken from option
‘&quot;preferRaster&quot;’: for the small print see ‘?image’.
</P></LI>

<LI><P>
‘str()’ now displays reference class objects and their fields, rather
than treating them as classical S4 classes.
</P></LI>

<LI><P>
New function ‘aregexec()’ in package ‘utils’ for finding the positions
of approximate string matches as well as all substrings corresponding
to parenthesized subexpressions of the given regular expression.
</P></LI>

<LI><P>
‘download.file()’ has an ‘extra’ argument to pass additional
command-line options to the non-default methods using command-line
utilities.
</P>
<P>
‘cacheOK = FALSE’ is now supported for ‘method = &quot;curl&quot;’.
</P></LI>

<LI><P>
‘interaction.plot(*, type = .)’ now also allows type ‘&quot;o&quot;’ or ‘&quot;c&quot;’.
</P></LI>

<LI><P>
‘axTicks(*, log=TRUE)’ did sometimes give more values than the ticks in
the corresponding ‘graphics::axis()’.  By default, it now makes use of
the new (‘graphics’-package independent) ‘axisTicks()’ which can make
use of a new utility ‘.axisPars()’.  Further, it now returns a
decreasing sequence (as for ‘log=FALSE’) when ‘usr’ is decreasing.
</P></LI>

</UL>
<H4>2.14.0 GRAPHICS DEVICES </H4>
<UL>
<LI><P>
The ‘pdf()’ device makes use of Flate compression: this is controlled
by the new logical argument ‘compress’, and is enabled by default.
</P></LI>

<LI><P>
Devices ‘svg()’, ‘cairo_pdf()’ and ‘cairo_ps()’ gain a ‘family’
argument.
</P>
<P>
On a Unix-alike ‘X11()’ gains a ‘family’ argument.  This is one of the
‘x11.options()’ and so can be passed as an argument to the ‘bmp()’,
‘jpeg()’, ‘png()’ and ‘tiff()’ devices.
</P>
<P>
Analogous changes have been made on Windows, so all built-in R graphics
devices now have a ‘family’ argument except ‘pictex()’ (which has no
means to change fonts).
</P></LI>

<LI><P>
The ‘bmp()’, ‘jpeg()’, ‘png()’ and ‘tiff()’ devices now make use of the
‘antialias’ argument for ‘type = &quot;quartz&quot;’.
</P></LI>

<LI><P>
There are several new built-in font mappings for ‘X11(type = &quot;Xlib&quot;)’:
see the help on ‘X11Fonts()’.
</P></LI>

<LI><P>
There is a new type ‘X11(type = &quot;dbcairo&quot;)’ which updates the screeen
less frequently: see its help page.
</P></LI>

<LI><P>
The ‘X11()’ device now makes use of cursors to distinguish its states.
The normal cursor is an arrow (rather than a crosshair); the crosshair
is used when the locator is in use, and a watch cursor is shown when
plotting computations are being done.  (These are the standard names
for X11 cursors: how they are actually displayed depends on the window
manager.)
</P></LI>

<LI><P>
New functions ‘dev.hold()’ and ‘dev.flush()’ for use with graphics
devices with buffering.  These are used for most of the high-level
graphics functions such as ‘boxplot()’, so that the plot is only
displayed when the page is complete.
</P>
<P>
Currently implemented for ‘windows(buffered = TRUE)’, ‘quartz()’ and
the cairographics-based ‘X11()’ types with buffering (which are the
default on-screen devices).
</P></LI>

<LI><P>
New function ‘dev.capture()’ for capture of bitmap snapshots of
image-based devices (a superset of the functionality provided by
‘grid.cap()’ in ‘grid’).
</P></LI>

<LI><P>
The default ‘colormodel’ for ‘pdf()’ and ‘postscript()’ is now called
‘&quot;srgb&quot;’ to more accurately describe it.  (Instead of ‘&quot;rgb&quot;’, and in
the case of ‘postscript()’ it no longer switches to and from the gray
colorspace, by default.)
</P>
<P>
The‘colormodel’ for ‘postscript()’ which does use both gray and sRGB
colorspace is now called ‘&quot;srgb+gray&quot;’.
</P>
<P>
Plots which are known to use only black/white/transparent can
advantageously use ‘colormodel = &quot;gray&quot;’ (just as before, but there is
now slightly more advantage in doing so).
</P></LI>

<LI><P>
‘postscript()’ with values ‘colormodel = &quot;rgb&quot;’ and ‘colormodel =
&quot;rgb-nogray&quot;’ give the behaviour prior to R 2.13.0 of uncalibrated RGB,
which under some circumstances can be rendered much faster by a viewer.
</P>
<P>
‘pdf(colormodel = &quot;rgb&quot;)’ gives the behaviour prior to R 2.13.0 of
uncalibrated RGB, which under some circumstances can be rendered faster
by a viewer, and the files will be smaller (by about 9KB if compression
is not used).
</P></LI>

<LI><P>
The ‘postscript()’ device only includes the definition of the sRGB
colorspace in the output file for the colormodels which use it.
</P></LI>

<LI><P>
The ‘postscript()’ and ‘pdf()’ devices now output greyscale raster
images (and not RGB) when ‘colormodel = &quot;gray&quot;’.
</P></LI>

<LI><P>
‘postscript(colormodel = &quot;gray&quot;)’ now accepts non-grey colours and uses
their luminance (as ‘pdf()’ long has).
</P></LI>

<LI><P>
‘colormodel = &quot;grey&quot;’ is allowed as an alternative name for
‘postscript()’ and ‘pdf()’.
</P></LI>

<LI><P>
‘pdf()’ in the default sRGB colorspace outputs many fewer changes of
colorspace, which may speed up rendering in some viewing applications.
</P></LI>

<LI><P>
There is a new function ‘dev.capabilities()’ to query the capabilities
of the current device.  The initial set of capabilities are support for
semi-transparent colours, rendering and capturing raster images, the
locator and for interactive events.
</P></LI>

<LI><P>
For ‘pdf()’, ‘maxRasters’ is increased as needed so the argument is no
longer used.
</P></LI>

</UL>
<H4>2.14.0 SWEAVE &amp; VIGNETTES </H4>
<UL>
<LI><P>
Options ‘keep.source = TRUE, figs.only = FALSE’ are now the default.
</P></LI>

<LI><P>
The way the type of user-defined options is determined has changed.
Previously they were all regarded as logical: now the type is
determined by the value given at first use.
</P></LI>

<LI><P>
The allowed values of logical options are now precisely those allowed
for character inputs to ‘as.logical()’: this means that ‘t’ and ‘f’ are
no longer allowed (although ‘T’ and ‘F’ still are).
</P></LI>

<LI><P>
The preferred location for vignette sources is now the directory
‘vignettes’ and not ‘inst/doc’: ‘R CMD build’ will now re-build
vignettes in directory ‘vignettes’ and copy the ‘.Rnw’ (etc) files and
the corresponding PDFs to ‘inst/doc’.  Further files to be copied to
‘inst/doc’ can be specified _via_ the file ‘vignettes/.install_extras’.
</P></LI>

<LI><P>
‘R CMD Sweave’ now supports a ‘--driver’ option to select the Sweave
driver: the default is equivalent to ‘--driver=RweaveLatex’.
</P></LI>

<LI><P>
‘R CMD Sweave’ and ‘R CMD Stangle’ support options ‘--encoding’ and
‘--options’.
</P></LI>

<LI><P>
The ‘Rtangle()’ driver allows ‘output = &quot;stdout&quot;’ or ‘output =
&quot;stderr&quot;’ to select the output or message connection.  This is
convenient for scripting using something like
</P>
<P>
    R CMD Stangle --options='output=&quot;stdout&quot;' foo.Rnw &gt; foo2.R
</P></LI>

<LI><P>
There is a new option ‘pdf.compress’ controlling whether PDF figures
are generated using Flate compression (they are by default).
</P></LI>

<LI><P>
‘R CMD Sweave’ now has a ‘--pdf’ option to produce a PDF version of the
processed Sweave document.
</P></LI>

