Siesta party

[Ivor Williams]

Just to continue the exuberance we had in Paris, how about a party to celebrate 
Siesta? Maybe when a certain milestone is complete.
Maybe when the london.pm list moves onto siesta.

I don't have an occasion. I don't have a date.

But I do have a venue:

http://openguides.org/london/index.cgi?Acapulco%2C_NW3_6ND

Let's bring the sombreros and see if we can drink the place out of tequila.

Okay, so it's North London, not Central London. But it is zone 2, and in easy reach of 
public transport.

What do you think? Are you up for it?

Re: london.pm digest, Vol 1 #1485 - 18 msgs

[Ivor Williams]

- Original Message - 
> Date: Mon, 7 Jul 2003 16:46:11 +0100
> Subject: Re: OT: More sybase related - IDENTIFIER TOO LONG
> Cc: <[EMAIL PROTECTED]>
> To: [EMAIL PROTECTED]
> From: David Hodgkinson <[EMAIL PROTECTED]>
> Reply-To: [EMAIL PROTECTED]
> 
> 
> On Monday, July 7, 2003, at 03:41 PM, Raf wrote:
> > Seems like a bit of poor error handling/reporting on the part of sybase
> > though.
> >
> 
> Get the SAMS Sybase Unleashed book.
> 
> Oh, and then read it.
> 
Any chance of the ISBN? I would very much like to buy a copy for myself for work,
but I can't find it on Amazon or samspublishing.com. Is the book still in print?

Ivor.

Expletives (was Eurocracy sucks)

[Ivor Williams]

- Original Message - 
> Date: Mon, 14 Jul 2003 15:50:41 +0100
> From: Earle Martin <[EMAIL PROTECTED]>
> To: [EMAIL PROTECTED]
> Subject: Eurocracy sucks.
> Organization: Strike Force for Indolence and Spiritual Beauty
> Reply-To: [EMAIL PROTECTED]
>
> Some of you may have heard this from me, some not
>

Interestingly enough,  this message, and at least four of the replies to it have 
tripped the corporate dodgy word detector on-site
where I am working (lurking). Whilst I fully concur with Earle's sentiments about the 
attitude of the French authorities towards his
wife, I am wondering whether to unsubscribe from this list at work.

I thought that this list was more or less work-safe and clean, now I am not so sure.

IvorW (@home).

[Announce] Central London treasure hunt: Today

[Ivor Williams]

Rendezvous at: "The Tattershall Castle" - a boat pub on the Thames at Embankment.

Time: 4 pm onwards
Entrance: £5 / person

Please mail me off-list if interested or if you want to know more.

Ivor

Re: [ANNOUNCE] Call For Participation: March Techincal Meeting

[Ivor Williams]

From: Mark Fowler <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Subject: [ANNOUNCE] Call For Participation: March Techincal Meeting
Date: Thu, 20 Feb 2003 13:03:20 + (GMT)
Reply-To: [EMAIL PROTECTED]

> The next London Perl Mongers Tech meet will be held mid march, and I'm
> looking for talks.
>
> The simplest kind of talk you can do is a lightning talk.  These talks
> typically last between five and seven minutes and serve just long enough
> to briefly introduce a topic but not go into any detail.  They're great
> for first time speakers (you only have to be standing for a few minutes)
> and allow many people to cover a lot of different ground.  Previously we
> had an entire tech talk made up of talks and it might give you ideas:
> http://london.pm.org/tech_talks/21_nov_2002/.
>
> The normal length talks typically last about twenty minutes.  These
> normally allow a speaker to explain most topics with a reasonable level of
> detail.  Finally there's also the ultra long forty minute.  These
> typically allow you start from the beginning of a topic assuming no prior
> knowledge and cover the subject in depth.
>
> So, has anyone got any material to present?  Remember, it doesn't have to
> be overly complicated - in fact it could just be as simple as a "war
> story"  and how you overcame a problem you were having with Perl.  Also,
> is there any subject that anyone would particularly like to hear
> discussed?  Remember kids, this month gives a good opportunity for people
> to give practice talks for the upcoming conference season.
>

As mentioned in the pub after the last tech meet, I intend to do a talk on VCS::Lite. 
I will make this a 5 minute lightning talk, as
that's all I have the energy to prepare at the monent.

Ivor.

Mailing list topping and tailing

[Ivor Williams]

I know that the subject of corporate disclaimers comes up from 
time to time. I was wondering whether a general solution might be a good idea. 
If the mailing list code looks for a particular  ... , 
and distributes only the text between the tags, that would be quite useful - 
certainly to me.
 
If the message contains no  pair, or the pair is 
syntactically incorrect, the entire message is distributed as per what happens 
at present. I am also stuck for a tag name.
 
My present client has a dickslammer which takes the biscuit, 
and includes the words: "If you are not an intended recipient please delete 
this e-mail and notify postmaster@$Corporate" . Yeouch!! 
Not only that, but the mails are also prefixed with a banner saying "Please read 
the disclaimer below".
 
I am currently subscribed in lurk mode at $Corporate, and 
digest mode at $Home; living in trepidation of inadvertantly replying on-list, 
and having the instructions taken literally.
 
From a legal perspective, the purpose of a mailing list 
address is to send the message to a _COMPUTER_, and the blurb about the 
intended recipient surely does not apply, as the recipient COMPUTER is not in 
itself capable of copying, disclosing, deleting or any other activity outside 
what it has been instructed to do. The fact that I have 
instructed the computer to extract a tagged snippet and forward just the snippet 
does not invalidate any confidences. It would be just the same if I had a 
human slave scribe sitting outside the corporate network, obeying my 
instructions to cut and paste the section and distribute this to a list of 
subscribers.
 
Has anybody done this already? How easy for someone to embrace 
the snake and patch Mailman for london.pm? What about other list managers, e.g. 
ezmlm?
 
What do others think? Am I tilting at 
wind(mill|ow)s?
 
Ivor.

RE: dragons

[Ivor Williams]

On Friday, January 31, 2003 1:56 PM, Ben [SMTP:[EMAIL PROTECTED]] wrote:
> Is anyone thinking of going to this tonight?
> 
> http://www.tateandegglive.com/event1_cai.html
> 

Suggested rendezvous: The Founder's Arms, 6:00 p.m. onwards

Ivor.

RE: Drink for Peace? [[was: [PUB] Spread Eagle, NW1]]

[Ivor Williams]

On Tuesday, January 28, 2003 1:35 PM, Andy Wardley [SMTP:[EMAIL PROTECTED]] 
wrote:
> > "london perl mongers for peace" sounds really cool but i would also
> > respect the view that in terms of Perl Advocacy and gaining mindshare in
> > corporate land, having Perl associated with political ideology could be
> > doing the wider community a disservice.
>
> How about:
>
>London Peace Mongers
>
> or
>
>JAPH: Just Another Peace Hacker
>
> or
>
>#!/usr/bin/peace
>

Reminds me of a rule we used to have in the Makefile defaults:

love:
echo "not war?"

Ivor.

RE: [OT] Oldest machine still running perl

[Ivor Williams]

On Tuesday, January 21, 2003 3:35 PM, Alex McLintock [SMTP:[EMAIL PROTECTED]] 
wrote:
> At 22:37 20/01/03, you wrote:
> >On Mon, 20 Jan 2003, Shevek wrote:
> >
> > > I just gave away a 386/8MHz which ran it. Slackware 4.0.
> >
> >386/8MHz ? Thee were lucky lad !
> >
> >When I were a boy I had to make do with an abacus !
>
> Just had a friendly argument with a hitchhikers fan as to whether or not an
> abacus was
> a digital or analogue computer.
>

Humm... A slide rule is a good analogy for an anologue computer.

This should appeal to ZZ9ers. If I can remember the quote:

Zaphod (about Eddie the shipboard computer): I think we'd be better off with a 
slide rule.

Potential new module

[Ivor Williams]

I've been reviewing what I have learned from the LWP stuff I have been doing 
for Grubstreet. This involved form filling and POST method.

I was struck with the feeling that boiling down the form data is something that 
has probably been done many times over - but a search didn't find anything 
obvious.

I feel a CPAN module coming on (unless it's already been done and I've missed 
it), but I'm stuck on which namespace to use: HTML::Formdata, HTTP::Formdata, 
LWP::HTMLForm, thoughts please.

>From the existing code that I have written, is a sub formdata, which takes the 
HTML page and form name as parameters. The form name is optional; if no form 
name is specified, the routine picks up the first form on the page.

sub formdata {
my ($html,$formname) = @_;

my $tp = HTML::TokeParser->new(\$html) or die "Bad HTML form";
while (my $form = $tp->get_tag('form')) {
last if !$formname || ($form->[1]{name} eq $formname);
$tp->get_text('/form');
}
my @form;
while (my $field = $tp->get_tag('input','select','textarea')) {
my ($tag,$attr) = @$field;
if ($tag eq 'textarea') {
my $text = $tp->get_text('/textarea');
push @form,$attr->{name},$text;
next;
}

if ($tag eq 'select') {
my $selected;
while (my $tok = $tp->get_token) {
last if $tok->[-1] =~ m(/select)i;
my ($typ,$tag,$att) = @$tok;
next unless $typ eq 'S' && $tag eq 'option';
$selected = $att->{value} if exists $att->{selected};
}

push @form,$attr->{name},$selected if defined $selected;
next;
}

if ($attr->{type} =~ /hidden|password|text/) {
push @form,$attr->{name},$attr->{value};
}

if ($attr->{type} =~ /radio|checkbox/ &&
exists $attr->{checked}) {
push @form,$attr->{name},$attr->{value};
}
}
@form;
}

The sub returns a list of key/value pairs. Thinking about it, I realised that 
if the calling code turns it into a hash, this could lose any duplicate keys.

At this point, the light of recognition came on in my mind. This was a very 
familiar concept, that of a CGI object.
I could make formdata return a CGI object or something inheriting from CGI, 
giving access to all the input fields via $form->param. Besides being capable 
of being submitted via a normal POST of encoding type 
application/x-www-form-urlencoded, I would also like the code to be able to 
handle file uploads and encoding type multipart/form-data.

