Richard Levitte [Sun, 18 Jun 2000 16:06:44 +0000 (16:06 +0000)]
Modifications for VMS.
Richard Levitte [Sun, 18 Jun 2000 15:59:04 +0000 (15:59 +0000)]
Add support for dynamically created and destroyed mutexes. This will
be needed in some ENGINE code, and might serve elsewhere as well.
Note that it's implemented in such a way that the locking itself is
done through the same CRYPTO_lock function as the static locks.
WARNING: This is currently experimental and untested code (it will get
tested soon, though :-)).
Richard Levitte [Sun, 18 Jun 2000 14:06:40 +0000 (14:06 +0000)]
First of all, with the current macros, we should never get any
type-specific stack function. Second, even when we don't build any of
those functions, DECLARE_STACK_OF lines should not find themselves
into $def.
Bodo Möller [Sat, 17 Jun 2000 23:41:44 +0000 (23:41 +0000)]
Using speaking "variable" names in macros so that e.g. grepping for
sk_whatever_insert and sk_whatever_set immediately reveals the subtle
difference in parameter order.
Change mkstack.pl so that safestack.h is not rewritten when
nothing has changed.
Dr. Stephen Henson [Fri, 16 Jun 2000 23:29:26 +0000 (23:29 +0000)]
Safe stack reorganisation in terms of function casts.
After some messing around this seems to work but needs
a few more tests. Working out the syntax for sk_set_cmp_func()
(cast it to a function that itself returns a function pointer)
was painful :-(
Needs some testing to see what other compilers think of this
syntax.
Also needs similar stuff for ASN1_SET_OF etc etc.
Richard Levitte [Fri, 16 Jun 2000 15:25:41 +0000 (15:25 +0000)]
Change to have a single library that works on both Win9x and WinNT.
As far as I understand, it still needs to be compiled on NT...
Contributed by Arne Ansper <arne@ats.cyber.ee>
Bodo Möller [Fri, 16 Jun 2000 14:35:05 +0000 (14:35 +0000)]
typo
Bodo Möller [Fri, 16 Jun 2000 14:29:51 +0000 (14:29 +0000)]
In EVP_BytesToKey, replace explicit "8" by "PKCS5_SALT_LEN".
Bodo Möller [Fri, 16 Jun 2000 14:24:37 +0000 (14:24 +0000)]
There are compilers that complain if a variable has the same name as a
label. (Reported by Alexei Bakharevski.)
Geoff Thorpe [Fri, 16 Jun 2000 10:45:36 +0000 (10:45 +0000)]
Currently the DSO_METHOD interface has one entry point to bind all
"symbols" including functions (of all prototypes( and variables. Whilst
casting any function type to another violates ANSI C (I believe), it is
a necessary evil in shared-library APIs. However, it is quite
conceivable that functions in general and data symbols could very well
be represented differently to each other on some systems, as Bodo said;
> Since the function/object distinction is a lot more likely to be
> important on real-life platforms supporting DSO *and* it can be quite
> easily done *and* it will silence compilers that don't like
> assignments from void pointers to function pointer variables, why
> not do it?
I agree. So this change splits the "dso_bind" handler in DSO_METHOD
into "dso_bind_var" and "dso_bind_func". Similarly the exported
function DSO_bind() has been split in two. I've also put together
changes for the various DSO_METHOD implementations, but so far only
DSO_dlfcn() has been tested. BTW: The prototype for dso_bind had been
a bit strange so I've taken the opportunity to change its shape (in
both variations).
Also, the README has been updated - particularly with a note about
using customised native name-translation for shared libraries (and that
you can't do it yet).
Dr. Stephen Henson [Thu, 15 Jun 2000 23:48:05 +0000 (23:48 +0000)]
Add support for the modified SGC key format used in IIS.
Richard Levitte [Thu, 15 Jun 2000 20:57:25 +0000 (20:57 +0000)]
This seems to work better with enums...
Richard Levitte [Thu, 15 Jun 2000 11:10:15 +0000 (11:10 +0000)]
Make sure that bs is not getting free'd again.
Reported by Robert Eiglmaier <robert.eiglmaier@ixos.de>
Richard Levitte [Wed, 14 Jun 2000 17:27:02 +0000 (17:27 +0000)]
Add info on what some other people are currently working on.
