Merge pull request #814 from jonathanberthias/content-type-no-space

Fix content-type when no space after semicolon

Author: p1c2u

openapi_core/templating/media_types/finders.py

@@ -1,6 +1,7 @@
 """OpenAPI core templating media types finders module"""
 
 import fnmatch
+import re
 from typing import Mapping
 from typing import Tuple
 
@@ -38,12 +39,32 @@ def find(self, mimetype: str) -> MediaType:
         raise MediaTypeNotFound(mimetype, list(self.content.keys()))
 
     def _parse_mimetype(self, mimetype: str) -> Tuple[str, Mapping[str, str]]:
-        mimetype_parts = mimetype.split("; ")
-        mime_type = mimetype_parts[0]
+        mimetype_parts = mimetype.split(";")
+        mime_type = mimetype_parts[0].lower().rstrip()
         parameters = {}
         if len(mimetype_parts) > 1:
             parameters_list = (
-                param_str.split("=") for param_str in mimetype_parts[1:]
+                self._parse_parameter(param_str)
+                for param_str in mimetype_parts[1:]
             )
             parameters = dict(parameters_list)
         return mime_type, parameters
+
+    def _parse_parameter(self, parameter: str) -> Tuple[str, str]:
+        """Parse a parameter according to RFC 9110.
+
+        See https://www.rfc-editor.org/rfc/rfc9110.html#name-parameters
+
+        Important points:
+            * parameter names are case-insensitive
+            * parameter values are case-sensitive
+                except "charset" which is case-insensitive
+                https://www.rfc-editor.org/rfc/rfc2046#section-4.1.2
+        """
+        name, value = parameter.split("=")
+        name = name.lower().lstrip()
+        # remove surrounding quotes from value
+        value = re.sub('^"(.*)"$', r"\1", value, count=1)
+        if name == "charset":
+            value = value.lower()
+        return name, value.rstrip()

tests/unit/templating/test_media_types_finders.py

@@ -21,10 +21,19 @@ def content(self, spec):
     def finder(self, content):
         return MediaTypeFinder(content)
 
-    def test_charset(self, finder, content):
-        mimetype = "text/html; charset=utf-8"
-
-        mimetype, parameters, _ = finder.find(mimetype)
+    @pytest.mark.parametrize(
+        "media_type",
+        [
+            # equivalent according to RFC 9110
+            "text/html;charset=utf-8",
+            'Text/HTML;Charset="utf-8"',
+            'text/html; charset="utf-8"',
+            "text/html;charset=UTF-8",
+            "text/html ; charset=utf-8",
+        ],
+    )
+    def test_charset(self, finder, content, media_type):
+        mimetype, parameters, _ = finder.find(media_type)
         assert mimetype == "text/*"
         assert parameters == {"charset": "utf-8"}