Stupid CPAN Tricks (MadMongers Perl Meeting 2014/02/12)

Stupid CPAN Tricks (MadMongers Perl Meeting 2014/02/12)

JT Smith’s presentation on various CPAN modules. (Apologies for the abrupt ending. Camera ran out of space. Main part of the presentation is there, though.)

About Editor

Comments

5 Responses to “Stupid CPAN Tricks (MadMongers Perl Meeting 2014/02/12)”
  1. Dave Jacoby says:

    This is a video from the Madison, WI Perl Mongers, on Stupid CPAN Tricks.
    Video has MadMongers administrivia for the first 3 and a half minutes, so
    feel free to skip ahead.

    Slides are here: http://files.meetup.com/501101/stupid%20cpan%20tricks.pdf

  2. Sean Burke says:

    I had no idea that anyone was this enthusiastic about my Unidecode module!
    or ever even just gave it a second glance for anything!

    Because of feeling like nobody much uses it, I’ve let the module’s data
    tables lie fallow for about a dozen years now; but the encouragement I’ve
    gotten from this video just bowls me over, so I’m going to try like the
    dickens to jump back into the Unicode 6.0 standard and actually pound out
    some coverage for the new characters that’ve been added over the past
    decade.
    I owe it all to you guys, seriously.
    (I’ve got something ready, with the final task of waiting for me to wrestle
    the ornery Makefile.PL into submission. I’ll do a Dear Lazyweb in a few
    days.)

    BTW, about tables: it’s stored as tables (as opposed to implemented with
    any rules), for purposes of the dist. (Or, ignoring the implementation:
    yeah, it’s basically one big hash.)

    As for as how those tables (sections of the “hash”) came about:

    All the kanji/hanzi stuff is autogenerated from the Unihan database that
    the Unicode Consortium provides.

    The rest of it, i.e., the alphabets and symbols:
    Done entirely manually.

    I spent weeks and weeks going through every character in the
    alphabets-and-symbols sections of Unicode (i.e., everything that’s not
    kanji). So I would get to the Armenian block, and for each character, I’d
    decide on a passable transliteration… then Cyrillic (including all its
    kookoo Turkic extensions), then Burmese… Cherokee… Georgian,…
    Maldivian… Coptic,… core Arabic and then jillion Indic languages’
    extensions to Arabic script…
    For a tatty old linguist, I never had so much fun!

  3. Steve Nolte says:

    Thanks for taping and sharing, I really enjoyed this.
    Seeing DBIx::Class::Helper::ResultSet::Random again and the brief
    discussion on how it works across DB backends pointed me to its dependency
    DBIx::Introspector which solves a problem I’ve been contemplating for a
    while! Thanks JT!

  4. Sean Burke says:

    BTW, for people who didn’t get the escapes, you can rewrite the last line
    as:

    say unidecode( “北亰” ); # Bei Jing

    Interested folks might like to read this article I wrote, about the
    creation story of Unidecode:
    interglacial.com/tpj/22/

    Thanks, JT and pals!

  5. etimespi says: