support for libdeflate

git-svn-id: https://svn.r-project.org/R/trunk@84880 00db46b3-68df-0310-9c12-caf00c1e9a41

configure

@@ -1114,6 +1114,7 @@ with_recommended_packages
 with_ICU
 enable_byte_compiled_packages
 with_static_cairo
+with_libdeflate_compression
 enable_static
 enable_shared
 with_pic
@@ -1926,6 +1927,9 @@ Optional Packages:
   --with-ICU              use ICU library (if available) [yes]
   --with-static-cairo     allow for the use of static cairo libraries [no, yes
                           on macOS]
+  --with-libdeflate-compression
+                          use libdeflate (if available) to (de)compress
+                          lazy-loaded R objects [no]
   --with-pic[=PKGS]       (libtool) try to use only PIC/non-PIC objects
                           [default=use both]
   --with-aix-soname=aix|svr4|both
@@ -5055,6 +5059,21 @@ if test "x$use_static_cairo" = xdefault; then
   esac
 fi
 
+
+# Check whether --with-libdeflate-compression was given.
+if test ${with_libdeflate_compression+y}
+then :
+  withval=$with_libdeflate_compression; use_libdeflate=${withval}
+else $as_nop
+  use_libdeflate=no
+fi
+
+if test "x$use_libdeflate" = xyes; then
+
+printf "%s\n" "#define USE_LIBDEFLATE 1" >>confdefs.h
+
+fi
+
 ### ** Precious variables.
 
 
@@ -27111,7 +27130,7 @@ _ACEOF
 # flags.
 r_save_CFLAGS=$CFLAGS
 CFLAGS="$CFLAGS $r_verb"
-(eval echo $as_me:27116: \"$ac_link\") >&5
+(eval echo $as_me:27135: \"$ac_link\") >&5
 r_c_v_output=`eval $ac_link 5>&1 2>&1 | grep -v 'Driving:'`
 echo "$r_c_v_output" >&5
 CFLAGS=$r_save_CFLAGS
@@ -27187,7 +27206,7 @@ _ACEOF
 # flags.
 r_save_CFLAGS=$CFLAGS
 CFLAGS="$CFLAGS $r_cv_prog_c_v"
-(eval echo $as_me:27192: \"$ac_link\") >&5
+(eval echo $as_me:27211: \"$ac_link\") >&5
 r_c_v_output=`eval $ac_link 5>&1 2>&1 | grep -v 'Driving:'`
 echo "$r_c_v_output" >&5
 CFLAGS=$r_save_CFLAGS
@@ -51298,6 +51317,85 @@ else
 fi
 
