Re: Perl 6
* Nathan Torkington ([EMAIL PROTECTED]) wrote: > > We're going to use RFCs for future additions to Perl, we just need to > find some good filters that will prevent them from consuming > everyone's time. how about adding a field on the RFC template such as, ``forum initially discussed in'' - then encourage people that before they submit an RFC they should of had it discussed in an open forum such as P5P, #perl or even their local (or favourite) Perl monger list. this has the ( devious ) side effect of gently spamming some people about issues for Perl 6 hence getting more pairs of eyeballs on the development and design of the beastie Greg -- Greg McCarroll http://www.mccarroll.uklinux.net
Re: Hiring
Steve Mynott <[EMAIL PROTECTED]> writes: > David Hodgkinson <[EMAIL PROTECTED]> writes: > > > Simon Wistow <[EMAIL PROTECTED]> writes: > > > > > I think > > > > Ta, > > How did he do then? You're a bad boy :-) -- Dave Hodgkinson, http://www.hodgkinson.org Editor-in-chief, The Highway Star http://www.deep-purple.com Apache, mod_perl, MySQL, Sybase hired gun for, well, hire -
Re: Hiring
David Hodgkinson <[EMAIL PROTECTED]> writes: > Simon Wistow <[EMAIL PROTECTED]> writes: > > > I think > > Ta, How did he do then? :-> -- 1024/D9C69DF9 steve mynott [EMAIL PROTECTED] microsoft: where do you want to go today? linux: where do you want to go tomorrow? bsd: are you guys coming or what?
Re: Hiring
Simon Wistow <[EMAIL PROTECTED]> writes: > I think Ta, -- Dave Hodgkinson, http://www.hodgkinson.org Editor-in-chief, The Highway Star http://www.deep-purple.com Apache, mod_perl, MySQL, Sybase hired gun for, well, hire -
Re: Hiring
David Hodgkinson wrote: > > Anyone got Richard Clamp's mobile no? He's due here now... 07720 298487 I think
Re: Perl 6
Piers Cawley writes: > > As Piers said, we are blocked on Larry. We're working on some > > interpreter design now, but some language issues really need to be > > nailed down before we know what we're going to be writing. > > Any idea how long we're going to stay blocked? None whatsoever. Many phone conversations with Larry have lead me to conclude that he'll complete it in his own time or not at all. The trick is deciding which it is :-) Nat
Re: Hiring
Anyone got Richard Clamp's mobile no? He's due here now... -- Dave Hodgkinson, http://www.hodgkinson.org Editor-in-chief, The Highway Star http://www.deep-purple.com Apache, mod_perl, MySQL, Sybase hired gun for, well, hire -
Re: Perl 6
Andy Wardley writes: > It might also draw the all-talk-and-no-trousers crowd away from the > serious perl6 development process. It would give us, er, I mean *them* > somewhere to rant without bothering too many people doing the real work > on crafting Perl 6. Yes, and no. The problem is that it's easy for the dumbasses to swamp the clueful by posting an infinite number of idiot RFCs. We're going to use RFCs for future additions to Perl, we just need to find some good filters that will prevent them from consuming everyone's time. Nat
Re: Perl 6
On Jan 9, 8:24am, Nathan Torkington wrote: > As Piers said, we are blocked on Larry. We're working on some > interpreter design now, but some language issues really need to be > nailed down before we know what we're going to be writing. Are there any plans to keep the RFC process going in the future? It occurs to me that people are always coming up with half-assed ideas about the next greatest thing that should go in Perl and an RFC process would allow them to air them for peer review. It might also draw the all-talk-and-no-trousers crowd away from the serious perl6 development process. It would give us, er, I mean *them* somewhere to rant without bothering too many people doing the real work on crafting Perl 6. A -- Andy Wardley <[EMAIL PROTECTED]> Signature regenerating. Please remain seated. <[EMAIL PROTECTED]> For a good time: http://www.kfs.org/~abw/
Re: Perl commandments
Piers Cawley sent the following bits through the ether: > This time. The discussion has been back and forth on various lists, > usually with benchmarks. Thou shalt optimise for programmer time unless absolutely necessary, when thou shalt Benchmark and quoth both the benchmark and the results. Leon -- Leon Brocard.http://www.astray.com/ yapc::Europehttp://yapc.org/Europe/ ... All new improved Brocard, now with Template Toolkit!