Bodo Möller [Wed, 14 Jun 2000 10:09:46 +0000 (10:09 +0000)]
Report "error" (usually just "File exists", which is harmless)
when symlink() fails.
Bodo Möller [Tue, 13 Jun 2000 15:07:39 +0000 (15:07 +0000)]
This probably fixes a BN_rshift bug.
Bodo Möller [Tue, 13 Jun 2000 13:50:42 +0000 (13:50 +0000)]
Use BN_CTX_end when exiting early from BN_mod_exp_mont_word because
BN_mod_exp_atalla could be used.
Geoff Thorpe [Tue, 13 Jun 2000 13:18:51 +0000 (13:18 +0000)]
The atalla functionality doesn't work with the "word" version of
BN_mod_exp. Call the regular atalla mod_exp operation instead.
Geoff Thorpe [Tue, 13 Jun 2000 12:59:38 +0000 (12:59 +0000)]
Enable DSO support on alpha (OSF1), cc and gcc.
Also, "make update" has added some missing functions to libeay.num,
updated the TABLE for the alpha changes, and updated thousands of
dependancies that have changed from recent commits.
Ulf Möller [Mon, 12 Jun 2000 19:45:38 +0000 (19:45 +0000)]
protoypes
Ulf Möller [Mon, 12 Jun 2000 18:50:59 +0000 (18:50 +0000)]
typo
Dr. Stephen Henson [Sun, 11 Jun 2000 15:43:17 +0000 (15:43 +0000)]
Fix evp_locl.h macros.
Documentation correction.
Dr. Stephen Henson [Sun, 11 Jun 2000 12:27:58 +0000 (12:27 +0000)]
Make EVP_SealInit() and EVP_OpenInit() check EVP_EncryptInit() and
EVP_DecryptInit() return values.
Update docs.
Dr. Stephen Henson [Sun, 11 Jun 2000 12:18:15 +0000 (12:18 +0000)]
Document EVP routines. Change EVP_SealInit() and EVP_OpenInit()
to support multiple calls.
New function to retrieve email address from certificates and
requests.
Bodo Möller [Sat, 10 Jun 2000 12:05:52 +0000 (12:05 +0000)]
In longer tests with g=2, DH exchange does not become quite as fast
as expected -- maybe it's the different processor, maybe my
previous timings were too inaccurate.
Bodo Möller [Sat, 10 Jun 2000 10:08:31 +0000 (10:08 +0000)]
BN_mod_exp_mont_word entry:
Don't give performance gain estimates that appear to be more precise
than they really are, especially when they are wrong
(2/(1/1.15 + 1) = ca. 1.0698).
Bodo Möller [Fri, 9 Jun 2000 16:43:49 +0000 (16:43 +0000)]
Comment about bcopy on SunOS 4.x.
Richard Levitte [Fri, 9 Jun 2000 10:41:35 +0000 (10:41 +0000)]
Using checks of the existence of HEADER_{foo}_H in other header files
was a really bad idea. For example, the following:
#include <x509.h>
#include <bio.h>
#include <asn1.h>
would make sure that things like ASN1_UTCTIME_print() wasn't defined
unless you moved the inclusion of bio.h to above the inclusion of
x509.h. The reason is that x509.h includes asn1.h, and the
declaration of ASN1_UTCTIME_print() depended on the definition of
HEADER_BIO_H. That's what I call an obscure bug.
Instead, this change makes sure that whatever header files are needed
for the correct process of one header file are included automagically,
and that the definitions of, for example, BIO-related things are
dependent on the absence of the NO_{foo} macros. This is also
consistent with the way parts of OpenSSL can be excluded at will.
Bodo Möller [Fri, 9 Jun 2000 09:11:30 +0000 (09:11 +0000)]
Comment for increased code clarity.
Bodo Möller [Thu, 8 Jun 2000 22:47:09 +0000 (22:47 +0000)]
Another attempt to allow compiling on SunOS 4.*.
Bodo Möller [Thu, 8 Jun 2000 22:41:03 +0000 (22:41 +0000)]
Harmonize indentation.
Bodo Möller [Thu, 8 Jun 2000 22:40:09 +0000 (22:40 +0000)]
Accept -F4 option in lower case, which is what the usage information
says one should use.
Bodo Möller [Thu, 8 Jun 2000 20:26:03 +0000 (20:26 +0000)]
Slightly faster DSA verification (BN_mod_exp2_mont),
marginally faster BN_mod_exp for 1024 bit exponents.
Bodo Möller [Thu, 8 Jun 2000 15:01:14 +0000 (15:01 +0000)]
Add entry that Richard forgot.
Bodo Möller [Thu, 8 Jun 2000 09:39:28 +0000 (09:39 +0000)]
Use the equivalent of a sliding window (without precomputation
because we're only handling words anyway) in BN_mod_exp_mont_word
making it a little faster for very small exponents,
and adjust the performance gain estimate in CHANGES according
to slightly more thorough measurements.
(15% faster than BN_mod_exp_mont for "large" base,
20% faster than BN_mod_exp_mont for small base.)
Bodo Möller [Wed, 7 Jun 2000 21:29:25 +0000 (21:29 +0000)]
Speed up DH with small generator.
Bodo Möller [Wed, 7 Jun 2000 21:28:15 +0000 (21:28 +0000)]
No need to abort if c_rehash fails here (e.g. because Perl is not where
it is expected).
Bodo Möller [Wed, 7 Jun 2000 19:43:44 +0000 (19:43 +0000)]
use consistent indentation
Ulf Möller [Tue, 6 Jun 2000 20:52:42 +0000 (20:52 +0000)]
Option "no-symlinks" to configure without creating the links (e.g.
for use with makefile.one)
Andy Polyakov [Tue, 6 Jun 2000 15:21:12 +0000 (15:21 +0000)]
Compaq C warns that "the expression 'p=scan_esc(p)' modifies the variable
'p' more than once without an intervening sequence point. This behavior
is undefined." What it essentially complains about is 'p=p+=1'. Now it's
changed to 'p=p+1'...
Andy Polyakov [Tue, 6 Jun 2000 14:13:16 +0000 (14:13 +0000)]
MT-support for IRIX 6.x and Alpha-Linux
Ulf Möller [Tue, 6 Jun 2000 02:45:18 +0000 (02:45 +0000)]
Increased consideration for stupid Linux users.
Andy Polyakov [Mon, 5 Jun 2000 14:39:47 +0000 (14:39 +0000)]
GCC 2.95.2 from IRIX 6.5 -mabi=64 compiler bug workaround:-(
Submitted by:
Reviewed by:
PR:
Bodo Möller [Mon, 5 Jun 2000 13:50:57 +0000 (13:50 +0000)]
int may be smaller than 32 bits.
Richard Levitte [Sun, 4 Jun 2000 07:17:28 +0000 (07:17 +0000)]
According to Gordon Atwood <gordon@cs.ualberta.ca>, GNU C on SunOS
4.1.4 uses libiberty to define strtoul and strerror.
Richard Levitte [Sun, 4 Jun 2000 07:12:20 +0000 (07:12 +0000)]
According to Gordon Atwood <gordon@cs.ualberta.ca>, stdlib.h is
needed, or size_t won't be defined on SunOS 4.1.4.
Richard Levitte [Sun, 4 Jun 2000 07:08:29 +0000 (07:08 +0000)]
Typo...
Ulf Möller [Sat, 3 Jun 2000 23:23:10 +0000 (23:23 +0000)]
CygWin32 support.
Submitted by: John Jarvie <jjarvie@newsguy.com>
Ulf Möller [Sat, 3 Jun 2000 23:21:43 +0000 (23:21 +0000)]
Yet another bc FAQ.
Ben Laurie [Sat, 3 Jun 2000 14:13:58 +0000 (14:13 +0000)]
EVP constification.
Richard Levitte [Thu, 1 Jun 2000 22:19:21 +0000 (22:19 +0000)]
There have been a number of complaints from a number of sources that names
like Malloc, Realloc and especially Free conflict with already existing names
on some operating systems or other packages. That is reason enough to change
the names of the OpenSSL memory allocation macros to something that has a
better chance of being unique, like prepending them with OPENSSL_.
This change includes all the name changes needed throughout all C files.
Ulf Möller [Thu, 1 Jun 2000 20:25:44 +0000 (20:25 +0000)]
Use NO_FP_API.
Ulf Möller [Thu, 1 Jun 2000 19:34:25 +0000 (19:34 +0000)]
Don't include <stdlib.h>. In the NO_FP_API case, don't include <stdio.h>.