<LI><P>
It is no longer allowed to have two vignettes with the same vignette
basename (e.g. ‘vig.Rnw’ and ‘vig.Snw’).  (Previously one vignette hid
the other in the ‘vignette()’ function.)
</P></LI>

</UL>
<H4>2.14.0 C-LEVEL FACILITIES </H4>
<UL>
<LI><P>
Function ‘R_tmpnam2’ has been added to the API to allow a temporary
filename to include a specified extension.
</P></LI>

</UL>
<H4>2.14.0 PACKAGE INSTALLATION </H4>
<UL>
<LI><P>
Package ‘DESCRIPTION’ file field ‘KeepSource’ forces the package to be
installed with ‘keep.source = TRUE’ (or ‘FALSE’).  (Suggestion of Greg
Snow.  Note that as all packages are lazy-loaded, this is now only
relevant at installation.)
</P>
<P>
There are corresponding options ‘--with-keep.source’ and
‘--without-keep.source’ for ‘R CMD INSTALL’.
</P></LI>

<LI><P>
‘R CMD INSTALL’ has a new option ‘--byte-compile’ to byte-compile the
packages during installation (since all packages are now lazy-loaded).
This can be controlled on a per-package basis by the optional field
‘ByteCompile’ in the ‘DESCRIPTION’ file.
</P></LI>

<LI><P>
A package R code but without a ‘NAMESPACE’ file will have a default one
created at ‘R CMD build’ or ‘R CMD INSTALL’ time, so all packages will
be installed with namespaces.  A consequence of this is that
‘.First.lib()’ functions need to be copied to ‘.onLoad()’ (usually) or
‘.onAttach()’.  For the time being, if there is an auto-generated
‘NAMESPACE’ file and no ‘.onLoad()’ nor ‘.onAttach()’ function is found
but ‘.First.lib()’ is, it will be run as the attach hook (unless the
package is one of a list of known exceptions, when it will be run as
the load hook).
</P></LI>

<LI><P>
A warning is given if test-loading a package changes a locked binding
in a package other than itself.  It is likely that this will be
disallowed in future releases.  (There are _pro tem_ some exceptions to
the warning.)
</P></LI>

<LI><P>
A dependency on SVN revision is allowed for ‘R’, e.g. ‘R (&gt;= r56550)’.
This should be used in conjunction with a version number, e.g. ‘R (&gt;=
2.14.0), R (&gt;= r56550)’ to distinguish beteen R-patched and R-devel
versions with the same SVN revision.
</P></LI>

<LI><P>
‘installed.packages()’ now hashes the names of its cache files to avoid
very rare problems with excessively long path names.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14669">PR#14669</a>)
</P></LI>

<LI><P>
A top-level ‘COPYING’ file in a package is no longer installed (file
names ‘LICENSE’ or ‘LICENCE’ having long been preferred).
</P></LI>

</UL>
<H4>2.14.0 UTILITIES </H4>
<UL>
<LI><P>
‘R CMD check’ now gives an error if the R code in a vignette fails to
run, unless this is caused by a missing package.
</P></LI>

<LI><P>
‘R CMD check’ now unpacks tarballs in the same way as ‘R CMD INSTALL’,
including making use of the environment variable ‘R_INSTALL_TAR’ to
override the default behaviour.
</P></LI>

<LI><P>
‘R CMD check’ performs additional code analysis of package startup
functions, and notifies about incorrect argument lists and (incorrect)
calls to functions which modify the search path or inappropriately
generate messages.
</P></LI>

<LI><P>
‘R CMD check’ now also checks compiled code for symbols corresponding
to functions which might terminate R or write to ‘stdout’/‘stderr’
instead of the console.
</P></LI>

<LI><P>
‘R CMD check’ now uses a ‘pdf()’ device when checking examples (rather
than ‘postscript()’).
</P></LI>