 
+## libdeflate headers and libraries.
+
+         for ac_header in libdeflate.h
+do :
+  ac_fn_c_check_header_compile "$LINENO" "libdeflate.h" "ac_cv_header_libdeflate_h" "$ac_includes_default"
+if test "x$ac_cv_header_libdeflate_h" = xyes
+then :
+  printf "%s\n" "#define HAVE_LIBDEFLATE_H 1" >>confdefs.h
+ have_libdeflate=yes
+else $as_nop
+  have_libdeflate=no
+fi
+
+done
+if test "${have_libdeflate}" = yes; then
+  { printf "%s\n" "$as_me:${as_lineno-$LINENO}: checking for libdeflate_alloc_compressor in -ldeflate" >&5
+printf %s "checking for libdeflate_alloc_compressor in -ldeflate... " >&6; }
+if test ${ac_cv_lib_deflate_libdeflate_alloc_compressor+y}
+then :
+  printf %s "(cached) " >&6
+else $as_nop
+  ac_check_lib_save_LIBS=$LIBS
+LIBS="-ldeflate  $LIBS"
+cat confdefs.h - <<_ACEOF >conftest.$ac_ext
+/* end confdefs.h.  */
+
+/* Override any GCC internal prototype to avoid an error.
+   Use char because int might match the return type of a GCC
+   builtin and then its argument prototype would still apply.  */
+char libdeflate_alloc_compressor ();
+#ifdef FC_DUMMY_MAIN
+#ifndef FC_DUMMY_MAIN_EQ_F77
+#  ifdef __cplusplus
+     extern "C"
+#  endif
+   int FC_DUMMY_MAIN() { return 1; }
+#endif
+#endif
+int
+main (void)
+{
+return libdeflate_alloc_compressor ();
+  ;
+  return 0;
+}
+_ACEOF
+if ac_fn_c_try_link "$LINENO"
+then :
+  ac_cv_lib_deflate_libdeflate_alloc_compressor=yes
+else $as_nop
+  ac_cv_lib_deflate_libdeflate_alloc_compressor=no
+fi
+rm -f core conftest.err conftest.$ac_objext conftest.beam \
+    conftest$ac_exeext conftest.$ac_ext
+LIBS=$ac_check_lib_save_LIBS
+fi
+{ printf "%s\n" "$as_me:${as_lineno-$LINENO}: result: $ac_cv_lib_deflate_libdeflate_alloc_compressor" >&5
+printf "%s\n" "$ac_cv_lib_deflate_libdeflate_alloc_compressor" >&6; }
+if test "x$ac_cv_lib_deflate_libdeflate_alloc_compressor" = xyes
+then :
+  have_libdeflate=yes
+else $as_nop
+  have_libdeflate=no
+fi
+
+fi
+if test "x${r_cv_have_libdeflate}" = xno; then
+  have_libdeflate=no
+fi
+if test "x${have_libdeflate}" = xyes; then
+  { printf "%s\n" "$as_me:${as_lineno-$LINENO}: result: yes" >&5
+printf "%s\n" "yes" >&6; }
+  LIBS="-ldeflate ${LIBS}"
+
+printf "%s\n" "#define HAVE_LIBDEFLATE 1" >>confdefs.h
+
+fi
+
+
 ## PCRE[12] headers and libraries.
 have_pcre2=no
 if test "x${use_pcre2}" = xyes; then

configure.ac

@@ -580,6 +580,14 @@ if test "x$use_static_cairo" = xdefault; then
   esac
 fi
 
+AC_ARG_WITH([libdeflate-compression],
+[AS_HELP_STRING([--with-libdeflate-compression],[use libdeflate (if available) to (de)compress lazy-loaded R objects @<:@no@:>@])],
+[use_libdeflate=${withval}],
+[use_libdeflate=no])
+if test "x$use_libdeflate" = xyes; then
+  AC_DEFINE(USE_LIBDEFLATE, 1, [Define to use libdefault rather than libz])
+fi
+
 ### ** Precious variables.
 
 AC_ARG_VAR([R_PRINTCMD],
@@ -2569,6 +2577,9 @@ R_BZLIB
 ## LZMA headers and libraries from xz-utils
 R_LZMA
 
+## libdeflate headers and libraries.
+R_LIBDEFLATE
+
 ## PCRE[12] headers and libraries.
 R_PCRE
 

doc/NEWS.Rd

@@ -32,7 +32,7 @@
       \item The \code{confint()} and \code{profile} methods for
       \code{"glm"} objects have gained a possibility to do profiling
       based on the Rao Score statistic in addition to the default
-      Likelihood Ratio. This is controlled by a new \code{test=}
+      Likelihood Ratio. This is controlled by a new \code{test =}
       argument.
 
       \item The \code{"glm"} method for \code{anova()} computes test
@@ -42,7 +42,7 @@
       \code{test} a false logical value.
 
       \item In \code{setRepositories()} the repositories can be set
-      using their names via \code{name=} instead of index \code{ind=}.
+      using their names via \code{name =} instead of index \code{ind =}.
 
       \item \code{methods()} and \code{.S3methods()} gain a
       \code{all.names} option for the (rare) case where functions starting
@@ -73,24 +73,40 @@
       \dQuote{binary} Windows installation \bold{if} a source \file{tests/}
       folder is present.
 
-      \item \code{range(<DT_with_Inf>, finite=TRUE)} now work for objects
+      \item \code{range(<DT_with_Inf>, finite = TRUE)} now work for objects
       of class \code{"Date"}, \code{"POSIXct"}, and \code{"POSIXlt"} with
       infinite entries, analogously to \code{range.default()}, as proposed
       by Davis Vaughan on R-devel.  Other \code{range()}-methods can make
       use of new \code{.rangeNum()}.
 
