1 files changed, 5 insertions, 28 deletions
diff --git a/src/hydrilla/proxy/http_messages.py b/src/hydrilla/proxy/http_messages.py
index 718022f..74f1f02 100644
--- a/src/hydrilla/proxy/http_messages.py
+++ b/src/hydrilla/proxy/http_messages.py
@@ -30,6 +30,7 @@
 """
 
 import re
+import cgi
 import dataclasses as dc
 import typing as t
 import sys
@@ -120,42 +121,18 @@ def make_parsed_url(url: t.Union[str, url_patterns.ParsedUrl]) \
     return url_patterns.parse_url(url) if isinstance(url, str) else url
 
 
-# For details of 'Content-Type' header's structure, see:
-# https://datatracker.ietf.org/doc/html/rfc7231#section-3.1.1.1
-content_type_reg = re.compile(r'''
-^
-(?P<mime>[\w-]+/[\w-]+)
-\s*
-(?:
-    ;
-    (?:[^;]*;)* # match possible parameter other than "charset"
-)
-\s*
-charset=        # no whitespace allowed in parameter as per RFC
-(?P<encoding>
-    [\w-]+
-    |
-    "[\w-]+"    # quotes are optional per RFC
-)
-(?:;[^;]+)*     # match possible parameter other than "charset"
-$               # forbid possible dangling characters after closing '"'
-''', re.VERBOSE | re.IGNORECASE)
-
 @dc.dataclass(frozen=True)
 class HasHeadersMixin:
     headers: IHeaders
 
     def deduce_content_type(self) -> tuple[t.Optional[str], t.Optional[str]]:
-        content_type = self.headers.get('content-type')
-        if content_type is None:
-            return (None, None)
-
-        match = content_type_reg.match(content_type)
-        if match is None:
+        content_type_header = self.headers.get('content-type')
+        if content_type_header is None:
             return (None, None)
 
-        mime, encoding = match.group('mime'), match.group('encoding')
+        mime, options = cgi.parse_header(content_type_header)
 
+        encoding = options.get('charset')
         if encoding is not None:
             encoding = encoding.lower()