From schinder@pobox.com

The usual failure

lib/anydbm...........#
# anydbm.t test 12 will fail when AnyDBM_File uses the combination of
# DB_File and Berkeley DB 2.4.10 (or greater).
# You are using DB_File 1.72 and Berkeley DB 2.4.14
#
# Berkeley DB 2 from version 2.4.10 onwards does not allow null keys.
# This feature will be reenabled in a future version of Berkeley DB.
#
FAILED at test 12

All others passed.

Site configuration information for perl 5.00564:

Configured by schinder at Wed Feb  2 08:22:02 EST 2000.

Summary of my perl5 (revision 5.0 version 5 subversion 640) configuration:
  Platform:
    osname=linux, osvers=2.2.14, archname=ppc-linux-thread-multi
    uname='linux c22234-c.scllg1.pa.home.com 2.2.14 #2 sun jan 9 20:34:00 est 2000 ppc unknown '
    config_args='-Dprefix=/usr/local -Dusethreads -Dcc=gcc -des'
    hint=recommended, useposix=true, d_sigaction=define
    usethreads=define use5005threads=undef useithreads=define
    usesocks=undef useperlio=undef d_sfio=undef
    use64bits=undef uselargefiles=define usemultiplicity=define
  Compiler:
    cc='gcc', optimize='-O2', gccversion=2.95.2 19991024 (release/franzo)
    cppflags='-D_REENTRANT -Dbool=char -DHAS_BOOL -fno-strict-aliasing -I/usr/local/include'
    ccflags ='-D_REENTRANT -Dbool=char -DHAS_BOOL -fno-strict-aliasing -I/usr/local/include'
    stdchar='char', d_stdstdio=define, usevfork=false
    intsize=4, longsize=4, ptrsize=4, doublesize=8
    d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=8
    alignbytes=8, usemymalloc=n, prototype=define
  Linker and Libraries:
    ld='gcc', ldflags =' -L/usr/local/lib'
    libpth=/usr/local/lib /lib /usr/lib
    libs=-lnsl -lndbm -lgdbm -ldb -ldl -lm -lpthread -lc -lposix -lcrypt
    libc=/lib/libc-2.1.3.so, so=so, useshrplib=false, libperl=libperl.a
  Dynamic Linking:
    dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun=undef, ccdlflags='-rdynamic'
    cccdlflags='-fpic', lddlflags='-shared -L/usr/local/lib'

Locally applied patches:
    


@INC for perl 5.00564:
    lib
    /usr/local/lib/perl5/5.5.640/ppc-linux-thread-multi
    /usr/local/lib/perl5/5.5.640
    /usr/local/lib/perl5/site_perl/5.5.640/ppc-linux-thread-multi
    /usr/local/lib/perl5/site_perl/5.5.640
    .


Environment for perl 5.00564:
    HOME=/usr/local/src
    LANG (unset)
    LANGUAGE (unset)
    LD_LIBRARY_PATH=/usr/local/lib:/usr/local/qt/lib:/lib:/usr/lib:/usr/X11R6/lib
    LOGDIR (unset)
    PATH=/usr/local/bin:/bin:/sbin:/usr/bin:/usr/etc:/usr/sbin:/usr/local/sbin:/usr/X11R6/bin:/var/qmail/bin:.
    PERL_BADLANG (unset)
    SHELL=/bin/sh

From [Unknown Contact. See original ticket]

schinder@​pobox.com wrote

lib/anydbm...........#
# anydbm.t test 12 will fail when AnyDBM_File uses the combination of
# DB_File and Berkeley DB 2.4.10 (or greater).
# You are using DB_File 1.72 and Berkeley DB 2.4.14
#
# Berkeley DB 2 from version 2.4.10 onwards does not allow null keys.
# This feature will be reenabled in a future version of Berkeley DB.
#
FAILED at test 12

Since this is a known result, and since the test has gone to the
trouble of determining that DB_File is in use and the version of
Berkeley DB (as evidenced by the message text), why doesn't it
quietly ignore in this case?

Mike Guy

From @TimToady

M.J.T. Guy writes​:
: Since this is a known result, and since the test has gone to the
: trouble of determining that DB_File is in use and the version of
: Berkeley DB (as evidenced by the message text), why doesn't it
: quietly ignore in this case?

Or we could patch around it with a key of "*!@​NuLl kEy@​!*" or such.

Larry

From [Unknown Contact. See original ticket]

Larry Wall (lists.p5p)​:

Or we could patch around it with a key of "*!@​NuLl kEy@​!*" or such.

Somebody, *somewhere*, and you know it...

--

God Save the Queen!
And let Satan take the Prime Minister...
  - Tanuki, in the monastery.

From @TimToady

simon@​brecon.co.uk writes​:
: Larry Wall (lists.p5p)​:
: >Or we could patch around it with a key of "*!@​NuLl kEy@​!*" or such.
:
: Somebody, *somewhere*, and you know it...

Well, sure, but it still wouldn't be our fault. :-)

And it would make it less buggy than it is.

Larry

From [Unknown Contact. See original ticket]

On 2 Feb 2000, Simon Cozens wrote​:

Larry Wall (lists.p5p)​:

Or we could patch around it with a key of "*!@​NuLl kEy@​!*" or such.

Somebody, *somewhere*, and you know it...

Use a Truely Random Bits source such as HotBits
http​://www.fourmilab.ch/hotbits/how.html and run it out to 4096 bits. It
will never[1] collide.

--
Benjamin Franz

[1] For values of never smaller than the heat death of the universe.

From [Unknown Contact. See original ticket]

Benjamin Franz writes​:

Use a Truely Random Bits source such as HotBits
http​://www.fourmilab.ch/hotbits/how.html and run it out to 4096 bits. It
will never[1] collide.

[1] For values of never smaller than the heat death of the universe.

I hope I will not cause the immediate heat death of the universe by
the following remark​:

a) Write a DB_File with Perl modified as you propose;
b) convert it to, say, Oracle by an independent tool;
c) Now read these Oracle and DB_File databases into one Perl application.

The keys would collide.

Ilya

From [Unknown Contact. See original ticket]

On Wed, 2 Feb 2000, Ilya Zakharevich wrote​:

Benjamin Franz writes​:

Use a Truely Random Bits source such as HotBits
http​://www.fourmilab.ch/hotbits/how.html and run it out to 4096 bits. It
will never[1] collide.

[1] For values of never smaller than the heat death of the universe.

I hope I will not cause the immediate heat death of the universe by
the following remark​:

a) Write a DB_File with Perl modified as you propose;
b) convert it to, say, Oracle by an independent tool;
c) Now read these Oracle and DB_File databases into one Perl application.

The keys would collide.

True. Can't be avoided as far as I can tell once you regard DB_File
interfaces as an open system. But I have a weasle-out on semantics since
"right now" still satisfies the specified inequality.[1] :)

--
Benjamin Franz
[1] And thus the universe is saved from an untimely death.

From [Unknown Contact. See original ticket]

Simon Cozens <simon@​brecon.co.uk> writes​:

Larry Wall (lists.p5p)​:

Or we could patch around it with a key of "*!@​NuLl kEy@​!*" or such.

Somebody, *somewhere*, and you know it...

We could use a Tcl-like hack that used a utf8-encoded 0 ;-)

--
Nick Ing-Simmons

From [Unknown Contact. See original ticket]

Nick Ing-Simmons writes​:

We could use a Tcl-like hack that used a utf8-encoded 0 ;-)

You mean some quantity which is not utf8-encoded *anything*. ;-)

Ilya