BUG: url regex in style_render does not pass colon and other valid

characters

URLs containing some valid characters such as colon in port numbers get
cut off when html-formatting. As a workaround, expanded the regex to
match a wider variety of URLs.

Author: kianelbo

pandas/io/formats/style_render.py

@@ -1589,7 +1589,7 @@ def _render_href(x, format):
             href = r"\href{{{0}}}{{{0}}}"
         else:
             raise ValueError("``hyperlinks`` format can only be 'html' or 'latex'")
-        pat = r"(https?:\/\/|ftp:\/\/|www.)[\w/\-?=%.]+\.[\w/\-&?=%.]+"
+        pat = r"(https?:\/\/|ftp:\/\/|www.)[\w/\-?=%.]+\.[\w/\-&?=%.,':;~!@#$*()\[\]]+"
         return re.sub(pat, lambda m: href.format(m.group(0)), x)
     return x
 

pandas/tests/io/formats/style/test_html.py

@@ -780,6 +780,8 @@ def test_hiding_index_columns_multiindex_trimming():
         ("ftp scheme: ftp://www.web", True, "ftp://www.web"),
         ("subdirectories: www.web.com/directory", True, "www.web.com/directory"),
         ("Multiple domains: www.1.2.3.4", True, "www.1.2.3.4"),
+        ("with port: http://web.com:80", True, "http://web.com:80"),
+        ("with anchor: http://web.com/anc#hor", True, "http://web.com/anc#hor"),
     ],
 )
 def test_rendered_links(type, text, exp, found):