support Telugu and probably other Asian scripts (#3)

This commit is contained in:
Egor Kislitsyn 2019-04-03 18:53:35 +07:00
parent e862ee9d21
commit 50882b427a
2 changed files with 16 additions and 1 deletions

View file

@ -44,7 +44,7 @@ defmodule AutoLinker.Parser do
# https://www.w3.org/TR/html5/forms.html#valid-e-mail-address
@match_email ~r"^[a-zA-Z0-9.!#$%&'*+\/=?^_`{|}~-]+@[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?(?:\.[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?)*$"u
@match_hashtag ~r/^(?<tag>\#[[:word:]_]*[[:alpha:]_·][[:word:]_·]*)/u
@match_hashtag ~r/^(?<tag>\#[[:word:]_]*[[:alpha:]_·][[:word:]_·\p{M}]*)/u
@prefix_extra [
"magnet:?",

View file

@ -195,6 +195,21 @@ defmodule AutoLinkerTest do
) == expected
end
test "support Telugu" do
text = "#చక్రం #కకకకక్ #కకకకాక #కకకక్రకకకక"
expected =
"<a href=\"/t/చక్రం\">#చక్రం</a> <a href=\"/t/కకకకక్\">#కకకకక్</a> <a href=\"/t/కకకకాక\">#కకకకాక</a> <a href=\"/t/కకకక్రకకకక\">#కకకక్రకకకక</a>"
assert AutoLinker.link(text,
hashtag: true,
hashtag_prefix: "/t/",
class: false,
rel: false,
new_window: false
) == expected
end
test "do not turn urls with hashes into hashtags" do
text = "google.com#test #test google.com/#test #tag"