Update.

1998-02-16 17:33  Ulrich Drepper  <drepper@cygnus.com>

	* elf/rtld.c (dl_main): Recognize --library-path parameter and
	pass value (or NULL) to _dl_init_paths.
	* elf/dl-load.c (_dl_init_paths): Change to take one parameter,
	replacing local variable llp.  If llp is NULL examine LD_LIBRARY_PATH
	environment variable.
	* elf/link.h: Change prototype for _dl_init_paths.
	* elf/dl-support.c: Pass NULL in _dl_init_paths call.

	* localedata/Makefile (distribute): Add test files.

1998-02-17  Andreas Jaeger  <aj@arthur.rhein-neckar.de>

	* localedata/tests/{test1.cm, test2.cm, test3.cm, test4.cm,
	test1.def, test2.def, test3.def, test4.def}: Simple input files
	for localedef. Contributed by Yung-Ching Hsiao
	<yhsiao@cae.wisc.edu>.

	* localedata/Makefile (tests): Call tst-locale.sh.

	* localedata/tst-locale.sh: New file, regression tests for some
	localedef problems.

1998-02-15  Thorsten Kukuk  <kukuk@vt.uni-paderborn.de>

	* nis/nss_nisplus/nisplus-alias.c: Use __stpncpy.

	* nis/nss_nisplus/nisplus-hosts.c: Make sure buffer is always NUL
	terminated.
	* nis/nss_nisplus/nisplus-network.c: Likewise.
	* nis/nss_nisplus/nisplus-proto.c: Likewise.
	* nis/nss_nisplus/nisplus-rpc.c: Likewise.
	* nis/nss_nisplus/nisplus-service.c: Likewise.

	Add more changes from TI-RPC 2.3 for rpcgen to fix include/C++ bug
	and support generating thread safe RPC code.
	* sunrpc/rpc_main.c: Add changes.
	* sunrpc/rpc_clntout.c: Likewise.
	* sunrpc/rpc_cout.c: Likewise.
	* sunrpc/rpc_hout.c: Likewise.
	* sunrpc/rpc_parse.c: Likewise.
	* sunrpc/rpc_sample.c: Likewise.
	* sunrpc/rpc_scan.c: Likewise.
	* sunrpc/rpc_svcout.c: Likewise.
	* sunrpc/rpc_util.c: Likewise.
	* sunrpc/rpc_util.h: Add new structs and prototypes.
	* sunrpc/proto.h: Remove prototypes for static functions.

1998-02-15  Andreas Schwab  <schwab@issan.informatik.uni-dortmund.de>

	* locale/programs/ld-messages.c (messages_finish): Don't skip
	error checking when being quiet.
	* locale/programs/ld-ctype.c (ctype_finish): Likewise.
	(set_class_defaults): Likewise.
	* locale/programs/charmap.c (parse_charmap): Likewise.
	* locale/programs/ld-collate.c (collate_finish): Likewise.
	* locale/programs/ld-monetary.c (monetary_finish): Likewise.
	* locale/programs/ld-time.c (time_finish): Likewise.
	* locale/programs/locfile.c (write_locale_data): Likewise.

	* locale/programs/ld-ctype.c (ctype_class_to): Silently ignore
	unknown characters and empty ranges.
	* locale/programs/ld-collate.c (collate_order_elem): When
	processing an ellipsis properly form a linked list in the result
	table, fix typo when allocating ordering array.
	[PR libc/419]

1998-02-13  Andreas Schwab  <schwab@issan.informatik.uni-dortmund.de>

	* elf/Makefile (ld-map): Define.
	(rtld-ldscript): Define.  Change all `$(objpfx)rtld-ldscript' to
	`$(rtld-ldscript)'.
	($(objpfx)ld.so): Combine the two versions of this rule.  Depend
	on $(ld-map).
	(rtld-link): Combine the two versions of this definition.  Fixed
	to make it work when no symbol versioning is used.

1998-02-16  Ulrich Drepper  <drepper@cygnus.com>

	* Makeconfig (build-program-cmd): Use --library-path parameter to
	ld.so instead of environment variable.

	* sunrpc/Makefile (rpcgen-cmd): Don't use -$ parameter.
	* sunrpc/rpc_main.c: Remove support for -$$ option again.

1998-02-16  Andreas Jaeger  <aj@arthur.rhein-neckar.de>

	* Make-dist: Respect with-cvs setting.
	* MakeTAGS (all-pot): Likewise.
	* sysdeps/sparc/sparc32/Makefile: Likewise.
	* sysdeps/mach/hurd/Makefile: Likewise.
	* stdlib/Makefile: Likewise.
	* posix/Makefile: Likewise.
	* intl/Makefile: Likewise.
	* po/Makefile (linguas): Likewise

1 files changed, 167 insertions, 0 deletions

diff --git a/manual/arith.texi b/manual/arith.texi
index 21b6380..3b06068 100644
--- a/manual/arith.texi
+++ b/manual/arith.texi
@@ -46,6 +46,7 @@ These functions are declared in the header files @file{math.h} and
 				 division.
 * Parsing of Numbers::          Functions for ``reading'' numbers
 			         from strings.
+* Old-style number conversion:: Low-level number to string conversion.
 @end menu
 
 @node Infinity
@@ -1324,3 +1325,169 @@ need not detect overflow and underflow errors.  The @code{atof} function
 is provided mostly for compatibility with existing code; using
 @code{strtod} is more robust.
 @end deftypefun
+
+
+@node Old-style number conversion
+@subsection Old-style way of converting numbers to strings
+
+The @w{System V} library provided three functions to convert numbers to
+strings which have a unusual and hard-to-be-used semantic.  The GNU C
+library also provides these functions together with some useful
+extensions in the same sense.
+
+Generally, you should avoid using these functions unless the really fit
+into the problem you have to to solve.  Otherwise it is almost always
+better to use @code{sprinf} since it's greater availability (it is an
+@w{ISO C} function).
+
+
+@comment stdlib.h
+@comment SVID, Unix98
+@deftypefun {char *} ecvt (double @var{value}, int @var{ndigit}, int *@var{decpt}, int *@var{sign})
+The function @code{ecvt} converts the floating-point number @var{value}
+to a string with at most @var{ndigit} decimal digits.  If @code{ndigit}
+is greater than the accuracy of the @code{double} floating-point type
+the implementation can shorten @var{ndigit} to a reasonable value. The
+returned string neither contains decimal point nor sign. The high-order
+digit of the string is non-zero (unless @var{value} is actually zero)
+and the low-order digit is rounded. The variable pointed to by
+@var{decpt} gets the position of the decimal character relative to the
+start of the string. If @var{value} is negativ, @var{sign} is set to a
+non-zero value, otherwise to 0.
+
+The returned string is statically allocated and overwritten by each call
+to @code{ecvt}.
+
+If @var{value} is zero, it's implementation defined if @var{decpt} is
+@code{0} or @code{1}.
+
+The prototype for this function can be found in @file{stdlib.h}.
+@end deftypefun
+
+As an example @code{ecvt (12.3, 5, &decpt, &sign)} returns @code{"12300"}
+and sets @var{decpt} to @code{2} and @var{sign} to @code{0}.
+
+@comment stdlib.h
+@comment SVID, Unix98
+@deftypefun {char *} fcvt (double @var{value}, int @var{ndigit}, int @var{decpt}, int *@var{sign})
+The function @code{fcvt} is similar to @code{ecvt} with the difference
+that @var{ndigit} specifies the digits after the decimal point.  If
+@var{ndigit} is less than zero, @var{value} is rounded to the left of
+the decimal point upto the reasonable limit (e.g., @math{123.45} is only
+rounded to the third digit before the decimal point, even if
+@var{ndigit} is less than @math{-3}).
+
+The returned string is statically allocated and overwritten by each call
+to @code{fcvt}.
+
+The prototype for this function can be found in @file{stdlib.h}.
+@end deftypefun
+
+@comment stdlib.h
+@comment SVID, Unix98
+@deftypefun {char *} gcvt (double @var{value}, int @var{ndigit}, char *@var{buf})
+The @code{gcvt} function also converts @var{value} to a NUL terminated
+string but does in a way similar to the @code{%g} format of
+@code{printf}.  It also does not use a static buffer but instead uses
+the user-provided buffer starting at @var{buf}.  It is the user's
+responsibility to make sure the buffer is long enough to contain the
+result.  Unlike the @code{ecvt} and @code{fcvt} function @code{gcvt}
