From hartzell at alerce.com Fri Mar 1 07:39:56 2013 From: hartzell at alerce.com (George Hartzell) Date: Fri, 1 Mar 2013 07:39:56 -0800 Subject: [sf-perl] problem use Regexp::Common, "Can't Regexp::Common::FIRSTKEY"... Message-ID: <20784.52172.92871.397633@gargle.gargle.HOWL> I'm trying to be a good boy and reuse regular expressions from the Regexp::Common world (ultimately $RE{time}{american}). I have things working, but as I was playing around in the debugger I discovered something "funny". It seems that any time I try to touch %RE I get an error, sometimes things Go Badly enough that I end up back at my shell prompt. It blows up with pretty much the same message in 5.14.2, 5.14.3 and 5.16.2 on a current Mac OS X system. (alacrity)[17:26:03]~>>perl -d -e '' Loading DB routines from perl5db.pl version 1.33 Editor support available. Enter h or `h h' for help, or `man perldebug' for more help. Debugged program terminated. Use q to quit or R to restart, use o inhibit_exit to avoid stopping after program termination, h q, h R or h o to get additional info. DB<1> use strict DB<2> use warnings DB<3> use Regexp::Common qw(number) DB<4> x $RE{num}{int} 0 Regexp::Common=HASH(0x7fad2c02d828) Can't Regexp::Common::FIRSTKEY at /Users/hartzell/perl5/perlbrew/perls/perl-5.14.3/lib/5.14.3/dumpvar.pl line 205 END failed--call queue aborted. at -e line 0 This doesn't seem to keep my code from working but it smells funny.... Can anyone shed any light on what's going on? g. From russt at releasetools.org Fri Mar 1 09:22:34 2013 From: russt at releasetools.org (Russ Tremain) Date: Fri, 1 Mar 2013 09:22:34 -0800 Subject: [sf-perl] yapop Message-ID: Yet Another Perl Optimization Puzzle. Okay, I admit it - I'm weird. I wrote a program called sqlpj[1] that uses Tim Bunce's JDBC module, which is his experimental wrapper to expose java.sql classes via Inline::Java. I like JDBC because it is easy and it is ubiquitous. Everything is in one jar for each database. Every vendor has one. I don't like DBI:: as much, because it means a couple days stuck in the CPAN mud to get it installed for various databases. Frankly, that has always been a barrier to me, so I never bothered to spend much serious time with it. (JDBC.pm has similar (cpan) problems because there is an impedance mismatch somewhere which prevents it from installing cleanly. Long story short: work-around is to use perl 5.8.9 or lower. However, you only have to install it once, for all databases.) Writing a front-end to SQL in perl is a joy compared to writing it in java. As I derived the perl program from an earlier java program, I have some knowledge on the issue. For example, you can treat JDBC meta calls as data, and voila, you have a full jdbc meta-data explorer for the price of a cut-and-paste from the javadocs. All was well until I did a query that returned a million rows. Then things got dull, as I waited for the cursor to return. As usual, my initial idea about what was wrong was, well wrong. What's the saying? Measure twice, optimize once? So I did, and not surprisingly[2], all my time was getting spent in the loop that fetched from the JDBC Result set (java.sql.ResultSet): while ($rset->next()) { push @allrows, [&getRow($rset, $self->getXmlDisplay(), @colmap)]; } Here "rset" is a reference to an instance of java.sql.ResultSet. As you can see, all the work is done in getRow. Here is the relevant loop there: sub getRow { ... my $m = $rset->getMetaData(); my $colcnt = $m->getColumnCount(); for (my $ii = 1; $ii <= $colcnt; $ii++) { next unless ($colmap[$ii-1]); #skip if column is not selected #note - you have to do the fetch first, #which sets wasNull() for the current column. my $str = $rset->getString($ii); #if we are displaying xml rowsets... if ($xmldisplay) { #...then set SQL NULL elements to undef: push @data, ($rset->wasNull() ? undef : $str); } else { #otherwise, we will display the string "(NULL)": push @data, ($rset->wasNull() ? "(NULL)" : $str); } ... } Geez, lots of opportunities for optimization here, where should I start? What would you do to improve the performance here? I will post what I actually did in a couple of days to give you a chance to think about it. Hint: anything derived from $rset is handled by Inline::Java, which means at least one write/read cycle to a socket connected java VM. cheers, -Russ [1] I will be posting the source for sqlpj soon, but email me if you are interested. [2] After the fact, it was not surprising, as little is. -------------- next part -------------- An HTML attachment was scrubbed... URL: From fobispo at isc.org Fri Mar 1 09:35:48 2013 From: fobispo at isc.org (Francisco Obispo) Date: Fri, 1 Mar 2013 09:35:48 -0800 Subject: [sf-perl] yapop In-Reply-To: References: Message-ID: <24EF25D8-FDE6-4123-86F6-69F560F6E0EA@isc.org> On Mar 1, 2013, at 9:22 AM, Russ Tremain wrote: > Geez, lots of opportunities for optimization here, where should I start? I would suggest, by removing all Java-related code. > > What would you do to improve the performance here? Learn and improve your code using DBIx::Class > > I will post what I actually did in a couple of days to give you a chance to think about it. > > Hint: anything derived from $rset is handled by Inline::Java, which means at least one write/read cycle to a socket connected java VM. > > cheers, > -Russ I have seen programs written in pure Java, fail miserably on high performance database applications, mostly IO bound on context changes. If you really like the java code, I would suggest you use java, and not try to mix it with perl. Francisco Obispo Director of Applications and Services - ISC email: fobispo at isc.org Phone: +1 650 423 1374 || INOC-DBA *3557* NOC PGP KeyID = B38DB1BE From russt at releasetools.org Fri Mar 1 09:57:12 2013 From: russt at releasetools.org (Russ Tremain) Date: Fri, 1 Mar 2013 09:57:12 -0800 Subject: [sf-perl] yapop In-Reply-To: <24EF25D8-FDE6-4123-86F6-69F560F6E0EA@isc.org> References: <24EF25D8-FDE6-4123-86F6-69F560F6E0EA@isc.org> Message-ID: I agree with you that JDBC.pm is not a practical idea for production database work, though it works remarkably well. This is more of an academic exercise, which was the reason Tim Bunce originally wrote JDBC.pm - to help explore the java model as a means to inform future DBI development. Think of this as a pure exercise in optimization, given a constrained problem. tgif. At 9:35 AM -0800 3/1/13, Francisco Obispo wrote: >On Mar 1, 2013, at 9:22 AM, Russ Tremain wrote: > >> Geez, lots of opportunities for optimization here, where should I start? > >I would suggest, by removing all Java-related code. > >> >> What would you do to improve the performance here? > >Learn and improve your code using DBIx::Class > >> >> I will post what I actually did in a couple of days to give you a chance to think about it. >> >> Hint: anything derived from $rset is handled by Inline::Java, which means at least one write/read cycle to a socket connected java VM. >> >> cheers, >> -Russ > >I have seen programs written in pure Java, fail miserably on high performance database applications, mostly IO bound on