From grant at mclean.net.nz Mon May 1 20:41:12 2006 From: grant at mclean.net.nz (Grant McLean) Date: Tue, 02 May 2006 15:41:12 +1200 Subject: [Wellington-pm] Meeting Next Tuesday Message-ID: <1146541272.22156.8.camel@localhost.localdomain> Hi Mongers The next meeting will be on Tuesday May the 9th. The following two speakers are lined up: * Michael Robinson - Plucene * Srdjan Jankovic - Using CPAN for Fun and Profit See you then. Grant From michael at diaspora.gen.nz Mon May 1 21:43:11 2006 From: michael at diaspora.gen.nz (michael at diaspora.gen.nz) Date: Tue, 02 May 2006 16:43:11 +1200 Subject: [Wellington-pm] CPAN and local installs (was Re: Meeting Next Tuesday) In-Reply-To: Your message of "Tue, 02 May 2006 15:41:12 +1200." <1146541272.22156.8.camel@localhost.localdomain> Message-ID: Grant McLean writes: >The next meeting will be on Tuesday May the 9th. The following two >speakers are lined up: > > * Michael Robinson - Plucene Awww, shite, that means I have a presentation to write between now and then :) > * Srdjan Jankovic - Using CPAN for Fun and Profit Can we include a section on how setting up CPAN with: 'makepl_arg' => q[PREFIX=~/] (on Linux) promptly breaks when you try and install a Module::Build module, and thereby drives one slightly bonkers? The above gave me a nice setup where everything went into: $HOME/lib/perl/5.8.7 and $HOME/share/perl/5.8.7 but the best I seem to be able to do with Module::Build is: $HOME/lib/i486-linux-gnu-thread-multi/ and $HOME/lib/i486-linux-gnu-thread-multi/auto/ Which is slightly frustrating. -- michael. From grant at mclean.net.nz Mon May 1 21:52:04 2006 From: grant at mclean.net.nz (Grant McLean) Date: Tue, 02 May 2006 16:52:04 +1200 Subject: [Wellington-pm] CPAN and local installs (was Re: Meeting Next Tuesday) In-Reply-To: References: Message-ID: <1146545524.22156.10.camel@localhost.localdomain> On Tue, 2006-05-02 at 16:43 +1200, michael at diaspora.gen.nz wrote: > Can we include a section on how setting up CPAN with: > > 'makepl_arg' => q[PREFIX=~/] > > (on Linux) promptly breaks when you try and install a Module::Build > module, and thereby drives one slightly bonkers? Is it still broken with the new Module::Build released this week? Grant From sam at vilain.net Mon May 1 21:58:14 2006 From: sam at vilain.net (Sam Vilain) Date: Tue, 02 May 2006 16:58:14 +1200 Subject: [Wellington-pm] CPAN and local installs (was Re: Meeting Next Tuesday) In-Reply-To: References: Message-ID: <4456E6E6.7010003@vilain.net> michael at diaspora.gen.nz wrote: >Can we include a section on how setting up CPAN with: > > 'makepl_arg' => q[PREFIX=~/] > >(on Linux) promptly breaks when you try and install a Module::Build >module, and thereby drives one slightly bonkers? > CPANPLUS is supposedly better. It deals with this Module::Build backwards incompatibility brain damage. On a test run of installing all the dependencies for MojoMojo, a Catalyst.pm wiki, it only barfed "a few" times, compared to CPAN.pm, which barfed "continuously". So, yes, don't use Module::Build or your users will hate you. Module::Install seems to be the way people are going these days. It also does nice things like bundle your dependencies (including itself!) when you 'make dist'. Sam. From michael at diaspora.gen.nz Mon May 1 21:58:32 2006 From: michael at diaspora.gen.nz (michael at diaspora.gen.nz) Date: Tue, 02 May 2006 16:58:32 +1200 Subject: [Wellington-pm] CPAN and local installs (was Re: Meeting Next Tuesday) In-Reply-To: Your message of "Tue, 02 May 2006 16:52:04 +1200." <1146545524.22156.10.camel@localhost.localdomain> Message-ID: Grant McLean writes: >On Tue, 2006-05-02 at 16:43 +1200, michael at diaspora.gen.nz wrote: >> Can we include a section on how setting up CPAN with: >> >> 'makepl_arg' => q[PREFIX=~/] >> >> (on Linux) promptly breaks when you try and install a Module::Build >> module, and thereby drives one slightly bonkers? > >Is it still broken with the new Module::Build released this week? I'm using the Ubuntu package, which is version 0.26-1. I'll give it a go with the latest from CPAN. The problem with the 0.26 version is that I get an error message denying all knowledge of the PREFIX command line argument from Module::Build; one is meant to use install_base and installdirs={site,core,vendor} instead. Which is understandable, really, just a pain in the ass; because if I'm building a tree of modules through the CPAN shell, the ones with Module::Build need different makepl_arg values from "normal" Extutils::MakeMaker ones. -- michael. From srdjan at catalyst.net.nz Mon May 1 22:08:11 2006 From: srdjan at catalyst.net.nz (Srdjan) Date: Tue, 02 May 2006 17:08:11 +1200 Subject: [Wellington-pm] CPAN and local installs (was Re: Meeting Next Tuesday) In-Reply-To: References: Message-ID: <4456E93B.10902@catalyst.net.nz> > Can we include a section on how setting up CPAN with: > > 'makepl_arg' => q[PREFIX=~/] > > (on Linux) promptly breaks when you try and install a Module::Build > module, and thereby drives one slightly bonkers? Absolutely not. I'm not stealing bread from the "Reporting perl bugs" presenter. But we can include "Going bonkers with perl for Fun and Profit" talk. S. From enkidu at cliffp.com Wed May 3 03:45:26 2006 From: enkidu at cliffp.com (Cliff Pratt) Date: Wed, 03 May 2006 22:45:26 +1200 Subject: [Wellington-pm] 'my' and 'our' Message-ID: <445889C6.6010007@cliffp.com> I'm confused about 'my' and 'our'. Can anyone explain the difference? Preferably in reasonably short words. I've read the section on 'our' in "Chapter 29: Functions" in the Camel book, but I'm not a lot wiser... When would you use 'my' and when would you use 'our'? I've used 'my' a squillion times before, but something tells me I should find out more about 'our', especially as I'm trying to write object-type stuff. I create a variable in a package/class which is not in any method and use 'my'. I can use it in any package/class method. Is this the correct way of doing it? Cheers, Cliff -- http://barzoomian.blogspot.com From jarich at perltraining.com.au Wed May 3 03:57:05 2006 From: jarich at perltraining.com.au (Jacinta Richardson) Date: Wed, 03 May 2006 20:57:05 +1000 Subject: [Wellington-pm] 'my' and 'our' In-Reply-To: <445889C6.6010007@cliffp.com> References: <445889C6.6010007@cliffp.com> Message-ID: <44588C81.2010609@perltraining.com.au> my is for lexical variables. That is variables that exist and can only be seen in their current scope. A variable's scope is defined as the smallest enclosing block (including file). our is for package variables (think globals). These are variables that exist and can be accessed outside of the scope in which they were declared, for example in different files. Consider this example: ---------------------- package Foo; my $bar; our $baz; { # open new scope my $gatx; print $gatx; # works } print $gatx; # doesn't work (error) ---------------------- # main program use Foo; print $Foo::baz; # works print $Foo::bar; # error, $bar can't be seen out here. ----------------------- our replaces the need for "use vars" from code prior to 5.6. It's a tiny bit more complicated than this, but not so much that it matters. If this doesn't help, please feel free to ask more questions. All the best, Jacinta -- ("`-''-/").___..--''"`-._ | Jacinta Richardson | `6_ 6 ) `-. ( ).`-.__.`) | Perl Training Australia | (_Y_.)' ._ ) `._ `. ``-..-' | +61 3 9354 6001 | _..`--'_..-_/ /--'_.' ,' | contact at perltraining.com.au | (il),-'' (li),' ((!.-' | www.perltraining.com.au | From enkidu at cliffp.com Wed May 3 13:00:02 2006 From: enkidu at cliffp.com (Cliff Pratt) Date: Thu, 04 May 2006 08:00:02 +1200 Subject: [Wellington-pm] 'my' and 'our' In-Reply-To: <44588C81.2010609@perltraining.com.au> References: <445889C6.6010007@cliffp.com> <44588C81.2010609@perltraining.com.au> Message-ID: <44590BC2.5010604@cliffp.com> Jacinta Richardson wrote: > my is for lexical variables. That is variables that exist and can only be seen > in their current scope. A variable's scope is defined as the smallest enclosing > block (including file). > > our is for package variables (think globals). These are variables that exist > and can be accessed outside of the scope in which they were declared, for > example in different files. > > Consider this example: > > ---------------------- > package Foo; > > my $bar; > our $baz; > > { # open new scope > > my $gatx; > print $gatx; # works > } > > print $gatx; # doesn't work (error) > > ---------------------- > # main program > use Foo; > > print $Foo::baz; # works > print $Foo::bar; # error, $bar can't be seen out here. > > ----------------------- > > our replaces the need for "use vars" from code prior to 5.6. It's a tiny bit > more complicated than this, but not so much that it matters. > Thanks Jacinta. So, if the variable ($bar) is not used outside the package, it can be declared with 'my'. If it is defined with 'our' it is available anywhere in the package (or scope) as simply $bar, but outside the package (or scope) it is available as $Foo::bar, then? Since all object variables should be accessed by an accessor method, in an object situation, the use of 'our' should be rare. Similarly in an ordinary package, you wouldn't normally expose a package variable unless you had a good reason, probably to configure something basic in the package. Am I correct? Cheers, Cliff -- http://barzoomian.blogspot.com From sam at vilain.net Wed May 3 16:02:04 2006 From: sam at vilain.net (Sam Vilain) Date: Thu, 04 May 2006 11:02:04 +1200 Subject: [Wellington-pm] 'my' and 'our' In-Reply-To: <44590BC2.5010604@cliffp.com> References: <445889C6.6010007@cliffp.com> <44588C81.2010609@perltraining.com.au> <44590BC2.5010604@cliffp.com> Message-ID: <4459366C.3030204@vilain.net> Cliff Pratt wrote: >So, if the variable ($bar) is not used outside the package, it can be >declared with 'my'. If it is defined with 'our' it is available anywhere >in the package (or scope) as simply $bar > > "Scope" is correct. If you declare a variable with "our" inside a sub, for instance, you will need to declare it in other subs in the same package to refer to it. ie: use strict; sub foo { our $bar; } sub baz { $bar; # error } sub frop { our $bar; # same $bar as in &foo } >Since all object variables should be accessed by an accessor method, in >an object situation, the use of 'our' should be rare. > > Indeed. It's generally considered bad form, especially where it can be easily avoided. >Similarly in an ordinary package, you wouldn't normally expose a package >variable unless you had a good reason, probably to configure something >basic in the package. > > Yes, and if you're writing OO properly, it's usually more appropriate to use "class methods" for those. ie, people need to write "ClassName->foo" to get the variables. Sam. >Am I correct? > >Cheers, > >Cliff > > > From jarich at perltraining.com.au Wed May 3 16:57:56 2006 From: jarich at perltraining.com.au (Jacinta Richardson) Date: Thu, 04 May 2006 09:57:56 +1000 Subject: [Wellington-pm] 'my' and 'our' In-Reply-To: <44590BC2.5010604@cliffp.com> References: <445889C6.6010007@cliffp.com> <44588C81.2010609@perltraining.com.au> <44590BC2.5010604@cliffp.com> Message-ID: <44594384.1040506@perltraining.com.au> Cliff Pratt wrote: > So, if the variable ($bar) is not used outside the package, it can be > declared with 'my'. If it is defined with 'our' it is available anywhere > in the package (or scope) as simply $bar, but outside the package (or > scope) it is available as $Foo::bar, then? This is correct. You can also export package variables using Exporter (just as you would other symbols) but since this is a terribly bad idea, most people just use their fully qualified name. Sam explained the scoping requirements when using "our". Most people ignore these by declaring any package variables at the top of their program, but there are good reasons to share a package variable with only a few subroutines. If you want to make a package variable in the main scope you can do that too: # main script; our $foo; # elsewhere package Foo; print $main::foo; # or (shorter, but the same) print $::foo; Really though, this suggests that you should be passing the value in as an argument to a subroutine or something like that. Manipulation of package variables is usually called "action from a distance" and can cause all sorts of maintainer nightmares. My personal preference for any package variable is to use them as read-only outside their declared scope. So if I were to do the above, then I'd only *change* $main::foo in the main script, and never in the Foo package. Likewise with Perl's real globals (the special variables), I always localise my changes: open(my $file, "<", $filename) or die $!; my $contents; { local $/; # slurp mode on $contents = <$file>; } because I'm not game to guarantee that a later part of my script (or a module I use) won't expect the default values. The exception to this is changes I make to %ENV in scripts running with taint mode turned on. > Since all object variables should be accessed by an accessor method, in > an object situation, the use of 'our' should be rare. It's rare, but certainly not unheard of. Most modules provide a $VERSION package variable for example. use DBI; use CGI; use File::Find; print $DBI::VERSION; print $CGI::VERSION; print $File::Find::VERSION; As Sam said, in a typical OO design; object variables and class variables are usually accessed by methods so the use of "our" in these cases is rare. > Similarly in an ordinary package, you wouldn't normally expose a package > variable unless you had a good reason, probably to configure something > basic in the package. Common uses for package variables are as follows: * Provide VERSION information * Turn on/off verbose or debug modes * Provide access to configuration information There are other good uses for package variables, but they all tend to be application specific. In a typical, well designed module, package variables (other than $VERSION) probably won't exist. Of course there are an awful lot of badly designed modules, which use package variables instead of subroutine arguments and returns; but let's not think about those. ;) all the best, J -- ("`-''-/").