Skip to content

Latest commit

 

History

History
230 lines (152 loc) · 4.15 KB

README.md

File metadata and controls

230 lines (152 loc) · 4.15 KB
Raw

Verbal Exprejon

VerbalExpressions with a lisp

travis travis



Usage

Create a Regex

(def url
  (->> (then "http")
       (maybe "s")
       (then "://")
       (maybe "www.")
       (anything-but " ")
       (end-of-line)))

(re-matches url "http://www.google.com")

Create your own midleware

A middleware is a function that takes a regex as its last parameter, and output a regex.

; Match a `domain` url like https://domain.com

(defpattern url?
  "Match if it is a url"
  [domain]
  (->>  (then "http")
        (maybe "s")
        (then "://")
        (maybe "www.")
        (then domain)
        (anything-but " ")))

Reuse your middlewares

Match stuff like https://www.facebook.com facebook or https://www.google.com google

(defpattern url-name
  "Match if it is a domain url followed by its name"
  [domain]
  (->>  (url? domain)
        (then " ")
        (then domain)))

(def match (matcher url-name "google"))
(println (match "https://www.google.com google"))

;; -> true

Examples:

(defpattern hour-pattern
  "Matches an hour"
  [] (->> (interval [\0 \9])
          (times [1 2])
          (any-blank)
          (maybe (OR ["h" (->> (then "hour") (maybe "s"))]))))

(defpattern minute-pattern
  "Matches a minute"
  [] (->> (interval [\0 \9])
          (times [1 2])
          (any-blank)
          (maybe (OR ["m" (->> (then "minute") (maybe "s"))]))))

(defpattern second-pattern
  "Matches a second"
  [] (->> (interval [\0 \9])
          (times [1 2])
          (any-blank)
          (maybe (OR ["s" (->> (then "second") (maybe "s"))]))))

(defpattern time-separator
  "Matches the separation between hours, minutes and seconds"
  [] (maybe (->> (any-blank)
                 (maybe (OR ["and" " " ", " ":"]))
                 (any-blank))))

(defpattern time-pattern
  "Matches a time"
  [] (->> (maybe (->> (hour)
                      (time-separator)))
          (maybe (->> (minute)
                      (time-separator)))
          (maybe (second))))

(def time? (matcher time-pattern))

(defn testit
  [time]
  (time? time))

(testit "12h")
(testit "12h5")
(testit "12h05")
(testit "12h30m")
(testit "12:07")
(testit "12hours")
(testit "12hours and 3 minutes")
(testit "12:5:4")
(testit "12h and 5 minutes")
(testit "5 hours")
(testit "12 minutes")
(testit "12 minutes and 5 seconds")
(testit "5s")

Reference

(then "string"):

Matches the string literally

(maybe "string"):

Matches the string if any

(anything):

Matches anything

(anything-but "string"):

Matches anything except the provided value

(one-or-more):

Matches the previous middleware one ore more times

(zero-or-more):

Matches the previous middleware zero or more times

(any "letters"):

Matches any letter from the provided string

(any-blank):

Matches any blank characters including line breaks, spaces and tabs

(end-of-line):

Matches the end of a line

(start-of-line):

Matches the start of a line

(line-break):

Matches a \n

(interval [characters]):

Matches the pairs ranges provided. Examples:

;; matches any letter between a-z and A-Z
(interval \a \z \A \Z)

(tab):

Matches a tabulation

(word):

Matches a word (case insensitive)

(OR [vector]):

Matches the first matching expression in the vector

(times [start end]):

Matches the previous middleware from start times to end times Example:

; [0-9]{1,3}

(interval \0 \9)
(times [1 3])

(words-split-by [delimiters]):

Matches a sequence of words split by any delimiters in the provided vector

(sentence):

Matches a sequence of words split by spaces

License

This has been done by a Clojure newbie.

Copyright © 2016 Guillaume Badi

Distributed under the MIT License