Richard Levitte [Thu, 1 Jun 2000 17:40:34 +0000 (17:40 +0000)]
Small documentation bug, probably a cut'n'paste, corrected.
Ulf Möller [Thu, 1 Jun 2000 14:24:59 +0000 (14:24 +0000)]
Bug fix for 64 bit HP-UX.
Submitted by: Karsten Spang <ks@bellesystems.com>
Richard Levitte [Thu, 1 Jun 2000 11:23:20 +0000 (11:23 +0000)]
Small fix to enable reading from stdin as well.
Contributed by Yoichiro Okabe <okabe@wizsoft.co.jp>
Geoff Thorpe [Thu, 1 Jun 2000 06:07:19 +0000 (06:07 +0000)]
"make update" + stripping the type-specific stack functions out of
libeay.num and ssleay.num.
Geoff Thorpe [Thu, 1 Jun 2000 05:13:52 +0000 (05:13 +0000)]
This change will cause builds (by default) to not use different STACK
structures and functions for each stack type. The previous behaviour
can be enabled by configuring with the "-DDEBUG_SAFESTACK" option.
This will also cause "make update" (mkdef.pl in particular) to
update the libeay.num and ssleay.num symbol tables with the number of
extra functions DEBUG_SAFESTACK creates.
The way this change works is to accompany each DECLARE_STACK_OF()
macro with a set of "#define"d versions of the sk_##type##_***
functions that ensures all the existing "type-safe" stack calls are
precompiled into the underlying stack calls. The presence or abscence
of the DEBUG_SAFESTACK symbol controls whether this block of
"#define"s or the DECLARE_STACK_OF() macro is taking effect. The
block of "#define"s is in turn generated and maintained by a perl
script (util/mkstack.pl) that encompasses the block with delimiting
C comments. This works in a similar way to the auto-generated error
codes and, like the other such maintenance utilities, is invoked
by the "make update" target.
A long (but mundane) commit will follow this with the results of
"make update" - this will include all the "#define" blocks for
each DECLARE_STACK_OF() statement, along with stripped down
libeay.num and ssleay.num files.
Geoff Thorpe [Thu, 1 Jun 2000 02:36:58 +0000 (02:36 +0000)]
The previous commit to crypto/stack/*.[ch] pulled the type-safety strings
yet tighter, and also put some heat on the rest of the library by
insisting (correctly) that compare callbacks used in stacks are prototyped
with "const" parameters. This has led to a depth-first explosion of
compiler warnings in the code where 1 constification has led to 3 or 4
more. Fortunately these have all been resolved to completion and the code
seems cleaner as a result - in particular many of the _cmp() functions
should have been prototyped with "const"s, and now are. There was one
little problem however;
X509_cmp() should by rights compare "const X509 *" pointers, and it is now
declared as such. However, it's internal workings can involve
recalculating hash values and extensions if they have not already been
setup. Someone with a more intricate understanding of the flow control of
X509 might be able to tighten this up, but for now - this seemed the
obvious place to stop the "depth-first" constification of the code by
using an evil cast (they have migrated all the way here from safestack.h).
Fortunately, this is the only place in the code where this was required
to complete these type-safety changes, and it's reasonably clear and
commented, and seemed the least unacceptable of the options. Trying to
take the constification further ends up exploding out considerably, and
indeed leads directly into generalised ASN functions which are not likely
to cooperate well with this.
Geoff Thorpe [Thu, 1 Jun 2000 02:15:40 +0000 (02:15 +0000)]
This is the first of two commits (didn't want to dump them all into the
same one). However, the first will temporarily break things until the
second comes through. :-)
The safestack.h handling was mapping compare callbacks that externally
are of the type (int (*)(type **,type **)) into the underlying callback
type used by stack.[ch], which is (int (*)(void *,void *)). After some
degree of digging, it appears that the callback type in the underlying
stack code should use double pointers too - when the compare operations
are invoked (from sk_find and sk_sort), they are being used by bsearch
and qsort to compare two pointers to pointers. This change corrects the
prototyping (by only casting to the (void*,void*) form at the moment
it is needed by bsearch and qsort) and makes the mapping in safestack.h
more transparent. It also changes from "void*" to "char*" to stay in
keeping with stack.[ch]'s assumed base type of "char".