Re: Perl 6
Nathan Torkington <[EMAIL PROTECTED]> writes: > David Cantrell writes: > > From what I can tell, there ain't a lot happening. > > As Piers said, we are blocked on Larry. We're working on some > interpreter design now, but some language issues really need to be > nailed down before we know what we're going to be writing. Any idea how long we're going to stay blocked? -- Piers
Re: Perl commandments
Leon Brocard <[EMAIL PROTECTED]> writes: > Greg McCarroll sent the following bits through the ether: > > > in my original rule it was all to do with good programming style, not > > eeking out every bit of performance, my reply was actually that i > > thought dave choose a very grey area in terms of programming style > > Indeed. And someone mentioned another bit of code being faster, > *without benchmarking it*. This time. The discussion has been back and forth on various lists, usually with benchmarks. -- Piers
Re: Perl 6
David Cantrell writes: > From what I can tell, there ain't a lot happening. As Piers said, we are blocked on Larry. We're working on some interpreter design now, but some language issues really need to be nailed down before we know what we're going to be writing. Nat
RE: Much Coolness XML Wise.
> > How does the output compare to XML::Simple::XMLout() ? > I found XML::Simple::XMLout() to be deeply yucky, as it kind of randomly chooses to go for sub elements or attributes based on phases of the moon*. Also, it will happily output invalid XML if you have invalid data in your datastructure. Also it dies if given recursive structures. And it doesn't do binary data. I've found XML::Writer to be OK, although it doesn't do an 'export this data structure as XML' thing. You could use it write your own though. Or use XML::Dumper, I guess. *Something like all single nested items are made into attributes, or, if you prefer, nothing is made into attributes. Ick.
Re: Perl commandments
Greg McCarroll sent the following bits through the ether: > in my original rule it was all to do with good programming style, not > eeking out every bit of performance, my reply was actually that i > thought dave choose a very grey area in terms of programming style Indeed. And someone mentioned another bit of code being faster, *without benchmarking it*. Leon -- Leon Brocard.http://www.astray.com/ yapc::Europehttp://yapc.org/Europe/ ... All new improved Brocard, now with Template Toolkit!
Re: Perl commandments
* Leon Brocard ([EMAIL PROTECTED]) wrote: > Greg McCarroll sent the following bits through the ether: > > > nope daves was a bad example > > Benchmark.pm - hard facts are better than guesses (hmm, I could do a > talk on this...) > in my original rule it was all to do with good programming style, not eeking out every bit of performance, my reply was actually that i thought dave choose a very grey area in terms of programming style -- Greg McCarroll http://www.mccarroll.uklinux.net
Re: Perl commandments
Greg McCarroll sent the following bits through the ether: > nope daves was a bad example Benchmark.pm - hard facts are better than guesses (hmm, I could do a talk on this...) Leon -- Leon Brocard.http://www.astray.com/ yapc::Europehttp://yapc.org/Europe/ ... All new improved Brocard, now with Template Toolkit!
Re: Perl commandments
* Piers Cawley ([EMAIL PROTECTED]) wrote:
> Greg McCarroll <[EMAIL PROTECTED]> writes:
>
> > * Piers Cawley ([EMAIL PROTECTED]) wrote:
> > > David Cantrell <[EMAIL PROTECTED]> writes:
> > > > On Tue, Jan 09, 2001 at 11:25:18AM +, Greg McCarroll wrote:
> > > > > 6.) regular expressions are not the only way to code, length and
> > > > > substr are in the language for a reason
> > > >
> > > > Also index. These two snippets are equivalent:
> > > > if($foo=~/foo/) { ... }
> > > > if(index($foo, 'foo')!=-1) { ... }
> > > > I always want to do just plain if(index(...)) though.
> > >
> > > ISTR that (for weird reasons), the regex version of that is faster.