-      \item New \code{.internalGenerics} complementing \code{.S3PrimitiveGenerics},
-      for documentation and low-level book keeping.
+      \item New \code{.internalGenerics} complementing
+      \code{.S3PrimitiveGenerics}, for documentation and low-level book
+      keeping.
 
       \item \code{grid()} now invisibly returns the x- and y- coordinates
       at which the grid-lines were drawn.
 
       \item \code{norm(., type)} now also works for complex matrices.
 
-      \item \code{kappa(., exact=TRUE, norm=*)} now works for all norms and
-      also for complex matrices.  In symmetric / triangular cases, new argument
-      \code{uplo = "U" | "L"} allows to specify the upper or lower
-      triangular part.
+      \item \code{kappa(., exact = TRUE, norm = *)} now works for all
+      norms and also for complex matrices.  In symmetric / triangular
+      cases, new argument \code{uplo = "U" | "L"} allows to specify the
+      upper or lower triangular part.
+
+      \item \code{memDecompress(type = "unknown")} recognizes compresion
+      in the defailt \sQuote{zlib} format as used by
+      \code{memCompress(type = "gzip")}.
+      
+      \item (\strong{Experimental:}) \code{memCompress()} and
+      \code{memDecompress()} have a new type \code{"libdeflate"} which
+      use the \code{libdeflate} library
+      (\url{https://github.com/ebiggers/libdeflate}) if installed.  This
+      uses the same type of compression as their default \code{type =
+	"gzip"} but is substantially faster on commpn plaftorms (it has
+      optimized code for \code{x86} and ARM CPUs, including their 64-bit
+      variants.)
+
+      This option may replace \code{type = "gzip"} before release.
     }
   }
 
@@ -221,14 +237,15 @@
       arguments list (for multi-line \verb{\item} labels in the Rd source).
 
       \item \code{kappa()} and \code{rcond()} work correctly in more cases;
-      \code{kappa(., norm="2")} now warns that it computes the 1-norm with
+      \code{kappa(., norm = "2")} now warns that it computes the 1-norm with
       (default) \code{exact = FALSE}; prompted by Mikael Jagan's quite
       comprehensive \PR{18543}.
 
       \item Rd skeletons generated by \code{prompt()} or
-      \code{promptData()} now use a dummy title (so R CMD build works).
-      \code{tools::checkRdContents()} has been updated to detect
-      such template leftovers, including from \code{promptPackage()}.
+      \code{promptData()} now use a dummy title (so \command{R CMD
+      build} works).  \code{tools::checkRdContents()} has been updated
+      to detect such template leftovers, including from
+      \code{promptPackage()}.
     }
   }
 }

doc/manual/R-admin.texi

@@ -3458,6 +3458,9 @@ uses @file{rpc/xdr.h} but that includes @file{netconfig.h} from the top
 @file{tirpc} directory.} need to be on the C include path or under
 @file{/usr/include/tirpc}.
 
+Library @code{libdeflate} (@url{https://github.com/ebiggers/libdeflate})
+is used by @code{memCompress()} and @code{memDecompress()} if available.
+
 Use of the X11 clipboard selection requires the @code{Xmu} headers and
 libraries.  These are normally part of an X11 installation (e.g.@: the
 Debian meta-package @samp{xorg-dev}), but some distributions have split
@@ -3487,12 +3490,12 @@ following packages and their development versions, and this may provide
 a useful checklist for other systems:
 @example
 bzip2 cairo fontconfig freetype fribidi gcc gcc-gfortran gcc-c++ glib2
-glibc harfbuzz libX11 libXext libXt libcurl libicu libjpeg libpng
-libtiff libtirpc libxcrypt ncurses pango pkgconf-pkg-config pcre2
-readline tcl tk xz zlib
+glibc harfbuzz lapack libX11 libXext libXt libcurl libdeflate libicu
+libjpeg libpng libtiff libtirpc libxcrypt ncurses pango
+pkgconf-pkg-config pcre2 readline tcl tk xz zlib
 @end example
 @noindent
-plus, preferably a TeX installation, Java and LAPACK >= 3.10.0.
+plus, preferably a TeX installation and Java.
 
 @menu
 * Tcl/Tk::

m4/R.m4

@@ -3469,6 +3469,27 @@ else
 fi
 ])# R_BZLIB
 
+## R_LIBDEFLATE
+## ------------
+## Try finding libdeflate library and headers.
+## We check that both are installed,
+AC_DEFUN([R_LIBDEFLATE],
+[
+  AC_CHECK_HEADERS(libdeflate.h, [have_libdeflate=yes], [have_libdeflate=no])
+if test "${have_libdeflate}" = yes; then
+  AC_CHECK_LIB(deflate, libdeflate_alloc_compressor, [have_libdeflate=yes], [have_libdeflate=no])
+fi
+if test "x${r_cv_have_libdeflate}" = xno; then
+  have_libdeflate=no
+fi
+if test "x${have_libdeflate}" = xyes; then
+  AC_MSG_RESULT([yes])
+  LIBS="-ldeflate ${LIBS}"
+  AC_DEFINE(HAVE_LIBDEFLATE, 1,
+            [Define to 1 if you have libdeflate headers and library.])
+fi
+])# R_LIBDEFLATE
+
 ## R_TRE
 ## -------
 ## Try finding tre library and headers.

src/gnuwin32/fixed/h/config.h

@@ -467,6 +467,12 @@
 /* Define if __libc_stack_end is visible. */
 /* #undef HAVE_LIBC_STACK_END */
 
+/* Define to 1 if you have libdeflate headers and library. */
+/* #undef HAVE_LIBDEFLATE */
+
+/* Define to 1 if you have the <libdeflate.h> header file. */
+/* #undef HAVE_LIBDEFLATE_H */
+
 /* Define to 1 if you have the `dl' library (-ldl). */
 /* #undef HAVE_LIBDL */
 
@@ -1218,6 +1224,9 @@
 /* Define to use internal time-zone code */
 #define USE_INTERNAL_MKTIME 1
 
+/* Define to use libdefault rather than libz */
+/* #undef USE_LIBDEFLATE */
+
 /* Define if the POSIX multithreading library can be used. (For intl) */
 /* #undef USE_POSIX_THREADS */
 

src/include/config.h.in

@@ -448,6 +448,12 @@
 /* Define if __libc_stack_end is visible. */
 #undef HAVE_LIBC_STACK_END
 
+/* Define to 1 if you have libdeflate headers and library. */
+#undef HAVE_LIBDEFLATE
+
+/* Define to 1 if you have the <libdeflate.h> header file. */
+#undef HAVE_LIBDEFLATE_H
+
 /* Define to 1 if you have the `dl' library (-ldl). */
 #undef HAVE_LIBDL
 
@@ -1147,6 +1153,9 @@
 /* Define to use internal time-zone code */
 #undef USE_INTERNAL_MKTIME
 
+/* Define to use libdefault rather than libz */
+#undef USE_LIBDEFLATE
+
 /* Define if the POSIX multithreading library can be used. (For intl) */
 #undef USE_POSIX_THREADS
 

src/library/base/R/connections.R

@@ -323,22 +323,23 @@ socketSelect <- function(socklist, write = FALSE, timeout = NULL) {
 }
 
 memCompress <-
-    function(from, type = c("gzip", "bzip2", "xz", "none"))
+    function(from, type = c("gzip", "bzip2", "xz", "libdeflate", "none"))
 {
     if(is.character(from))
         from <- charToRaw(paste(from, collapse = "\n"))
     else if(!is.raw(from)) stop("'from' must be raw or character")
-    type <- match(match.arg(type), c("none", "gzip", "bzip2", "xz"))
+    type <- match(match.arg(type), c("none", "gzip", "bzip2", "xz",
+                                     "libdeflate"))
     .Internal(memCompress(from, type))
 }
 
 memDecompress <-
     function(from,
-             type = c("unknown", "gzip", "bzip2", "xz", "none"),
+             type = c("unknown", "gzip", "bzip2", "xz", "libdeflate", "none"),
              asChar = FALSE)
 {
     type <- match(match.arg(type),
-                  c("none", "gzip", "bzip2", "xz", "unknown"))
+                  c("none", "gzip", "bzip2", "xz", "unknown", "libdeflate"))
     ans <- .Internal(memDecompress(from, type))
     if(asChar) rawToChar(ans) else ans
 }