context changes. If you really like the java code, I would suggest you use java, and not try to mix it with perl. > > >Francisco Obispo >Director of Applications and Services - ISC >email: fobispo at isc.org >Phone: +1 650 423 1374 || INOC-DBA *3557* NOC >PGP KeyID = B38DB1BE From quinn at pgexperts.com Fri Mar 1 10:55:29 2013 From: quinn at pgexperts.com (Quinn Weaver) Date: Fri, 1 Mar 2013 10:55:29 -0800 Subject: [sf-perl] problem use Regexp::Common, "Can't Regexp::Common::FIRSTKEY"... In-Reply-To: <20784.52172.92871.397633@gargle.gargle.HOWL> References: <20784.52172.92871.397633@gargle.gargle.HOWL> Message-ID: I don't have a full answer, but the immediate problem you're seeing is that the regex is a tied hash and one of the magic hash methods (FIRSTKEY) is not implemented. Why that bothers the debugger I don't know. If you want to read the underlying regular expression, try 'perldoc -l Regexp::Common' (using the same Perl as your program uses, of course), then chop the .pm off the result and look in that directory for time.pm or american.pm. Be warned: this may not help. Some regexes in Regexp::Common (e.g., profanity.pm) are so optimized that they look more like a series of letters and |'s than a regex you or I would write. If you find the Common regex unreadable, you might be better off writing your own after all. Good luck! Hope that helps a little. Regards, -- Quinn Weaver PostgreSQL Experts, Inc. http://pgexperts.com/ 1-888-743-9778 (my extension: 510) Sent from my phone; pardon my brevity. On Mar 1, 2013, at 7:39 AM, George Hartzell wrote: > > I'm trying to be a good boy and reuse regular expressions from the > Regexp::Common world (ultimately $RE{time}{american}). I have things > working, but as I was playing around in the debugger I discovered > something "funny". > > It seems that any time I try to touch %RE I get an error, sometimes > things Go Badly enough that I end up back at my shell prompt. > > It blows up with pretty much the same message in 5.14.2, 5.14.3 and > 5.16.2 on a current Mac OS X system. > > (alacrity)[17:26:03]~>>perl -d -e '' > > Loading DB routines from perl5db.pl version 1.33 > Editor support available. > > Enter h or `h h' for help, or `man perldebug' for more help. > > Debugged program terminated. Use q to quit or R to restart, > use o inhibit_exit to avoid stopping after program termination, > h q, h R or h o to get additional info. > DB<1> use strict > > DB<2> use warnings > > DB<3> use Regexp::Common qw(number) > > DB<4> x $RE{num}{int} > 0 Regexp::Common=HASH(0x7fad2c02d828) > Can't Regexp::Common::FIRSTKEY at /Users/hartzell/perl5/perlbrew/perls/perl-5.14.3/lib/5.14.3/dumpvar.pl line 205 > END failed--call queue aborted. > at -e line 0 > > This doesn't seem to keep my code from working but it smells funny.... > > Can anyone shed any light on what's going on? > > g. > _______________________________________________ > SanFrancisco-pm mailing list > SanFrancisco-pm at pm.org > http://mail.pm.org/mailman/listinfo/sanfrancisco-pm From quinn at pgexperts.com Fri Mar 1 11:32:55 2013 From: quinn at pgexperts.com (Quinn Weaver) Date: Fri, 1 Mar 2013 11:32:55 -0800 Subject: [sf-perl] problem use Regexp::Common, "Can't Regexp::Common::FIRSTKEY"... In-Reply-To: References: <20784.52172.92871.397633@gargle.gargle.HOWL> Message-ID: <89A08DCA-7211-41BB-8C43-5DF380172008@pgexperts.com> On Mar 1, 2013, at 10:55 AM, Quinn Weaver wrote: > I don't have a full answer, but the immediate problem you're seeing is that the regex is a tied hash and one of the magic hash methods (FIRSTKEY) is not implemented. Why that bothers the debugger I don't know. Oh, wait, I think I get it. I bet the debugger's x command is trying to display the tied hash by iterating over its elements, but it can't since FIRSTKEY is not implemented. Anyway, the most straightforward solution is still the same: read the regex from source. > > If you want to read the underlying regular expression, try 'perldoc -l Regexp::Common' (using the same Perl as your program uses, of course), then chop the .pm off the result and look in that directory for time.pm or american.pm. > > Be warned: this may not help. Some regexes in Regexp::Common (e.g., profanity.pm) are so optimized that they look more like a series of letters and |'s than a regex you or I would write. If you find the Common regex unreadable, you might be better off writing your own after all. > > Good luck! Hope that helps a little. > > Regards, > > -- > Quinn Weaver > PostgreSQL Experts, Inc. http://pgexperts.com/ > 1-888-743-9778 (my extension: 510) > Sent from my phone; pardon my brevity. > > On Mar 1, 2013, at 7:39 AM, George Hartzell wrote: > >> >> I'm trying to be a good boy and reuse regular expressions from the >> Regexp::Common world (ultimately $RE{time}{american}). I have things >> working, but as I was playing around in the debugger I discovered >> something "funny". >> >> It seems that any time I try to touch %RE I get an error, sometimes >> things Go Badly enough that I end up back at my shell prompt. >> >> It blows up with pretty much the same message in 5.14.2, 5.14.3 and >> 5.16.2 on a current Mac OS X system. >> >> (alacrity)[17:26:03]~>>perl -d -e '' >> >> Loading DB routines from perl5db.pl version 1.33 >> Editor support available. >> >> Enter h or `h h' for help, or `man perldebug' for more help. >> >> Debugged program terminated. Use q to quit or R to restart, >> use o inhibit_exit to avoid stopping after program termination, >> h q, h R or h o to get additional info. >> DB<1> use strict >> >> DB<2> use warnings >> >> DB<3> use Regexp::Common qw(number) >> >> DB<4> x $RE{num}{int} >> 0 Regexp::Common=HASH(0x7fad2c02d828) >> Can't Regexp::Common::FIRSTKEY at /Users/hartzell/perl5/perlbrew/perls/perl-5.14.3/lib/5.14.3/dumpvar.pl line 205 >> END failed--call queue aborted. >> at -e line 0 >> >> This doesn't seem to keep my code from working but it smells funny.... >> >> Can anyone shed any light on what's going on? >> >> g. >> _______________________________________________ >> SanFrancisco-pm mailing list >> SanFrancisco-pm at pm.org >> http://mail.pm.org/mailman/listinfo/sanfrancisco-pm > _______________________________________________ > SanFrancisco-pm mailing list > SanFrancisco-pm at pm.org > http://mail.pm.org/mailman/listinfo/sanfrancisco-pm From doom at kzsu.stanford.edu Fri Mar 1 16:05:29 2013 From: doom at kzsu.stanford.edu (Joseph Brenner) Date: Fri, 1 Mar 2013 16:05:29 -0800 Subject: [sf-perl] setlist for "Module show and tell" Message-ID: A list of the speakers at the "Module show and tell" session at the last meeting: (1) Eric Wolf Yote perl/js web app servers -- schema-less madyote.com coyocanid at gmail.com (2) Earl C. Ruby "Making Great Looking Applications" perl/js techniques for slick display of JSON/XML data (Mason, jquery, jwizard, droppy.js, jqGrid, d3.js) webcdr.com (3) Tatsuhiko Miyagawa cpanminus 1.6 New features: very flexible version specification Requesting an old version accesses BackPAN transparently Can install modules directly from github (4) George Hartzell Using Sub::Exporter to implement