> >
> > but of course we don't (shouldn't) program perl for program time
> > optimization but for programmer time optimization ;-)
>
> So the regex wins on all counts then. Faster, clearer, shorter, easier
> to maintain. The list goes on.
>
nope daves was a bad example
--
Greg McCarroll http://www.mccarroll.uklinux.net
Re: Perl commandments
Greg McCarroll <[EMAIL PROTECTED]> writes:
> * Piers Cawley ([EMAIL PROTECTED]) wrote:
> > David Cantrell <[EMAIL PROTECTED]> writes:
> > > On Tue, Jan 09, 2001 at 11:25:18AM +, Greg McCarroll wrote:
> > > > 6.) regular expressions are not the only way to code, length and
> > > > substr are in the language for a reason
> > >
> > > Also index. These two snippets are equivalent:
> > > if($foo=~/foo/) { ... }
> > > if(index($foo, 'foo')!=-1) { ... }
> > > I always want to do just plain if(index(...)) though.
> >
> > ISTR that (for weird reasons), the regex version of that is faster.
>
> but of course we don't (shouldn't) program perl for program time
> optimization but for programmer time optimization ;-)
So the regex wins on all counts then. Faster, clearer, shorter, easier
to maintain. The list goes on.
--
Piers
Re: Perl 6
Greg McCarroll sent the following bits through the ether: > has anyone been keeping up with the Perl 6 mailing lists and the Perl 6 > project in general? i'd like to have a look into some of the proposals > and i've sort of lost touch with it all. The RFC's are available at http://dev.perl.org/ and a wonderful way to try out the ideas is to implement them in Perl 5. lathos's stuff is getting way scary though: http://search.cpan.org/search?dist=B-Generate Leon -- Leon Brocard.http://www.astray.com/ yapc::Europehttp://yapc.org/Europe/ ... All new improved Brocard, now with Template Toolkit!
Re: Perl commandments
* Piers Cawley ([EMAIL PROTECTED]) wrote:
> David Cantrell <[EMAIL PROTECTED]> writes:
> > On Tue, Jan 09, 2001 at 11:25:18AM +, Greg McCarroll wrote:
> > > 6.) regular expressions are not the only way to code, length and
> > > substr are in the language for a reason
> >
> > Also index. These two snippets are equivalent:
> > if($foo=~/foo/) { ... }
> > if(index($foo, 'foo')!=-1) { ... }
> > I always want to do just plain if(index(...)) though.
>
> ISTR that (for weird reasons), the regex version of that is faster.
>
but of course we don't (shouldn't) program perl for program time optimization
but for programmer time optimization ;-)
--
Greg McCarroll http://www.mccarroll.uklinux.net
Re: Perl commandments
David Cantrell <[EMAIL PROTECTED]> writes:
> On Tue, Jan 09, 2001 at 11:25:18AM +, Greg McCarroll wrote:
> > 6.) regular expressions are not the only way to code, length and
> > substr are in the language for a reason
>
> Also index. These two snippets are equivalent:
> if($foo=~/foo/) { ... }
> if(index($foo, 'foo')!=-1) { ... }
> I always want to do just plain if(index(...)) though.
ISTR that (for weird reasons), the regex version of that is faster.
Re: Perl 6
On Tue, Jan 09, 2001 at 11:57:46AM +, Greg McCarroll wrote: > has anyone been keeping up with the Perl 6 mailing lists and the Perl 6 > project in general? i'd like to have a look into some of the proposals > and i've sort of lost touch with it all. >From what I can tell, there ain't a lot happening. -- David Cantrell | [EMAIL PROTECTED] | http://www.cantrell.org.uk/david Any technology distinguishable from magic is insufficiently advanced
Re: Much Coolness XML Wise.
On Mon, 8 Jan 2001, Dean S Wilson wrote: >I've been involved in the debates on this list about Perl and Java and >I'd thought I'd say something nice for once. > >I've just spent three hours playing with XML::Generator and md5 >checksums, I was having problems building hash's of hash's with the >module (Seems to only add the last one in a foreach loop to the XML. >Gotta be me :)) and then with a quick glance at cpan I found >XML::Dumper. I don't know if I'm the only one here who didn't know >about this module but its Data::Dumper that outputs well formed xml. >Two minutes to read the docs and make a quick test script and tada! >Well formed xml files of my structures ready to be used. In under five >additional lines of code. How does the output compare to XML::Simple::XMLout() ? Cheers, Mike -- Mike Wyer <[EMAIL PROTECTED]> || "Woof?" http://www.doc.ic.ac.uk/~mw || Gaspode the Wonder Dog Work: +44 020 7594 8440|| Mobile: +44 07879 697119|| ICQ: 43922064
Re: Perl commandments
On Tue, Jan 09, 2001 at 11:25:18AM +, Greg McCarroll wrote:
> * Piers Cawley ([EMAIL PROTECTED]) wrote:
>
> > Thou shalt ensure that everything knows as little as possible about
> >anything else. (There's *got* to be a way to condense that...)
Thou shalt encapsulate?
> ok heres my 10 rules
>
> 3.) use warnings and use strict, even if you have to sprinkle no
> strict and use strict block throughout the code
And 3b) always document why you are using no strict.
> 6.) regular expressions are not the only way to code, length and
> substr are in the language for a reason
Also index. These two snippets are equivalent:
if($foo=~/foo/) { ... }
if(index($foo, 'foo')!=-1) { ... }
I always want to do just plain if(index(...)) though.
--
David Cantrell | [EMAIL PROTECTED] | http://www.cantrell.org.uk/david
Any technology distinguishable from magic is insufficiently advanced
Re: Perl 6
Greg McCarroll <[EMAIL PROTECTED]> writes: > has anyone been keeping up with the Perl 6 mailing lists and the Perl 6 > project in general? i'd like to have a look into some of the proposals > and i've sort of lost touch with it all. It's kind of blocked on Larry at the moment. There's some stuff going on in 'internals' about possible approaches given some of his hints, and there's quite a bit of licensing issue discussions and worries about library dependencies as well. -- Piers
Perl 6
has anyone been keeping up with the Perl 6 mailing lists and the Perl 6 project in general? i'd like to have a look into some of the proposals and i've sort of lost touch with it all. -- Greg McCarroll http://www.mccarroll.uklinux.net
Re: Perl commandments
* Piers Cawley ([EMAIL PROTECTED]) wrote: > David Hodgkinson <[EMAIL PROTECTED]> writes: > > > Was it here that we discussed the Perl commandments? You remember, > > things like "thou shalt use strict and -w" and so on... > > I don't think we were. But: > > Thou shalt not write the same thing twice. > Thou shalt not use map in a void context. > Thou shalt ensure that everything knows as little as possible about >anything else. (There's *got* to be a way to condense that...) > Thou shalt use DWIMmery where appropriate > Thou shalt not write C/COBOL/Basic/FORTRAN in Perl. Unless that's the >point of the joke. > ok heres my 10 rules 1.) There is a time for perl golf and its not writing production code 2.) Before writing a module check search.cpan.org, and if you don't find that module check it again just to be sure. 3.) use warnings and use strict, even if you have to sprinkle no strict and use strict block throughout the code 4.) one package per file with an appropriate name and directory structure 5.) if you are writing a large module (set of modules) write them as if they were going onto CPAN, documented with a Makefile.PL, test files, the whole works 6.) regular expressions are not the only way to code, length and substr are in the language for a reason 7.) program extra defensively in perl, you don't have types so you have to be extra careful some idiot is not using your module or routine in the wrong way 8.) create a Programmers::.pm module for yourself and the rest of your team, use it as a place to put those little useful routines that dont belong anywhere yet, then periodically have an afternoon of everyone comparing and contrasting there areas to see what can come out as new modules to be used generally - this