___..--''"`-._ | Jacinta Richardson | `6_ 6 ) `-. ( ).`-.__.`) | Perl Training Australia | (_Y_.)' ._ ) `._ `. ``-..-' | +61 3 9354 6001 | _..`--'_..-_/ /--'_.' ,' | contact at perltraining.com.au | (il),-'' (li),' ((!.-' | www.perltraining.com.au | From grant at mclean.net.nz Mon May 8 14:44:07 2006 From: grant at mclean.net.nz (Grant McLean) Date: Tue, 09 May 2006 09:44:07 +1200 Subject: [Wellington-pm] Reminder: meeting tonight Message-ID: <1147124647.24593.3.camel@localhost.localdomain> Hi Mongers Wellington Perl Mongers are meeting tonight (Tuesday May 9th). The following two speakers are lined up: * Michael Robinson - Plucene * Srdjan Jankovic - Using CPAN for Fun and Profit See you then. Grant From grant at mclean.net.nz Mon May 8 14:48:33 2006 From: grant at mclean.net.nz (Grant McLean) Date: Tue, 09 May 2006 09:48:33 +1200 Subject: [Wellington-pm] [Fwd: O'Reilly UG Program News: DSUG Discount Changes] Message-ID: <1147124913.24593.7.camel@localhost.localdomain> Just to let you know that O'Reilly have recently increased the discount they give to user group members. If people do want to order books, it might be worth coordinating at tonight's meeting to get the even bigger discount and save on shipping. Regards Grant -------- Forwarded Message -------- > From: Marsee Henon > To: perlmongers at catalyst.net.nz > Subject: O'Reilly UG Program News: DSUG Discount Changes > Date: Wed, 03 May 2006 12:16:34 -0700 > Hello, > > Get 30% off a single book or 35% off two or more books from O'Reilly, No > Starch, Paraglyph, PC Publishing, Pragmatic Bookshelf, SitePoint, or > Syngress books you purchase directly from O'Reilly. Just use code DSUG > when ordering online or by phone 800-998-9938. Free ground shipping on > orders $29.95 or more in the US. > > > Other benefits you receive when you buy directly from O'Reilly include: > > *100% Satisfaction Guarantee* > If, for any reason, you're not completely satisfied with your purchase, > return it to us and get your money back. A return shipping label is > included with every direct purchase, and directions are posted online in > case you've misplaced it: > . > > *Safari Enabled* > Whenever possible, our books are "Safari Enabled." This means you can > access your book for free online for 45 days through the O'Reilly Safari > Bookshelf. How do you know if your book is Safari Enabled? Turn your > book over and look for the "Safari Enabled" logo on the bottom right of > the page. If it's there, flip through the last couple pages of your book > until you find directions for accessing your book online. > > *Booktech* > Have a question about your book? O'Reilly is the only publisher that > offers tech support for books. Send an email to > and we'll help you out. Be specific: Include > the book title and page number. It's also a good idea to include the > ISBN so we know what edition you have. > > *Reader Reviews* > Our reader reviews are read by most people at O'Reilly, including Tim > O'Reilly, all our editors, as well as sales, marketing, and PR. So if > you have praise, a gripe, or ideas for improvement, writing a reader > review on oreilly.com is a sure way for your voice to be heard. Just go > to your book's catalog page on oreilly.com and click the "Write a > Review" button. > > *Book Registration* > Register your book online and we'll notify you when the book has been > updated or a new edition is available. You can also win books and other > prizes. Haven't registered your books? Just go to > . > > *Newsletters* > Our newsletters keep you updated on the latest articles, books, news, > and events. A complete list of newsletters and lists can be found at > . > > We're working on a slew of additional benefits to serve you even better > so stay tuned. > > As always, thanks for your help spreading the word. > > Marsee Henon > > ================================================================ > O'Reilly > 1005 Gravenstein Highway North > Sebastopol, CA 95472 > http://ug.oreilly.com/ http://ug.oreilly.com/creativemedia/ > ================================================================ From grant at mclean.net.nz Tue May 9 14:33:12 2006 From: grant at mclean.net.nz (Grant McLean) Date: Wed, 10 May 2006 09:33:12 +1200 Subject: [Wellington-pm] Reminder: meeting tonight In-Reply-To: <1147124647.24593.3.camel@localhost.localdomain> References: <1147124647.24593.3.camel@localhost.localdomain> Message-ID: <1147210392.14708.20.camel@localhost.localdomain> Hi Mongers Well it looks like the mailing list has started delivering messages again. Apologies if anyone missed the meeting because the reminder was delayed (but really you ought to just mark your calendar :-)) Thanks to both our speakers for interesting presentations. Michael if you want to send me your slides I'll put them up on the web site (perhaps without the footnotes). For Win32 CPAN users, here's a link with details of alternative PPM repositories and the download URL for the free 'nmake' utility: http://perl-xml.sourceforge.net/faq/#win32_expat We need a volunteer or two to speak at next month's meeting. I thought I might put together an introduction to Test-Driven Development with Perl. Other volunteers or suggestions are welcome. On an administrative note, it looks like we need to institute policy for deciding the order of speakers in the infrequent case where there is competition for a particular slot. I suggest that from now on the rule is: the person who was first to volunteer gets to chose whether they speak first or second. If I'd thought of that rule last night, then Michael would have got the first slot. The date for the next meeting is the 13th of June. Regards Grant On Tue, 2006-05-09 at 09:44 +1200, Grant McLean wrote: > Hi Mongers > > Wellington Perl Mongers are meeting tonight (Tuesday May 9th). The > following two speakers are lined up: > > * Michael Robinson - Plucene > * Srdjan Jankovic - Using CPAN for Fun and Profit > > See you then. > Grant > > _______________________________________________ > Wellington-pm mailing list > Wellington-pm at pm.org > http://mail.pm.org/mailman/listinfo/wellington-pm From srdjan at catalyst.net.nz Tue May 9 14:41:17 2006 From: srdjan at catalyst.net.nz (Srdjan) Date: Wed, 10 May 2006 09:41:17 +1200 Subject: [Wellington-pm] Reminder: meeting tonight In-Reply-To: <1147210392.14708.20.camel@localhost.localdomain> References: <1147124647.24593.3.camel@localhost.localdomain> <1147210392.14708.20.camel@localhost.localdomain> Message-ID: <44610C7D.5010502@catalyst.net.nz> But there was an overriding rule: Oracle always wins S. Grant McLean wrote: > Hi Mongers > > Well it looks like the mailing list has started delivering messages > again. Apologies if anyone missed the meeting because the reminder was > delayed (but really you ought to just mark your calendar :-)) > > Thanks to both our speakers for interesting presentations. Michael if > you want to send me your slides I'll put them up on the web site > (perhaps without the footnotes). > > For Win32 CPAN users, here's a link with details of alternative PPM > repositories and the download URL for the free 'nmake' utility: > > http://perl-xml.sourceforge.net/faq/#win32_expat > > We need a volunteer or two to speak at next month's meeting. I thought > I might put together an introduction to Test-Driven Development with > Perl. Other volunteers or suggestions are welcome. > > On an administrative note, it looks like we need to institute policy for > deciding the order of speakers in the infrequent case where there is > competition for a particular slot. I suggest that from now on the rule > is: the person who was first to volunteer gets to chose whether they > speak first or second. If I'd thought of that rule last night, then > Michael would have got the first slot. > > The date for the next meeting is the 13th of June. > > Regards > Grant > > On Tue, 2006-05-09 at 09:44 +1200, Grant McLean wrote: >> Hi Mongers >> >> Wellington Perl Mongers are meeting tonight (Tuesday May 9th). The >> following two speakers are lined up: >> >> * Michael Robinson - Plucene >> * Srdjan Jankovic - Using CPAN for Fun and Profit >> >> See you then. >> Grant >> >> _______________________________________________ >> Wellington-pm mailing list >> Wellington-pm at pm.org >> http://mail.pm.org/mailman/listinfo/wellington-pm > > _______________________________________________ > Wellington-pm mailing list > Wellington-pm at pm.org > http://mail.pm.org/mailman/listinfo/wellington-pm > > From grant at mclean.net.nz Tue May 9 14:50:48 2006 From: grant at mclean.net.nz (Grant McLean) Date: Wed, 10 May 2006 09:50:48 +1200 Subject: [Wellington-pm] Reminder: meeting tonight In-Reply-To: <44610C7D.5010502@catalyst.net.nz> References: <1147124647.24593.3.camel@localhost.localdomain> <1147210392.14708.20.camel@localhost.localdomain> <44610C7D.5010502@catalyst.net.nz> Message-ID: <1147211448.14708.22.camel@localhost.localdomain> On Wed, 2006-05-10 at 09:41 +1200, Srdjan wrote: > But there was an overriding rule: > Oracle always wins Except when it comes to benchmarks - in which case we're not allowed to say who won. From michael at diaspora.gen.nz Tue May 9 15:06:10 2006 From: michael at diaspora.gen.nz (michael at diaspora.gen.nz) Date: Wed, 10 May 2006 10:06:10 +1200 Subject: [Wellington-pm] Reminder: meeting tonight In-Reply-To: Your message of "Wed, 10 May 2006 09:33:12 +1200." <1147210392.14708.20.camel@localhost.localdomain> Message-ID: Grant McLean writes: >Thanks to both our speakers for interesting presentations. Michael if >you want to send me your slides I'll put them up on the web site >(perhaps without the footnotes). Definitely without the footnotes :) >On an administrative note, it looks like we need to institute policy for >deciding the order of speakers in the infrequent case where there is >competition for a particular slot. I suggest that from now on the rule >is: the person who was first to volunteer gets to chose whether they >speak first or second. If I'd thought of that rule last night, then >Michael would have got the first slot. As it turns out, the wife wasn't upset at all, so apologies for the aggro, Srdjan :/. I've noticed the rule in previous meetings has been that whoever has actually got their presentation in a presentable state goes first; the other hacks away on a laptop to finish the presentation. That seems like a reasonable rule, that relies on human nature; it's just that Srdjan and I are both obviously overachievers who prepare things before time. As it was, there was at least one slide that was written during Srdjan's presentation (and yes, the footnotes as well), so I didn't suffer as much as I might have made out :). -- michael. From srdjan at catalyst.net.nz Tue May 9 15:35:10 2006 From: srdjan at catalyst.net.nz (Srdjan) Date: Wed, 10 May 2006 10:35:10 +1200 Subject: [Wellington-pm] Reminder: meeting tonight In-Reply-To: References: Message-ID: <4461191E.80102@catalyst.net.nz> Quite contrary my friend, I should apologise. As I was gently reminded it was *Me* who mislead you to believe that it would take 15 minutes. Of course, no one mentioned *helpers* to the gross overrun, but you know - we all share our victories, but in defeats we're all alone... How about deciding the order by a lucky draw? A module Presenter::Random? michael at diaspora.gen.nz wrote: > Grant McLean writes: >> Thanks to both our speakers for interesting presentations. Michael if >> you want to send me your slides I'll put them up on the web site >> (perhaps without the footnotes). > > Definitely without the footnotes :) > >> On an administrative note, it looks like we need to institute policy for >> deciding the order of speakers in the infrequent case where there is >> competition for a particular slot. I suggest that from now on the rule >> is: the person who was first to volunteer gets to chose whether they >> speak first or second. If I'd thought of that rule last night, then >> Michael would have got the first slot. > > As it turns out, the wife wasn't upset at all, so apologies for the aggro, > Srdjan :/. > > I've noticed the rule in previous meetings has been that whoever has > actually got their presentation in a presentable state goes first; the > other hacks away on a laptop to finish the presentation. > > That seems like a reasonable rule, that relies on human nature; it's > just that Srdjan and I are both obviously overachievers who prepare > things before time. > > As it was, there was at least one slide that was written during Srdjan's > presentation (and yes, the footnotes as well), so I didn't suffer as > much as I might have made out :). > > -- michael. > _______________________________________________ > Wellington-pm mailing list > Wellington-pm at pm.org > http://mail.pm.org/mailman/listinfo/wellington-pm > > From michael at diaspora.gen.nz Tue May 9 16:55:13 2006 From: michael at diaspora.gen.nz (michael at diaspora.gen.nz) Date: Wed, 10 May 2006 11:55:13 +1200 Subject: [Wellington-pm] Prizes! (was Re: Reminder: meeting tonight) In-Reply-To: Your message of "Wed, 10 May 2006 10:35:10 +1200." <4461191E.80102@catalyst.net.nz> Message-ID: Srdjan writes: >How about deciding the order by a lucky draw? A module Presenter::Random? Well, given that there's only ever two presenters (barring when the fearful dictator runs the lightning talks), a binary decision of such a sort is usually decided by either: 1. a token of some monetary value, or 2. a sophisticated cryptographically strong choice protocol, such as rock paper scissors. I'd go for the latter, myself. First person to: * code up a rock paper scissors game * playable over the web or on the presentation PC * for two (or more) people, all physically colocated with each other, using the same PC * in Perl * for the purpose of deciding speaking order wins either: * one (1) chocolate fish (the "normal person" prize), or * a commitment to talk on a topic of my choice at the July 2006 meeting (the "Grant" prize :) Solutions must be demonstrated at the June 2006 meeting of wellington.pm.org. In the event of more than one solution being presented, I'll award prizes as I see fit, based principally on the humor value of the solution. (Note: the principal problem an implementor will face is trying to keep the participants from knowing each others solution. Limiting to only two participants is permissible. I'd suggest something like a key based submission protocol; ie, hit a key, submit via Javascript or similar (Ajax, maybe?) each participant does this, then the final answer is displayed.) If you do code up a solution, it'd probably make a good presentation. *And* it'd make Grant happy. -- michael. From matt at catalyst.net.nz Tue May 9 17:05:50 2006 From: matt at catalyst.net.nz (Matthew Hunt) Date: Wed, 10 May 2006 12:05:50 +1200 Subject: [Wellington-pm] Unbleach Message-ID: <1147219550.1764.1.camel@localhost.localdomain> And for those that have used too much Acme::Bleach on their perl code, Google yields a solution in the form of unbleach.pl http://qs321.pair.com/~monkads/?node_id=85918 Matt. -- Matthew Hunt From sweetpea-perl at tentacle.net Wed May 10 00:19:30 2006 From: sweetpea-perl at tentacle.net (Kevin Dorne) Date: Wed, 10 May 2006 00:19:30 -0700 Subject: [Wellington-pm] Prizes! (was Re: Reminder: meeting tonight) In-Reply-To: References: <4461191E.80102@catalyst.net.nz> Message-ID: <20060510071930.GA38951@chaos.tentacle.net> On Wed, May 10, 2006 at 11:55:13AM +1200, michael at diaspora.gen.nz wrote: [...] > First person to: > > * code up a rock paper scissors game > * playable over the web or on the presentation PC > * for two (or more) people, all physically colocated with each > other, using the same PC > * in Perl > * for the purpose of deciding speaking order > > wins either: > > * one (1) chocolate fish (the "normal person" prize), or > * a commitment to talk on a topic of my choice at the July 2006 > meeting (the "Grant" prize :) How about they always get to go first, thereby nullifying any advantage gained by coding up the game, breaking the utilito-causal chain and sending the world into a spiral of paradox? -- Peace, love & happiness to you from me =) -Sweetpea http://sweetpea.tentacle.net "You get a taste of the elixir of life, but you sign away your soul to get it." - BD -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 187 bytes Desc: not available Url : http://mail.pm.org/pipermail/wellington-pm/attachments/20060510/99347420/attachment.bin From grant at mclean.net.nz Sun May 14 18:19:48 2006 From: grant at mclean.net.nz (Grant McLean) Date: Mon, 15 May 2006 13:19:48 +1200 Subject: [Wellington-pm] Wierd and wierder [was: RE: Adam Quill] Message-ID: <1147655988.14224.3.camel@localhost.localdomain> I received the following message which apparently relates to this message I forwarded last August. http://mail.pm.org/pipermail/wellington-pm/2005-August/000379.html Odd. -------- Forwarded Message -------- > From: Richard Cary > To: perlmongers at catalyst.net.nz > Subject: RE: Adam Quill > Date: Sun, 14 May 2006 11:39:23 +0800 > Hi guys, you may have received a request from an Adam Quill about help > for a perl script for a game. On behalf of the game's players and > administration, we strongly urge you to deny and perl help to this > person. His, mission being to destroy the game that thousands of > people play online all over the world. Refusal of his request would be > greatly appreciated by all. Thankyou > > ______________________________________________________________________ From srdjan at catalyst.net.nz Sun May 14 18:44:30 2006 From: srdjan at catalyst.net.nz (Srdjan) Date: Mon, 15 May 2006 13:44:30 +1200 Subject: [Wellington-pm] A talk suggestion In-Reply-To: <1147655988.14224.3.camel@localhost.localdomain> References: <1147655988.14224.3.camel@localhost.localdomain> Message-ID: <4467DCFE.1060501@catalyst.net.nz> No I'm not suggesting a competition for an ultimate destroyer script that will destroy any sub-culture, name(s) supplied as argument(s). I was wondering if anyone used exception raise/catch paradigm for error handling/propagation. To clarify a bit: I usually use objects with methods that return false in case of an error, and a method to get the error string. So every time I make a call I check return value - get error - set error - return false. I'm not complaining, that's just what I do (very C-ish, I know...). So I'm wondering if anyone with die - eval experience would like to step up and share his views on how those two paradigms compare, in a form of an engaging talk. Thank you. Srdjan From jarich at perltraining.com.au Sun May 14 18:58:36 2006 From: jarich at perltraining.com.au (Jacinta Richardson) Date: Mon, 15 May 2006 11:58:36 +1000 Subject: [Wellington-pm] A talk suggestion In-Reply-To: <4467DCFE.1060501@catalyst.net.nz> References: <1147655988.14224.3.camel@localhost.localdomain> <4467DCFE.1060501@catalyst.net.nz> Message-ID: <4467E04C.2060006@perltraining.com.au> Srdjan wrote: > So I'm wondering if anyone with die - eval experience would like to step up and > share his views on how those two paradigms compare, in a form of an engaging talk. ^^^ Or her views. ;) Actually I'd love to, but I won't be in Wellington in the near future. All the best, Jacinta -- ("`-''-/").___..--''"`-._ | Jacinta Richardson | `6_ 6 ) `-. ( ).`-.__.`) | Perl Training Australia | (_Y_.)' ._ ) `._ `. ``-..-' | +61 3 9354 6001 | _..`--'_..-_/ /--'_.' ,' | contact at perltraining.com.au | (il),-'' (li),' ((!.-' | www.perltraining.com.au | From grant at mclean.net.nz Sun May 14 19:00:23 2006 From: grant at mclean.net.nz (Grant McLean) Date: Mon, 15 May 2006 14:00:23 +1200 Subject: [Wellington-pm] A talk suggestion In-Reply-To: <4467DCFE.1060501@catalyst.net.nz> References: <1147655988.14224.3.camel@localhost.localdomain> <4467DCFE.1060501@catalyst.net.nz> Message-ID: <1147658423.14224.10.camel@localhost.localdomain> On Mon, 2006-05-15 at 13:44 +1200, Srdjan wrote: > ... > So I'm wondering if anyone with die - eval experience would like to step up and > share his views on how those two paradigms compare, in a form of an engaging talk. What an 'exception'ally good idea. Someone planning to do such a thing could even make their job easier by borrowing from this talk: http://axkit.org/docs/presentations/tpc2002/exceptions.pdf (warning, slow link). BTW, I'm sure Srdjan didn't intend to exclude persons of the female persuasion from sharing their views too. Cheers Grant From grant at mclean.net.nz Tue May 16 14:35:29 2006 From: grant at mclean.net.nz (Grant McLean) Date: Wed, 17 May 2006 09:35:29 +1200 Subject: [Wellington-pm] [Fwd: Newsletter from O'Reilly UG Program, May 15] Message-ID: <1147815329.20199.0.camel@localhost.localdomain> -------- Forwarded Message -------- > From: Marsee Henon > Subject: Newsletter from O'Reilly UG Program, May 15 > Date: Mon, 15 May 2006 12:07:33 -0700 > ================================================================ > O'Reilly News for User Group Members > May 15, 2006 > ================================================================ > ---------------------------------------------------------------- > New Releases > ---------------------------------------------------------------- > -The Art of Software Process Improvement > -Build Your Own Website The Right Way Using HTML & CSS > -Building Scalable Web Sites > -DNS and BIND, Fifth Edition > -Enterprise JavaBeans 3.0, Fifth Edition > -Enterprise SOA > -iMovie 6 & iDVD: The Missing Manual > -IPv6 Essentials, Second Edition > -It's Never Done That Before > -Java I/O, Second Edition > -Learning PHP and MySQL > -MCSE Core Required Exams in a Nutshell, Third Edition > -Perl Hacks > -Rails Recipes > -Scripting for Testers > -Statistics Hacks > -Syngress IT Security Project Management Handbook > -VB 2005 Black Book > -Video Conferencing over IP > ---------------------------------------------------------------- > Upcoming Events > ---------------------------------------------------------------- > -O'Reilly at SANE 2006, Delft, The Netherlands--May 15-19 > -O'Reilly at XTech 2006: "Building Web 2.0," Amsterdam--May 16-19 > -O'Reilly at JavaOne, San Francisco, CA--May 16-19 > -Peter Krogh at the American Society of Picture Professionals, > Fairfax, VA--May 17 > -Peter Morville at Enterprise Search Summit 2006, New York--May 23-24 > -Stephen Few at Technology Transfer, Italy--May 24-26 > -Eddie Tapp Photo Workshop, Palo Alto, CA--May 24 > ---------------------------------------------------------------- > Conference News > ---------------------------------------------------------------- > -Register for OSCON, July 24-28--Portland,OR > -Register for the Where 2.0 Conference, June 13-14--San Jose, CA > ---------------------------------------------------------------- > News > ---------------------------------------------------------------- > -Scan, Open, & Print Money with dekePod > -Announcing the 2006 O?Reilly Photoshop Cook-Off > -For SARS Press 1, for Bird Flu Press 2... > -BSD Certification Group Inc. Announces UG Fundraising Competition > -The Chicago Perl Mongers are hosting Yet Another Perl Conference > North America--June 26-28 > -Creating a Dual-Boot Windows XP and Ubuntu Laptop > -Grabbing iTMS Preview Tracks the Geek Way > -Mac FTP: A Guided Tour > -Build a Web-Based Bug Tracking App > -Ensuring Application Compatibility in Vista > -AJAX and Screenreaders: When Can it Work? > -Rounded Corners with Clean HTML and No JavaScript > -Search Indexing Limits: Where Do Robots Stop? > -Real-Time Java: An Introduction > -Configuration Management in Java EE Applications Using Subversion > -Summer Projects > -The Maker Faire > -Make Video Podcast: Andrew Filo's Rocket Belt > > ---------------------------------------------------------------- > New Releases--Books, PDFs, and Rough Cuts > ---------------------------------------------------------------- > Get 30% off a single book or 35% off two or more books from O'Reilly, > No Starch, Paraglyph, PC Publishing, Pragmatic Bookshelf, SitePoint, > or Syngress books you purchase directly from O'Reilly. > Just use code DSUG when ordering online or by phone 800-998-9938. > > > Free ground shipping on orders $29.95 or more. > For more details, go to: > > > Did you know you can request a free book to review for your > group? Ask your group leader for more information. > > For book review writing tips and suggestions, go to: > > > ***The Art of Software Process Improvement > Publisher: O'Reilly > Rough Cuts Version (limited Review copies available) > "The Art of Software Process Improvement" combines the foundation needed > to understand process improvement theory with the best practices to help > individuals implement process improvement initiatives in their > organization. The three leading programs: ISO 9001:2000, CMMI, and Six > Sigma-amidst the buzz and hype-tend to get lumped together under a > common label. This book delivers a combined guide to all three programs, > compares their applicability, and then sets the foundation for further > exploration. It's a one-stop-shop designed to give you a working > orientation to what the field is all about. > > > > ***Build Your Own Website The Right Way Using HTML & CSS > Publisher: Sitepoint > ISBN: 0975240293 > This book introduces you to HTML and CSS as you follow along with the > author, step-by-step, to create a fully functional web site. Along the > way, you'll learn how to apply web standards to make your site fast > loading, easy to maintain, viewable in almost every web browser, and > accessible to disabled users. > > > > ***Building Scalable Web Sites > Publisher: O'Reilly > ISBN: 0596102356 > This comprehensive guide covers the design of software and hardware > systems for web applications. Using scores of examples and leading-edge > tips, it details proven methods for scaling web applications to millions > of users. Topics include application architecture, development > practices, technologies, Unicode, and general infrastructure work. Ideal > for anyone ready to realize the cost and performance benefits available > to web applications today. > > > > ***DNS and BIND, Fifth Edition > Publisher: O'Reilly > ISBN: 0596100574 > The fifth edition covers BIND 9.3.2, the most recent release of the BIND > 9 series, as well as BIND 8.4.7. BIND 9.3.2 contains further > improvements in security and IPv6 support, and important new features > such as internationalized domain names, ENUM (electronic numbering), and > SPF (the Sender Policy Framework). Whether you're an administrator > involved with DNS on a daily basis or a user who wants to be more > informed about the Internet and how it works, you'll find that this book > is essential reading. > > > > ***Enterprise JavaBeans 3.0, Fifth Edition > Publisher: O'Reilly > ISBN: 059600978X > This new edition of "Enterprise JavaBeans 3.0", written by Bill Burke > and Richard Monson-Haefel. has been updated to capture the very latest > need-to-know Java technologies in the same award-winning fashion that > drove the success of the previous four editions. Its easy-to-follow > style and hundreds of practical examples help you simplify the complex > world of EJB--without the costly trial and error. > > > > ***Enterprise SOA > Publisher: O'Reilly > ISBN: 0596102380 > Based on extensive research with experts from the German software > company SAP, this definitive book is ideal for architects, developers, > and other IT professionals who want to understand the technology and > business relevance of enterprise SOA in a detailed way--especially those > who want to move on the technology now, rather than in the next year or > two. > > > > ***iMovie 6 & iDVD: The Missing Manual > Publisher: O'Reilly > ISBN: 0596527268 > Movie 6 is all about the ease of moviemaking, but if you really want to > learn its capabilities, Apple documentation won't make the cut. The new > edition of this Missing Manual is the ideal third-party authority that > covers all of these changes through an objective lens, from choosing and > using a digital camcorder to burning the finished work onto DVDs or > posting it online. > > > > ***IPv6 Essentials, Second Edition > Publisher: O'Reilly > ISBN: 0596100582 > "IPv6 Essentials, Second Edition" provides a succinct, in-depth tour of > all the new features and functions in IPv6. It guides you through > everything you need to know to get started, including how to configure > IPv6 on hosts and routers and which applications currently support IPv6. > Aimed at system and network administrators, engineers, network > designers, and IT managers, this book will help you understand, plan > for, design, and integrate IPv6 into your current IPv4 infrastructure. > > > > ***It's Never Done That Before > Publisher: No Starch Press > ISBN: 1593270763 > "It's Never Done That Before" is a guide to Windows XP troubleshooting > techniques, solving the most common problems in Windows XP, and finding > and fixing more obscure ones. The book provides pointers to explanations > of BIOS beep codes and blue screen errors, instructions for using the > troubleshooting tools supplied with Windows XP such as Safe Mode and the > Recovery Console, and advice for dealing with device drivers, the ROM > BIOS, and the Windows Registry. > > > > ***Java I/O, Second Edition > Publisher: O'Reilly > ISBN: 0596527500 > All of Java's Input/Output (I/O) facilities are based on streams, which > provide simple ways to read and write data of different types. "Java I/O, > 2nd Edition" tells you all you need to know about the four main > categories of streams and uncovers less-known features to help make your > I/O operations more efficient. You'll also learn how to control number > formatting, use characters aside from the standard ASCII character set, > and get a head start on writing truly multilingual software. > > > > ***Learning PHP and MySQL > Publisher: O'Reilly > ISBN: 0596101104 > Featuring basic concepts explained in plain English, "Learning PHP and > MySQL" is the ideal guide for newcomers attracted to the popular PHP and > MySQL combination. Learn how to generate dynamic web content in a slow, > easy-to-follow fashion. Also covers error handling, security, HTTP > authentication, and more. > > > > ***MCSE Core Required Exams in a Nutshell, Third Edition > Publisher: O'Reilly > ISBN: 0596102283 > Written by the premier author in Windows administration, William Stanek, > and addressing the needs of Windows 2003 administrators preparing for > the Microsoft Certified Systems Engineer (MCSE) exams, "MCSE Core > Required Exams in a Nutshell" is invaluable. Not only does this book > provide the resources administrators need to succeed on the exams, but > to succeed in the real world as well. They can think of this book as the > notes they would have highlighted and then recorded for every essential > nugget of information related to the skills measured in Exams 70-290, > 70-291, 70-293, and 70-294 (and by association Exams 70-292 and > 70-296). > > > > ***Perl Hacks > Publisher: O'Reilly > ISBN: 0596526741 > "Perl Hacks" taps into the collective wisdom of the world's most > creative Perl gurus, so you can learn from their experiences. It's the > perfect book for experienced developers looking for time-saving > practical tips or dabblers who are simply curious about Perl's many cool > capabilities. Topics include user interaction, data munging, working > with modules, object hacks, and debugging. > > > > ***Rails Recipes > Publisher: Pragmatic Bookshelf > ISBN: 0977616606 > You've read the tutorials and watched the online videos. You have a > strong grasp of all of the ingredients that make up a successful Rails > application. But ingredients don't just turn themselves into a meal. > Chad Fowler's "Rails Recipes" is a collection of recipes that will take > you step by step through the most cutting edge, modern Rails techniques, > mixing the ingredients to create world-class web applications. Learn how > to do it, and how to do it right. > > > > ***Scripting for Testers > Publisher: Pragmatic Bookshelf > ISBN: 0977616614 > "Scripting for Testers" is divided into two parts: Talking the Talk and > Walking the Walk. In the first, you'll learn the Ruby scripting language > and the overall craft of scripting. In the second, you'll see how that > knowledge can be applied to solve common testing problems. > > > > ***Statistics Hacks > Publisher: O'Reilly > ISBN: 0596101643 > Want to calculate the probability that an event will happen? Be able to > spot fake data? Prove beyond doubt whether one thing causes another? Or > learn to be a better gambler? You can do that and much more with 75 > practical and fun hacks packed into "Statistics Hacks." These cool tips, > tricks, and mind-boggling solutions from the world of statistics, > measurement, and research methods will not only amaze and entertain you, > but will give you an advantage in several real-world > situations-including business. > > > > ***Syngress IT Security Project Management Handbook > Publisher: Syngress > ISBN: 1597490768 > The definitive work for IT professionals responsible for the management > of the design, configuration, deployment, and maintenance of enterprise > wide security projects. Provides specialized coverage of key project > areas including Penetration Testing, Intrusion Detection and Prevention > Systems, and Access Control Systems. > > > > ***VB 2005 Black Book > Publisher: Paraglyph Press > ISBN: 1933097086 > "VB 2005 Programming Black Book" is one of the first comprehensive books > that cover the new version of Visual Basic and the development features > of Microsoft's .NET platform in depth. The book explains the major > changes to VB and provides numerous tips and practical solutions for > developing applications. "VB 2005 Programming Black Book" approaches > every topic methodically: the In Depth section covers all the concepts > being introduced, followed by the Immediate Solutions section that > provide hands-on real-world techniques. This book will guide programmers > through all new features of VB 2005 and provide detailed coverage of all > the new controls, language enhancements, and new architecture features. > > > > ***Video Conferencing over IP > Publisher: Syngress > ISBN: 1597490636 > Until recently, the reality of video conferencing didn't live up to the > marketing hype. That's all changed. The network infrastructure and > broadband capacity is now in place to deliver clear, real time video and > voice feeds between multiple points of contacts, with market leaders > such as Cisco and Microsoft continuing to invest heavily in development. > In addition, newcomers Skype and Google are poised to launch services > and products targeting this market. "Video Conferencing over IP" is the > perfect guide to getting up and running with video teleconferencing for > small to medium size enterprises. > > > > ***MAKE Magazine Subscriptions > The annual subscription price for four issues is $34.95. When you > subscribe with this link, you'll get a free issue--the first one plus > four more for $34.95. So subscribe for yourself or friends with this > great offer for charter subscribers: five volumes for the cost of four. > Subscribe at: > > > ================================================ > Upcoming Events > ================================================ > ***For more events, please see: > http://events.oreilly.com/ > > > ***O'Reilly at SANE 2006, Delft, The Netherlands--May 15-19 > Stop by our bookstall and peruse our latest batch of System > Administration and Networks books and more at the 5th System > Administration and Network Engineering Conference. > > > > ***O'Reilly at XTech 2006: "Building Web 2.0," Amsterdam--May 16-19 > Meet Kendall Clark (XML.com) and our authors Edd Dumbill ("Mono: A > Developer's Notebook and Linux Unwired"), Paul Graham ("Hackers and > Painters"), Eric van der Vlist ("Relax NG" and "XML Schema") at XTech > 2006: "Building Web 2.0." > > > > ***O'Reilly at JavaOne, San Francisco, CA--May 16-19 > Stop by our booth (#413) and to hear authors Bill Burke ("Enterprise > JavaBeans 3.0, 5th Ed") and Kent Beck ("JUnit Pocket Guide") talk about > the latest developments in Java and how they affect you. While you're > there pick up the new edition of ("Java I/O"). > > > > ***Peter Krogh at the American Society of Picture Professionals, > Fairfax, VA--May 17 > Author Peter Krogh ("The DAM Book: Digital Asset Management for > Photographers") will be giving an evening presentation on DAM at the > Washington D.C. Chapter of ASPP. > > > > ***Peter Morville at Enterprise Search Summit 2006, > New York, NY--May 23-24 > Author Peter Morville ("Ambient Findability") is speaking at Enterprise > Search Summit 2006 on Ambient Findability. > > > > ***Stephen Few at Technology Transfer, Italy--May 24-26 > Author Stephen Few ("Information Dashboard Design") conducts three > workshops: "Show Me the Numbers" and "Information Visualization for > Discovery and Analysis," and "Dashboard Design for Immediate Insight." > > > > ***Eddie Tapp Photo Workshop, Palo Alto, CA--May 24 > Eddie Tapp, author of the upcoming book, "Photoshop Workflow Setups: > Eddie Tapp on Digital Photography" will be conducting