Tiny::DSL https://github.com/hartzell/dsl-tiny https://metacpan.org/module/Sub::Exporter (5) David Storrs Copper (think "pipes") Source/Sink architecture using Moose objects (6) Jeremy Zawodny Shows off a Mojolicious-based hack to broadcast stdout (7) Mike Doherty (visiting from Canada) Talked up his utf8::all, (and fielded the always plentiful questions about unicode) (8) Xiong Changnian Conway's Smart::Comments and his own Devel::Comments (a variant that can write to any filehandle, not just STDERR) (9) David Sharnoff Steam::Aggregate (for weblog processing, etc). From josh at agliodbs.com Wed Mar 6 11:53:29 2013 From: josh at agliodbs.com (Josh Berkus) Date: Wed, 06 Mar 2013 11:53:29 -0800 Subject: [sf-perl] April 16: Postgres+Perl Message-ID: <51379EB9.9070907@agliodbs.com> Folks, SF Postgres' April meetup will involve Perl webapps heavily. We have plenty of space, so feel free to sign up. http://www.meetup.com/postgresql-1/events/107756262/ -- Josh Berkus PostgreSQL Experts Inc. http://pgexperts.com From peter at thoeny.org Wed Mar 6 13:03:27 2013 From: peter at thoeny.org (Peter Thoeny) Date: Wed, 6 Mar 2013 13:03:27 -0800 Subject: [sf-perl] [meeting] Regular Expression talk at SVPerl tomorrow Thu 7pm Message-ID: Hi SF folks, I'll be giving a talk on Regular Expressions tomorrow at the Plug and Play Tech Center for SVPerl. Please join if interested, already 70+ people signed up. RSVP at http://www.meetup.com/SVPerl/events/89342932/ Cheers, Peter -- * Peter Thoeny - Peter[at]Thoeny.org * http://TWiki.org - is your team already TWiki enabled? * Knowledge cannot be managed, it can be discovered and shared * This e-mail is: (_) private (_) ask first (x) public From mehryar at mehryar.com Wed Mar 13 13:59:27 2013 From: mehryar at mehryar.com (mehryar) Date: Wed, 13 Mar 2013 13:59:27 -0700 (PDT) Subject: [sf-perl] Google Trends for Perl Message-ID: Sadly: http://www.google.com/trends/explore#q=perl but maybe Perl is really picking up in India :-) cheers, -Mehryar From peter at thoeny.org Wed Mar 13 15:05:20 2013 From: peter at thoeny.org (Peter Thoeny) Date: Wed, 13 Mar 2013 15:05:20 -0700 Subject: [sf-perl] Google Trends for Perl In-Reply-To: References: Message-ID: <9F1F92ED-65FF-4051-BF07-EECF8A4D6B7C@thoeny.org> No base for alarm, it's all relative - compare to other languages such as http://www.google.com/trends/explore#q=perl%2C%20php%2C %20javascript&cmpt=q Cheers, Peter On Mar 13, 2013, at 1:59 PM, mehryar wrote: > > Sadly: > > http://www.google.com/trends/explore#q=perl > > but maybe Perl is really picking up in India :-) > > cheers, > -Mehryar > > _______________________________________________ > SanFrancisco-pm mailing list > SanFrancisco-pm at pm.org > http://mail.pm.org/mailman/listinfo/sanfrancisco-pm -- * Peter Thoeny - Peter[at]Thoeny.org * http://TWiki.org - is your team already TWiki enabled? * Knowledge cannot be managed, it can be discovered and shared * This e-mail is: (_) private (x) ask first (_) public From fred at redhotpenguin.com Wed Mar 13 15:06:32 2013 From: fred at redhotpenguin.com (Fred Moyer) Date: Wed, 13 Mar 2013 15:06:32 -0700 Subject: [sf-perl] Google Trends for Perl In-Reply-To: References: Message-ID: On Wed, Mar 13, 2013 at 1:59 PM, mehryar wrote: > > Sadly: > > http://www.google.com/trends/explore#q=perl Even more sad is when I missed ET's visit in May 2007! http://www.google.com/trends/explore#q=extraterrestrials Just goes to show you that because it's on Google, doesn't mean it's true in the context you might initially believe. > > but maybe Perl is really picking up in India :-) > > cheers, > -Mehryar > > _______________________________________________ > SanFrancisco-pm mailing list > SanFrancisco-pm at pm.org > http://mail.pm.org/mailman/listinfo/sanfrancisco-pm From frimicc at gmail.com Wed Mar 13 21:08:27 2013 From: frimicc at gmail.com (Mike Friedman) Date: Wed, 13 Mar 2013 21:08:27 -0700 Subject: [sf-perl] Google Trends for Perl In-Reply-To: <9F1F92ED-65FF-4051-BF07-EECF8A4D6B7C@thoeny.org> References: <9F1F92ED-65FF-4051-BF07-EECF8A4D6B7C@thoeny.org> Message-ID: <687C00D2-50D6-44F6-833E-82095E9F4C86@gmail.com> Yep. Google explains that it's based on the number of searches vs. total number of Google searches. As long as Google traffic goes up overall, *any* term with steady query volume will go down over time. http://support.google.com/trends/answer/87285?hl=en I don't think anyone would argue that queries of the term "perl" are becoming more frequent faster than all Google traffic is growing. -- Mike ______________________________ Michael Friedman frimicc at gmail.com On Mar 13, 2013, at 3:05 PM, Peter Thoeny wrote: > No base for alarm, it's all relative - compare to other languages such as > http://www.google.com/trends/explore#q=perl%2C%20php%2C%20javascript&cmpt=q > > Cheers, > Peter > > > On Mar 13, 2013, at 1:59 PM, mehryar wrote: > >> >> Sadly: >> >> http://www.google.com/trends/explore#q=perl >> >> but maybe Perl is really picking up in India :-) >> >> cheers, >> -Mehryar >> >> _______________________________________________ >> SanFrancisco-pm mailing list >> SanFrancisco-pm at pm.org >> http://mail.pm.org/mailman/listinfo/sanfrancisco-pm > > -- > * Peter Thoeny - Peter[at]Thoeny.org > * http://TWiki.org - is your team already TWiki enabled? > * Knowledge cannot be managed, it can be discovered and shared > * This e-mail is: (_) private (x) ask first (_) public > > > > _______________________________________________ > SanFrancisco-pm mailing list > SanFrancisco-pm at pm.org > http://mail.pm.org/mailman/listinfo/sanfrancisco-pm -------------- next part -------------- An HTML attachment was scrubbed... URL: From not.com at gmail.com Thu Mar 14 05:20:31 2013 From: not.com at gmail.com (yary) Date: Thu, 14 Mar 2013 08:20:31 -0400 Subject: [sf-perl] Google Trends for Perl In-Reply-To: <687C00D2-50D6-44F6-833E-82095E9F4C86@gmail.com> References: <9F1F92ED-65FF-4051-BF07-EECF8A4D6B7C@thoeny.org> <687C00D2-50D6-44F6-833E-82095E9F4C86@gmail.com> Message-ID: On Thu, Mar 14, 2013 at 12:08 AM, Mike Friedman wrote: >As long as Google traffic goes up overall, *any* > term with steady query volume will go down over time. The decline of programming-language queries requires a more subtle explanation. It's a continuing trend of non-tech people using the 'web which has been going on since the start. Additional "whole-world" searches diminish the rank of "tech" searches. I discovered that when I added "python" to the graph and realized that most of those searches were for the animal, not the language! It's pretty close to "snake" in Google-search-history terms. http://www.google.com/trends/explore#q=perl%2C%20snake%2C%20python&cmpt=q And how many programming-Perl searches are typed as "Pearl", and how many people looking for pearls type "perl"? Still, Perl can use more positive publicity. My comments should not be construed as complacency. From bobrere at jumpmark.com Thu Mar 14 08:55:12 2013 From: bobrere at jumpmark.com (bobrere at jumpmark.com) Date: Thu, 14 Mar 2013 08:55:12 -0700 (PDT) Subject: [sf-perl] Google Trends for Perl In-Reply-To: References: <9F1F92ED-65FF-4051-BF07-EECF8A4D6B7C@thoeny.org> <687C00D2-50D6-44F6-833E-82095E9F4C86@gmail.com> Message-ID: > Still, Perl can use more positive publicity. My comments should not be > construed as complacency. agreed... it feels to me a lot of en vogue tech has really just been successful at creating hype and really isn't that mature or even fundamentally that good. how can this happen for perl? is it in any perl dependent company's interest to promote it, or does it make more sense to corner the market in perl jobs? how do you promote something that's decidedly not new? -scott From earl at ruby.org Thu Mar 14 11:01:12 2013 From: earl at ruby.org (Earl Ruby) Date: Thu, 14 Mar 2013 11:01:12 -0700 Subject: [sf-perl] Google Trends for Perl In-Reply-To: References: <9F1F92ED-65FF-4051-BF07-EECF8A4D6B7C@thoeny.org> <687C00D2-50D6-44F6-833E-82095E9F4C86@gmail.com> Message-ID: Meanwhile, the growth in people searching for me grows as Google grows: http://www.google.com/trends/explore#q=ruby%2C%20earl&cmpt=q Some might claim that this is due to my name consisting of two common words. On Thu, Mar 14, 2013 at 8:55 AM, wrote: >> Still, Perl can use more positive publicity. My comments should not be >> construed as complacency. > > > agreed... it feels to me a lot of en vogue tech has really just been > successful at creating hype and really isn't that mature or even > fundamentally that good. how can this happen for perl? is it in any perl > dependent company's interest to promote it, or does it make more sense to > corner the market in perl jobs? how do you promote something that's > decidedly not new? > > -scott > > > _______________________________________________ > SanFrancisco-pm mailing list > SanFrancisco-pm at pm.org > http://mail.pm.org/mailman/listinfo/sanfrancisco-pm -- Earl Ruby http://earlruby.org/ http://www.linkedin.com/in/earlruby @earlruby From fred at redhotpenguin.com Wed Mar 20 14:11:01 2013 From: fred at redhotpenguin.com (Fred Moyer) Date: Wed, 20 Mar 2013 14:11:01 -0700 Subject: [sf-perl] [meeting] Pinto -- New And Notable Message-ID: Join us next Tuesday for our March meeting at Mother Jones. Jeff Thalhammer will be speaking on his latest work with Pinto and Stratopan. Pinto is a full-service application for curating custom repositories of Perl modules.I'll be talking about the newest features and enhancements in Pinto. I'll also give you anupdate on http://stratopan.com, the hosted solution based on Pinto. Finally I'm going to stir up some conversation about the CPAN ecosystem and how we can do to make it better. If you're a CPAN author or use a lot of CPAN modules, this will be a great session for you. RSVP at Meetup - http://www.meetup.com/San-Francisco-Perl-Mongers/events/110234022/ Jeff Thalhammer on CPAN - https://metacpan.org/author/THALJEF Stratopan - http://stratopan.com From josh at agliodbs.com Fri Mar 22 15:11:31 2013 From: josh at agliodbs.com (Josh Berkus) Date: Fri, 22 Mar 2013 15:11:31 -0700 Subject: [sf-perl] Google Trends for Perl In-Reply-To: References: Message-ID: <514CD713.5090404@agliodbs.com> On 03/13/2013 01:59 PM, mehryar wrote: > > Sadly: > > http://www.google.com/trends/explore#q=perl > > but maybe Perl is really picking up in India :-) Frankly, I wouldn't read too much into that. PostgreSQL is more popular than it's ever been, but if you went by Google Trends, we'd be at 1/5 the level we were at in 2005. -- Josh Berkus PostgreSQL Experts Inc. http://pgexperts.com From biztos at mac.com Fri Mar 22 20:24:31 2013 From: biztos at mac.com (Kevin Frost) Date: Sat, 23 Mar 2013 04:24:31 +0100 Subject: [sf-perl] Google Trends for Perl In-Reply-To: <514CD713.5090404@agliodbs.com> References: <514CD713.5090404@agliodbs.com> Message-ID: <07279CD8-B6EB-46A8-9E35-545E9BF88558@mac.com> That would seem to imply that clued-in folks are searching elsewhere. For instance, I generally try Stack Blabla for tech stuff, even if I might try it on GOOG first. My deeper searches stay in the experts zone. Pure speculation this, but with GOOG getting ever spammier and ad-ier maybe the nerds are already moving on, if not yet to a common place. This matters: the first GOOG became important because the nerds loved it. The second might too. Makes me wanna hack together a search startup, using PG of course... And Perl... -- frosty On Mar 22, 2013, at 11:11 PM, Josh Berkus wrote: > On 03/13/2013 01:59 PM, mehryar wrote: >> >> Sadly: >> >> http://www.google.com/trends/explore#q=perl >> >> but maybe Perl is really picking up in India :-) > > Frankly, I wouldn't read too much into that. PostgreSQL is more popular > than it's ever been, but if you went by Google Trends, we'd be at 1/5 > the level we were at in 2005. > > -- > Josh Berkus > PostgreSQL Experts Inc. > http://pgexperts.com > _______________________________________________ > SanFrancisco-pm mailing list > SanFrancisco-pm at pm.org > http://mail.pm.org/mailman/listinfo/sanfrancisco-pm From quinn at pgexperts.com Fri Mar 22 20:36:34 2013 From: quinn at pgexperts.com (Quinn Weaver) Date: Fri, 22 Mar 2013 20:36:34 -0700 Subject: [sf-perl] Google Trends for Perl In-Reply-To: <07279CD8-B6EB-46A8-9E35-545E9BF88558@mac.com> References: <514CD713.5090404@agliodbs.com> <07279CD8-B6EB-46A8-9E35-545E9BF88558@mac.com> Message-ID: <11969F19-89C4-46CB-AA3D-C2014A148E09@pgexperts.com> On Mar 22, 2013, at 8:24 PM, Kevin Frost wrote: > Makes me wanna hack together a search startup, using PG of course... And Perl... DuckDuckGo uses both, and is rather good. It doesn't track or bubble users, and it has support for custom domains (s.g., search Perl-related things) and a Tor exit enclave. From greg at blekko.com Fri Mar 22 22:25:30 2013 From: greg at blekko.com (Greg Lindahl) Date: Fri, 22 Mar 2013 22:25:30 -0700 Subject: [sf-perl] Google Trends for Perl In-Reply-To: <11969F19-89C4-46CB-AA3D-C2014A148E09@pgexperts.com> References: <514CD713.5090404@agliodbs.com> <07279CD8-B6EB-46A8-9E35-545E9BF88558@mac.com> <11969F19-89C4-46CB-AA3D-C2014A148E09@pgexperts.com> Message-ID: <20130323052530.GA9661@bx9.net> On Fri, Mar 22, 2013 at 08:36:34PM -0700, Quinn Weaver wrote: > On Mar 22, 2013, at 8:24 PM, Kevin Frost wrote: > > > Makes me wanna hack together a search startup, using PG of course... And Perl... > > DuckDuckGo uses both, and is rather good. It doesn't track or bubble users, and it has support for custom domains (s.g., search Perl-related things) and a Tor exit enclave. And there's always blekko, which wrote its own NoSQL database in perl, has its own crawl and index, a good privacy policy, and a "wow"- inducing tablet search app named izik. Funny that the two most interesting alternative search engines both use perl! -- greg From doomvox at gmail.com Fri Mar 22 22:31:35 2013 From: doomvox at gmail.com (Joseph Brenner) Date: Fri, 22 Mar 2013 22:31:35 -0700 Subject: [sf-perl] Google Trends for Perl In-Reply-To: <11969F19-89C4-46CB-AA3D-C2014A148E09@pgexperts.com> References: <514CD713.5090404@agliodbs.com> <07279CD8-B6EB-46A8-9E35-545E9BF88558@mac.com> <11969F19-89C4-46CB-AA3D-C2014A148E09@pgexperts.com> Message-ID: Quinn Weaver wrote: > Kevin Frost wrote: > >> Makes me wanna hack together a search startup, using PG of course... And Perl... > > DuckDuckGo uses both, and is rather good. It doesn't track or bubble users, and it has support for custom domains (s.g., search Perl-related things) and a Tor exit enclave. Yes, I thought duckduckgo was pretty good when I tried it some time back (though they'd junked up the UI with javascript a bit much for my taste). I've been using the local perl-based effort blekko.com almost exclusively for some years now-- so yeah, I guess I'm not part of The Zeitgeist according to google. blekko's gimmick is "slash tags" to narrow the domain of the search, so you can do stuff like "mouse /perl". From jeff at imaginative-software.com Sat Mar 23 01:54:38 2013 From: jeff at imaginative-software.com (Jeffrey Ryan Thalhammer) Date: Sat, 23 Mar 2013 01:54:38 -0700 Subject: [sf-perl] Pinto on FLOSS Weekly Next Wednesday Message-ID: <237C56A6-4D6C-4D7A-B7B4-B967E80AF99A@imaginative-software.com> I'll be doing a live webcast about Pinto with Randal Schwartz for FLOSS Weekly. Tune in to http://live.twit.tv/ next Wednesday, March 27 at 08:30 (Pacific Time) to the see the show. Your can send in your questions in real-time via the #twitlive channel on irc.twit.tv. See you then! -Jeff From quinn at pgexperts.com Mon Mar 25 09:06:17 2013 From: quinn at pgexperts.com (Quinn Weaver) Date: Mon, 25 Mar 2013 09:06:17 -0700 Subject: [sf-perl] Google Trends for Perl In-Reply-To: References: <514CD713.5090404@agliodbs.com> <07279CD8-B6EB-46A8-9E35-545E9BF88558@mac.com> <11969F19-89C4-46CB-AA3D-C2014A148E09@pgexperts.com> Message-ID: <98CCC16C-AB7A-4BCA-848F-1C231E08D0EF@pgexperts.com> On Mar 22, 2013, at 10:31 PM, Joseph Brenner wrote: > Quinn Weaver wrote: >> Kevin Frost wrote: >> >>> Makes me wanna hack together a search startup, using PG of course... And Perl... >> >> DuckDuckGo uses both, and is rather good. It doesn't track or bubble users, and it has support for custom domains (s.g., search Perl-related things) and a Tor exit enclave. > > Yes, I thought duckduckgo was pretty good when I tried it some > time back (though they'd junked up the UI with javascript a bit > much for my taste). > > I've been using the local perl-based effort blekko.com almost > exclusively for some years now-- so yeah, I guess I'm not part of > The Zeitgeist according to google. > > blekko's gimmick is "slash tags" to narrow the domain of the > search, so you can do stuff like "mouse /perl". It's a good idea. DuckDuckGo has a similar feature: !bang categories, so you can do, e.g., "!perl parser" or "!postgres parser". That's what I meant by my somewhat ill-worded reference to "custom domains" above. Anyway, thanks for the reminder about blekko. I've been meaning to give it a try. PS: I don't know what parts of the DDG UI you object to, but one trick that sometimes works for me is to spoof the User-Agent header of a mobile browser. That way I get the minimal version of a website, which is often much more usable than the original! -- Quinn Weaver PostgreSQL Experts, Inc. http://pgexperts.com/ 1-888-743-9778 (my extension: 510) From jeff at imaginative-software.com Mon Mar 25 10:34:06 2013 From: jeff at imaginative-software.com (Jeffrey Ryan Thalhammer) Date: Mon, 25 Mar 2013 10:34:06 -0700 Subject: [sf-perl] Vote for Fred Moyer for White Camel Award Message-ID: <5E51A87A-F53B-4B1F-BD25-A81ED3073A47@imaginative-software.com> Hey everyone- Nominations are now being accepted for the White Camel Award: http://whitecamelawards.mobrater.com/list#sort_by=score I suggest voting for our own Fred Moyer. I think Fred has done a fantastic job running SF.pm. We've had very regular and productive meetings, and the attendance is always strong. That's really hard to accomplish in a place like San Francisco where we have to compete with so many other technology meetups. Running a PM group is no small task, and that kind of grass-roots effort is key to the future success of Perl. I hope you'll join me and vote for Fred. -Jeff -------------- next part -------------- An HTML attachment was scrubbed... URL: From hartzell at alerce.com Mon Mar 25 10:39:17 2013 From: hartzell at alerce.com (George Hartzell) Date: Mon, 25 Mar 2013 10:39:17 -0700 Subject: [sf-perl] Vote for Fred Moyer for White Camel Award In-Reply-To: <5E51A87A-F53B-4B1F-BD25-A81ED3073A47@imaginative-software.com> References: <5E51A87A-F53B-4B1F-BD25-A81ED3073A47@imaginative-software.com> Message-ID: <20816.35781.316483.585489@gargle.gargle.HOWL> Jeffrey Ryan Thalhammer writes: > Hey everyone- > > Nominations are now being accepted for the White Camel Award: > > http://whitecamelawards.mobrater.com/list#sort_by=score > > I suggest voting for our own Fred Moyer. I think Fred has done a fantastic job running SF.pm. We've had very regular and productive meetings, and the attendance is always strong. That's really hard to accomplish in a place like San Francisco where we have to compete with so many other technology meetups. Running a PM group is no small task, and that kind of grass-roots effort is key to the future success of Perl. I hope you'll join me and vote for Fred. > > -Jeff_______________________________________________ > SanFrancisco-pm mailing list > SanFrancisco-pm at pm.org > http://mail.pm.org/mailman/listinfo/sanfrancisco-pm Hear hear! And you can vote here: http://whitecamelawards.mobrater.com/ g. From frimicc at gmail.com Mon Mar 25 10:46:54 2013 From: frimicc at gmail.com (Michael Friedman) Date: Mon, 25 Mar 2013 10:46:54 -0700 Subject: [sf-perl] Vote for Fred Moyer for White Camel Award In-Reply-To: <20816.35781.316483.585489@gargle.gargle.HOWL> References: <5E51A87A-F53B-4B1F-BD25-A81ED3073A47@imaginative-software.com> <20816.35781.316483.585489@gargle.gargle.HOWL> Message-ID: <06A1B428-7B36-474E-B197-568C20DC4777@gmail.com> Note that, somehow, Fred got cloned. There are two entries in the list for Fred with slightly different descriptions. To be safe, I voted for both of him! -- Mike On Mar 25, 2013, at 10:39 AM, George Hartzell wrote: > Jeffrey Ryan Thalhammer writes: >> Hey everyone- >> >> Nominations are now being accepted for the White Camel Award: >> >> http://whitecamelawards.mobrater.com/list#sort_by=score >> >> I suggest voting for our own Fred Moyer. I think Fred has done a fantastic job running SF.pm. We've had very regular and productive meetings, and the attendance is always strong. That's really hard to accomplish in a place like San Francisco where we have to compete with so many other technology meetups. Running a PM group is no small task, and that kind of grass-roots effort is key to the future success of Perl. I hope you'll join me and vote for Fred. >> >> -Jeff_______________________________________________ >> SanFrancisco-pm mailing list >> SanFrancisco-pm at pm.org >> http://mail.pm.org/mailman/listinfo/sanfrancisco-pm > > > Hear hear! > > And you can vote here: > > http://whitecamelawards.mobrater.com/ > > g. > _______________________________________________ > SanFrancisco-pm mailing list > SanFrancisco-pm at pm.org > http://mail.pm.org/mailman/listinfo/sanfrancisco-pm From quinn at pgexperts.com Mon Mar 25 11:00:12 2013 From: quinn at pgexperts.com (Quinn Weaver) Date: Mon, 25 Mar 2013 11:00:12 -0700 Subject: [sf-perl] Vote for Fred Moyer for White Camel Award In-Reply-To: <5E51A87A-F53B-4B1F-BD25-A81ED3073A47@imaginative-software.com> References: <5E51A87A-F53B-4B1F-BD25-A81ED3073A47@imaginative-software.com> Message-ID: <24A9854B-D901-4A01-AE37-AD9C1D408C5A@pgexperts.com> On Mar 25, 2013, at 10:34 AM, Jeffrey Ryan Thalhammer wrote: > Hey everyone- > > Nominations are now being accepted for the White Camel Award: > > http://whitecamelawards.mobrater.com/list#sort_by=score > > I suggest voting for our own Fred Moyer. Done! > I think Fred has done a fantastic job running SF.pm. We've had very regular and productive meetings, and the attendance is always strong. That's really hard to accomplish in a place like San Francisco where we have to compete with so many other technology meetups. Running a PM group is no small task, and that kind of grass-roots effort is key to the future success of Perl. I hope you'll join me and vote for Fred. I agree! I know from experience that running SF.pm is a lot more work than it appears. Fred et al. have really taken it to the next level. I wonder if there's some way we can modify this entry to include others who are involved in SF.pm, such as Joe Brenner, who has been involved in finding many of the speakers. Regards, -- Quinn Weaver PostgreSQL Experts, Inc. http://pgexperts.com/ 1-888-743-9778 (my extension: 510) From jpbida at ambrosevaast.com Mon Mar 25 14:32:51 2013 From: jpbida at ambrosevaast.com (jpbida) Date: Mon, 25 Mar 2013 14:32:51 -0700 Subject: [sf-perl] Managing Perl Projects Message-ID: What are your favorite set-ups for managing multiple developers contributing to a single perl code base? What is a good resource for doing this right? I'm really interested in projects using Dancer/Catalyst Frameworks. 1) *Revision Systems* svn 2) *Code Review Tools and Release Management* http://www.reviewboard.org 3) *Automated Testing * Test::More Jenkins Interface testing?? is there something out there that can simulate clicks? 4)* Requirements Management* trac, How to write a requirement? 5) *Managing Test, UAT, & production environments* or anything else that you would include in a good large project set-up. ~Jp -------------- next part -------------- An HTML attachment was scrubbed... URL: From jeff at imaginative-software.com Mon Mar 25 14:50:40 2013 From: jeff at imaginative-software.com (Jeffrey Ryan Thalhammer) Date: Mon, 25 Mar 2013 14:50:40 -0700 Subject: [sf-perl] Managing Perl Projects In-Reply-To: References: Message-ID: <56A2B1F7-B1F1-435E-BBEF-F901E5C485E5@imaginative-software.com> On Mar 25, 2013, at 2:32 PM, jpbida wrote: > What are your favorite set-ups for managing multiple developers contributing to a single perl code base? What is a good resource for doing this right? I'm really interested in projects using Dancer/Catalyst Frameworks. No simple answer here. Aside from a few Perl-centric tools (like maybe Smolder) none of these are any better or worse just because you're using Perl. If you want to know what people's favorite version control system is, or what they like for bug tracking, that is a slightly different question. You've identified most of the major types of tools it the development stack, so you've already won half the battle. So now, you just make sure you use them :) -Jeff -------------- next part -------------- An HTML attachment was scrubbed... URL: From hartzell at alerce.com Mon Mar 25 15:01:05 2013 From: hartzell at alerce.com (George Hartzell) Date: Mon, 25 Mar 2013 15:01:05 -0700 Subject: [sf-perl] Managing Perl Projects In-Reply-To: References: Message-ID: <20816.51489.109573.685875@gargle.gargle.HOWL> I'll echo Jeff T's comment. It matters a lot more that you put some tools in place AND USE THEM than it does finding an optimal set. jpbida writes: > What are your favorite set-ups for managing multiple developers > contributing to a single perl code base? What is a good resource for doing > this right? I'm really interested in projects using Dancer/Catalyst > Frameworks. > > 1) *Revision Systems* > svn or git. Knowing git a la GitHub can be really useful for collaborating with many of the popular CPAN package authors. > 2) *Code Review Tools and Release Management* > http://www.reviewboard.org > 3) *Automated Testing * > Test::More > Jenkins > Interface testing?? is there something out there that can simulate I've used Jenkins in the past, but can't say much more than it works. You "should" include some quality-ish tests along with your real tests, e.g. running perlcritic, possibly perltidy, etc.... You "should" probably be building your dists automagically. Dist::Zilla makes this reasonable. And then you should be testing the building process too. I used to have jenkins (hudson, back in the day) configured to run a simple shell fragment that did a 'dzil test' in the project root, this tested the build and (with some extra env. variables) ran through all of the tests (t, xt, ...) too. > clicks? > 4)* Requirements Management* > trac, How to write a requirement? > 5) *Managing Test, UAT, & production environments* Pinto. cpanm. > or anything else that you would include in a good large project set-up. emacs (there! If nothing above started a discussion, *that* ought to...). g. From frimicc at gmail.com Mon Mar 25 15:45:10 2013 From: frimicc at gmail.com (Michael Friedman) Date: Mon, 25 Mar 2013 15:45:10 -0700 Subject: [sf-perl] Managing Perl Projects In-Reply-To: References: Message-ID: For automated interface (Web GUI) testing, Polyvore is using a combination of tools: * Test::WWW::Mechanize for web pages that don't involve Javascript. * Test::WWW::Selenium for web pages that do involve Javascript, but work on Linux Firefox. * Sauce (an external service) for web pages that involve Javascript, but work on other browsers. If you're doing something that doesn't have a Web GUI, I'm not sure where you would turn. For our test environment we have a single script that sets up everything for the tests: creates a local instance of MySQL, populates initial data, initializes Selenium, and returns the important connecting information (port numbers, for example) so that the actual tests can use them. This script runs before the tests run every time (automatically in Jenkins) so that each test run has a blank slate. For testing non-blank-slates, developers don't run the setup script every time, so we get old testing cruft in our systems. That way the automated tests don't break because of it -- our test runs do and we can fix things before committing them. It's a pretty cool setup, I must say. -- Mike Friedman On Mar 25, 2013, at 2:32 PM, jpbida wrote: > What are your favorite set-ups for managing multiple developers contributing to a single perl code base? What is a good resource for doing this right? I'm really interested in projects using Dancer/Catalyst Frameworks. > > 1) Revision Systems > svn > 2) Code Review Tools and Release Management > http://www.reviewboard.org > 3) Automated Testing > Test::More > Jenkins > Interface testing?? is there something out there that can simulate clicks? > 4) Requirements Management > trac, How to write a requirement? > 5) Managing Test, UAT, & production environments > > or anything else that you would include in a good large project set-up. > > ~Jp > > > > > > > > _______________________________________________ > SanFrancisco-pm mailing list > SanFrancisco-pm at pm.org > http://mail.pm.org/mailman/listinfo/sanfrancisco-pm From doomvox at gmail.com Tue Mar 26 11:04:45 2013 From: doomvox at gmail.com (Joseph Brenner) Date: Tue, 26 Mar 2013 11:04:45 -0700 Subject: [sf-perl] Vote for Fred Moyer for White Camel Award In-Reply-To: <24A9854B-D901-4A01-AE37-AD9C1D408C5A@pgexperts.com> References: <5E51A87A-F53B-4B1F-BD25-A81ED3073A47@imaginative-software.com> <24A9854B-D901-4A01-AE37-AD9C1D408C5A@pgexperts.com> Message-ID: Quinn Weaver wrote: > I wonder if there's some way we can modify this entry to include > others who are involved in SF.pm, such as Joe Brenner, who has been > involved in finding many of the speakers. And some of the worst venues. Just vote for Fred, I say. From fred at redhotpenguin.com Tue Mar 26 11:08:37 2013 From: fred at redhotpenguin.com (Fred Moyer) Date: Tue, 26 Mar 2013 11:08:37 -0700 Subject: [sf-perl] [meeting] Pinto -- New And Notable In-Reply-To: References: Message-ID: Just a heads up, Pinto is tonight at 7pm! Please RSVP by 5pm if you plan to attend - http://www.meetup.com/San-Francisco-Perl-Mongers/events/110234022/ On Wed, Mar 20, 2013 at 2:11 PM, Fred Moyer wrote: > Join us next Tuesday for our March meeting at Mother Jones. Jeff > Thalhammer will be speaking on his latest work with Pinto and > Stratopan. > > Pinto is a full-service application for curating custom repositories > of Perl modules.I'll be talking about the newest features and > enhancements in Pinto. I'll also give you anupdate on > http://stratopan.com, the hosted solution based on Pinto. Finally I'm > going to stir up some conversation about the CPAN ecosystem and how we > can do to make it better. If you're a CPAN author or use a lot of CPAN > modules, this will be a great session for you. > > RSVP at Meetup - > http://www.meetup.com/San-Francisco-Perl-Mongers/events/110234022/ > > Jeff Thalhammer on CPAN - https://metacpan.org/author/THALJEF > > Stratopan - http://stratopan.com From jeff at imaginative-software.com Tue Mar 26 22:09:03 2013 From: jeff at imaginative-software.com (Jeffrey Ryan Thalhammer) Date: Tue, 26 Mar 2013 22:09:03 -0700 Subject: [sf-perl] Thanks For Coming Out Tonight Message-ID: Hi everyone- Thanks for coming out to the PM meeting tonight -- I had a great time. Special thanks to Fred and Joe for organizing, and Mother Jones for hosting. And don't forget to tune into my live FLOSS Weekly interview with Randal Schwartz tomorrow at 8:30am (Pacific Time). The webcast will be available in the archives after a couple days. -Jeff -------------- next part -------------- An HTML attachment was scrubbed... URL: From jeff at imaginative-software.com Wed Mar 27 12:51:09 2013 From: jeff at imaginative-software.com (Jeffrey Ryan Thalhammer) Date: Wed, 27 Mar 2013 12:51:09 -0700 Subject: [sf-perl] Pinto on FLOSS Weekly Next Wednesday In-Reply-To: <237C56A6-4D6C-4D7A-B7B4-B967E80AF99A@imaginative-software.com> References: <237C56A6-4D6C-4D7A-B7B4-B967E80AF99A@imaginative-software.com> Message-ID: My FLOSS Weekly interview on Pinto is now online: http://twit.tv/show/floss-weekly/246 From jpbida at ambrosevaast.com Thu Mar 28 10:10:35 2013 From: jpbida at ambrosevaast.com (jpbida) Date: Thu, 28 Mar 2013 10:10:35 -0700 Subject: [sf-perl] Thanks For Coming Out Tonight In-Reply-To: References: Message-ID: We had talked about what would make CPAN better at the meet-up. Last summer I analyzed github data to try to figure out what github usage patterns are correlated with open source project growth as measured by project downloads. A couple interesting observations: 1) There was a strong correlation between diverse team compositions (in terms of what programming languages the team members used) and how many downloads the project would get. 2) The amount of code submitted was inversely proportional to the success of the project. Maybe one thing CPAN/stratopan could improve is ways to integrate with other programming language development tools or workflows and encourage a diverse group of programmers to work with perl. I tried to attach a presentation but it was blocked by the list. If you are interested in seeing it I can pass it along on an individual basis. Cheers, ~Jp On Tue, Mar 26, 2013 at 10:09 PM, Jeffrey Ryan Thalhammer < jeff at imaginative-software.com> wrote: > Hi everyone- > > Thanks for coming out to the PM meeting tonight -- I had a great time. > Special thanks to Fred and Joe for organizing, and Mother Jones for > hosting. > > And don't forget to tune into my live FLOSS Weekly interview with Randal > Schwartz tomorrow at 8:30am (Pacific Time). > > The webcast will be available in the archives after > a couple days. > > -Jeff > > > _______________________________________________ > SanFrancisco-pm mailing list > SanFrancisco-pm at pm.org > http://mail.pm.org/mailman/listinfo/sanfrancisco-pm > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From hartzell at alerce.com Thu Mar 28 17:16:49 2013 From: hartzell at alerce.com (George Hartzell) Date: Thu, 28 Mar 2013 17:16:49 -0700 Subject: [sf-perl] Prototypes, List::MoreUtils and subroutine references, OH MY! Message-ID: <20820.56689.995903.704893@gargle.gargle.HOWL> I have a bunch of subs that are line for line identical, except that they use a different one of List::MoreUtils qw(any all) [etc...]. No problem, I thought, I'll just create a generic version and pass in a reference to the list function that I'd like to use. Given something like this my $any = \&any; I've discovered that I need to call it passing in a code ref as the first item in a list: &$any( sub { ... }, @things ); where when I call any directly I pass in a block any { ...