fix: improve convert_urls

This commit is contained in:
selberhad 2022-10-03 14:50:18 -04:00
parent f9ba117680
commit 80878fb124
2 changed files with 26 additions and 15 deletions

View file

@ -250,13 +250,13 @@ class TestTextToHTMLparser(TestCase):
def test_url_scheme_ftp(self):
self.assertEqual(
self.parser.convert_urls("ftp.example.com"),
'<a href="ftp.example.com" target="_blank">ftp.example.com</a>',
'<a href="//ftp.example.com" target="_blank">ftp.example.com</a>',
)
def test_url_scheme_www(self):
self.assertEqual(
self.parser.convert_urls("www.example.com"),
'<a href="www.example.com" target="_blank">www.example.com</a>',
'<a href="//www.example.com" target="_blank">www.example.com</a>',
)
def test_url_scheme_ftpproto(self):
@ -280,7 +280,7 @@ class TestTextToHTMLparser(TestCase):
def test_url_chars_slash(self):
self.assertEqual(
self.parser.convert_urls("www.example.com/homedir"),
'<a href="www.example.com/homedir" target="_blank">www.example.com/homedir</a>',
'<a href="//www.example.com/homedir" target="_blank">www.example.com/homedir</a>',
)
def test_url_chars_colon(self):
@ -313,22 +313,16 @@ class TestTextToHTMLparser(TestCase):
' target="_blank">https://groups.google.com/forum/?fromgroups#!categories/evennia/ainneve</a>',
)
def test_url_edge_leadingw(self):
self.assertEqual(
self.parser.convert_urls("wwww.example.com"),
'w<a href="www.example.com" target="_blank">www.example.com</a>',
)
def test_url_edge_following_period_eol(self):
self.assertEqual(
self.parser.convert_urls("www.example.com."),
'<a href="www.example.com" target="_blank">www.example.com</a>.',
'<a href="//www.example.com" target="_blank">www.example.com</a>.',
)
def test_url_edge_following_period(self):
self.assertEqual(
self.parser.convert_urls("see www.example.com. "),
'see <a href="www.example.com" target="_blank">www.example.com</a>. ',
'see <a href="//www.example.com" target="_blank">www.example.com</a>. ',
)
def test_url_edge_brackets(self):
@ -356,3 +350,9 @@ class TestTextToHTMLparser(TestCase):
'</span><a href="http://example.com/" target="_blank">'
'http://example.com/</a><span class="red">',
)
def test_non_url_with_www(self):
self.assertEqual(
self.parser.convert_urls('Awwww.this should not be highlighted'),
'Awwww.this should not be highlighted'
)

View file

@ -88,8 +88,9 @@ class TextToHTMLparser(object):
re.S | re.M | re.I,
)
re_url = re.compile(
r'(?<!=")((?:ftp|www|https?)\W+(?:(?!\.(?:\s|$)|&\w+;)[^"\',;$*^\\(){}<>\[\]\s])+)(\.(?:\s|$)|&\w+;|)'
r'(?<!=")(\b(?:ftp|www|https?)\W+(?:(?!\.(?:\s|$)|&\w+;)[^"\',;$*^\\(){}<>\[\]\s])+)(\.(?:\s|$)|&\w+;|)'
)
re_protocol = re.compile(r'^(?:ftp|https?)://')
re_mxplink = re.compile(r"\|lc(.*?)\|lt(.*?)\|le", re.DOTALL)
re_mxpurl = re.compile(r"\|lu(.*?)\|lt(.*?)\|le", re.DOTALL)
@ -147,9 +148,19 @@ class TextToHTMLparser(object):
text (str): Processed text.
"""
# -> added target to output prevent the web browser from attempting to
# change pages (and losing our webclient session).
return self.re_url.sub(r'<a href="\1" target="_blank">\1</a>\2', text)
m = self.re_url.search(text)
if m:
href = m.group(1)
label = href
# if there is no protocol (i.e. starts with www) prefix with // so the link isn't treated as relative
if not self.re_protocol.match(href):
href = "//" + href
rest = m.group(2)
# -> added target to output prevent the web browser from attempting to
# change pages (and losing our webclient session).
return text[:m.start()] + f'<a href="{href}" target="_blank">{label}</a>{rest}' + text[m.end():]
else:
return text
def sub_mxp_links(self, match):
"""