Also - the "const" situation was that safestack.h was throwing away
"const"s, and to compound the problem - a close examination of stack.c
showed that (const char **) is not really achieving what it is supposed
to when the callback is being invoked, what is needed is
(const char * const *). So the underlying stack.[ch] and the mapping
macros in safestack.h have all been altered to correct this.
What will follow are the vast quantities of "const" corrections required
in stack-dependant code that was being let "slip" through when
safestack.h was discarding "const"s. These now all come up as compiler
warnings.
Ulf Möller [Thu, 1 Jun 2000 00:27:59 +0000 (00:27 +0000)]
is needed.
Bodo Möller [Wed, 31 May 2000 23:20:10 +0000 (23:20 +0000)]
typo
Bodo Möller [Wed, 31 May 2000 23:07:48 +0000 (23:07 +0000)]
dh and gendh have been obsoleted by dhparam.
Ulf Möller [Wed, 31 May 2000 22:25:44 +0000 (22:25 +0000)]
#include <stdio.h> not needed.
Ulf Möller [Wed, 31 May 2000 22:13:59 +0000 (22:13 +0000)]
#include <stdlib.h> is not needed.
Richard Levitte [Wed, 31 May 2000 21:35:38 +0000 (21:35 +0000)]
Result of "make update"
Richard Levitte [Wed, 31 May 2000 20:11:11 +0000 (20:11 +0000)]
This seems to become a FAQ...
Geoff Thorpe [Wed, 31 May 2000 17:41:34 +0000 (17:41 +0000)]
sk_***_new_null() seems to be there to avoid exactly this sort of thing
which is a cast between NULL and a function pointer.
Geoff Thorpe [Wed, 31 May 2000 17:35:11 +0000 (17:35 +0000)]
sk_value was also suffering from de-const-ification.
Also, add in a couple of missing declarations in pkcs7 code.
Richard Levitte [Wed, 31 May 2000 17:06:10 +0000 (17:06 +0000)]
config can now detect ARM Linux automagically.
Contributed by Jeremy Norris <jeremy.norris@rebel.com>
Bodo Möller [Wed, 31 May 2000 16:44:22 +0000 (16:44 +0000)]
strtoul is not used anywhere.
Bodo Möller [Wed, 31 May 2000 16:42:47 +0000 (16:42 +0000)]
When compiling with /opt/SUNWspro/SC4.2/bin/cc on Solaris, __svr4__ is
not defined, but __SVR4 is.
Geoff Thorpe [Wed, 31 May 2000 15:28:01 +0000 (15:28 +0000)]
All the little functions created by the IMPLEMENT_STACK_OF() macro will
cast their type-specific STACK into a real STACK and call the underlying
sk_*** function. The problem is that if the STACK_OF(..) parameter being
passed in has a "const *" qualifier, it is discarded by the cast.
I'm currently implementing a fix for this but in the mean-time, this is
one case I noticed (a few type-specific sk_**_num() functions pass in
const type-specific stacks). If there are other errors in the code where
consts are being discarded, we will similarly not notice them. yuck.
Bodo Möller [Wed, 31 May 2000 12:48:35 +0000 (12:48 +0000)]
Move Windows seeding functions into a separate file.
They have nothing to do with the particular PRNG (md_rand.c).
Bodo Möller [Wed, 31 May 2000 09:51:55 +0000 (09:51 +0000)]
Add "FIXME" comment.
Bodo Möller [Tue, 30 May 2000 21:44:36 +0000 (21:44 +0000)]
Improve PRNG robustness.
Dr. Stephen Henson [Tue, 30 May 2000 18:26:22 +0000 (18:26 +0000)]
More EVP cipher revision.
Change EVP_SealInit() and EVP_OpenInit() to
handle cipher parameters.
Make it possible to set RC2 and RC5 params.
Make RC2 ASN1 code use the effective key bits
and not the key length.
TODO: document how new API works.
Geoff Thorpe [Tue, 30 May 2000 12:59:43 +0000 (12:59 +0000)]
"make update"
Also, corrects the linux-elf-arm config string, it was previously setting
$des_obj = dlfcn :-)
Richard Levitte [Tue, 30 May 2000 08:01:24 +0000 (08:01 +0000)]
You must have an empty line between =item's
Dr. Stephen Henson [Tue, 30 May 2000 02:21:15 +0000 (02:21 +0000)]
Fourth phase EVP revision.