Has anything like this been done before? Please let me know if I am duplicating 
effort here.

Ivor.

FOSDEM 2003

[Ivor Williams]

Free & Open Source Developers Europe Meeting, 8th-9th February in Brussels.

http://www.fosdem.org/

Anyone interested in going?

Ivor.

RE: hello to london perl mongers

[Ivor Williams]

On Wednesday, January 08, 2003 3:11 PM, [EMAIL PROTECTED] 
[SMTP:[EMAIL PROTECTED]] wrote:
>
>
> is anyone interested in giving me some feedback on some software projects i'm
> sketching out? if yes, would you get in touch?

Are you free on Thursday Evening, If so, why not come to Penderel's Oak and 
meet us in the flesh.

Details are on the website:

http://london.pm.org/meetings

RE: Mounting a CDROM case-insensitively

[Ivor Williams]

On Thursday, December 19, 2002 11:00 AM, Jonathan Peterson 
[SMTP:[EMAIL PROTECTED]] wrote:
>
> > but isn't appearing to here. What's the magic?!
>
> The real magic would be a case insensitive filesystem for *nix. Has
> anyone written a kernel mod for Linux that gives you a case preserving
> case insensitive filesystem? I'd love one of those. Does anyone know of
> any *nixes that had the option for a case insensitive fs??

I'm just wondering what the likes of Samba and NFS do. As Samba is open source, 
there may be some clues here.

NFS can talk to systems that have case insensitive file systems, such as VMS. 
IIRC, NFS munges the filenames on VMS by adding $ signs.

Sysadmin story: Cannot send mail over more than 500 miles

[Ivor Williams]

http://www.robotthoughts.com/print.php?sid=107

Enjoy.

Apologies if this one has already done the rounds.

Ivor.

RE: Crazy maths proof

[Ivor Williams]

On Monday, December 09, 2002 10:11 AM, Jonathan Peterson 
[SMTP:[EMAIL PROTECTED]] wrote:
>
>
> Chris Ball wrote:
> > So, another maths exercise.  I'll award a pint at the January social
> > meet for the first correct post with the next number in the sequence,
> > and another for an explanation of the sequence itself.
> >
> >2 3 5 7 11 13 17 19 23 29 31 37 41 43 47 53 59 60
>
> The next number is 61, then 67
>
> These are orders of simple groups, only a newbie would be sidetracked by
> the whole prime numbers thing.
>
> I don't know what I'm talking about but the sequence appears in google,
> so it must be right. See:
>
In case anyone is interested in what a simple group is, check out 
http://www.wikipedia.org/wiki/Simple_group. Google is my friend to.

Party invite on behalf of Gordon Tytler

[Ivor Williams]

Hi all,

Gordon (who was around at last Thursday's social meet) has asked me to post an 
invite to the list, for his party on Saturday. He's not subscribed himself 
(shame!)

===

Saturday 14th December 7pm
14 Mayfield Gardens, Hanwell, W7
Ealing Broadway Tube - E1 bus

Shake your floor subwoofer
Loud R&B dancing, Auzzie ex-patriot and geek party with posh nibbles and nice 
booze.
===

RE: re-animating regexes

[Ivor Williams]

On Wednesday, November 27, 2002 6:50 PM, Mark [SMTP:[EMAIL PROTECTED]] wrote:
> I checked if similar would happen with Data::Dumper.  From a quick test,
> looks like this may allow you to do what you want, if appropriate bits
> added to the freeze/thaw subs.
>
> [mmorgan]$ perl -MData::Dumper -le '$a=qr/a+b/; $b = eval( Dumper $a );
> print /$b/ ? "yep" : "nope" foreach ( "acb", "abc" )'
> nope
> yep
>
> The version of Data::Dumper being used is 2.102.  This was only a quick
> test, and I've not looked into it too deeply, so YMMV. :)
>

Not sure if this would prove anything as $b does not have to be a Regexp object 
- it would work just as well as a string.

Another niggle: /me avoids $a and $b like the plague, as they have special 
meaning to sort.

Ivor.

RE: glaziers

[Ivor Williams]

On Wednesday, November 27, 2002 10:25 AM, Ben [SMTP:[EMAIL PROTECTED]] wrote:
> Hi,
>
> (Not that I use london.pm as a Source Of All Knowledge or anything)
>
> I'm in need of a glazier to repair my living room windows that were damaged
> in the recent-ish storms (yes, I really am that lazy). Can anyone recommend
> someone who's reliable, works in London and not stupidly expensive?
>

In the past I've used Acorn Glass and Glazing in Acton (they're local to me). 
They were quite reasonable.

I don't have the phone number to hand, but I guess they are in the Yellow 
Pages.

RE: re-animating regexes

[Ivor Williams]

On Wednesday, November 27, 2002 2:44 AM, Paul Makepeace 
[SMTP:[EMAIL PROTECTED]] wrote:
> use Disclaimer qw(late_night_post);
>
> Is it possible to Storable a qr// ?
>
> $ cat storable_re.pl
>   #!/usr/bin/perl -w
>   use strict;
>   use Storable qw(freeze thaw dclone);
>
>   my $r = qr/hello/i;
>   for my $mushy(0,1) {
>   $r = thaw freeze $r if $mushy;
>   print "(\$r is $r)\n";
>   print "Hi!\n" if $ARGV[0] =~ /$r/;
>   }
>
>
> $ perl storable_re.pl Hello
>   ($r is (?i-xsm:hello))
>   Hi!
>   ($r is Regexp=SCALAR(0x8100028))
>
> $
>
> WTF is a Regexp=SCALAR(0x) anyway, and how could I play with it?
>

use strict;
use warnings;

use Devel::Peek;

Dump qr/hello/i;

__OUTPUT__
SV = RV(0x1bc6ff8) at 0x1bdf03c
  REFCNT = 1
  FLAGS = (TEMP,ROK)
  RV = 0x1bdf0f0
  SV = PVMG(0x1d3f968) at 0x1bdf0f0
REFCNT = 1
FLAGS = (OBJECT,RMG)
IV = 0
NV = 0
PV = 0
MAGIC = 0x1bc0114
  MG_VIRTUAL = 0x2808c138
  MG_TYPE = 'r'
  MG_OBJ = 0x1bdb540
STASH = 0x1bdf120   "Regexp"

Looks like a regexp is implemented as an object - I can't find any Regexp:: 
namespace though.

Perl coding standards

[Ivor Williams]

The following thread on Perlmonks seems quite timely following our discussion 
last night in the pub on use strict;

http://perlmonks.org/index.pl?node_id=214741

The Peon's Guide To Secure System Development

[Ivor Williams]

I've just come across this:

http://m.bacarella.com/papers/secsoft/html/
Although it's not targetted at Perl people, it seems like quite a good 
reference document.

RE: law

[Ivor Williams]

On Wednesday, November 13, 2002 5:16 PM, Michael Stevens 
[SMTP:[EMAIL PROTECTED]] wrote:
> Can anyone in the UK recommend a good solicitor for employment law,
> preferably one with free time in the next 24 hours :)
>

I can't recommend a solicitor, but for useful information, check out 
http://ourworld.compuserve.com/homepages/pntodd/labour/labour.htm

btw Mike, what has happened?

Ivor.

RE: st peter's ale

[Ivor Williams]

On Wednesday, November 13, 2002 10:16 AM, Billy Abbott 
[SMTP:[EMAIL PROTECTED]] wrote:
> This seems to be true, and they even happily admit it on their website.
>
> http://www.stpetersbrewery.co.uk/index.htm?p=develop&t=htm
>
> dunno about the outsourcing bit - they claim to brew their beer in a manor
> house in surrey.
>

Actually the brewery is in Suffolk, and it exists, as my brother has visited 
it. Don't know what Camra's line is on St. Peter's Brewery - will check my GBG 
when I get home.

RE: st peter's ale

[Ivor Williams]

On Wednesday, November 13, 2002 9:57 AM, Simon Wistow 
[SMTP:[EMAIL PROTECTED]] wrote:
> On another mailing list (I was thinking of you! honest! they meant
> nothing) where we were having the perpetual "What pub shallw e go to
> next" discussion I mentioned The Jerusalem Tavern and said that I liked
> St Peter's ale. I got this reply
>
> - Begin forwarded message -
> > That St. Peter's the branding exercise, right? Boy Jackson has the odd
> > tale to tell...
>
> Hmm. I read a depressing article in a weekend supplement a while back. St.
> Peters was created by the former head of Interbrand (they came up with the
> PriceWaterhouseCoopers - phew, that took some getting out - concept, among
> other pieces of genius...).
>
> This guy went on about how he'd always wanted to own a "brand", and stated
> that all it took was a couple of pubs, a brewery, and some nice bottles.
> There was a comment something like "What's in the bottle doesn't matter,
> just what it looks like", and some stuff about outsourcing most of the
> production to Scottish and Newcastle or similar.
>
> I quite like some of the beer, and I've never been to one of the pubs, but
> my attitude them has never quite been the same since I read the article...
> - End forwarded message -
>
>
> That makes me feel dirty - even though I love the ale. Anybody know
> anything else?

Go check out http://www.stpetersbrewery.co.uk/ for the full story. I don't 
disapprove of what they have done.

Ivor.

RE: My First CPAN Module... but comments first, please?

[Ivor Williams]

On Tuesday, November 12, 2002 11:09 AM, Dave Hinton at home 
[SMTP:[EMAIL PROTECTED]] wrote:
>
>
> I've written a module which I intend to release onto CPAN.  I've never done
> this before, I haven't even registered on PAUSE yet.  So before I do that,
> I'd appreciate it if people would have a quick glance at what I've written,
> and let me know what major boo-boos my inexperience has caused :-)
>
> The module is called Tie::Deref, but I'm having second thoughts and wondering
>
> whether Alias::Lexical might not be a better name.
>
> The documentation is at http://thereaction.co.uk/dah/Tie-Deref-man
>
> The module itself is at http://thereaction.co.uk/dah/Tie-Deref-pm
>
Have you written any tests?

These are a good idea when submitting to CPAN, and are useful giving an insight 
into your thoughts on all the possible scenarios.

Ivor.

RE: Seeking Perl Advocate in Africa

[Ivor Williams]

On Tuesday, November 12, 2002 1:59 AM, George Woolley 
[SMTP:[EMAIL PROTECTED]] wrote:
> I am a squire of the Oakland Perl Mongers.
> I strive to make the world a little better
> and to, someday, become a knight.
>
> I am on a quest.
> To fulfill the last part of this quest I seek contact
> with a Perl advocate in Africa.
> Perhaps you are that person or can direct me to such a one.
> I request your aid in this.
>

Have you looked at Perlmonks? They have a monk map with people entered by 
longitude and lattitude, check out:

http://tinymicros.com/pm/index.php?goto=MonkMap

Not all monks have entered their lat + long, but you might find some who are 
close, who have.

Ivor.

RE: Looking for a group in the North of England

[Ivor Williams]

On Monday, November 11, 2002 8:09 PM, john imrie 
[SMTP:[EMAIL PROTECTED]] wrote:
> Does any one know of a Perl group in North of England?
>
> I got made redundent a month back and I'd like to stay up hear where housing
> is cheep :-)

Just curious, where abouts upnorth?

RE: mine data fast (was Re: [JOB] I Need One)

[Ivor Williams]

On Sunday, November 10, 2002 4:41 PM, Nicholas Clark [SMTP:[EMAIL PROTECTED]] 
wrote:
> On Sat, Nov 09, 2002 at 08:52:48PM -, Dean Wilson wrote:
>
> > who know all the requirements. If you saw a job for a MOD_PEARL developer
> > that you were interested in wouldn't you go for it?[0]
> >
> >   Dean
> > [0] Cue rants about not working for idiots. A jobs a job.
>
> true, but not all jobs are equal.
> It would depend on the circumstances. One is told that having errors on a CV
> gives a very bad impression. This works both ways.
>
> I much prefer working for non-idiots. Working for idiots is made more
> bearable
> by having an escape plan, but I find I conserve my meager reserves of sanity
> better by not being in a hole in the first place.
>
> Actually, the first thing I'd think on seeing a PEARL job is "bah, I've got
> to get past some clueless moron first before getting to someone interesting".
> But this still doesn't solve the real problem of firms being disreputable
> (or incompetent (or startup, which means that they don't really have a
> as much of a choice)) and recruiting people only to make them redundant next
> week (or month)

One sad thing is that often it is the PHB of the HR department that puts 
together adverts, who wouldn't know Perl from PEARL. It might be that the 
employer/client is perfectly OK, but Chinese whispers have crept into the 
advert. However, this says something about the bureaucracy and/or politics of 
the organisation.

Ivor.

RE: (no subject)

[Ivor Williams]

On Friday, November 08, 2002 5:28 PM, Kevin Gurney [SMTP:[EMAIL PROTECTED]] 
wrote:
>
> Anyone able to shed some light on how i can compare 2 arrays.
>
> I've tried ==, and this don't seem to work.
>

This may be overkill for what you want, but check out Algorithm::Diff on CPAN.

http://search.cpan.org/search?query=Algorithm::Diff

RE: Perl is someone else's bitch

[Ivor Williams]

On Friday, November 08, 2002 12:35 AM, Lusercop 
[SMTP:`the.lusercop'@lusercop.net] wrote:
> David Adler is responsible for the ';' -> 'vanderbilt' transformation
> (as helpfully enacted by the Hon M. Schwern, Esq.).

I wonder, is this related to the Schwarzian transform, or perhaps the 
Gutta-Percha transform ;-)

RE: Usernames?

[Ivor Williams]

On Wednesday, November 06, 2002 11:39 AM, Andrew Wilson 
[SMTP:[EMAIL PROTECTED]] wrote:
>
> I don't like DCL and my admins don't have enough clue to have installed
> any decent editors so I'm stuck with eve or edt (yuck!!!).  Arse to DCL!
> (Now I want to be ill all over the keyboard (yuck, yuck, yuck!))
>
You don't have to be a privileged user to install and set up your own editor, 
just write permission to a disk. A vi emulator called vile is available on 
Decus - Emacs has also been ported to VMS.

In terms of DCL - I have abandoned this in favour of Perl ;-) The VMS port 
works quite well, once you have got to know the few gotchas: long lines, 
binmode, flock, no symlinks, %ENV, etc.

RE: Usernames?

[Ivor Williams]

On Wednesday, November 06, 2002 10:04 AM, Billy Abbott 
[SMTP:[EMAIL PROTECTED]] wrote:
> On Tue, 5 Nov 2002, Adam Turoff wrote:
>
> > Early versions of unix supported monocased terminals.  I remember a
> > professor telling me that if you logged in and your username was
> > allcaps, the tty would go into a compatability mode AND DISPLAY
> > EVERYTHING IN UPPERCASE, EXCEPT FOR CAPITAL LETTERS LIKE \A AND \B
> > WHICH WOULD APPEAR WITH A PRECEDING BACKSLASH.
>
> VMS isn;t quite that nasty (although it fills my days with a constant
> nagging pain), but it has no idea of case, does most everythign in caps,
> and ignores illegal characters typed in passwords without giving you a
> warning, as well as ignoring case in passwords...

VMS++
I &heart; VMS

It also uppercases the echoing of characters, when the application program 
wants uppercase data. Nice.

You do get some interesting effects when VMS connects to an NFS mount on a Unix 
box. IIRC they use $ signs in the filespec to indicate a case override (such as 
an upper case character on the Unix side).

>
> it does have built in file versioning and an advanced beard growing plugin
> though...

Some Unixes , AIX I believe, have file versioning that can be enabled as a 
feature. But this is usually turned off for compatibility. I remember what it 
was like to have many versions of a file - What did it look like before I 
started fiddling with it ;-), Let's see if we can track the last version that 
worked. Lazy practices admittedly, but it saved me needing to go back to backup 
on many occasion.

I also remember that VMS files have more date/times on them than on Unix - 
Created date, Modified date, Date last backed up.

RE: Proposal: Filesystem abstraction layer

[Ivor Williams]

On Thursday, October 31, 2002 11:15 AM, Shevek [SMTP:[EMAIL PROTECTED]] wrote:

> Questions:
>
> I intend to write this module.
>
> 1) Is there any code I should look at already? I'm assuming that I will
> inherit IO::File and IO::Dir into my own File and Dir objects.

Take a look at the File::Spec and other modules in the File:: namespace. Also 
read perldoc perlport, as this presents the current guidelines for writing 
portable code.

>
> 2) What should I call it and where should it live in the namespace?
>

How about File::System or File::Abstract

Online communities

[Ivor Williams]

I've discovered this essay by 'chromatic', which may be of interest:

http://www.oreillynet.com/pub/a/network/2002/10/21/community.html

RFC: Usemod wiki enhancement for searches

[Ivor Williams]

 
The primary application for this is Grub Street (http://grault.net/grubstreet) 
though I am sending it to the london.pm list because I am interested in whether 
any mongers can pick any security holes in it.AFAIK the grub street list is a 
subset of london.pm.

Earle/Kake,

Assuming the script is OK, any chance of giving it a go on Grub Street. Please 
let me know what you think.

How to use it is in the small pod. How to install is in the comments at the 
top.

Ivor.



supersearch.pl
Description: Binary data

RE: Books on london.pm.org (was Re: applying patterns)

[Ivor Williams]

On Friday, October 11, 2002 4:17 PM, Simon Wistow [SMTP:[EMAIL PROTECTED]] 
wrote:
> On Fri, Oct 11, 2002 at 04:02:21PM +0100, Ivor Williams said:
> > Maybe introduce some kind of XP style voting system, like on Perlmonks or
> > Everything2.
>
> Alternatively we could just do it and then, if anybody complains, deal
> with it then rather than our current modus operandi of burning our
> bridges before we come to them (I know that doesn't actually make sense
> but I just liked the imagery)
>
[snip]
>
> Carpe the tuits! If you script it THEY WILL COME! They drew first blood!
> I'll be back! Friends, Romans, Perl Mongers! etc etc ad infinitum ad
> nauseam.
>
> Having said that I'm currently in no position to write the code to back
> my polemic up so I should probably shut up.
>
Perhaps we don't need to write code. It may just be a case of downloading and 
installing the Everything Engine. Anybody got a machine on which to host it?

[JOB] Rational Clearcase expert

[Ivor Williams]

Sorry if details seem a bit vague. Sorry again, this is not a Perl job.

My employers (Sopra Group) are looking for a Clearcase person.

We are looking for someone skilled in:

*   Installing and configuring Clearcase
*   Clearcase administration
*   Configuration Management
*   Development workflows and release cycles

Also, Unix skills are needed (goes without saying).

Please reply off-list to me directly if interested.

Ivor.

RE: Books on london.pm.org (was Re: applying patterns)

[Ivor Williams]

On Friday, October 11, 2002 3:49 PM, Simon Wilcox [SMTP:[EMAIL PROTECTED]] 
wrote:
> On Fri, 11 Oct 2002, Simon Batistoni wrote:
>
> > I think personal bios still have a problem, in that there has to be
> > some criteria for "bio-worthiness", which doesn't wind up looking
> > elitist. I have a sinking feeling that the two things are
> > incompatible, and that nice idea as it is, it really wouldn't work in
> > practice.
>
> Funnily enough, one of the things I liked about the "handwriting" site[1]
> was the biog page[2][3].
>
> My criteria for getting on it would probably be similar to the rules for
> voting in last years leadership contest, that is:
>
> * Must be a poster to the list
> * Or a regular on IRC
> * Or a regular at the pub and/or technical meets.
>
Maybe introduce some kind of XP style voting system, like on Perlmonks or 
Everything2.

Just a thought,

Ivor.

Perl monger funding (was RE: applying patterns)

[Ivor Williams]

On Friday, October 11, 2002 11:59 AM, Barbie [SMTP:[EMAIL PROTECTED]] wrote:
> From: "Nicholas Clark" <[EMAIL PROTECTED]>
>
> > might be somewhat excessive for our needs (linking to 4 different vendors)
> > but it does have the advantage that it uses an Amazon affiliates code that
> > raises money for the perl mongers (if you buy via Amazon)
> >
> > Might we want to do that, to try to funnel some revenue to the perl
> mongers?
>
> This was recently suggested for Birmingham.pm, but haven't had time to join
> any. Apparently there are quite a few affiliate programmes around these
> days, so could be quite useful.
>

Perhaps this is something that can be co-ordinated through pm.org itself, as 
this may avoid conflicts of interest between regional monger groups.