a 2-hour workshop > on digital workflow. > > > ================================================ > Conference News > ================================================ > ***OSCON, July 24-28--Portland,OR > OSCON, the O'Reilly Open Source Convention, is still where open source > rubber meets the road. OSCON happens July 24-28, 2006 in open source > hotspot Portland, Oregon, and registration has just opened. Hundreds of > sessions and tutorials. Thousands of open source mavericks, brainiacs, > hackers, activists, scientists, and their admirers, some in > business-casual disguise. Read all about it. > > > User Group members who register before June 5, 2006 get a double > discount. Use code "os06dsug" when you register, and receive 15% off > the early registration price. > > To register for the conference, go to: > > > > ***Register for the Where 2.0 Conference, June 13-14--San Jose, CA > The Where 2.0 Conference brings together the people, projects, and > issues leading the charge into the location-based technology frontier. > Join the developers, innovators, and business people behind the new era > of geospatial technology as they come together--because everything > happens somewhere, and it's all happening here at Where 2.0. > > > Use code "whr06dsug" when you register, and receive 15% off > the registration price. > > To register for the conference, go to: > > > ================================================ > News From O'Reilly & Beyond > ================================================ > --------------------- > General News > --------------------- > ***Scan, Open, & Print Money with dekePod > Deke McClelland Digital imaging guru Deke McClelland just introduced the > pilot for his new show "dekePod," a video podcast with a hip, visually > compelling style. The pilot episode, "It's Your Money, Scan It!", shows > viewers a workaround in Photoshop to scan and open money (under legal > guidelines). > > > > ***Announcing the 2006 O?Reilly Photoshop Cook-Off > Win some great prizes from Adobe, Pantone, Lowepro, GretagMacbeth, > iStock Photo, Lynda.com, and more in the 2006 O?Reilly Photoshop > Cook-Off. Just take up to three of your own photos and manipulate them > with Adobe Photoshop, using recipes from any of the five O?Reilly > Photoshop Cookbooks: "Photoshop Retouching Cookbook for Digital > Photographers," "Photoshop Blending Modes Cookbook for Digital > Photographers," "Photoshop Photo Effects Cookbook," "Photoshop Filter > Effects Encyclopedia," and "Photoshop Fine Art Effects Cookbook." The > contest will be judged by a panel of experts from the industry?s A-list: > Bert Monroy, Katrin Eismann, Deke McClelland, Tim Grey, and Mikkel > Aaland, among others. Entries may be submitted from May 15--August 15, > 2006. For details and contest rules, go to: > > > > ***For SARS Press 1, for Bird Flu Press 2... > One of the key things we can do during a pandemic is to move in-person > congregations and meetings online, creating electronic surrogates for > daily social activities such as classes, business meetings, and social > outings. Brian McConnell examines how organizations can use the latest > telecom technology to help keep society functioning and diminish the > spread of disease if this kind of a disaster takes place. > > > --------------------- > Open Source > --------------------- > ***BSD Certification Group Inc. Announces UG Fundraising Competition > Calling all User Groups! The BSD Certification Group Inc. just announced > a fundraising competition at BSDCan 2006 (http://www.bsdcan.org). > If you're a member of a BUG, LUG, UUG, or sysadmin user group, here's > your chance to help the BSD Certification Group > (http://www.bsdcertification.org) realize their goal of creating the > standard certification for system administrators of the BSD family of > operating systems. This is a great opportunity to assist in a worthy > cause while showcasing the generosity and creative spirit found within > your user group. For contest details and deadlines, go to: > > > > ***The Chicago Perl Mongers are hosting Yet Another Perl Conference > North America--June 26-28 > YAPC::NA conference takes place June 26th-28th and will be followed by > open Perl programming classes on the 29th and 30th. Come hear Perl > luminaries such as Larry Wall ("Programming Perl"), Damian Conway ("Perl > Best Practices" & "Perl Hacks"), Randal Schwartz ("Learning Perl" & > "Intermediate Perl"), brian d foy ("Learning Perl" and "Intermediate > Perl"), chromatic ("Perl Testing: A Developer's Notebook" & "Perl > Hacks"), Allison Randal ("Perl 6 & Parrot Essentials"), and Andy Lester > ("Mac OS X Tiger in a Nutshell"), and more. Find out what is new in Perl > 5 and what is happening with Perl 6 and Parrot. > > > > ***Creating a Dual-Boot Windows XP and Ubuntu Laptop > Running a Windows-only laptop is hardly ideal for people who do > considerable work in the Linux environment. When Cygwin and ssh aren't > enough, consider at least dual-booting into the free software world. > Kevin Farnham recently converted his new laptop into a half-Windows, > half-Ubuntu GNU/Linux machine. Here's how. > > > --------------------- > Mac > --------------------- > ***Grabbing iTMS Preview Tracks the Geek Way > Ever tried to copy a preview track from the iTMS onto your iPod? Hint: > you can't. iTunes won't add those 30-second free previews. Why would you > want to? Some great audio and video are waiting for you to enjoy. Plus, > there's the sheer thrill of geek accomplishment--the "oh, so that's how > to do it" satisfaction. > > > > ***Mac FTP: A Guided Tour > (S)FTP has a valuable place in the hearts of web builders and > developers, and is still one of the most practical methods of getting > files from one place to another in a secure manner. In this article, > Giles Turnbull surveys six FTP clients for the Mac platform and shows > you the major characteristics of each. > > > --------------------- > Windows/.NET > --------------------- > ***Build a Web-Based Bug Tracking App > Having problems tracking bugs every time you create a new project? Jesse > Liberty has the solution. Here, he shows you how to build a Web-based > bug-tracking application using ASP.NET. > > > > ***Ensuring Application Compatibility in Vista > What should you do if your enterprise has mission-critical > line-of-business apps that simply must continue running properly once > you've upgraded your desktops to Vista? Mitch Tulloch, author of Windows > Server Hacks, offers insight and advice. > > > --------------------- > Web > --------------------- > ***AJAX and Screenreaders: When Can it Work? > In this insightful report, James reveals the results of independent > tests he has conducted using AJAX scripts in a variety of screen reader > software. The results are sure to surprise you! > > > > ***Rounded Corners with Clean HTML and No JavaScript > 'Spanky Corner' is an experimental technique for using only CSS to > produced 'round-cornered content boxes' with semantically pure markup. > It does not require JavaScript to work. > > > > ***Search Indexing Limits: Where Do Robots Stop? Ever wondered how much > of each of your pages is being crawled by the search engines? Serge has, > which is why he conducted an experiment to test the exact page size that > could be crawled by the search bots. > > > --------------------- > Java > --------------------- > ***Real-Time Java: An Introduction > "Real-Time" Java doesn't mean "really fast," but it does mean "really > predictable," and that's especially important in many fields where an > unpredictable response time, usually caused by the Java Virtual > Machine's garbage collector, can cost money or lives. Peter Mikhalenko > looks at the Real-Time Specification for Java and Sun's first > implementation of the spec. > > > > ***Configuration Management in Java EE Applications Using Subversion > Does your enterprise Java application need to store not only complex > objects or documents but a history of changes to them? Many developers > try to solve this problem with database wizardry, but it's probably a > better option to hand the job over to a configuration management system > that is built for the task. In this article, Swaminathan Radhakrishnan > shows how you can implement requirements for change tracking by using a > Subversion repository from your Java application, by way of the JavaSVN > library. > > > --------------------- > Podcasts > --------------------- > ***Summer Projects > Are you looking for something fun or interesting to do this summer? This > week, we begin with an interview with Julieanne Kost, whose pictures > from airplanes are published in the book Window Seat. Then Google's > Chris Dibona talks to us about the upcoming second Summer of Code. > Finally, Dale Dougherty reads from his article in Make magazine on > Natalie Jeremijenko and her robot dogs. (DTF 05-08-2006: 24 minutes 35 > seconds) > > > > ***The Maker Faire > Thousands of makers met up at the San Mateo Fairgrounds on April 22 and > 23 for the first-ever Maker Faire. We talk to Make magazine publisher > Dale Dougherty about the Faire and hear from some of the attendees and > exhibitors that made this event so successful. (DTF 05-01-2006: 19 > minutes 55 seconds) > > > > ***Make Video Podcast: Andrew Filo's Rocket Belt > > > For more Maker Faire coverage, go to: > > > > Until next time-- > > Marsee Henon > > > ================================================================ > O'Reilly > 1005 Gravenstein Highway North > Sebastopol, CA 95472 > http://ug.oreilly.com/ http://ug.oreilly.com/creativemedia/ > ================================================================ From LRW at clear.net.nz Mon May 22 21:11:34 2006 From: LRW at clear.net.nz (Lesley Walker) Date: Tue, 23 May 2006 16:11:34 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? Message-ID: <1148357495.31782.28.camel@localhost.localdomain> Hi folks, I have a CGI script which currently (under Debian Woody) runs with setgid unix permissions in order to read mail logs (so that certain people can trace where emails have gone without needing login access to certain servers). After an upgrade to Sarge, however, it doesn't run, and I get this error in the apache logs: Can't do setuid (cannot exec sperl) Googling finds a solution, which is to install perl-suid. However, the package decription says: Usage of this program is now strongly deprecated upstream and support (along with this package) will probably be removed in 5.10 Therefore I'd like to find a solution OTHER than perl-suid. Is there some other package or module that can be used instead? I've tried setting $) and it has no effect. You can assume I've thought of these already: making copies of the log files giving the apache user account admin permissions changing the ownership/permissions of the log files None of them are really suitable. LesleyW From pspam at paradise.net.nz Mon May 22 21:19:09 2006 From: pspam at paradise.net.nz (Peter C. Kelly) Date: Tue, 23 May 2006 16:19:09 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: <1148357495.31782.28.camel@localhost.localdomain> References: <1148357495.31782.28.camel@localhost.localdomain> Message-ID: <44728D3D.1060707@paradise.net.nz> Hi Lesley Walker wrote: >You can assume I've thought of these already: > making copies of the log files > giving the apache user account admin permissions > changing the ownership/permissions of the log files >None of them are really suitable. > > How about using sudo? Cheers, Peter From LRW at clear.net.nz Mon May 22 21:21:32 2006 From: LRW at clear.net.nz (Lesley Walker) Date: Tue, 23 May 2006 16:21:32 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: <44728D3D.1060707@paradise.net.nz> References: <1148357495.31782.28.camel@localhost.localdomain> <44728D3D.1060707@paradise.net.nz> Message-ID: <1148358092.31782.32.camel@localhost.localdomain> On Tue, 2006-05-23 at 16:19 +1200, Peter C. Kelly wrote: > How about using sudo? How exactly would I go about doing that? This is a CGI script. L. From pspam at paradise.net.nz Mon May 22 21:34:31 2006 From: pspam at paradise.net.nz (Peter C. Kelly) Date: Tue, 23 May 2006 16:34:31 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: <1148358092.31782.32.camel@localhost.localdomain> References: <1148357495.31782.28.camel@localhost.localdomain> <44728D3D.1060707@paradise.net.nz> <1148358092.31782.32.camel@localhost.localdomain> Message-ID: <447290D7.5030403@paradise.net.nz> Hi Lesley Walker wrote: >On Tue, 2006-05-23 at 16:19 +1200, Peter C. Kelly wrote: > > >>How about using sudo? >> >> >How exactly would I go about doing that? This is a CGI script. > > I must concede I haven't done a proof of concept. But I'd imagine that you do the following: have a shell script, running as apache or whatever the web server runs as; has that run sudo /usr/local/bin/your_script.pl; and have an /etc/sudoers line that allows apache to run /usr/local/bin/your_script.pl as the target user without a password. (see man sudoers) Peter From ewen at naos.co.nz Mon May 22 21:42:35 2006 From: ewen at naos.co.nz (Ewen McNeill) Date: Tue, 23 May 2006 16:42:35 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: Message from Lesley Walker of "Tue, 23 May 2006 16:21:32 +1200." <1148358092.31782.32.camel@localhost.localdomain> Message-ID: <20060523044235.64A26868A93@wat.la.naos.co.nz> In message <1148358092.31782.32.camel at localhost.localdomain>, Lesley Walker writes: >On Tue, 2006-05-23 at 16:19 +1200, Peter C. Kelly wrote: >> How about using sudo? > >How exactly would I go about doing that? This is a CGI script. www-data ALL = (mail) NOPASSWD: /usr/local/bin/analyzelogs Which says the "www-data" user (which is what Apache runs as in Debian, at least by default), can run the command "/usr/local/bin/analyzelogs" as the "mail" user, without providing a password. (The "ALL =" bit means that it can do this on any host that the entry is present in the /etc/sudoers file.) Then the script would call: sudo -u mail /usr/local/bin/analyzelogs Sudo would probably be my suggestion in this situation too; I've used it for other batch-script privilege-required automation quite successfully. The next best solution is a C program which can be setuid, either to do the whole task or to run the perl program as the appropriate user (with taint turned on, etc). Ewen PS: sudo tip -- sudo processes the configuration file from top to bottom overwriting abilities as it goes, so the last rule wins. This is confusing if one is expecting, eg, the most specific rule to win. So place exceptions like the above at the bottom of the /etc/sudoers file. From grant at mclean.net.nz Mon May 22 21:48:23 2006 From: grant at mclean.net.nz (Grant McLean) Date: Tue, 23 May 2006 16:48:23 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: <1148358092.31782.32.camel@localhost.localdomain> References: <1148357495.31782.28.camel@localhost.localdomain> <44728D3D.1060707@paradise.net.nz> <1148358092.31782.32.camel@localhost.localdomain> Message-ID: <1148359703.18743.18.camel@localhost.localdomain> On Tue, 2006-05-23 at 16:21 +1200, Lesley Walker wrote: > On Tue, 2006-05-23 at 16:19 +1200, Peter C. Kelly wrote: > > How about using sudo? > > How exactly would I go about doing that? This is a CGI script. Sudo is the right answer. The fact that it's a CGI script means that you'll need two scripts. Move that code that needs special permissions into its own script and then your CGI script can run that script via sudo and collect its output. For example your CGI script might include this line: my $log_data = `/usr/bin/sudo -u loguser /path/to/priviliged/script $args`; The sudoers file might include a line like this: www-data ALL = (loguser) NOPASSWD: /path/to/priviliged/script The net result is that the www-data user has the ability to run the privileged script as the user 'loguser'. The www-data user can't modify that script and is therefore limited to doing whatever that script allows. Using the setuid or setgid approach, the www-data user would have the ability to do anything that the user or group permissions allowed. Good luck Grant From LRW at clear.net.nz Mon May 22 21:53:04 2006 From: LRW at clear.net.nz (Lesley Walker) Date: Tue, 23 May 2006 16:53:04 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: <20060523044235.64A26868A93@wat.la.naos.co.nz> References: <20060523044235.64A26868A93@wat.la.naos.co.nz> Message-ID: <1148359984.31782.44.camel@localhost.localdomain> Thanks guys, much appreciated. I think I've got it now. The man page for sudoers was making my head hurt so I appreciate the examples especially. Lesley. From LRW at clear.net.nz Mon May 22 23:19:35 2006 From: LRW at clear.net.nz (Lesley Walker) Date: Tue, 23 May 2006 18:19:35 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: <1148359984.31782.44.camel@localhost.localdomain> References: <20060523044235.64A26868A93@wat.la.naos.co.nz> <1148359984.31782.44.camel@localhost.localdomain> Message-ID: <1148365176.31782.55.camel@localhost.localdomain> On Tue, 2006-05-23 at 16:53 +1200, Lesley Walker wrote: > Thanks guys, much appreciated. I think I've got it now. Argh! My script is now able to be called through the wrapper, but doing it this way appears to have caused something to break. We use some kind of magic involving the browser id string to identify users - should I expect that stuff to get passed through the wrapper to the real script, or do I have to do something specific to avoid having it dropped into the bit bucket? Sorry, I know this is getting a bit off-topic. If you think it's more appropriate to reply privately, please do. Lesley. From ewen at naos.co.nz Tue May 23 01:01:22 2006 From: ewen at naos.co.nz (Ewen McNeill) Date: Tue, 23 May 2006 20:01:22 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: Message from Lesley Walker of "Tue, 23 May 2006 18:19:35 +1200." <1148365176.31782.55.camel@localhost.localdomain> Message-ID: <20060523080122.D2C19868A93@wat.la.naos.co.nz> NOTE: I've replied to the list even though this is technically off-topic because it might be useful for someone else trying to do the same thing. In message <1148365176.31782.55.camel at localhost.localdomain>, Lesley Walker writes: >We use some kind of magic involving the browser id string to identify >users - should I expect that stuff to get passed through the wrapper to >the real script, or do I have to do something specific to avoid having >it dropped into the bit bucket? Assuming you're using the Debian sudo then beware that the security patched versions clense the environment significantly, allowing through only a restricted set of blessed variables. If you need specific things passed through either force them onto the command line for the privileged script to pick up, or "bless" those enviroment variables in the /etc/sudoers file. See the advisory for details: http://www.debian.org/security/2006/dsa-946 Ewen From grant at mclean.net.nz Tue May 23 01:05:33 2006 From: grant at mclean.net.nz (Grant McLean) Date: Tue, 23 May 2006 20:05:33 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: <1148365176.31782.55.camel@localhost.localdomain> References: <20060523044235.64A26868A93@wat.la.naos.co.nz> <1148359984.31782.44.camel@localhost.localdomain> <1148365176.31782.55.camel@localhost.localdomain> Message-ID: <1148371533.12965.7.camel@localhost.localdomain> On Tue, 2006-05-23 at 18:19 +1200, Lesley Walker wrote: > On Tue, 2006-05-23 at 16:53 +1200, Lesley Walker wrote: > > Thanks guys, much appreciated. I think I've got it now. > > Argh! > > My script is now able to be called through the wrapper, but doing it > this way appears to have caused something to break. > > We use some kind of magic involving the browser id string to identify > users - should I expect that stuff to get passed through the wrapper to > the real script, or do I have to do something specific to avoid having > it dropped into the bit bucket? Whether sudo passes environment variables through or not will depend on the configuration in the sudoers file and the compile-time options when sudo was built. I have a feeling that the Sarge build of sudo is fairly restrictive by default. If you wanted to pass an environment variable called (for example) REMOTE_ADDRESS then you could add an entry like this to sudoers: Defaults:www-data env_keep += REMOTE_ADDRESS Having said that, it might be safest to do all the environment and argument parsing in the CGI script and pass anything to the other script via command-line arguments. It would certainly be easier to debug. Regards Grant From LRW at clear.net.nz Tue May 23 14:42:08 2006 From: LRW at clear.net.nz (Lesley Walker) Date: Wed, 24 May 2006 09:42:08 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: <20060523080122.D2C19868A93@wat.la.naos.co.nz> References: <20060523080122.D2C19868A93@wat.la.naos.co.nz> Message-ID: <1148420528.30988.12.camel@localhost.localdomain> Thanks guys, This has pretty much turned out to be an Apache question, so I'll try that keep-env thing, and if that doesn't do what I need I'll go and search some Apache mailing lists. Much appreciated. LesleyW From grant at mclean.net.nz Tue May 23 15:26:31 2006 From: grant at mclean.net.nz (Grant McLean) Date: Wed, 24 May 2006 10:26:31 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: <1148420528.30988.12.camel@localhost.localdomain> References: <20060523080122.D2C19868A93@wat.la.naos.co.nz> <1148420528.30988.12.camel@localhost.localdomain> Message-ID: <1148423191.2642.12.camel@localhost.localdomain> On Wed, 2006-05-24 at 09:42 +1200, Lesley Walker wrote: > Thanks guys, > This has pretty much turned out to be an Apache question, Are you sure? Can the CGI script not get the info you need? > so I'll try that keep-env thing, and if that doesn't do what > I need I'll go and search some Apache mailing lists. Using keep-env wouldn't be my preferred approach. My vote would be for the CGI script to get the info it needs and pass it to the priviliged script on the command line. Passing things via the environment is non-obvious enough without also having to rely on some obscure entry in the sudoers file. It may seem like the path of least resistance was Peter's suggestion of not changing the CGI script but just adding a CGI wrapper that invokes the original script via sudo. But from a security perspective it would be safer to keep the bulk of your code in the non-priviliged CGI script and only extract the specific part which needs special permissions. If you just need to read a file then script invoked via sudo would be very short and very easy to audit. Cheers Grant From srdjan at catalyst.net.nz Tue May 23 15:34:20 2006 From: srdjan at catalyst.net.nz (Srdjan) Date: Wed, 24 May 2006 10:34:20 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: <1148423191.2642.12.camel@localhost.localdomain> References: <20060523080122.D2C19868A93@wat.la.naos.co.nz> <1148420528.30988.12.camel@localhost.localdomain> <1148423191.2642.12.camel@localhost.localdomain> Message-ID: <44738DEC.3090806@catalyst.net.nz> But what if you have sensitive data you don't want to be seen on the command line, like passwords? Grant McLean wrote: > On Wed, 2006-05-24 at 09:42 +1200, Lesley Walker wrote: >> Thanks guys, >> This has pretty much turned out to be an Apache question, > > Are you sure? Can the CGI script not get the info you need? > >> so I'll try that keep-env thing, and if that doesn't do what >> I need I'll go and search some Apache mailing lists. > > Using keep-env wouldn't be my preferred approach. My vote would be for > the CGI script to get the info it needs and pass it to the priviliged > script on the command line. Passing things via the environment is > non-obvious enough without also having to rely on some obscure entry in > the sudoers file. > > It may seem like the path of least resistance was Peter's suggestion of > not changing the CGI script but just adding a CGI wrapper that invokes > the original script via sudo. But from a security perspective it would > be safer to keep the bulk of your code in the non-priviliged CGI script > and only extract the specific part which needs special permissions. If > you just need to read a file then script invoked via sudo would be very > short and very easy to audit. > > Cheers > Grant > > > > _______________________________________________ > Wellington-pm mailing list > Wellington-pm at pm.org > http://mail.pm.org/mailman/listinfo/wellington-pm > > From grant at mclean.net.nz Tue May 23 15:57:56 2006 From: grant at mclean.net.nz (Grant McLean) Date: Wed, 24 May 2006 10:57:56 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: <44738DEC.3090806@catalyst.net.nz> References: <20060523080122.D2C19868A93@wat.la.naos.co.nz> <1148420528.30988.12.camel@localhost.localdomain> <1148423191.2642.12.camel@localhost.localdomain> <44738DEC.3090806@catalyst.net.nz> Message-ID: <1148425077.2642.19.camel@localhost.localdomain> On Wed, 2006-05-24 at 10:34 +1200, Srdjan wrote: > Grant McLean wrote: > > Passing things via the environment is > > non-obvious enough without also having to rely on some obscure entry in > > the sudoers file. > > But what if you have sensitive data you don't want to be seen on the command > line, like passwords? Yes, passing a password between processes via command line argument would be insecure, since the command line arguments are visible via 'ps'. On the other hand, passing a password between processes via the environment would also be insecure, since a process' environment is also visible via 'ps'. Sending it into the STDIN of the child process via a pipe would be one more secure approach. Cheers Grant From sam at vilain.net Tue May 23 17:15:30 2006 From: sam at vilain.net (Sam Vilain) Date: Wed, 24 May 2006 12:15:30 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: <1148357495.31782.28.camel@localhost.localdomain> References: <1148357495.31782.28.camel@localhost.localdomain> Message-ID: <4473A5A2.5090109@vilain.net> Lesley Walker wrote: >Googling finds a solution, which is to install perl-suid. However, the >package decription says: > Usage of this program is now strongly deprecated upstream and > support (along with this package) will probably be removed in > 5.10 > >Therefore I'd like to find a solution OTHER than perl-suid. Is there >some other package or module that can be used instead? > > I know this issue is already mostly gone, but the FastCGI module for Apache has not succumed to this notion that "suid is just bad, m'kay?". So long as you haven't tied your application to the Apache API, it's a fairly easy switch. See FastCGISuExec in the mod_fastcgi manual page. Unlike mod_perl, with fastcgi you can get high performance without depending on a specific webserver with notoriously fragile configuration method, and without sacrificing the security of unix user seperation. Sam. From sam at vilain.net Tue May 23 17:16:30 2006 From: sam at vilain.net (Sam Vilain) Date: Wed, 24 May 2006 12:16:30 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: <4473A5A2.5090109@vilain.net> References: <1148357495.31782.28.camel@localhost.localdomain> <4473A5A2.5090109@vilain.net> Message-ID: <4473A5DE.9010009@vilain.net> Sam Vilain wrote: >So long as you haven't tied your application to the Apache API, it's a >fairly easy switch. See FastCGISuExec in the mod_fastcgi manual page. > > ...and you have to 'use CGI::Fast' instead of 'use CGI' From LRW at clear.net.nz Tue May 23 17:41:48 2006 From: LRW at clear.net.nz (Lesley Walker) Date: Wed, 24 May 2006 12:41:48 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: <1148423191.2642.12.camel@localhost.localdomain> References: <20060523080122.D2C19868A93@wat.la.naos.co.nz> <1148420528.30988.12.camel@localhost.localdomain> <1148423191.2642.12.camel@localhost.localdomain> Message-ID: <1148431308.30988.30.camel@localhost.localdomain> On Wed, 2006-05-24 at 10:26 +1200, Grant McLean wrote: > On Wed, 2006-05-24 at 09:42 +1200, Lesley Walker wrote: > > Thanks guys, > > This has pretty much turned out to be an Apache question, > > Are you sure? Can the CGI script not get the info you need? I'm still trying to figure out (1) which variables I need and (2) how I would get the POST data through. This seems very messy. > It may seem like the path of least resistance was Peter's suggestion of > not changing the CGI script but just adding a CGI wrapper that invokes > the original script via sudo. But from a security perspective it would > be safer to keep the bulk of your code in the non-priviliged CGI script > and only extract the specific part which needs special permissions. If > you just need to read a file then script invoked via sudo would be very > short and very easy to audit. I think I prefer this approach as well - ie instead of doing "open LOG, $logfile" I need to open a pipe from a command, right? But I'm having trouble getting this to work as well. I have written a little script catmaillog (which obviously chooses the correct log and cats it), and I can "sudo catmaillog" from the command line as www-data and it works correctly. But when I run myscript.pl, either from the command line or via Apache, I get "cannot open: Permission denied". Oh well, at least it's a little more Perl-relevant now... LesleyW From LRW at clear.net.nz Tue May 23 18:03:23 2006 From: LRW at clear.net.nz (Lesley Walker) Date: Wed, 24 May 2006 13:03:23 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: <4473A5A2.5090109@vilain.net> References: <1148357495.31782.28.camel@localhost.localdomain> <4473A5A2.5090109@vilain.net> Message-ID: <1148432604.30988.35.camel@localhost.localdomain> Thanks Sam, I'll check that out. Even though I'm stubbornly trying to make my script work without suid, there are other affected scripts which will eventually need to be fixed too. On Wed, 2006-05-24 at 12:15 +1200, Sam Vilain wrote: > I know this issue is already mostly gone, but the FastCGI module for > Apache has not succumed to this notion that "suid is just bad, m'kay?". > > So long as you haven't tied your application to the Apache API, it's a > fairly easy switch. See FastCGISuExec in the mod_fastcgi manual page. > > Unlike mod_perl, with fastcgi you can get high performance without > depending on a specific webserver with notoriously fragile configuration > method, and without sacrificing the security of unix user seperation. From LRW at clear.net.nz Tue May 23 21:23:36 2006 From: LRW at clear.net.nz (Lesley Walker) Date: Wed, 24 May 2006 16:23:36 +1200 Subject: [Wellington-pm] perl-suid deprecated, in favour of what? In-Reply-To: <1148431308.30988.30.camel@localhost.localdomain> References: <20060523080122.D2C19868A93@wat.la.naos.co.nz> <1148420528.30988.12.camel@localhost.localdomain> <1148423191.2642.12.camel@localhost.localdomain> <1148431308.30988.30.camel@localhost.localdomain> Message-ID: <1148444616.30988.38.camel@localhost.localdomain> On Wed, 2006-05-24 at 12:41 +1200, Lesley Walker wrote: > I think I prefer this approach as well - ie instead of doing "open LOG, > $logfile" I need to open a pipe from a command, right? > > But I'm having trouble getting this to work as well. I'll take that back - it's working fine. (I had another open command that was failing, which caused me some confusion.) Lesley. From grant at mclean.net.nz Wed May 24 01:26:06 2006 From: grant at mclean.net.nz (Grant McLean) Date: Wed, 24 May 2006 20:26:06 +1200 Subject: [Wellington-pm] Roger, the penguin Message-ID: <1148459166.14034.1.camel@localhost.localdomain> http://www.perlmonks.org/index.pl?node_id=494927 Well it made me laugh. From george at metaart.org Wed May 24 01:56:40 2006 From: george at metaart.org (George Woolley) Date: Wed, 24 May 2006 01:56:40 -0700 Subject: [Wellington-pm] Roger, the penguin In-Reply-To: <1148459166.14034.1.camel@localhost.localdomain> References: <1148459166.14034.1.camel@localhost.localdomain> Message-ID: <200605240156.41046.george@metaart.org> Thanks, it made me chuckle too. I've added it to http://wiki.oreillynet.com/usergroups/index.cgi?LanguageFun There's a fair amount listed uner Perl there. Is there anything else I should add? George On Wednesday 24 May 2006 01:26, Grant McLean wrote: > http://www.perlmonks.org/index.pl?node_id=494927 > > Well it made me laugh. > > _______________________________________________ > Wellington-pm mailing list > Wellington-pm at pm.org > http://mail.pm.org/mailman/listinfo/wellington-pm From grant at mclean.net.nz Wed May 24 19:52:00 2006 From: grant at mclean.net.nz (Grant McLean) Date: Thu, 25 May 2006 14:52:00 +1200 Subject: [Wellington-pm] Still need a volunteer for the June meeting Message-ID: <1148525520.26176.8.camel@localhost.localdomain> Hi Mongers The next meeting of Wellington Perl Mongers is two and a half weeks away on June 13th. I'm planning to do a talk on Test-Driven Development but that still leaves room for another speaker. Srdjan suggested he would like to have someone convince him he ought to be using exceptions rather than mucking around with checking return values. That sounds like an excellent idea to me and wouldn't need to take more than 10-15 minutes. Please speak up now if you'd like to volunteer to speak on that, or any other, topic. Also, if there's anyone out there interested in Ruby on Rails, note that there's a group looking to set up regular meetings to share info on that topic: http://weblog.rubyonrails.org/articles/2006/05/19/wellington-rails-users-group Cheers Grant From douglas at paradise.net.nz Wed May 24 20:23:03 2006 From: douglas at paradise.net.nz (Douglas Bagnall) Date: Thu, 25 May 2006 15:23:03 +1200 Subject: [Wellington-pm] Still need a volunteer for the June meeting In-Reply-To: <1148525520.26176.8.camel@localhost.localdomain> References: <1148525520.26176.8.camel@localhost.localdomain> Message-ID: <44752317.4090802@paradise.net.nz> Grant McLean wrote: > Hi Mongers > > The next meeting of Wellington Perl Mongers is two and a half weeks away > on June 13th. I think Sam was saying he'd like to talk about FastCGI, comparing it to modperl and other methods such as plain CGI, SCGI and http proxying to a perl server. but maybe that was a wishful misreading. douglas From sam at vilain.net Wed May 24 20:32:06 2006 From: sam at vilain.net (Sam Vilain) Date: Thu, 25 May 2006 15:32:06 +1200 Subject: [Wellington-pm] Still need a volunteer for the June meeting In-Reply-To: <44752317.4090802@paradise.net.nz> References: <1148525520.26176.8.camel@localhost.localdomain> <44752317.4090802@paradise.net.nz> Message-ID: <44752536.3090400@vilain.net> Douglas Bagnall wrote: >Grant McLean wrote: > > >>Hi Mongers >> >>The next meeting of Wellington Perl Mongers is two and a half weeks away >>on June 13th. >> >> > >I think Sam was saying he'd like to talk about FastCGI, comparing it to >modperl and other methods such as plain CGI, SCGI and http proxying to a >perl server. > >but maybe that was a wishful misreading. > > I could mention it in passing in a Catalyst.pm talk, perhaps? Sam. From finlay at catalyst.net.nz Wed May 24 20:34:42 2006 From: finlay at catalyst.net.nz (Finlay Thompson) Date: Thu, 25 May 2006 15:34:42 +1200 Subject: [Wellington-pm] Still need a volunteer for the June meeting In-Reply-To: <44752536.3090400@vilain.net> References: <1148525520.26176.8.camel@localhost.localdomain> <44752317.4090802@paradise.net.nz> <44752536.3090400@vilain.net> Message-ID: <447525D2.4040604@catalyst.net.nz> I'd be interested in a Catalyst.pm talk (as in the MVC) Finlay Sam Vilain wrote: >Douglas Bagnall wrote: > > > >>Grant McLean wrote: >> >> >> >> >>>Hi Mongers >>> >>>The next meeting of Wellington Perl Mongers is two and a half weeks away >>>on June 13th. >>> >>> >>> >>> >>I think Sam was saying he'd like to talk about FastCGI, comparing it to >>modperl and other methods such as plain CGI, SCGI and http proxying to a >>perl server. >> >>but maybe that was a wishful misreading. >> >> >> >> > >I could mention it in passing in a Catalyst.pm talk, perhaps? > >Sam. >_______________________________________________ >Wellington-pm mailing list >Wellington-pm at pm.org >http://mail.pm.org/mailman/listinfo/wellington-pm > > > From michael at diaspora.gen.nz Wed May 24 20:50:48 2006 From: michael at diaspora.gen.nz (michael at diaspora.gen.nz) Date: Thu, 25 May 2006 15:50:48 +1200 Subject: [Wellington-pm] Still need a volunteer for the June meeting In-Reply-To: Your message of "Thu, 25 May 2006 15:34:42 +1200." <447525D2.4040604@catalyst.net.nz> Message-ID: Finlay Thompson writes: >I'd be interested in a Catalyst.pm talk (as in the MVC) Andrew Ruthven talked about in last November, didn't he? -- michael. From andrew.ruthven at catalyst.net.nz Wed May 24 20:54:31 2006 From: andrew.ruthven at catalyst.net.nz (Andrew Ruthven) Date: Thu, 25 May 2006 15:54:31 +1200 Subject: [Wellington-pm] Still need a volunteer for the June meeting In-Reply-To: References: Message-ID: <1148529271.531.1.camel@dirk.catalyst.net.nz> On Thu, 2006-05-25 at 15:50 +1200, michael at diaspora.gen.nz wrote: > Finlay Thompson writes: > >I'd be interested in a Catalyst.pm talk (as in the MVC) > > Andrew Ruthven talked about in last November, didn't he? Yes, yes I did. But things have change a bit in the Catalyst world... -- Andrew Ruthven, Wellington, New Zealand Catalyst IT Limited --> http://www.catalyst.net.nz At work: andrew.ruthven at catalyst.net.nz At home: andrew at etc.gen.nz GPG fpr: 34CA 12A3 C6F8 B156 72C2 D0D7 D286 CE0C 0C62 B791 From LRW at clear.net.nz Wed May 24 21:09:43 2006 From: LRW at clear.net.nz (Lesley Walker) Date: Thu, 25 May 2006 16:09:43 +1200 Subject: [Wellington-pm] Still need a volunteer for the June meeting In-Reply-To: <1148529271.531.1.camel@dirk.catalyst.net.nz> References: <1148529271.531.1.camel@dirk.catalyst.net.nz> Message-ID: <1148530183.30988.82.camel@localhost.localdomain> On Thu, 2006-05-25 at 15:54 +1200, Andrew Ruthven wrote: > On Thu, 2006-05-25 at 15:50 +1200, michael at diaspora.gen.nz wrote: > > Finlay Thompson writes: > > >I'd be interested in a Catalyst.pm talk (as in the MVC) > > > > Andrew Ruthven talked about in last November, didn't he? > > Yes, yes I did. But things have change a bit in the Catalyst world... I must have missed that one. Sounds interesting. LW From enkidu at cliffp.com Sat May 27 22:39:23 2006 From: enkidu at cliffp.com (Cliff Pratt) Date: Sun, 28 May 2006 17:39:23 +1200 Subject: [Wellington-pm] Referencing arrays in an object Message-ID: <4479378B.8010009@cliffp.com> Hi all, Can someone please clarify something for me? I have an object for which I am creating accessor methods using Class::Accessor::Fast. One of the