Declare ciphers in terms of macros. This reduces
the amount of code and places each block cipher EVP
definition in a single file instead of being spread
over 4 files.
Dr. Stephen Henson [Tue, 30 May 2000 02:10:57 +0000 (02:10 +0000)]
Fourth phase EVP revision.
Declare ciphers in terms of macros. This reduces
the amount of code and places each block cipher EVP
definition in a single file instead of being spread
over 4 files.
Richard Levitte [Mon, 29 May 2000 16:18:37 +0000 (16:18 +0000)]
Configuration for linux on ARM (contributed by Jeremy Norris
<jeremy.norris@rebel.com>)
Geoff Thorpe [Mon, 29 May 2000 15:52:21 +0000 (15:52 +0000)]
Minor corrections to documentation.
* speed processes any/all options passed to it, not just one.
* DH and DSA have no "_get_method()" functions, only RSA does.
* typos.
Geoff Thorpe [Mon, 29 May 2000 04:20:52 +0000 (04:20 +0000)]
another warning.
Geoff Thorpe [Mon, 29 May 2000 03:50:37 +0000 (03:50 +0000)]
It seems that mktime does what is required here. Certainly timegm() can
not be used because it is not available on all systems (most notably,
win32).
Geoff Thorpe [Mon, 29 May 2000 03:17:45 +0000 (03:17 +0000)]
This declaration seems to have been added into the header file accidently.
There's no trace of it being implemented and it doesn't seem to have been
intended given that it is prototyped with a BIO yet there was a BIO-
specific version added in at the same time.
Geoff Thorpe [Mon, 29 May 2000 02:42:57 +0000 (02:42 +0000)]
oops.
Dr. Stephen Henson [Sun, 28 May 2000 12:44:46 +0000 (12:44 +0000)]
Third phase of EVP cipher overhaul.
Remove duplicated code in EVP.
Bodo Möller [Sat, 27 May 2000 22:25:01 +0000 (22:25 +0000)]
Bugfix: clear error queue after ignoring ssl_verify_cert_chain result.
Dr. Stephen Henson [Sat, 27 May 2000 12:38:43 +0000 (12:38 +0000)]
Second phase of EVP cipher overhaul.
Change functions like EVP_EncryptUpdate() so they now return a
value. These normally have software only implementations
which cannot fail so this was acceptable. However ciphers
can be implemented in hardware and these could return errors.
Dr. Stephen Henson [Fri, 26 May 2000 23:51:35 +0000 (23:51 +0000)]
Beginnings of EVP cipher overhaul. This should eventually
enhance and tidy up the EVP interface.
This patch adds initial support for variable length ciphers
and changes S/MIME code to use this.
Some other library functions need modifying to support use
of modified cipher parameters.
Also need to change all the cipher functions that should
return error codes, but currenly don't.
And of course it needs extensive testing...
Geoff Thorpe [Thu, 25 May 2000 13:20:11 +0000 (13:20 +0000)]
cvs update
Bodo Möller [Thu, 25 May 2000 11:24:42 +0000 (11:24 +0000)]
typo
Bodo Möller [Thu, 25 May 2000 09:50:40 +0000 (09:50 +0000)]
Implement SSL_OP_TLS_ROLLBACK_BUG for servers.
Call dh_tmp_cb with correct 'is_export' flag.
Avoid tabs in CHANGES.
Richard Levitte [Wed, 24 May 2000 22:00:59 +0000 (22:00 +0000)]
Add the target system linux-m68k
Dr. Stephen Henson [Wed, 24 May 2000 13:09:59 +0000 (13:09 +0000)]
Add DSA library string. Workaround for IIS .key file invalid
ASN1 encoding.
Ralf S. Engelschall [Tue, 23 May 2000 18:33:44 +0000 (18:33 +0000)]
Fix Blowfish URL.
Submitted by: Arnaud De Timmerman <Arnaud.De.Timmerman@branchur.fr>
Ben Laurie [Sun, 21 May 2000 15:21:55 +0000 (15:21 +0000)]
Yet more typesafety.
Bodo Möller [Sun, 21 May 2000 14:21:24 +0000 (14:21 +0000)]
Avoid sprintf
Bodo Möller [Sun, 21 May 2000 14:17:01 +0000 (14:17 +0000)]
Avoid sprintf.