+includes the sign and the decimal point character (which is determined
+according to the current locale) in the result.  Therefore there are yet
+less reasons to use this function instead of @code{printf}.
+
+The return value is @var{buf}.
+
+The prototype for this function can be found in @file{stdlib.h}.
+@end deftypefun
+
+
+All these three functions have in common that they use @code{double}
+values as the parameters.  Calling these functions using @code{long
+double} values would mean a loss of precision due to the implicit
+rounding.  Therefore the GNU C library contains three more functions
+with similar semantic which take @code{long double} values.
+
+@comment stdlib.h
+@comment GNU
+@deftypefun {char *} qecvt (long double @var{value}, int @var{ndigit}, int *@var{decpt}, int *@var{sign})
+This function is equivalent to the @code{ecvt} function except that it
+takes an @code{long double} value for the first parameter.
+
+This function is a GNU extension.  The prototype can be found in
+@file{stdlib.h}.
+@end deftypefun
+
+@comment stdlib.h
+@comment GNU
+@deftypefun {char *} qfcvt (long double @var{value}, int @var{ndigit}, int @var{decpt}, int *@var{sign})
+This function is equivalent to the @code{fcvt} function except that it
+takes an @code{long double} value for the first parameter.
+
+This function is a GNU extension.  The prototype can be found in
+@file{stdlib.h}.
+@end deftypefun
+
+@comment stdlib.h
+@comment GNU
+@deftypefun {char *} qgcvt (long double @var{value}, int @var{ndigit}, char *@var{buf})
+This function is equivalent to the @code{gcvt} function except that it
+takes an @code{long double} value for the first parameter.
+
+This function is a GNU extension.  The prototype can be found in
+@file{stdlib.h}.
+@end deftypefun
+
+
+@cindex gcvt_r
+As said above the @code{ecvt} and @code{fcvt} function along with their
+@code{long double} equivalents have the problem that they return a value
+located in a static buffer which is overwritten by the next call of the
+function.  This limitation is lifted in yet another set of functions
+which also are GNU extensions.  These reentrant functions can be
+recognized by the by the conventional @code{_r} ending.  Obviously there
+is no need for a @code{gcvt_r} function.
+
+@comment stdlib.h
+@comment GNU
+@deftypefun {char *} ecvt_r (double @var{value}, int @var{ndigit}, int *@var{decpt}, int *@var{sign}, char *@var{buf}, size_t @var{len})
+The @code{ecvt_r} function is similar to the @code{ecvt} function except
+that it places its result into the user-specified buffer starting at
+@var{buf}.
+
+This function is a GNU extension.  The prototype can be found in
+@file{stdlib.h}.
+@end deftypefun
+
+@comment stdlib.h
+@comment SVID, Unix98
+@deftypefun {char *} fcvt_r (double @var{value}, int @var{ndigit}, int @var{decpt}, int *@var{sign}, char *@var{buf}, size_t @var{len})
+The @code{fcvt_r} function is similar to the @code{fcvt} function except
+that it places its result into the user-specified buffer starting at
+@var{buf}.
+
+This function is a GNU extension.  The prototype can be found in
+@file{stdlib.h}.
+@end deftypefun
+
+@comment stdlib.h
+@comment GNU
+@deftypefun {char *} qecvt_r (long double @var{value}, int @var{ndigit}, int *@var{decpt}, int *@var{sign}, char *@var{buf}, size_t @var{len})
+The @code{qecvt_r} function is similar to the @code{qecvt} function except
+that it places its result into the user-specified buffer starting at
+@var{buf}.
+
+This function is a GNU extension.  The prototype can be found in
+@file{stdlib.h}.
+@end deftypefun
+
+@comment stdlib.h
+@comment GNU
+@deftypefun {char *} qfcvt (long double @var{value}, int @var{ndigit}, int @var{decpt}, int *@var{sign}, char *@var{buf}, size_t @var{len})
+The @code{qfcvt_r} function is similar to the @code{qfcvt} function except
+that it places its result into the user-specified buffer starting at
+@var{buf}.
+
+This function is a GNU extension.  The prototype can be found in
+@file{stdlib.h}.
+@end deftypefun