Add support for Zone Identifiers from RFC 6874

Refs #2

docs/source/index.rst

@@ -3,7 +3,8 @@
 =========
 
 |rfc3986| is a Python implementation of :rfc:`3986` including validation and 
-authority parsing.
+authority parsing. This module also supports :rfc:`6874` which adds support
+for zone identifiers to IPv6 Addresses.
 
 The maintainers strongly suggest using `pip`_ to install |rfc3986|. For 
 example,

src/rfc3986/abnf_regexp.py

@@ -31,6 +31,10 @@ NON_PCT_ENCODED_SET = RESERVED_CHARS_SET.union(UNRESERVED_CHARS_SET).union('%')
 # We need to escape the '-' in this case:
 UNRESERVED_RE = 'A-Za-z0-9._~\-'
 
+# Percent encoded character values
+PERCENT_ENCODED = PCT_ENCODED = '%[A-Fa-f0-9]{2}'
+PCHAR = '([' + UNRESERVED_RE + SUB_DELIMITERS_RE + ':@]|%s)' % PCT_ENCODED
+
 # NOTE(sigmavirus24): We're going to use more strict regular expressions
 # than appear in Appendix B for scheme. This will prevent over-eager
 # consuming of items that aren't schemes.
@@ -111,7 +115,16 @@ IPv_FUTURE_RE = 'v[0-9A-Fa-f]+.[%s]+' % (
     UNRESERVED_RE + SUB_DELIMITERS_RE + ':'
 )
 
-IP_LITERAL_RE = '\[({0}|{1})\]'.format(IPv6_RE, IPv_FUTURE_RE)
+
+# RFC 6874 Zone ID ABNF
+ZONE_ID = '(?:[' + UNRESERVED_RE + ']|' + PCT_ENCODED + ')+'
+IPv6_ADDRZ_RE = IPv6_RE + '%25' + ZONE_ID
+
+IP_LITERAL_RE = '\[({0}|(?:{1})|{2})\]'.format(
+    IPv6_RE,
+    IPv6_ADDRZ_RE,
+    IPv_FUTURE_RE,
+)
 
 # Pattern for matching the host piece of the authority
 HOST_RE = HOST_PATTERN = '({0}|{1}|{2})'.format(
@@ -128,10 +141,6 @@ PORT_RE = '[0-9]{1,5}'
 
 # See http://tools.ietf.org/html/rfc3986#section-3.3 for more information
 # about the path patterns defined below.
-
-# Percent encoded character values
-PERCENT_ENCODED = PCT_ENCODED = '%[A-Fa-f0-9]{2}'
-PCHAR = '([' + UNRESERVED_RE + SUB_DELIMITERS_RE + ':@]|%s)' % PCT_ENCODED
 segments = {
     'segment': PCHAR + '*',
     # Non-zero length segment

src/rfc3986/misc.py

@@ -51,7 +51,7 @@ URI_MATCHER = re.compile(abnf_regexp.URL_PARSING_RE)
 
 SUBAUTHORITY_MATCHER = re.compile((
     '^(?:(?P<userinfo>{0})@)?'  # userinfo
-    '(?P<host>{1}?)'  # host
+    '(?P<host>{1})'  # host
     ':?(?P<port>{2})?$'  # port
     ).format(abnf_regexp.USERINFO_RE,
              abnf_regexp.HOST_PATTERN,

tests/conftest.py

@@ -7,16 +7,30 @@ import pytest
 SNOWMAN = b'\xe2\x98\x83'
 
 valid_hosts = [
-    '[21DA:00D3:0000:2F3B:02AA:00FF:FE28:9C5A]', '[::1]',
-    '[21DA:D3:0:2F3B:2AA:FF:FE28:9C5A]', '[FE80::2AA:FF:FE9A:4CA2]',
-    '[FF02::2]', '[FF02:3::5]', '[FF02:0:0:0:0:0:0:2]',
-    '[FF02:30:0:0:0:0:0:5]', '127.0.0.1', 'www.example.com', 'localhost',
+    '[21DA:00D3:0000:2F3B:02AA:00FF:FE28:9C5A]',
+    '[::1]',
+    '[::1%25lo]',  # With ZoneID
+    '[FF02:0:0:0:0:0:0:2%25en01]',  # With ZoneID
+    '[FF02:30:0:0:0:0:0:5%25en1]',  # With ZoneID
+    '[21DA:D3:0:2F3B:2AA:FF:FE28:9C5A]',
+    '[FE80::2AA:FF:FE9A:4CA2]',
+    '[FF02::2]',
+    '[FF02:3::5]',
+    '[FF02:0:0:0:0:0:0:2]',
+    '[FF02:30:0:0:0:0:0:5]',
+    '127.0.0.1',
+    'www.example.com',
+    'localhost',
     'http-bin.org',
     ]
 
 invalid_hosts = [
     '[FF02::3::5]',  # IPv6 can only have one ::
     '[FADF:01]',  # Not properly compacted (missing a :)
+    '[FADF:01%en0]',  # Not properly compacted (missing a :), Invalid ZoneID
+    '[FADF::01%en0]',  # Invalid ZoneID separator
+    '[FADF::01%]',  # Invalid ZoneID separator and no ZoneID
+    '[FADF::01%25]',  # Missing ZoneID
     'localhost:80:80:80',  # Too many ports
     '256.256.256.256',  # Invalid IPv4 Address
     SNOWMAN.decode('utf-8')