NAME

Lingua::EN::Titlecase - Titlecase English words by traditional editorial rules.

VERSION

0.14

SYNOPSIS

use Lingua::EN::Titlecase;
my $tc = Lingua::EN::Titlecase->new("CAN YOU FIX A TITLE?");
print $tc->title(), $/;

$tc->title("and again but differently");
print $tc->title(), $/;

$tc->title("cookbook don't work, do she?");
print "$tc\n";

DESCRIPTION

Editorial titlecasing in English is the initial capitalization of regular words minus inner articles, prepositions, and conjunctions.

This is one of those problems that is somewhat easy to solve for the general case but impossible to solve for all cases. Hence the lack of module till now. This module takes an optimistic approach, assuming that some words, unless there are clues to the contrary, are likely to be correct already. Most titlecase implementations, for example, convert everything to lowercase first. This is obviously flawed for many common cases like proper names and abbreviations.

Simple techniques like--

$data =~ s/(\w+)/\u\L$1/g;

Fail on words like "can't" and don't always take into account editorial rules or cases like--

compound words -- Perl-like
abbreviations -- USA
mixedcase and proper names -- eBay: nEw KEyBOArD
all caps -- SHOUT ME DOWN

Lingua::EN::Titlecase attempts to cater to the general cases and provide hooks to address the special.

INTERFACE

$tc = Lingua::EN::Titlecase->new

The string to be titlecased can be set three ways. Single argument to new. The "original" hash element to new. With the title method.

$tc = Lingua::EN::Titlecase->new("this should be titlecased");

$tc = Lingua::EN::Titlecase->new(original => "no, this is",
                                 mixed_threshold => 0.5);

$tc->title("i beg to differ");

The last is to be able to reuse the Titlecase object.

Lingua::EN::Titlecase objects stringify to their processed titlecase, if they have a string, the ref of the object otherwise.

$tc->original

Returns the original string.

$tc->title

Set the original string, returns the titlecased version. Both can be done at once.

print $tc->title("did you get that thing i sent?"), "\n";
$tc->titlecase

Returns the titlecased string. Croaks if there is no original set via the constructor or the method title.

$tc->uppercase

Returns the list of uppercase letters found. Includes those mixed case letters. Chiefly used internally for determining if string has exceeded the set threshold to be considered "all caps."

$tc->word_punctuation(qr/['-]/)

Sets the regex which will be used to allow punctuation inside words. The default is "[:punct:]." This is more reasonable that it might sound as word boundaries generally have either a space or more than one piece of punctuation. Any instance of the word_punctuation is allowed inside a "word" if it is surrounded by [:alpha:]s. E.g., [:punct:] makes all these one "word" for titlecasing--

can't
cpan.org
cow-catcher
M!M

Set on construction or reset it to change the behavior--

Lingua::EN::Titlecase->new(word_punctuation => "['-]");

$tc->word_punctuation(qr/['-]/)
# "can't" and "cow-catcher" are still one word
# "cpan.org" is now two and the "Org" will get titlecased
$tc->lexer

Get/set the lexer sub ref. You should probably ignore this. If you think otherwise, read the source for more.

STRATEGIES

One of the hardest parts of properly titlecasing input is knowing if part of it is already correct and should not be clobbered. E.g.--

Old MacDonald had a farm

Is partly right and the proper name MacDonald should be left alone. Lowercasing the whole string and then title casing would yield--

Old Macdonald Had a Farm

So, to determine when to flatten a title to lowercase before processing, we check the ratio of mixedcase and the ratio of caps.

$tc->mixed_threshold

Set/get. The ratio of mixedcase to letters which triggers lowercasing the whole string before trying to titlecase. The built-in threshold to clobber is 0.25. Example breakpoints.

0.09 --> Old Macdonald Had a Farm
0.10 --> Old MacDonald Had a Farm

0.14 --> An Ipod with Low Ph on Ebay
0.15 --> An iPod with Low pH on eBay
$tc->uc_threshold

Same as mixed but for "all" caps. Default threshold is 0.95.

$tc->mixed_case

Scalar context returns count of mixedcase letters found. All caps and initial caps are not counted. List context returns the letters. E.g.--

my $tc = Lingua::EN::Titlecase->new();
$tc->title("tHaT pROBABly Will nevEr BE CorrectlY hanDled");
printf "%d, %s\n",
    scalar($tc->mixedcase),
    join(" ", $tc->mixedcase);

Yields--

11, H T R O B A B E C Y D

This is useful for determining if a string is overly mixed. Substrings like "pH" crop up now and then but they should never compose a high percentage of a properly cased title.

$tc->wc

"Word" count. Scalar context returns count of "words." List returns them.

$tc->mixedcase

Count/list of mixedcase letters found.

$tc->whitespace

Count/list of whitespace -- \s+ -- found.

DIAGNOSTICS

No diagnostics for you!

[Non-existent description of error here]

TODO

Dictionary hook to allow BIG lists of proper names and lc to be applied.

Handle internal punctuation like an em-dash as the equivalent of "--"?

Handle hypens; user hooks.

Move to Mouse or Moose?

Handle classes of things to be left alone if of a case. Like Roman numerals? Better to have it be rule based where each rule is used to find a thing, apply a threshold map, possibly convert lc/uc, and then titlecase or accept. This could get much messier than a dictionary and might cause problems with overlap like i v I.

Allow a grammar parser object (on demand, if available) to correctly identify a word's part of speech before applying casing. "To" might be a proper name, for example, and "A" might be a grade.

Debug ability. Log object or to carp?

Recipes. Including TT2 "plugin" recipe. Mini-scripts to test strings or accomplish custom configuration goals.

Take out Class::Accessor...? For having it all in one place, checking args, and slight speed gain.

Add ignore classes? Like \bhttp://...

Bigger test suite.

SEE ALSO

Lingua::EN::Titlecase::HTML for titlecasing text with markup.

RECIPES

Passing L::E::T object to TT2

use Template;
use CGI "header";
use Lingua::EN::Titlecase;
my @titles = (
              "orphans of the sky",
              "childhood's end",
              "the many-colored land",
              "llana of gathol",
              );

print header(-content_type => "text/plain");
my $tt2 = Template->new();

$tt2->process(\*DATA,
              { tc => Lingua::EN::Titlecase->new(),
                title => \@titles }
              );
__DATA__
[%-USE format %]
[%-pretty_print = format('%30s : %s') %]
[%-FOR t IN title %]
[% pretty_print( t, tc.title(t) ) %]
[%-END %]

CONFIGURATION AND ENVIRONMENT

Lingua::EN::Titlecase requires no configuration files or environment variables.

DEPENDENCIES

Perl 5.6 or better to support POSIX regex classes.

BUGS AND LIMITATIONS

Please report any bugs or feature requests to bug-lingua-en-titlecase@rt.cpan.org, or through the web interface at http://rt.cpan.org.

AUTHOR

Ashley Pond V <ashley@cpan.org>.

LICENSE AND COPYRIGHT

Copyright (c) 2008-2009, Ashley Pond V <ashley@cpan.org>.

This module is free software; you can redistribute it and modify it under the same terms as Perl itself. See perlartistic.

DISCLAIMER OF WARRANTY

Because this software is licensed free of charge, there is no warranty for the software, to the extent permitted by applicable law. Except when otherwise stated in writing the copyright holders and other parties provide the software "as is" without warranty of any kind, either expressed or implied, including, but not limited to, the implied warranties of merchantability and fitness for a particular purpose. The entire risk as to the quality and performance of the software is with you. Should the software prove defective, you assume the cost of all necessary servicing, repair, or correction.

In no event unless required by applicable law or agreed to in writing will any copyright holder, or any other party who may modify or redistribute the software as permitted by the above license, be liable to you for damages, including any general, special, incidental, or consequential damages arising out of the use or inability to use the software (including but not limited to loss of data or data being rendered inaccurate or losses sustained by you or third parties or a failure of the software to operate with any other software), even if such holder or other party has been advised of the possibility of such damages.