Davorg care to comment?

RE: similarity detection

[Ivor Williams]

On Tuesday, October 08, 2002 11:50 AM, Tim Sweetman 
[SMTP:[EMAIL PROTECTED]] wrote:
> nemesis wrote:
> >
> > Hello again,
> >
> > I have a database (mySQL) full of variable length text fields (average
> > about
> > 1500 characters, 250 words).  Curently there are about 250 fields, but I
> > hope
> > this to expand to as many as possible (it is an online joke archive).
>
> Well...
>
> Firstly, AI notwithstanding, this is going to be a heuristic fudgy mess.
> Therefore you should not expect your solution to work reliably, and you
> will need manual techniques for merging/cross-relating similar jokes.

This is the classic search engine problem.

> Secondly, my best guess as to how to do this would be along the lines of
> "have you heard the one about the chicken and the mobius strip?"... blow
> away punctuation and stop-words, remove duplicates, sort the words that
> appear ...
>
> chicken,cross,mobius,strip <- heuristic key for "why did the chicken
> cross the mobius strip?"

Sounds like you could use code from the Eliza algorithm. Detect keywords, 
derive combination key from keywords.

> You will probably want to look up jokes with most of the same words.
>
> Tweak/refine/complicate as necessary, with thesauri etc

Sounds an interesting project. Could have more applications than jokes. Jeeves 
with attitude?

Ivor.

RE: Straw poll

[Ivor Williams]

-Original Message-
From:   Nicholas Clark [SMTP:[EMAIL PROTECTED]]
Sent:   Sunday, October 06, 2002 10:54 AM
To: [EMAIL PROTECTED]
Subject:Straw poll

And if a talk introducing the perl debugger were given at a YAPC, how many
people would think it worthy of going to?

[Ivor]  I'm a great fan of debugger systems, and the perl5db in particular. 
I grew up with the VMS debugger, which is very powerful. When I discovered 
that Perl had this facility, I used it extensively - instead of adding 
extra prints and Dumper calls to find out what is happening.

In terms of the talk, I would probably not learn anything new (but there 
again, I might be surprised), but I would be happy to critique and give 
feedback for the talk if it is given at a london.pm tech meet. Also, how 
about covering Devel::ptkdb. This is a graphical Tk debugger, and very 
handy for debugging cgi scripts (just alter the shebang line to add 
-d:ptkdb. Even works on Windows).

Ivor.

RE: Unix test

[Ivor Williams]

-Original Message-
From:   Chris Devers [SMTP:[EMAIL PROTECTED]]
Sent:   Monday, October 07, 2002 1:41 PM
To: London Perl Mongers
Subject:Re: Unix test

On Mon, 7 Oct 2002, Ivor Williams wrote:

> I have been asked at work, to come up with a Unix technical test

As in an exam, where you give a sheet full of questions & ash the
candidate to respond?
[Ivor]  That's the idea. The candidate will have of the order of 1 hour to 
complete the questions.

If so, then how about coming up with say 6 or 10
typical scenarios (general usage, programming situations, etc) and then
set up a series of open ended questions around these scenarios, asking the
candidate how she would approach the situation. One benefit of doing
things this way is that it'll allow the exam subject to choose their own
tools -- and if they go with Korn & C answers then great but if they can
reach desirable solutions with Perl or another language then okay.
[Ivor]  Good idea. I could even get them to write some code. Also, I 
thought of giving examples, with "what's wrong with this?" or "How can this 
be improved on?"

If you want a longer test, pad it out with another dozen or so simple
recognition questions -- "what is $foo", "compare & contrast $bar & $baz",
"why would one use $argle" -- that sort of thing. Put any questions to
this effect at the beginning so you can guage basic comprehension, and be
sure that at least some mention tools you want the person to know (Korn,
C, etc). Then the open ended ones above can show more creative thinking.
[Ivor]  That's also what I had in mind. Some easy starters to get the 
candidate going.

Unix test

[Ivor Williams]

I have been asked at work, to come up with a Unix technical test for use in 
interviewing prospective candidates, and I was wondering if there already 
was such a test in existence. To search the list, I tried a google advanced 
search on pm.org - without any avail.

I did ask whether there was any preference as to shell, and what the role 
involves. I was told that the role is a Unix developer with C/C++; the 
client has a bias towards Korn shell. No perl there - shame!

It's not that I intend to crib the entire test, more that it will give me 
ideas as to the kind of questions to ask.

Any URLs would be appreciated.

Thanks in advance,

Ivor.

Re: dim sum, charing cross road

[Ivor Williams]

- Original Message -
From: "Ivor Williams" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: 02 October 2002 11:02
Subject: RE: dim sum, charing cross road


>
>
> -Original Message-
> From: Greg McCarroll [SMTP:[EMAIL PROTECTED]]
> Sent: Wednesday, October 02, 2002 10:55 AM
> To: London.pm
> Subject: dim sum, charing cross road
>
> does anyone fancy dim sum this thursday? i'm currently working on CCR
> so it would be rather handy ;-) i'm guessing the time would be 1pm and
> the location would be the new world, was all the normal rules applying
>
> [Ivor]  I'm game. I'm on semi free time ATM, so I can take a reasonable lunch break.

I will not be dim summing after all. Nor will I be at tonight's pm social as I have a 
head full of cold and I am phoning in sick to
work.

RE: dim sum, charing cross road

[Ivor Williams]

-Original Message-
From:   Greg McCarroll [SMTP:[EMAIL PROTECTED]]
Sent:   Wednesday, October 02, 2002 10:55 AM
To: London.pm
Subject:dim sum, charing cross road
 
does anyone fancy dim sum this thursday? i'm currently working on CCR
so it would be rather handy ;-) i'm guessing the time would be 1pm and
the location would be the new world, was all the normal rules applying

[Ivor]  I'm game. I'm on semi free time ATM, so I can take a reasonable lunch break. 

Buffy fans meet on meetup.com

[Ivor Williams]

I know many of you are Buffy fans. check out http://buffy.meetup.com/?localeId=853 
FWIW.

Ivor.

Re: -tap tap tap-

[Ivor Williams]

Yo babe, it rocks dude!

- Original Message - 
From: "alex" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: 04 June 2002 13:58
Subject: -tap tap tap-


> 
> is this thing on?
> 
> 
> 
> 
> 
> 

[humour] How Google rates pages

[IVOR WILLIAMS]

Check out http://www.google.com/technology/pigeonrank.html

Security hole in Apache + PHP

[Ivor Williams]

see http://www.apacheweek.com/issues/02-03-01

May be useful info if any of you are running any PHP.

Or may be useful ammunition in the arguments in favour of Perl.

Ivor Williams
Sopra Mentor Consultant
LIFF? Core Systems Development
Extn: 2436   Mobile: 07752 234832



---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: Lightweight template module?

[Ivor Williams]

Leo Lapworth [mailto:[EMAIL PROTECTED]] wrote:

> Personally I think Template Toolkit is easy to install and there
> is only one other module that you need before it will installl
> (andy's AppConfig). Ok, it's not small, but then if you ever
> do want to do more complex stuff it's available.

It's easy to install if you happen to be installing it on a Unix box. I have
offered to Andy to fix the portability, but this has not happened due to a
lack of a circular tuit.

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: nms article online

[Ivor Williams]

Good one Dave!

Though I did at first wonder what software I need to download to read a
document in linux format. (Dohh!)

Then I wondered where the Acrobat zoom control was to correct for the fuzzy
text. I'm sure that Acrobat is keeping many opticians in business... or one
eyed kings possibly ;-)

But I realised that I was looking at a gif and you have OCR'ed the text
below (Dohh! again)

Still, I've not come across the programming language called Pen before :-)

Nice write up though. How much take-up is there so far on NMS? Are there any
webcounters on the site?

It would be interesting to see if the magazine improves the hit rate.

Ivor.

-Original Message-
From: Dave Cross [mailto:[EMAIL PROTECTED]]
Sent: 27 February 2002 16:05
To: [EMAIL PROTECTED]; london.pm
Subject: nms article online


***A T T E N T I O N***
This email originates from the Internet and therefore may not
be from the apparent sender.

If you have any doubts about the origin or content of the email please 
contact PC Support on x2288.
***

[x-posted. careful with those replies]

See .

Dave...

-- 

  Drugs are just bad m'kay


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: Bolloxia

[Ivor Williams]

I unearthed this little Bolloxia gem on Enron... on Perlmonks of all places.

http://perlmonks.org/index.pl?node_id=110244

Seems quite appropriate with hindsight. :-)

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: Supermarkets [Was: Tea]

[Ivor Williams]

Kate L Pugh [mailto:[EMAIL PROTECTED]] wrote
> I try to shop at local shops rather than supermarkets partly because I
> used to live somewhere where the local shops were pretty crap and I'm
> fairly sure that this was because there was a nearby out-of-town
> supermarket that all the people who could afford to have cars went to
> shop at.
[snip]
> So I want to support my local shops, because driving to a supermarket
> is not an option for everybody.  This is possibly mainly a matter of
> principle in Hammersmith -- there is a "Sainsbury's Local" within easy
> reach of my house at least, but that seems to sell mostly expensive
> posh food anyway.

Isn't there a Safeways in the mall @ H'smith?

In my experience, only 1 1/2 miles up the road at Acton, I find the local
shops are always changing hands, and are a complete rip-off with a limited
range of useful produce. And the milk is off when I get it home. There is a
Safeways and an Iceland (and a Netto but I don't use it) within 15 minutes
walk, and I can always get the bus, yes _bus_, to Ealing for a bigger range.

Anyway, back to Hammersmith, Brackenbury Village has some interesting shops
- not cheap but good wholesome organic stuff.