} @things; It seems that somehow the prototype magic doesn't travel along with the code ref. Is this the way it is or am I screwing something up and getting lucky? g. From jeff at imaginative-software.com Thu Mar 28 18:11:30 2013 From: jeff at imaginative-software.com (Jeffrey Ryan Thalhammer) Date: Thu, 28 Mar 2013 18:11:30 -0700 Subject: [sf-perl] Prototypes, List::MoreUtils and subroutine references, OH MY! In-Reply-To: <20820.56689.995903.704893@gargle.gargle.HOWL> References: <20820.56689.995903.704893@gargle.gargle.HOWL> Message-ID: <63A9678E-DF36-4B9A-8BE7-BDD83C03FEDF@imaginative-software.com> On Mar 28, 2013, at 5:16 PM, George Hartzell wrote: > Is this the way it is or am I screwing something up and getting lucky? Dunno. As an alterative approach, maybe consider auto-generating subs with a little eval magic. Sub::Quote is useful for that. -Jeff -------------- next part -------------- An HTML attachment was scrubbed... URL: From jeff at imaginative-software.com Fri Mar 29 01:19:02 2013 From: jeff at imaginative-software.com (Jeffrey Ryan Thalhammer) Date: Fri, 29 Mar 2013 01:19:02 -0700 Subject: [sf-perl] Thanks For Coming Out Tonight In-Reply-To: References: Message-ID: Those *are* some really interesting observations. > 1) There was a strong correlation between diverse team compositions (in terms of what programming languages the team members used) and how many downloads the project would get. > And what do you think the story is there? Programmers with diverse language experience write better code? Or do the projects themselves incorporate multiple languages, which means they have wider applicability? > 2) The amount of code submitted was inversely proportional to the success of the project. Curious. How did you define success? And what does "submitted" mean? A pull request? > > Maybe one thing CPAN/stratopan could improve is ways to integrate with other programming language development tools or workflows and encourage a diverse group of programmers to work with perl. Yes, I'll meditate on that for a while. -Jeff -------------- next part -------------- An HTML attachment was scrubbed... URL: From hartzell at alerce.com Fri Mar 29 09:12:27 2013 From: hartzell at alerce.com (George Hartzell) Date: Fri, 29 Mar 2013 09:12:27 -0700 Subject: [sf-perl] Prototypes, List::MoreUtils and subroutine references, OH MY! In-Reply-To: <63A9678E-DF36-4B9A-8BE7-BDD83C03FEDF@imaginative-software.com> References: <20820.56689.995903.704893@gargle.gargle.HOWL> <63A9678E-DF36-4B9A-8BE7-BDD83C03FEDF@imaginative-software.com> Message-ID: <20821.48491.453494.83762@gargle.gargle.HOWL> Jeffrey Ryan Thalhammer writes: > > On Mar 28, 2013, at 5:16 PM, George Hartzell wrote: > > > Is this the way it is or am I screwing something up and getting lucky? > > Dunno. As an alterative approach, maybe consider auto-generating > subs with a little eval magic. Sub::Quote is useful for that. Interesting thought. Been playing with it in the context of Moo, could be useful here. Thanks! g. From cweyl at alumni.drew.edu Fri Mar 29 09:21:08 2013 From: cweyl at alumni.drew.edu (Chris Weyl) Date: Fri, 29 Mar 2013 09:21:08 -0700 Subject: [sf-perl] Prototypes, List::MoreUtils and subroutine references, OH MY! In-Reply-To: <20820.56689.995903.704893@gargle.gargle.HOWL> References: <20820.56689.995903.704893@gargle.gargle.HOWL> Message-ID: On Thu, Mar 28, 2013 at 5:16 PM, George Hartzell wrote: > Is this the way it is or am I screwing something up and getting lucky? > That's the way it is. Prototypes are only used when a sub is invoked as a function, not from a coderef or method; this make sense in their context of providing the capacity to emulate built in functions. -Chris -- Chris Weyl Ex astris scientia -------------- next part -------------- An HTML attachment was scrubbed... URL: From hartzell at alerce.com Fri Mar 29 09:30:48 2013 From: hartzell at alerce.com (George Hartzell) Date: Fri, 29 Mar 2013 09:30:48 -0700 Subject: [sf-perl] Prototypes, List::MoreUtils and subroutine references, OH MY! In-Reply-To: References: <20820.56689.995903.704893@gargle.gargle.HOWL> Message-ID: <20821.49592.385366.715249@gargle.gargle.HOWL> Chris Weyl writes: > On Thu, Mar 28, 2013 at 5:16 PM, George Hartzell wrote: > > > Is this the way it is or am I screwing something up and getting lucky? > > > > That's the way it is. Prototypes are only used when a sub is invoked as a > function, not from a coderef or method; this make sense in their context of > providing the capacity to emulate built in functions. > > -Chris Makes sense. Thanks! Funny things happen when you take a reference to e.g. map and try the same trick: DB<1> $m = \&map DB<2> x $m 0 CODE(0x7ffa00a33520) -> &CODE(0x7ffa00a33520) in ??? DB<3> x map {$_ < 2} (1,2,3) 0 1 1 '' 2 '' DB<4> x &$m {$_ < 2} (1,2,3) syntax error at (eval 26)[/Users/hartzell/perl5/perlbrew/perls/perl-5.16.2/lib/5.16.2/perl5db.pl:646] line 2, near "$m {" DB<5> x &$m(sub {$_ < 2}, (1,2,3)) Undefined subroutine &main::map called at (eval 28)[/Users/hartzell/perl5/perlbrew/perls/perl-5.16.2/lib/5.16.2/perl5db.pl:646] line 2. DB<6> x $m->(sub {$_ < 2}, (1,2,3)) Undefined subroutine &main::map called at (eval 29)[/Users/hartzell/perl5/perlbrew/perls/perl-5.16.2/lib/5.16.2/perl5db.pl:646] line 2. But that only surprised me a little bit. g. From jeff at imaginative-software.com Fri Mar 29 09:42:18 2013 From: jeff at imaginative-software.com (Jeffrey Ryan Thalhammer) Date: Fri, 29 Mar 2013 09:42:18 -0700 Subject: [sf-perl] Prototypes, List::MoreUtils and subroutine references, OH MY! In-Reply-To: References: <20820.56689.995903.704893@gargle.gargle.HOWL> Message-ID: On Mar 29, 2013, at 9:21 AM, Chris Weyl wrote: > That's the way it is. Prototypes are only used when a sub is invoked as a function, not from a coderef or method; this make sense in their context of providing the capacity to emulate built in functions. I think of prototypes as a compile-time thing. Dereferencing a subroutine ref is a run-time event. Perl knows nothing about prototypes at that point. -Jeff -------------- next part -------------- An HTML attachment was scrubbed... URL: From jeff at imaginative-software.com Fri Mar 29 09:44:31 2013 From: jeff at imaginative-software.com (Jeffrey Ryan Thalhammer) Date: Fri, 29 Mar 2013 09:44:31 -0700 Subject: [sf-perl] Prototypes, List::MoreUtils and subroutine references, OH MY! In-Reply-To: <20821.48491.453494.83762@gargle.gargle.HOWL> References: <20820.56689.995903.704893@gargle.gargle.HOWL> <63A9678E-DF36-4B9A-8BE7-BDD83C03FEDF@imaginative-software.com> <20821.48491.453494.83762@gargle.gargle.HOWL> Message-ID: <7C8A241F-B3F9-4627-B2E2-12E94D346867@imaginative-software.com> On Mar 29, 2013, at 9:12 AM, George Hartzell wrote: >> Dunno. As an alterative approach, maybe consider auto-generating >> subs with a little eval magic. Sub::Quote is useful for that. > > Interesting thought. Been playing with it in the context of Moo, > could be useful here. And perlcritic be damned :) -Jeff -------------- next part -------------- An HTML attachment was scrubbed... URL: From hartzell at alerce.com Fri Mar 29 09:46:27 2013 From: hartzell at alerce.com (George Hartzell) Date: Fri, 29 Mar 2013 09:46:27 -0700 Subject: [sf-perl] Prototypes, List::MoreUtils and subroutine references, OH MY! In-Reply-To: <7C8A241F-B3F9-4627-B2E2-12E94D346867@imaginative-software.com> References: <20820.56689.995903.704893@gargle.gargle.HOWL> <63A9678E-DF36-4B9A-8BE7-BDD83C03FEDF@imaginative-software.com> <20821.48491.453494.83762@gargle.gargle.HOWL> <7C8A241F-B3F9-4627-B2E2-12E94D346867@imaginative-software.com> Message-ID: <20821.50531.954549.740514@gargle.gargle.HOWL> Jeffrey Ryan Thalhammer writes: > > On Mar 29, 2013, at 9:12 AM, George Hartzell wrote: > > >> Dunno. As an alterative approach, maybe consider auto-generating > >> subs with a little eval magic. Sub::Quote is useful for that. > > > > Interesting thought. Been playing with it in the context of Moo, > > could be useful here. > > > And perlcritic be damned :) Well, that's a cross it'll just have to bear. It's not like it'll be the first time.... g.