methods is called 'grid' initialised so: $self->grid([]) ; I can access the elements of the array referred to by $self->grid by the following: $xxxx = $self->grid->[$i] ; # Within a method of the class I'm not using any accessor method here, am I? I theenk that the accessor method will be something like the following: $xxxx = $self->grid([$i]) ; # Will this work? Obviously I'm going to give it a crack, but a little insight from one of the gurus would be appreciated. Cheers, Cliff -- http://barzoomian.blogspot.com From grant at mclean.net.nz Sun May 28 00:35:46 2006 From: grant at mclean.net.nz (Grant McLean) Date: Sun, 28 May 2006 19:35:46 +1200 Subject: [Wellington-pm] Referencing arrays in an object In-Reply-To: <4479378B.8010009@cliffp.com> References: <4479378B.8010009@cliffp.com> Message-ID: <1148801746.8427.14.camel@localhost.localdomain> On Sun, 2006-05-28 at 17:39 +1200, Cliff Pratt wrote: > One of the methods is called 'grid' initialised so: > > $self->grid([]) ; > > I can access the elements of the array referred to by $self->grid by the > following: > > $xxxx = $self->grid->[$i] ; # Within a method of the class > > I'm not using any accessor method here, am I? Yes, you are. $self->grid is calling the 'grid' accessor which is returning an array reference. You are then indexing into the referenced array by appending ->[$i] If you wanted to foreach through all the elements in the array you could do it like this: foreach my $cell ( @{ $self->grid } ) { # do something with $cell } > I theenk that the accessor > method will be something like the following: > > $xxxx = $self->grid([$i]) ; # Will this work? No, in this case you're calling the grid method and passing it a reference to a newly created array with one element, which contains the current value of $i. From what you said earlier, the effect will be overwriting the whole 'grid' array. You could conceivably write your own accessor method that took the grid index as an argument: $self->grid_cell($i) It's hard to say whether that would offer any advantage in your case. Regards Grant From enkidu at cliffp.com Sun May 28 01:39:13 2006 From: enkidu at cliffp.com (Cliff Pratt) Date: Sun, 28 May 2006 20:39:13 +1200 Subject: [Wellington-pm] Referencing arrays in an object In-Reply-To: <1148801746.8427.14.camel@localhost.localdomain> References: <4479378B.8010009@cliffp.com> <1148801746.8427.14.camel@localhost.localdomain> Message-ID: <447961B1.7080404@cliffp.com> Grant McLean wrote: > On Sun, 2006-05-28 at 17:39 +1200, Cliff Pratt wrote: > >>One of the methods is called 'grid' initialised so: >> >>$self->grid([]) ; >> >>I can access the elements of the array referred to by $self->grid by the >>following: >> >>$xxxx = $self->grid->[$i] ; # Within a method of the class >> >>I'm not using any accessor method here, am I? > > Yes, you are. $self->grid is calling the 'grid' accessor which is > returning an array reference. You are then indexing into the referenced > array by appending ->[$i] > Ah, thanks for clarifying that. > > If you wanted to foreach through all the elements in the array you could > do it like this: > > foreach my $cell ( @{ $self->grid } ) { > # do something with $cell > } > > >>I theenk that the accessor >>method will be something like the following: >> >>$xxxx = $self->grid([$i]) ; # Will this work? > > No, in this case you're calling the grid method and passing it a > reference to a newly created array with one element, which contains the > current value of $i. From what you said earlier, the effect will be > overwriting the whole 'grid' array. > Ugh! That's not what I want! > > You could conceivably write your own accessor method that took the grid > index as an argument: > > $self->grid_cell($i) > > It's hard to say whether that would offer any advantage in your case. > None really. That Class::Accessor::Fast module is pretty cool! Cheers, Cliff -- http://barzoomian.blogspot.com From jarich at perltraining.com.au Mon May 29 01:16:26 2006 From: jarich at perltraining.com.au (Jacinta Richardson) Date: Mon, 29 May 2006 18:16:26 +1000 Subject: [Wellington-pm] Open Source Developers' Conference 2006 - Call for papers Message-ID: <447AADDA.6000003@perltraining.com.au> I invite you all to join us at the conference this year! http://www.osdc.com.au/papers/cfp06.html The Open Source Developers' Conference is an Australian conference designed for developers, by developers. It covers numerous programming languages across a range of operating systems. We're seeking papers on Open Source languages, technologies, projects and tools as well as topics of interest to Open Source developers. The conference will be held in Melbourne, Victoria (Monash University's Caulfield Campus) from the 6th to the 8th of December, 2006. Last year's conference had about 160 people and around 60 presentations on a range of topics - see http://osdc2005.cgpublisher.com/proposals/ for a list. This list might also be useful if you're looking for ideas on what sort of thing would be appropriate. If you have any questions, or have never submitted a paper proposal before, please read our FAQ page at http://www.osdc.com.au/faq/index.html If you don't find an answer there, please contact richard at osdc.com.au To submit a proposal, follow the instructions at http://www.osdc.com.au/papers/cfp06.html This year we're also going to run a day of tutorials. See the CFP for more information. The deadline for proposals is 12th July 2006. Hope to see you there! The OSDC 2006 committee. From michael at diaspora.gen.nz Tue May 30 18:24:38 2006 From: michael at diaspora.gen.nz (michael at diaspora.gen.nz) Date: Wed, 31 May 2006 13:24:38 +1200 Subject: [Wellington-pm] Presentations next month. Message-ID: I went through the page at http://wellington.pm.org/archive/index.html, and extracted the names of past presenters. After cleaning up some duplicates courtesy of mis-spellings, we end up with the following list, counted. I think it's clear that others need to step up to the plate, or that we should rename ourselves the Grant McLean Fan Club :). (Unfortunately, I can't help, as I really really don't have the time this month. And I presented last month. So there. Maybe next month, if I can think of anything.) -- michael. The list: 19 Grant McLean 8 Sam Vilain 6 Michael Robinson 4 Finlay Thompson 3 Geoff Cant 2 Peter Love 2 Srdjan Jankovic 1 Finaly Thompson 1 Dan Morrison 1 Dave Moskovitz 1 Douglas Bagnall 1 Simon Green 1 Andrew MacMillan 1 Ewen McNeill 1 Don Christie 1 Martin Langhoff 1 Matt Hunt 1 Andrew McMillan 1 Steve Baker 1 Stephen Judd 1 Andrew Ruthven 1 Leslie Walker 1 Peter Kelly And the code: #!/usr/bin/perl use LWP::Simple; use strict; my $str = get('http://wellington.pm.org/archive/'); my %presenters; while ($str =~ m{ ^ \s* # start of line, with optional whitespace ]* > # td tag, with optional attributes \s* (\S+) # CAPTURED: first name [ \t]+ # whitespace, but not a newline (\S+) # CAPTURED: last name \s* # closing tag \s* $ # end of line }xmsg) { next if $1 eq " $presenters{$a} } keys %presenters; print "\n"; From matt at catalyst.net.nz Tue May 30 18:37:14 2006 From: matt at catalyst.net.nz (Matthew Hunt) Date: Wed, 31 May 2006 13:37:14 +1200 Subject: [Wellington-pm] Presentations next month. In-Reply-To: References: Message-ID: <1149039434.1771.10.camel@localhost.localdomain> On Wed, 2006-05-31 at 13:24 +1200, michael at diaspora.gen.nz wrote: > I think it's clear that others need to step up to the plate, or that we > should rename ourselves the Grant McLean Fan Club :). The latter sounds easier to me. All in favour, email 'Aye' Can we get badges? Matt. -- Matthew Hunt Catalyst IT Limited, PO Box 11053, Wellington 6034 Phone: +64 (4) 499 2267, Direct: +64 (4) 803 2216, Fax: +64 (4) 499 5596 http://catalyst.net.nz/ From srdjan at catalyst.net.nz Tue May 30 18:59:53 2006 From: srdjan at catalyst.net.nz (Srdjan) Date: Wed, 31 May 2006 13:59:53 +1200 Subject: [Wellington-pm] Presentations next month. In-Reply-To: <1149039434.1771.10.camel@localhost.localdomain> References: <1149039434.1771.10.camel@localhost.localdomain> Message-ID: <447CF899.3060601@catalyst.net.nz> Arrrr, arrr, me harties, the Great Captain McLean's ship is sailing out... Matthew Hunt wrote: > On Wed, 2006-05-31 at 13:24 +1200, michael at diaspora.gen.nz wrote: >> I think it's clear that others need to step up to the plate, or that we >> should rename ourselves the Grant McLean Fan Club :). > > The latter sounds easier to me. All in favour, email 'Aye' > > Can we get badges? > > Matt. From grant at mclean.net.nz Tue May 30 19:06:07 2006 From: grant at mclean.net.nz (Grant McLean) Date: Wed, 31 May 2006 14:06:07 +1200 Subject: [Wellington-pm] Presentations next month. In-Reply-To: References: Message-ID: <1149041167.15742.3.camel@localhost.localdomain> On Wed, 2006-05-31 at 13:24 +1200, michael at diaspora.gen.nz wrote: > I went through the page at http://wellington.pm.org/archive/index.html, > and extracted the names of past presenters. After cleaning up some > duplicates courtesy of mis-spellings ... Oh dear, hopefully the spelling crimes have all now been corrected (and the HTML tidied up too). Let me know if you spot any more. Cheers Grant From Peter.Love at netkno.com Tue May 30 19:22:44 2006 From: Peter.Love at netkno.com (Peter Love) Date: Wed, 31 May 2006 14:22:44 +1200 Subject: [Wellington-pm] Presentations next month. In-Reply-To: References: Message-ID: <447CFDF4.6070003@netkno.com> If there's interest, I'd be happy to give a talk about "How I make a living using Perl". It would probably be partly about running a business (so not so Perlish) and/or partly about what it is I produce (which is written in Perl) and also be a blatant advertisement to see if anyone out there wants some work. Or something like that. Peter From grant at mclean.net.nz Tue May 30 19:38:09 2006 From: grant at mclean.net.nz (Grant McLean) Date: Wed, 31 May 2006 14:38:09 +1200 Subject: [Wellington-pm] Presentations next month. In-Reply-To: <447CFDF4.6070003@netkno.com> References: <447CFDF4.6070003@netkno.com> Message-ID: <1149043089.15759.6.camel@localhost.localdomain> On Wed, 2006-05-31 at 14:22 +1200, Peter Love wrote: > If there's interest, I'd be happy to give a talk about "How I make a > living using Perl". > > It would probably be partly about running a business (so not so Perlish) > and/or partly about what it is I produce (which is written in Perl) and > also be a blatant advertisement to see if anyone out there wants some work. All of which sounds good. Could you do it on the 13th or are you volunteering for a later month? Cheers Grant From Peter.Love at netkno.com Tue May 30 19:57:47 2006 From: Peter.Love at netkno.com (Peter Love) Date: Wed, 31 May 2006 14:57:47 +1200 Subject: [Wellington-pm] Presentations next month. In-Reply-To: <1149043089.15759.6.camel@localhost.localdomain> References: <447CFDF4.6070003@netkno.com> <1149043089.15759.6.camel@localhost.localdomain> Message-ID: <447D062B.5020300@netkno.com> > All of which sounds good. Could you do it on the 13th or are you > volunteering for a later month? 13th June is fine. From grant at mclean.net.nz Tue May 30 20:01:12 2006 From: grant at mclean.net.nz (Grant McLean) Date: Wed, 31 May 2006 15:01:12 +1200 Subject: [Wellington-pm] Presentations next month. In-Reply-To: <447D062B.5020300@netkno.com> References: <447CFDF4.6070003@netkno.com> <1149043089.15759.6.camel@localhost.localdomain> <447D062B.5020300@netkno.com> Message-ID: <1149044472.17131.8.camel@localhost.localdomain> On Wed, 2006-05-31 at 14:57 +1200, Peter Love wrote: > > All of which sounds good. Could you do it on the 13th or are you > > volunteering for a later month? > > 13th June is fine. Excellent - that's now a confirmed booking. Cheers Grant From grant at mclean.net.nz Wed May 31 03:45:56 2006 From: grant at mclean.net.nz (Grant McLean) Date: Wed, 31 May 2006 22:45:56 +1200 Subject: [Wellington-pm] CPAN Mirror (and shell) Message-ID: <1149072357.25650.9.camel@localhost.localdomain> Hi Mongers I'm don't recall if there's been an announcement here that Catalyst is now hosting a local CPAN mirror, but since this follows on from a question raised during Srdjan's CPAN talk, I thought I'd mention it here ... The URL of the mirror is: http://cpan.catalyst.net.nz/CPAN/ It is publicly accessible, but it hasn't yet been added to the official mirrors file. To configure the standard CPAN shell to use this mirror, use these three simple steps: 1. Start the shell using either this command: cpan or if that doesn't work, this one: perl -MCPAN -e shell 2. Use this command to select the mirror for the current session o conf urllist = http://cpan.catalyst.net.nz/CPAN/ 3. Use this command to save the changes o conf commit (Note: those last two commands each start with a lower case letter 'o' followed by a space - intuitive huh?) Cheers Grant