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: [ANNOUNCE] Tanker 0.01

[Ivor Williams]

Simon Wistow [mailto:[EMAIL PROTECTED]] wrote:

> So you could have IRC+Jabber+Email->Speech if you wanted. I'd quite ike
> to write a webswerver :)

Ever thought of getting the RNIB to sponsor it?

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: select() operates globally?

[Ivor Williams]

Jonathan Peterson [mailto:[EMAIL PROTECTED]] wrote

> So far as I can see select operates globally across all package
> namespaces. Which is annoying.
> 
> Thus:
> 
> $obj = Class->new;
> $obj->some_method();  # Calls a select() within package Class
> print "wibble;# prints to modified file handle, not
STDOUT!!
> 
> Is there some way of making it not do so?
{
local *STDOUT;
$obj->some_method()
}

should do the trick.

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: Tea (was Re: Bolloxia)

[Ivor Williams]

Simon Wistow [mailto:[EMAIL PROTECTED]] wrote

> I *heart* borough market - I just wish I could go there more often. I
> can't wait for summer again so I can go and then go and sit by the
> Anchor and look over the river and have a pint and ... 

But the Anchor is closed for a refurb :-(

> 'Professor Aleksander said the ethical problem of "killing" a conscious
> computer by switching it off would not arise. The machine would not feel
> threatened by such action.

But biology doesn't have persistence or backups. At least not until there is
a working cloning machine and braindump facility.

> "A conscious computer would be so different from a conscious living
> thing," said the professor.'

> Say he's wrong and that turning off an AI is like killing it or
> whatever. Is it morally wrong to run it on unstable OSs like Windows,
> MacOS.  or Linux 

> Ignoring the troll, that is a problem, doncha think?

I think that AI has a long way to go with producing something useable, and
we are not in a position to draw conclusions from the state of the
technology at present.

I'm not knocking AI, just certain systems and research areas that seem to be
overhyped. I think that the situation could well change within 5 years let
alone 10.

The state of technology does not stop us thinking about and discussing the
philosophical aspect. Mind u, perhaps we should continue this discussion on
::scr.

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: Developing Perl based products for windows.

[Ivor Williams]

Tommie M. Jones [mailto:[EMAIL PROTECTED]] wrote:

> I am a Perl developer who mostly works with Unix.  I have a perl program
> that I would like to sell to windows users.

> I was wondering if anyone has any experience creating easy to install CD's
> for perl based product.  I assume that I will have to include perl itself
> and all the basic modules.  Also I will need a GUI IFC.  Does anyone have
> any suggestions for a Gui on Windows?  TK maybe.  Any input would be
> appreciated.

It's probably worth you looking at http://www.activestate.com and their perl
offerings - to see how they package them, and steal a few good ideas.

In terms of GUIs, there's also Win32::GUI, which gives a more M$oft look and
feel - less X-windowy. But this is a non-portable module.

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: Social meet kick up the arse [Was: Return to the Yorke]

[Ivor Williams]

Steve Keay [mailto:[EMAIL PROTECTED]] wrote

> It is quite a difficult thing to walk into a pub and land in a room
> full of strange people (people you don't know) who are all engaged in
> varying levels of intense conversation.

I would like to draw on some personal experience - running a successful
social club for 20 somethings and 30 somethings that met weekly(we had
between 15 and 30 people attending). This is what I was doing 10 years ago.
The issue of huddles of strange people forming cliques and in-jokes is an
issue that _any_ social organisation has to deal with. We had two solutions:
one is to get everyone to own the problem and encourage all members to talk
to newbies. The other is to appoint one or more hosts/hostesses to fill this
role. The first approach seemed to work better, as the others don't get the
feeling "it's not my job to talk to people".

On the whole, IMO london.pm doesn't do too badly. When I started attending
in November last year, I did feel welcome, and I did feel I had something to
offer. The first meeting I attended was actually a tech meet.

> I advocate more speeches involving the whole crowd - even if they're 2
> minute efforts along the lines of, "Buy a tee shirt." or, "Come to the
> next technical meet." or, "Here's our new fearless leader to say a few
> words about what we're here for."

IMO FWIW we don't want to go overboard with speeches. That is what tech
meetings are for. However, having an introductory or valedictory speech from
Paul sounds like a good idea. One problem with this though, is timing. With
social meets, there is an informality that we would potentially lose - we
would have to have a start time and stick to it.

> IMHO those moments give people a sense of, well, all coming together
> to suffer the same thing and do a great deal for actually meeting
> people you don't know.  

"How long have you been a Perl hacker?" ... "I keep trying to resist, but I
can't get away from cutting code." ... "I tried some OO the other day and it
was nice, but I don't touch the hard stuff - XS and all that" ;-)

On the subject of free booze, might be a good idea as a one off. There is a
danger that we could get into arguments about who's going to pay for it.
There's also the issue of who counts as a newbie. A suggestion is to pick a
special date - N year anniversary - and lay on some champagne for everyone.

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: Bolloxia

[Ivor Williams]

Rob Partington [mailto:[EMAIL PROTECTED]] wrote

> But it's being changed to fit in with a more homogenised global ideal.
> They're being changed because it's a whole lot easier dealing with one
> brand name.  I read a rationale somewhere about the SuperMop (I think)
> that pointed out that having one brand over all of Europe meant they
> could launch the product 6-9 months earlier and at about 50% of the cost.

IMO this is snake oil! This is EXACTLY what the likes of Bill Gates want us
to think: computers => Intel, operating_system => Windows. Total monopoly
and world domination.

Mongers stick together. Dare to be different. Insist on quality. Spread
information about alternatives. Break monopolies.

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: Bolloxia

[Ivor Williams]

Barbie [mailto:[EMAIL PROTECTED]] wrote
> From: "Simon Wistow" <[EMAIL PROTECTED]>
> 
> > Giving up Marathons wasn't so hard but Opal Fruits was more of a
problem.
> 
> Up until last year I could still ask for a "Marathon" and the shopkeeper
> wouldn't bat an eyelid and hand over the correct chocolate bar [1]. Last
> year was the first time I got "Sorry? we don't sell those. Are they new?".

> [1] It's current name always makes me think of disgusting dog-eared
trainers
> ... eugh!

Rumour has it that there is some grafitti on a wall somewhere in Slough near
the Mars factory which reads "Snickers to you too!".

Ivor. 


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

Matts scripts discussion hits perlmonks

[Ivor Williams]

See http://perlmonks.org/index.pl?lastnode_id=6364&node_id=146058

Perpetrator script here is the cookie counter.

Ivor Williams
Sopra Mentor Consultant
LIFFE Core Systems Development
Extn: 2436   Mobile: 07752 234832



---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

Book Reviews: Cross-Platform Perl

[Ivor Williams]

Hi Folks,

Anybody seen or got a copy of "Cross-Platform Perl" by Eric Foster-Johnson?

http://www.pconline.com/~erc/perlbook.htm

Is it any good? 
Tom Christiansen gave it 4 camels.

The chapter headings seem to imply this is a "teach yourself perl from
square zero" style book. 

This book may be useful in the war against regal myopia. (Davorg - I'd be
interested in your opinion)

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: User Input at speed

[Ivor Williams]

Philip Newton [mailto:[EMAIL PROTECTED]] wrote:
> Chris Devers wrote:
> > I'm sorry but I'm quite sure I've never seen ALTGR on a US 
> > keyboard. Ever.
> [snip]
> > but it isn't functionality that most US users are aware of or
> > would miss. 
>
> And I think that's the reason -- most US users don't need to produce
> accented letters (and many are probably not even aware they exist in other
> languages).

I bet this really pisses off the Hispanics and French Canadians.

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: User Input at speed

[Ivor Williams]

Chris Devers [mailto:[EMAIL PROTECTED]] wrote
> That or our keyboards have no AltGr key. Typical standard bottom row goes:
>
>   CTRL [WIN] ALT SPACEBAR ALT [[WIN]] [CTXT] CTRL

Betcha the second "ALT" right of the space bar is really an "ALTGR".

Try this key and 4 inside a MS application like Word. See if this gives you
a Euro symbol.

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: Rindolf

[Ivor Williams]

Steve Mynott [mailto:[EMAIL PROTECTED]] wrote:

> It's fascinating to compare new and old languages and see how things
> have changed (usually for the better it seems).  It's almost like a
> process of evolution from something dinosaur like BCPL to modern
> languages like Ruby, Java and Perl 6 via C.  There are probably some
> features in old languages which have been forgotten which could now be
> useful.

A good example is Ruby's "call by name" parameter passing. Straight from the
Algol standard. Intervening descendents of Algol used call by reference,
which is nowhere near as powerful.

> The name is lame as well. 
I agree. It might appeal if you are a JRR Tolkien fan, but this could become
hobbit forming. 

> But generally I don't see anything wrong in the idea of doing yet another
> language and its quite possible the author could come up with some
> good ideas which could influence Perl 6 so good luck to him.

Agreed. Perhaps we should be posting ideas and language features wishlists
to him - encouragement rather than criticism.

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: User Input at speed

[Ivor Williams]

David Cantrell [mailto:[EMAIL PROTECTED]] wrote:
> This 'ere keyboard (Sun type 6) has grey Tab, Caps Lock, Shift, Control,
> Alt, Meta, Compose, Alt-graph, Return, Backspace, Esc, F1-12, plus the
> blocks of keys on either side of the QWERTY block.  That's pretty normal
> AFAIK.

Ahh COMPOSE! I miss that one. Where's my VT420?

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: User Input at speed

[Ivor Williams]

Philip Newton [mailto:[EMAIL PROTECTED]] wrote:
> > "Alt Graphics" is correct. This was a "reserved for future 
> > use" key, that MS pressed into service when the Euro currency
> > symbol was invented. (AltGr + 4)

> It's been in use for years in Germany -- the only way to obtain { [ ] } \
~
> | on a standard German computer keyboard. (As well as less common
characters
> such as ² ³ µ.)

That must make writing perl quite painful. !-)
I suppose you would get used to it though. AltGr becomes just another shift
key.