<LI><P>
‘R CMD check’ now reports as a NOTE what look like methods documented
with their full names even if there is a namespace and they are
exported.  In almost all cases they are intended to be used only as
methods and should use the ‘\method’ markup.  In the other rare cases
the recommended form is to use a function such as ‘coefHclust’ which
would not get confused with a method, document that and register it in
the ‘NAMESPACE’ file by ‘s3method(coef, hclust, coefHclust)’.
</P></LI>

<LI><P>
The default for the environment variable ‘_R_CHECK_COMPACT_DATA2_’ is
now true: thus if using the newer forms of compression introduced in R
2.10.0 would be beneficial is now checked (by default).
</P></LI>

<LI><P>
‘R CMD Rd2dvi’ allows the use of LaTeX package ‘inputenx’ rather than
‘inputenc’: the value of the environment variable ‘RD2DVI_INPUTENC’ is
used.  (LaTeX package ‘inputenx’ is an optional install which provides
greater coverage of the UTF-8 encoding.)
</P></LI>

<LI><P>
‘Rscript’ on a Unix-alike now accepts file names containing spaces
(provided these are escaped or quoted in the shell).
</P></LI>

<LI><P>
‘R CMD build’ on a Unix-alike (only) now tries to preserve dates on
files it copies from its input directory.  (This was the undocumented
behaviour prior to R 2.13.0.)
</P></LI>

</UL>
<H4>2.14.0 DEPRECATED AND DEFUNCT </H4>
<UL>
<LI><P>
‘require()’ no longer has a ‘save’ argument.
</P></LI>

<LI><P>
The ‘gamma’ argument to ‘hsv()’, ‘rainbow()’, and ‘rgb2hsv()’ has been
removed.
</P></LI>

<LI><P>
The ‘--no-docs’ option for ‘R CMD build --binary’ is defunct: use
‘--install-args’ instead.
</P></LI>

<LI><P>
The option ‘--unsafe’ to ‘R CMD INSTALL’ is defunct: use the identical
option ‘--no-lock’ instead.
</P></LI>

<LI><P>
The entry point ‘pythag’ formerly in ‘Rmath.h’ is defunct: use instead
the C99 function ‘hypot’.
</P></LI>

<LI><P>
‘R CMD build --binary’ is formally defunct: ‘R CMD INSTALL --build’ has
long been the preferred alternative.
</P></LI>

<LI><P>
‘zip.file.extract()’ is now defunct: use ‘unzip()’ or ‘unz()’ instead.
</P></LI>

<LI><P>
‘R CMD Rd2dvi’ without the ‘--pdf’ option is now deprecated: only PDF
output will be supported in future releases (since this allows the use
of fonts only supported for PDF).
</P></LI>

<LI><P>
Options such as ‘--max-nsize’ and the function ‘mem.limits()’ are now
deprecated: these limits are nowadays almost never used, and are
reported by ‘gc()’ when they are in use.
</P></LI>

<LI><P>
Forms like ‘binomial(link = &quot;link&quot;)’ for GLM families deprecated since
R 2.4.0 are now defunct.
</P></LI>

<LI><P>
The ‘declarativeOnly’ argument to ‘loadNamespace()’ (not relevant since
R 2.13.0) has been removed.
</P></LI>

<LI><P>
Use of ‘library.dynam()’ without specifying all the first three
arguments is deprecated.  (It is often called from a namespace, and the
defaults are only appropriate to a package.)
</P></LI>

</UL>
<H4>2.14.0 INSTALLATION </H4>
<UL>
<LI><P>
The base and recommended packages are now byte-compiled (equivalent to
‘make bytecode’ in R 2.13.x).
</P></LI>

<LI><P>
Configure option ‘--with-system-zlib’ now only makes use of the basic
interface of ‘zlib’ and not the C function ‘gzseek’ which has shown
erroneous behaviour in ‘zlib’ 1.2.4 and 1.2.5.
</P></LI>

<LI><P>
The ‘zlib’ in the R sources is now version 1.2.5.  (This is safe even
on 32-bit Linux systems because only the basic interface is now used.)
</P></LI>

<LI><P>
The ‘.afm’ files in package ‘grDevices’ are now installed as compressed
files (as long done on Windows), saving ca 2MB on the installed size.
</P></LI>

<LI><P>
The non-screen cairo-based devices are no longer in the X11 module and
so can be installed without X11.  (We have never seen a Unix-alike
system with cairographics installed but not X11, but a user might
select ‘--without-x’.)
</P></LI>

<LI><P>
Configure will try to use ‘-fobjc-exceptions’ for the Objective-C
compiler (if present) to ensure that even compilers that do not enable
exceptions by default (such as vanilla gcc) can be used.  (Objective-C
is currently only used on Mac OS X.)
</P></LI>

<LI><P>
The system call ‘times’ is required.
</P></LI>

<LI><P>
The C99 functions ‘acosh’, ‘asinh’, ‘atanh’, ‘snprintf’ and ‘vsnprintf’
are now required.
</P></LI>

<LI><P>
There is no longer support for making DVI manuals _via_ ‘make dvi’,
‘make install-dvi’ and similar.  Only PDF manuals are supported (to
allow the use of fonts which are only available for PDF.)
</P></LI>

<LI><P>
The ‘configure’ arguments used during configuration of R are included
as a comment in ‘Makeconf’ for informative purposes on Unix-alikes in a
form suitable for shell execution.  Note that those are merely
command-line arguments, they do not include environment variables (one
more reason to use configure variables instead) or site configuration
settings.
</P></LI>

<LI><P>
Framework installation now supports ‘DESTDIR’ (Mac OS X only).
</P></LI>

</UL>
<H4>2.14.0 BUG FIXES </H4>
<UL>
<LI><P>
The locale category ‘LC_MONETARY’ was only being set on startup on
Windows: that is now done on Unix-alikes where supported.
</P></LI>

<LI><P>
Reference class utilities will detect an attempt to modify methods or
fields in a locked class definition (e.g., in a namespace) and generate
an error.
</P></LI>

<LI><P>
The formula methods for ‘lines()’, ‘points()’ and ‘text()’ now work
even if package ‘stats’ is not on the search path.
</P></LI>

<LI><P>
In principle, S4 classes from different packages could have the same
name.  This has not previously worked.  Changes have now been installed
that should allow such classes and permit methods to use them.  New
functions ‘className()’ and ‘multipleClasses()’ are related tools for
programming.
</P></LI>

<LI><P>
Work around an issue in Linux (a system ‘select’ call resetting ‘tv’)
which prevented internet operations from timing out properly.
</P></LI>

<LI><P>
Several stack trampling and overflow issues have been fixed in TRE,
triggered by ‘agrep’ and friends with long patterns.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14627">PR#14627</a>.)
</P></LI>

<LI><P>
(“design infelicity”) Field assignments in reference classes are now
consistent with slots in S4 classes: the assigned value must come from
the declared class (if any) for the field or from a subclass.
</P></LI>

<LI><P>
The methods objects constructed for ‘&quot;coerce&quot;’ and ‘&quot;coerce&lt;-&quot;’ were
lacking some essential information in the ‘generic’, ‘defined’ and
‘target’ slots; ‘as()’ did not handle duplicate class definitions
correctly.
</P></LI>

<LI><P>
The parser no longer accepts the digit ‘8’ in an octal character code
in a string, nor does it accept unterminated strings in a file.
(Reported by Bill Dunlap.)
</P></LI>

<LI><P>
The ‘print()’ method for class ‘&quot;summary.aov&quot;’ did not pass on argument
‘digits’ when ‘summary()’ was called on a single object, and hence used
more digits than documented.
</P></LI>

<LI><P>
X11 cairo back-end produced incorrect capture snapshot images on
big-endian machines.
</P></LI>

<LI><P>
‘loglin()’ gave a spurious error when argument ‘margin’ consists of a
single element of length one.  (<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14690">PR#14690</a>)
</P></LI>

<LI><P>
‘loess()’ is better protected against misuse, e.g. zero-length ‘span’.
(<a href="https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14691">PR#14691</a>)
</P></LI>

</UL>]]></description>  
</item>
     </channel>
</rss>