https://github.com/alabamenhu/regexfuzzytoken

A Raku (née Perl 6) module enabling a fuzzy token in grammars and regexen
https://github.com/alabamenhu/regexfuzzytoken

Last synced: about 1 month ago
JSON representation

A Raku (née Perl 6) module enabling a fuzzy token in grammars and regexen

• Host: GitHub
• URL: https://github.com/alabamenhu/regexfuzzytoken
• Owner: alabamenhu
• License: artistic-2.0
• Created: 2019-12-13T22:18:55.000Z (almost 5 years ago)
• Default Branch: master
• Last Pushed: 2022-02-22T04:39:49.000Z (over 2 years ago)
• Last Synced: 2023-03-08T10:08:28.921Z (over 1 year ago)
• Language: Raku
• Homepage:
• Size: 795 KB
• Stars: 3
• Watchers: 1
• Forks: 2
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
![Regex::FuzzyToken for Raku](resources/logo.png)

A Raku module enabling the use of fuzzy tokens in regexen and grammars:

    use Regex::FuzzyToken;

    my @fruits = ;

    "My favorite fruit is a bnana" ~~ /My favorite fruit is a /;

    say $.fuzz; # bnana

    say $.Str;  # banana

Support for more flexible capturing options is forthcoming.  The signature for

the fuzzy token are the following:

    fuzzy(*@words,

          :$i  = False,

          :$m  = False,

          :$ws = True,

          :$q  = 33,

          :$capture = (@words.tail ~~ Regex

                         ?? @words.tail

                         !! /\w+/ )

    )

Basically, you should provide a list of strings to be the “goal” words you want to find,

but will accept if they're slightly misspelled.  The `i`, `m`, and `ws` options mimic

the regex matching behavior, and will make comparisons ignore differences in case (**i**),

marks (**m**) or white space (**ws**).  The `q` option is the minimum Q-gram score desired to

allow for a match.  The default of 33 is fine for most cases, but through testing you

may find it necessary to increase or decrease the sensitivity (100 = only match exact,

0 = match everything).

The final option of :$capture allows you to specify the capture regex to use.  By

default it will only capture a sequence of word characters, but that will cause

problems if you need it to match spaces/apostrophes.  While you *can* make things

explicit with `:capture(/foo/)`, the signature was designed to allow you to

specify the final item as the capture regex, and so the following are equivalent:

    

    

# To do

It could be interesting to allow for a more complex capture, for example, that

matches only as many characters as it needs using `.match($capture, :exhaustive)` on a

substring from the current `.pos`.   That would require some tuning of the Q gram algorithm

and in many cases could take exponentially more time, but would be more accurate / usable.

# Licenses

All code is and documentation is licensed under the Artistic License 2.0, included with

the module.  The image used on Github is based on

[this butterfly](https://www.piqsels.com/en/public-domain-photo-frbsj)

which is licensed under CC-0 and modified in accordance with that license and released

under [CC-BY 4.0](https://creativecommons.org/licenses/by/4.0/)