> And the euro is AltGr+E in most Microsoft layouts IIRC. (It is in the
German
> layout.) Ah, right: see
> http://www.eu.microsoft.com/typography/faq/faq12.htm#kbd . The only ones
> who're different are "Greek Latin", "Hungarian", "Irish", "Latvia QWERTY",
> "Polish", "United Kingdom", and "US International", probably because they
> already had something on AltGr+E.

UK layout has AltGr+E = E acute accent. as does AltGr+vowel

I wonder how to get the other accents on a UK keyboard. I know CharMapper
and ALT+nnn do, but that's cheating. ;-)

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: User Input at speed

[Ivor Williams]

Philip Newton [mailto:[EMAIL PROTECTED]] wrote
> What I'd like to know is what "Alt Gr" stands for. I used to think it was
> "Alt German", but I think also other national keyboards have it. "Alt
> Graphic", perhaps?

"Alt Graphics" is correct. This was a "reserved for future use" key, that MS
pressed into service when the Euro currency symbol was invented. (AltGr + 4)


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

RE: User Input at speed

[Ivor Williams]

Philip Newton [mailto:[EMAIL PROTECTED]] wrote

> Especially since you *will* have to remain familiar with QWERTY as you
will
> have to use keyboards that are not your own at some point (typing on a
> co-worker's machine; on a computer whose keyboard driver isn't working; at
> an Internet café; whatever -- it's bound to happen).

I quite agree. I have enough difficulty adjusting to the miniature keys on a
laptop.

At one previous employer, I remember a few supposedly ergonomic keyboards
which were QWERTY, but split in the middle into two halves at an angle. I
found these keyboards impossible to use. The guys that had adapted to these
keyboards soon regretted it, as they were stuck if they had to use anyone
else's machine, or if the keyboard ended its useful life. I think that there
is a pile of a half dozen of these weird keyboards gathering dust in a
cupboard in this company.

I am also looking to track down a contrasty PC keyboard - black keys
labelled in white, or white keys labelled in black (as opposed to dark brown
keys labelled in light brown). Any ideas?

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of 
LIFFE Holdings Plc or any of its subsidiary companies.
---

Low budget applications

[Ivor Williams]

Just a thought...

Thinking about Perl usage in the M25 radius area, I wonder how much takeup
there is at the low budget end of the spectrum, in particular:

Education
Health
Charities

These seem to me to be ideal markets for Perl to become established in,
where the advantages of open source software over commercial packages could
be realised to the full.

It would be encouraging to see some _application_ code in these areas
appearing on CPAN.

Any perlmongers working in this area?

I was once a sysadmin in a hospital IT department - long before I knew perl.
Perl would have been very useful for many aspects of this job.

Ivor Williams
Sopra Mentor Consultant
LIFFE Core Systems Development
Extn: 2436   Mobile: 07752 234832



---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

Diversity [was Dan Sugalski calling all mongers groups]

[Ivor Williams]

Paul Mison [mailto:[EMAIL PROTECTED]] writes:

> Random brainfart: is the diversity of Perl communities (use.perl.org,
> perl.com (to some extent), rhizo #perl, efnet #perl,
> comp.lang.perl.misc, perlmonks, the multiple PM groups) a strength or a
> weakness? Do people end up in the right place for them or just the first
> one they stumble on?

I think that the answer to the second question is 'Yes', taking the 'or' as
an 'inclusive or'. Most of us have discovered Perl at some stage in our
career and found a suitable list (I am using the term list here generically
to include mailing lists, discussion groups, IRCs, etc). The list was right
(hopefully) at the time for the job required, e.g. CGI scripts. However,
subscribing to a list exposes the human to a number of Perl community
'tentacles', and they find themselves subscribing to other lists.

If the human is boring, busy or under boss surveilance (3 b's), they will
probably just stay on the original list. But these are not the people who
contribute directly to the community, they are the audience for the perl
quality crusade. If the list is of poor quality, when the tentacle crosses
the list, it is the duty of conscientious perl mon(ger|k)s to rectify the
situation (nice work with CGI101, Dave! :-)).

Watch the Principle of Emergence, or Process of Evolution in action!

If we lose a few lists through lack of interest, this will not harm the
community, but will focus people on the extant active lists. There is no
need for executive action to terminate or merge active lists - besides, this
goes against the grain as we are a bottom-up community.

A general lack of interest in all the Perl lists would indicate that the
language is dead, and we are all programming in something else, such as
Perlby-Jython++. :-)

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: Callbacks in class definition woe

[Ivor Williams]

From: Jonathan Peterson [mailto:[EMAIL PROTECTED]] wrote
> ... In the meantime, does anyonw know why
> WWW::Robot goes VERY slowly? It seems to take 30-60 seconds between
> deciding a link is worth following and retrieving the contents of the
> page. I'm testing on a bog standard apache serving flat files on the
> local machine, so I've no idea what the hold-up is... Crafting my own
> LWP retrieval functions operates at the expected high speed

It may be worth using DProf or SmallProf to profile your application.

Try changing the shebang line:

#!/usr/local/bin/perl -d:DProf 

or 

#!/usr/local/bin/perl -d:SmallProf


This will make the application run MUCH SLOWER, but will be able to tell you
where it is spending its time.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: Callbacks in class definition woe

[Ivor Williams]

Jonathan Peterson [mailto:[EMAIL PROTECTED]] wrote:

>   $self->{rb}->addHook('invoke-on-contents',
\$self->invoke_on_contents);
> I get
> 
> WWW::Robot: SCALAR(0x526e3c) is not a function reference; Ignoring it
> 
> And if I try
> 
>   $self->{rb}->addHook('invoke-on-contents',
> \&{$self->invoke_on_contents});

You need a wrapper for your object method. addHook is expecting a code ref,
but does not know anything about the first (implied) argument, i.e. the
object reference.

I suggest

$self->{rb}->addHook('invoke-on-contents',
sub {my $foo = $self; $foo->invoke_on_contents});

The intermediary $foo makes the anonymous sub into a closure.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: One-Eyed Kings

[Ivor Williams]

Nic Gibson [mailto:[EMAIL PROTECTED]] wrote:
> On Tue, Jan 29, 2002 at 06:49:58PM +, Dave Cross wrote:
> > This person even goes by the name "The Eye".
> > 
> > 
> 
> This would be the site you mentioned in your talk, yes?
> Whilst, I'm in slack/CFT mode I'll be reading.

I wonder... is CGI101 a reference to room 101. Seems somewhat appropriate
methinks. ;-)


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

Saving Time

[Ivor Williams]

A timeless lesson on how consultants can make a difference for an
organisation...

Last week, we took some friends out to a new restaurant, and noticed that
the waiter who took our order carried a spoon in his shirt pocket. It seemed
a little strange, but I ignored it. However, when the busboy brought out
water and utensils, I noticed he also had a spoon in his shirt pocket, then
looked around the room and saw that all the staff had spoons in their
pockets.

When the waiter came back to serve our soup I asked, "Why the spoon?"

"Well," he explained, "the restaurant's owners hired Ernst & Young
Consulting experts in efficiency, in order to revamp all our processes.
After several months of statistical analysis, they concluded that customers
drop their spoons 73.84 percent more often than any other
utensil. This represents a drop frequency of approximately 3 spoons per
table per hour. If our personnel is prepared to deal with that contingency,
we can reduce the number of trips back to the kitchen and save 15 man-hours
per shift."

As luck would have it I dropped my spoon and he was able to replace it with
his spare spoon. "I'll get another spoon next time I go to the kitchen
instead of making an extra trip to get it right now." I was rather
impressed.

The waiter served our main course and I continued to look around. I then
noticed that there was a very thin string hanging out of the waiter's fly.
Looking around, I noticed that all the waiters had the same string hanging
from their flies. My curiosity got the better of me and before he walked
off, I asked the waiter, "Excuse me, but can you tell me why you have that
string right there?"

"Oh, certainly!" he answered, lowering his voice. "Not everyone is as
observant as you. That consulting firm I mentioned also found out that we
can save time in the toilet."

"How so?"

"See," he continued, "by tying this string to the head of you know what, we
can pull it out over the urinal without touching it and that way eliminate
the need to wash the hands, shortening the time spent in the toilet by 76.39
percent."

"Okay, that makes sense, but . . . if the string helps you get it out, how
do you put it back in?"

"Well," he whispered, lowering his voice even further, "I don't know about
the others, but I use the 
spoon."

Enjoy your meal!



Ivor Williams
Sopra Mentor Consultant
LIFF? Core Systems Development
Extn: 2436   Mobile: 07752 234832



---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: [OT] 3 light bulbs problem

[Ivor Williams]

Greg McCarroll [mailto:[EMAIL PROTECTED]] wrote:
*SPOILERS*

.
.
.

> you turn them all off for a while, have a cold beer and relax, then
> turn one on till you reckon its mad hot and turn another on and run
> down as fast as you can and touch the two lit bulbs, hopefully the one
> thats been on for a while is going to fry your hand

as near as dammit.

Actually, it's easier to test the temperature when the bulbs are off.


The solution I have is as follows:

Switch on A and B (C is off).

Wait 5 minutes.

Switch off B.

Go downstairs.

A bulb is on.
B bulb is off and hot.
C bulb is off and cold.

Ivor.

No more mail please on this off list!


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

[OT] 3 light bulbs problem

[Ivor Williams]

Time for one of those lateral thinking problems.

Imagine you have a cellar in which there are 3 light bulbs on different
circuits. The light switches are all at ground level, at the top of the
staircase, and it is not possible to see from here, which light(s) are on.

Using just one journey downstairs, determine which switch controls which
bulb.

You do not need any equipment other than your hands and eyes to solve this.

If you want to verify a solution without giving it away to others, please
mail me off list.

Ivor Williams
Sopra Mentor Consultant
LIFFE Core Systems Development
Extn: 2436   Mobile: 07752 234832



---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: vim+hardlinks

[Ivor Williams]

Redvers Davies [mailto:[EMAIL PROTECTED]] wrote

> > I'm doing some rudimentary source management with cp -al foo foo-hck
> > and expecting that when I edit foo-hck/bar.pm that it'll do what emacs
> > does, write a new file called bar.pm in foo-hck, but instead it
> > doesn't disturb the hardlink at all and so modifies the pristine
> > version of the file too.
> 
> > Yes I know the simplest answer is to use emacs because it's sane, but
> 
> I would have thought that the behaviour you describe would be considered
> insane.
> 
> You set a hard link for a reason - that reason being that when
> you read and write to any of those "files" they all point to the same
> area of disk and therefor all change.  You are effectivly asking the
> editor to ignore the policy descision you made already.

I beg to differ. I have had exposure to a number of O/S platforms and 
editors, and the behaviour you are implying is quirky to say the least.
The alternative behaviour of creating a new file, not overwriting the 
one that was linked to is more what is expected.

Consider what the editor is doing. It has been asked to write out its
editing buffer to a file (which may or may not exist). Frankly, the 
behaviour of "only clobber the file if there are no other links to it"
seems correct. The fact that SOMEONE ELSE may kave linked to that file, 
means that they are interested in the contents of the file now, not 
after I have frobbed it with my new version.

If the other person wanted always the latest version of my file, s/he 
should use a symbolic link instead. That's what symbolic links are for!

There is a perl analogy:

$a = "foo";
$b = $a;
$a = "bar";
print $b;   # prints "foo"

#===

$a = "foo";
*b = *a;
$a = "bar";
print $b;   # prints "bar"



Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

Save the Camel campaign [was bad nasty evil thread]

[Ivor Williams]

Dave Cross [mailto:[EMAIL PROTECTED]] wrote:

> ... But not only is Perl perceived as being "that web
> language", it's perceived as being "that out of date web language".

So. What are we going to do about it apart from slagging off Matt wright?

Picking up on some points from Dave's talk last night, I have a few
suggestions:

1. When educating colleagues, pick on the sysadmins. They are a good
receptive audience for the perl quality argument, especially as it is their
machines that potentially get cracked. Are there any good Perl books in this
line? 

2. Can we improve on perl's innate ability to detect bad code (use strict;
taint, perl -w etc.)?

3. How about an exam with a qualification. The Perl Community could define
its own without relying on vendors with alterior motives.

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: bad nasty evil thread

[Ivor Williams]

Chris Devers [mailto:[EMAIL PROTECTED]] wrote
 
> I think the worst case scenario is that Perl could end up being like Cobol
> is today -- old, ugly, and unloved, but a lot of working systems depend on
> it years after they were written, so the need to maintain it (and to have
> developers that understand it) will remain into the forseeable future,
> until such time that these systems can be reimplemented in something newer
> (which may not be worth the effort in the first place). 

Sad, but could happen. However the silver lining to that is that it becomes
a 
rare "sought-after skill" and rates go through the roof.

I think though that C is more of a model as its predecessor, in the mode of
"you really can do anything in C, including abusing the language and O/S 
completely. It's possible but optional to write portable, human readable and
supportable code in C". 


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

Implementing continuations in perl

[Ivor Williams]

Just wondering about some more magical pieces of the jigsaw.

Is there a way of getting a coderef for where you are inside caller?
Is there also a way of tricking the call stack into making it look as if you
have come from there?
Also, is there a way of generating extra stack frames including their
lexical variable space (i.e. reverse PadWalker)?

If the answer to all 3 is Yes, then I can see a way of doing continuations
in perl.

How I imagine continuations happening is as follows:

All function calls are LexWrapped into a piece of code which does the
following:

*   Create a stack object
*   Probe and save in stack object all the lexicals of caller using
PadWalker
*   Save the "return address" in the stack object
*   Put in a back link to the previously called stack object (may need
global _head and _tail variables in the stack package's namespace) 
*   go to the original function

Execution continues as normal, but catching each function call and creating
stack objects.

Now, some code invokes the create_continuation method, which, as it is a
call, has created a new stack object for its caller. The method returns said
stack object.

At some later point, the continuation is invoked. The call stack is now
faked to look as if all the intervening functions were called, with their
lexical variable spaces, and the code returns to this point.

Whaddya think? 

Ivor Williams
Sopra Mentor Consultant
LIFFE Core Systems Development
Extn: 2436   Mobile: 07752 234832



---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

English language [was call LEVEL oddness]

[Ivor Williams]

Jonathan Peterson [mailto:[EMAIL PROTECTED]] wrote:

> I hate to poke my head above the parapet, but just where did English get
> this 'difficult to learn' reputation? English is not tonal, not
> gendered, and not declined. It has one, small, 26 character alphabet
> with no accents.

Many people (foreigners) would argue that declension and gender make a
language easier to understand.
However, English can be wonderfully ambiguous. This makes it a rich language
for poetry and humour.

"Time flies like an arrow" can parsed in 5 different ways.

See also http://www.english-express.com/English-OnLine/FunFacts/Index.htm


Ivor.

"This is the kind of grammar, up with which I will not put"
- Churchill


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: Pub

[Ivor Williams]

This is the youngs pub on the south bank by the swaying bridge - no?

-Original Message-
From: Dave Cross [mailto:[EMAIL PROTECTED]]
Sent: 21 January 2002 13:39
To: [EMAIL PROTECTED]
Subject: Pub


Adam Turoff is in town and a few of us are meeting him for a drink or three
this evening.

We'll be in the Founder's Arms from 6:30pm. Feel free to join us.

Dave...

-- 

  Drugs are just bad m'kay



---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: I want it now ...

[Ivor Williams]

Merijn Broeren [mailto:[EMAIL PROTECTED]] wrote
> > Maybe you can ask the powers that be whether the ^=~ operator has a
> > name?
> 
> I've already named it last year, and you just have to follow suit:
>
> the dragon operator
>
> See, it's a point headed dragon that breathes fire to the right. 
> See old Smaug, defending his hoard. 

Like the dragon variation, Sicilian defence.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: Erm, Hello?

[Ivor Williams]

Jonathan Peterson [mailto:[EMAIL PROTECTED]] wrote

 
> > what? .. with  *4* spaces ?? come come mr McCarroll, we all know it is 2
> > spaces ... :)

> Spaces? SPACES?!!! It's called a tab, Tee Ay Bee. Code is indented with
> TABs, not spaces. Only bad bad people use spaces to indent.

... or bad editors or IDEs that zap tabs on the way in/out.

... or even crappier terminal emulators that don't understand tab stops.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

[OT] Exam Question

[Ivor Williams]

Thought this might amuse the Perl mongers. :->
I.

Subject: EXAM QUESTION

The following is an actual question given on a University of Washington
engineering mid term exam.The answer was so "profound" that the Professor
shared it with colleagues, which is why we now have the pleasure of enjoying
it as well.

Bonus Question: Is Hell exothermic (gives off heat) or Endothermic (absorbs
heat)?

Most of the students wrote Proofs of their beliefs using Boyle's Law, (gas
cools off when it expands and heats up when it is compressed) or some
variant. One student, however, wrote the following:

First, we need to know how the mass of Hell is changing in time. So we need
to know the rate that souls are moving into Hell and the rate they are
leaving. I think that we can safely assume that once a soul gets to Hell, it
will not leave. Therefore, no souls are leaving.

As for how many souls are entering Hell, let's look at the different
religions that exist in the world today. Some of these religions state that
if you are not a member of their religion, you will go to Hell. Since there
are more than one of these religions and since people do not belong to more
than one religion, we can project that all souls go to Hell. With birth and
death rates as they are, we can expect the number of souls in Hell to
increase exponentially.

Now, we look at the rate of change of the volume in Hell because Boyle's Law
states that in order for the temperature and pressure in Hell to stay the
same, the volume of Hell has to expand as souls are added. This gives two
possibilities:

1. If Hell is expanding at a slower rate than the rate at which souls enter
Hell, then the temperature and pressure in Hell will increase until all Hell
breaks loose.

2. Of course, if Hell is expanding at a rate faster than the increase of
souls in Hell, then the temperature and pressure will drop until Hell
freezes over.

So which is it?

If we accept the postulate given to me by Teresa Banyan during my Freshman
year, "...that it will be a cold day in Hell before I sleep with you." and
take into account the fact that I still have not succeeded in having sexual
relations with her, then, #2 cannot be true,
and thus I am sure that Hell is exothermic and will not freeze."

This student received the only A.



---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: Of meetings technical and social

[Ivor Williams]

Dave Cross [mailto:[EMAIL PROTECTED]] wrote

> Codix is explained at 

There is no link to this page!

> Cittie of Yorke is on the same road as Penderels. A bit to the east
(towards
> Chancery Lane tube) and on the other side of the road. We'll be in the
> cellar bar I think.

Any chance the website meetings page could be updated, adding entries for
Codix and the Cittie of York.

Who looks after the website? Is it still Leo?

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: Erm, Hello?

[Ivor Williams]

Chris Devers [mailto:[EMAIL PROTECTED]] wrote:
 
> ...a job, actually. I just accepted a contract as a Python coder! :)

Does Python have any advantages over Perl?

I have noticed that although Python is more obscure, it seems to have more
commercial take-up in certain places. My present client are using it for
test scripts.

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: Erm, Hello?

[Ivor Williams]

Chris Benson [mailto:[EMAIL PROTECTED]] wrote

> Been quiet all day tho' ... I was wondering if Penderel were poorly:
> obviously not.
 
As a matter of interest, whom do we tell if Penderel, hence london.pm.org
does go off the air?
Who has keys to buildings and/or root or sysadmin passwords?

This has happened a few times recently.

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: Threads availability and support

[Ivor Williams]

Nicholas Clark [mailto:[EMAIL PROTECTED]] wrote on  15 January 2002 21:42:

> I believe that ithreads were "experimental" in 5.6
> (in that there was no perl level interface to make a thread, but the
> underlying ithreads mechanism is what is used to fake fork() on Windows)

Fake fork() sounds useful for getting round the nasty NT nested scripts
problem (which a couple of years ago stopped me installing Perl directly on
an NT box - I used a Win95 workstation to do the install, putting the files
on a shared drive).

Does this enable perl system() and backticks to work properly on NT?

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: Year 2K2 problem

[Ivor Williams]

* at 15/01 11:58 + robin szemeti said:
> 
> A 7-Page Interpretive Report created for your unique type. Simple to take 
> 93-question test online in the privacy of your home or office. Results 
> typically within 48 hours. Report in Adobe .pdf format. 

Anybody curious enough might like to try the "Enneagram test" based on 9
personality types.

Try http://www.duniho.com/fergus/enneagram/test/ for a test with 28
questions in a proper web app that gives you instant results.

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

Threads availability and support

[Ivor Williams]

Hi there,

Anyone know what stage the threads work is at?

CPAN lists perl threads as a "big project", and I gather that some
_experimental_ stuff went into 5.6.1.

Anyone know what date and/or release this is targetted towards?

Ivor Williams
Sopra Mentor Consultant
LIFFE Core Systems Development
Extn: 2436   Mobile: 07752 234832



---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: (void) OSCON

[Ivor Williams]

-Original Message-
From: Nicholas Clark [mailto:[EMAIL PROTECTED]]
Sent: 14 January 2002 16:46
To: [EMAIL PROTECTED]
Subject: Re: (void) OSCON


> On Sat, Jan 12, 2002 at 12:11:11PM +, robin szemeti wrote:
> > On Friday 11 January 2002 19:14, David H. Adler wrote:

> > > Hey!  The east coast is *still* waiting, you know... :-)
> 
> > I'd like to second DHA's sentiments .. somewhere on the east coast would
be 
> > an ideal location for OSCON. 
> > 
> > Skegness would be nice, or Bridlington.  
> > 
> > Grimsby has possibilities too  but tends to smell of fish.

> Lowestoft! You can't get much more east than that without getting wet

Looks like we've got a Mornington Crescent super extended version (ease
coast variant) rolling.

Great Yarmouth.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: Just thinking (like Greg)

[Ivor Williams]

-Original Message-
From: Nicholas Clark [mailto:[EMAIL PROTECTED]]
Sent: 10 January 2002 16:59
To: [EMAIL PROTECTED]
Subject: Just thinking (like Greg)

[snip]

> So what's stopping any and every enterprising steganographer from getting
> one of these files, one that you don't own the licence to play, and
> substituting other encrypted data in place of the music data? You'd need
to
> have all the substitute data (including any file header it might need)
> encrypted in some way so that it all looks like random binary data, then
> chop out exactly that number of bytes from the music file and splice in
your
> data.

Normally the medium includes a signature encrypted with the vendor's private
key, which can be verified by Jo Public using the vendor's public key.

Usually the body of the data is checksummed, and the checksum is built into
the signature. As such, if the body data has been tampered with, this does
not match the signature.

People inside the record company who have access to the private key would be
in a position to generate a new signature. 

> At which point, you have a file that appears to be a music file on casual
> inspection (because it has the header).  But you don't own the licence to
> play it, so if you give it to the player it will recognise it as a music
> file, and ask if you want to buy it.  And only when you hand over money
can
> you actually get the player to decrypt it, at which point you can tell
that
> it's not music (assuming your ears and speakers don't get fried by digital
> noise)

The player may well be able to detect an invalid checksum without you having
a valid licence. If the signature is encrypted with the vendor's public key,
this would be the case. The software gives you an "invalid medium" window
straight away before checking your key.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: Co-lo, was Re: London.pm List Weekly Summary 2001-12-31

[Ivor Williams]

Tom Hukins [mailto:[EMAIL PROTECTED]] wrote:

> That's fine if you're working with a typical filesystem, but what if
> you're working with database tables?  I'd normally dump the tables to
> a file and backup the dump.  However, I guess you'd end up rsyncing
> the whole database if you did this.  I guess you could archive
> transaction logs, but for a frequently changing database the logs will
> get quite large.

> Any alternatives?

The database vendor will provide tools to sync and backup databases.
Depending on the level of sophistication required, this can be done by
offlining a mirror copy of the database - needed for 24x7 availability if
this is a requirement.

The database usually also provides open transaction recovery journalling,
and after image journalling. The former permits the database to be rolled
back to a consistent state in the event of a system failure, and the latter
allows replay of transactions from a sync point (a backup snapshot).

Don't know if this helps. I suggest looking on the DB vendor's website for
more technical info.

Ivor.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

Patterns in Perl : decorator model

[Ivor Williams]

Nigel

I have been looking at the "decorator" model, and I would do it differently.

Instead of using inheritance, I would use AUTOLOAD to implement
quasi-inheritance. The relation is "Emulates" rather than "ISA". Does your
design model have to be language independent, or can we use perl specific
stuff like AUTOLOAD?

Attached is an example (factor.pl, bignum.pm, strnum.pm) of this
quasi-inheritance. As an exercise and alternative to Math::BigInt, I wanted
a quickly implementable representation for arbitrarily large positive
integers. I knocked this up first as strnum (objects are digit strings), but
wanted something more fast and juicy: bignum (objects are arrays of ints),
which required more human programming effort.

Having strnum available first meant that I could prototype all the methods
needed for a bignum object, and code them one by one, while having a working
application at all stages. This is called the evo methodology. bignum has an
AUTOLOAD method which makes a strnum object, calls the corresponding method
and unwraps the return value if this is an object reference.

There are a few assumptions here, but it would be possible to do something
more generic, particularly for methods which pass and return different
object of different classes, but you get the idea.

Coming back to your "decorator" model. I suggest implementing your debug
class as follows. This is generic and can be used with any classes, and
copes with arbitrarily complex arguments and return values.

#---

# calling code (all you need to do for each class)

package foo::Debug;
use MethodTracer;
@ISA = qw(MethodTracer);

   .
   .

$fred = new foo:Debug(@blah);   # was new foo(@blah)
foo:Debug->debug(1) # Turn on debugging of foo objects

#---

MethodTracer.pm

# This is called via inheritance to construct an object
# of class blah::blah::Debug. This assumes that the "real"
# class is called blah::blah

package MethodTracer;
use Data::Dumper;

sub new
{
my ($pkg) = @_;

$pkg =~ /\:\:\w+$/;
my $underly = $`;

$underly::new(@_);  # relies on constructor respecting package
param
}

sub debug   # get and set debug at a class level
{
my $pkg = shift;

@_ ? (${$pkg}::_debug = shift) : ${$pkg}::_debug;
}

sub DESTROY
{
1;  # to stop AUTOLOAD being called
}

sub AUTOLOAD
{
my ($self,@args) = @_;

my $method = $AUTOLOAD;
$method =~ s/\:\:Debug//;
my $pkg = ref $self;

print "called $method. Parameters:",Dumper(@args) if
${$pkg}::_debug;

if (wantarray)
{
my @rv = &$method(@_);
print "Return from $method. List context,
values:",Dumper(@rv)
if ${$pkg}::_debug;
@rv;
} else
my $rv = &$method(@_);
print "Return from $method. Scalar context,
returned:",Dumper($rv)
if ${$pkg}::_debug;
$rv;
}
}

1;

#--

I intend to go to tonight's meeting.

C U There.

Ivor.

-Original Message-
From: Nigel Wetters [mailto:[EMAIL PROTECTED]]
Sent: 08 January 2002 17:12
To: [EMAIL PROTECTED]
Subject: Patterns in Perl reminder


Study group starts tomorrow. For details, see 
http://london.pm.org/pipermail/london.pm/2002-January/007171.html



_
Chat with friends online, try MSN Messenger: http://messenger.msn.com




---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---



bignum.pm
Description: Binary data


factor.pl
Description: Binary data


strnum.pm
Description: Binary data

RE: Compiling strings into code refs

[Ivor Williams]

of course eval "sub..." will do it.

Dohh!

Sorry guys.


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: Compiling strings into code refs

[Ivor Williams]

Richard Clamp <[EMAIL PROTECTED]> wrote:
> eval "sub { $some_code }"

This is missing the point, as eval runs it now, I want to run it later.



my $foo = new Object::foo;

my $grunge = sub {
print "Performing grunge";
}

$foo->method({
action=>$grunge});

This is the normal hardcoded way of doing a callback, the message
"performing grunge" only appears when the foo class decides to do so.

What I want is for the code for $grunge to be read from a persistent store
or a .pl file and passed to my foo object. Below is what it would be if the
compile verb existed.


my $foo = new Object::foo;

my @input = ();

my $grunge = compile(join '',@input);

# $grunge is now a CODE ref.

$foo->method({
action=>$grunge});


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

Compiling strings into code refs

[Ivor Williams]

I've got an interesting issue with code refs. These are often used for
callbacks, many standard methods expect them as code refs, often retaining
internal values as closures: there are many examples in IO::Socket, Tk etc.

I want to be able to generate callbacks on the fly from code which is in
strings.

Normally, with on-the-fly code, you just eval it to get what you want. Here,
I don't want to run it, but to generate an anonymous code ref.

What would be handy is a compile verb a bit like eval. This would compile it
in the inline context of where compile was called (allowing closures, etc.),
and return a code ref.

Any thoughts?

Ivor Williams
Sopra Mentor Consultant
LIFFE Core Systems Development
Tel:+44 (0)20 7379 2436   
Mobile: +44(0)7752 234832



---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---

RE: attention risk players

[Ivor Williams]

Folks!

As a relative lon.pm newcomer, how do I get to this mythical
irc.rhizomatic.net?

I have hear wonderful things about it, the FAQ page on london.pm.org gives a
hint but no instructions. 

Anybody help?

Ivor.

-Original Message-
From: Greg McCarroll [mailto:[EMAIL PROTECTED]]
Sent: 07 January 2002 17:24
To: London.pm
Subject: attention risk players



i just had a trial run with Dave H, of xfrisk (i'm sure it used to
just be called xrisk years ago) and it isn't half bad

so if anyone fancies a game at around say 7.30pm (GMT), why not
download/compile the program, its at

 http://morphy.iki.fi/xfrisk/

and pop on irc.rhizomatic.net, #london.pm around then and we can enjoy
world domination

greg

-- 
Greg McCarroll http://217.34.97.146/~gem/


---
The information contained in this e-mail is confidential and solely 
for the intended addressee(s). Unauthorised reproduction, disclosure, 
modification, and/or distribution of this email may be unlawful. If you 
have received this email in error, please notify the sender immediately 
and delete it from your system. The views expressed in this message 
do not necessarily reflect those of LIFFE (Holdings) Plc or any of its 
subsidiary companies.
---