prevents you ending up with a new module everytime someone wants a routine that does not fit elsewhere 9.) program to 75% (or less) of your ability, other people need to understand the code and even if they are of the same ability they may not want to understand the logic of the code while reading code at the top of their understanding level 10.) lastly, achieve something everyday in a programming project, get some bit of functionality out, perl is a powerful language and any project where you cant make a difference in an hour is either not well defined or not well analysed into achievable sub projects and modules /me gets off the soapbox -- Greg McCarroll http://www.mccarroll.uklinux.net
Re: Perl commandments
David Hodgkinson <[EMAIL PROTECTED]> writes: > Was it here that we discussed the Perl commandments? You remember, > things like "thou shalt use strict and -w" and so on... I don't think we were. But: Thou shalt not write the same thing twice. Thou shalt not use map in a void context. Thou shalt ensure that everything knows as little as possible about anything else. (There's *got* to be a way to condense that...) Thou shalt use DWIMmery where appropriate Thou shalt not write C/COBOL/Basic/FORTRAN in Perl. Unless that's the point of the joke. -- Piers
Perl commandments
Was it here that we discussed the Perl commandments? You remember, things like "thou shalt use strict and -w" and so on... -- Dave Hodgkinson, http://www.hodgkinson.org Editor-in-chief, The Highway Star http://www.deep-purple.com Apache, mod_perl, MySQL, Sybase hired gun for, well, hire -
Re: Hiring
David Hodgkinson <[EMAIL PROTECTED]> writes: > Picture painted? Man, I REALLY need to check Reply-To's...! -- Dave Hodgkinson, http://www.hodgkinson.org Editor-in-chief, The Highway Star http://www.deep-purple.com Apache, mod_perl, MySQL, Sybase hired gun for, well, hire -
Re: Hiring
Piers Cawley <[EMAIL PROTECTED]> writes: > David Hodgkinson <[EMAIL PROTECTED]> writes: > > > Given another spate of layoffs last week, I'd just like to point out > > that Reading Room are hiring good Perl, Linux/Solaris/Anything, > > MySQL/Oracle/DBS/Anything hackers. > > Where is Reading Room? And are you talking permies or contractors. And > if only permies, what can you do to tempt a contractor? Halfway up Dean St and the bottom of Wardour. Permies. Probably nothing, some cute graphic designers maybe? The technology is erm, pragmatic. Cheap-ass client sites get boshed out in Cold Fusion, the big sites whatever the 500lb gorilla says. This has been Websphere (boo!), Silverstream (boo!), Interwoven (boo!), Autonomy (yay!), Oracle (boo!), MySQL (yay-ish) and a few others. I'd like to see a couple of "real programmers" and a sysadmin nazi in there before things get really out of hand. The programming slaves are pretty bright but largely unled. Picture painted? Cheers, Dave -- Dave Hodgkinson, http://www.hodgkinson.org Editor-in-chief, The Highway Star http://www.deep-purple.com Apache, mod_perl, MySQL, Sybase hired gun for, well, hire -
Re: Much Coolness XML Wise.
* Simon Wistow ([EMAIL PROTECTED]) wrote: > > > There are a few crypto modules on CPAN, although they all use C for the > > hard work. Whether they're available in PPMs for Activestate or not, > > I don't know. > > I'm doing a pure WML script implementation of MD5 at the moment that'd > be dead easy to port to Perl. Especially since Perl has things like > arrays which I always find useful for this sort of thing, grumble, > grumble, Wap rhymes with cr MIND THE WAP -- Greg McCarroll http://www.mccarroll.uklinux.net
Re: Much Coolness XML Wise.
> There are a few crypto modules on CPAN, although they all use C for the > hard work. Whether they're available in PPMs for Activestate or not, > I don't know. I'm doing a pure WML script implementation of MD5 at the moment that'd be dead easy to port to Perl. Especially since Perl has things like arrays which I always find useful for this sort of thing, grumble, grumble, Wap rhymes with cr
Re: Hiring
David Hodgkinson <[EMAIL PROTECTED]> writes: > Given another spate of layoffs last week, I'd just like to point out > that Reading Room are hiring good Perl, Linux/Solaris/Anything, > MySQL/Oracle/DBS/Anything hackers. Where is Reading Room? And are you talking permies or contractors. And if only permies, what can you do to tempt a contractor? -- Piers
