Created
August 3, 2021 02:04
-
-
Save thundergnat/87a0ab428bbc80a09cbe8acb9f8dd614 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
my %*SUB-MAIN-OPTS = :named-anywhere; | |
use Text::Sorensen :sorensen; | |
use JSON::Fast; | |
my $hashfile = './Sorenson-chars.json'; | |
unit sub MAIN ( $phrase, $head = 10, :$ge = 0.5 ); | |
my %out; | |
if !$hashfile.IO.e { | |
%out = (^0x1FFF0).race.map: { | |
my $n = .chr.uniname; | |
next if $n.lc.contains: any <private surrogate ideograph reserved hangul>; | |
$n => bi-gram $n | |
} | |
$hashfile.IO.spurt(%out.&to-json); | |
} | |
my %hash = +%out ?? %out !! $hashfile.IO.slurp.&from-json; | |
for sorensen( $phrase, %hash, :$ge ).head($head) { | |
printf "%s : 0x%X : 0d%d : %s : SDI %%%g\n", .[1].uniparse, .[1].uniparse.ord, .[1].uniparse.ord, .[1], .[0]*100; | |
} | |
#`[ | |
Will find "closest matching" characters. % match is the Sorensen-Dice index. If | |
you are searching on a phrase that is a small part of the name, you may need to | |
adjust the --ge=0.?? parameter down (0.5 by default) to allow for the search | |
phrase being a smaller fraction of the name. Returns (up to) the top 10 best | |
matches by default; pass in a different integer for more (or fewer) matches | |
within the --ge threshold. | |
Builds (and saves) the hash file on the first run. Will reload on subsequent | |
runs. It's a large file so it taks a few seconds to parse. | |
Assuming you saved the script as sdi.raku: | |
Try | |
raku sdi.raku 'greek omega' | |
or | |
raku sdi.raku butterfly | |
or | |
raku sdi.raku ' face' 100 --ge=.25 | |
] |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Cool!
One note: you can make the creation of the JSON file about 1 second faster by specifying
:!pretty
in theto-json
call. This will also save about .3 seconds on parsing the JSON (well, on my machine, anyway)