Apply UTF-8 globally (#78)

lilya/_internal/_parsers.py

@@ -52,7 +52,7 @@ def _user_safe_decode(src: bytes, codec: str) -> str:
     try:
         return src.decode(codec)
     except (UnicodeDecodeError, LookupError):
-        return src.decode("latin-1")
+        return src.decode("utf-8")
 
 
 class MultiPartException(Exception):
@@ -172,8 +172,8 @@ async def parse(self) -> FormData:
                 elif message_type == FormMessage.FIELD_DATA:
                     field_value += message_bytes
                 elif message_type == FormMessage.FIELD_END:
-                    name = unquote_plus(field_name.decode("latin-1"))
-                    value = unquote_plus(field_value.decode("latin-1"))
+                    name = unquote_plus(field_name.decode("utf-8"))
+                    value = unquote_plus(field_value.decode("utf-8"))
                     items.append((name, value))
 
         return FormData(items)
@@ -434,7 +434,7 @@ def _parse_content_type_header(self, params: Any) -> None:
             params (Any): Parameters from the Content-Type header.
         """
         charset = params.get(b"charset", "utf-8")
-        self._charset = charset.decode("latin-1") if isinstance(charset, bytes) else charset
+        self._charset = charset.decode("utf-8") if isinstance(charset, bytes) else charset
 
     def _get_multipart_boundary(self, params: dict[bytes, bytes]) -> bytes:
         """

lilya/datastructures.py

@@ -163,13 +163,13 @@ def parse_headers(self, value: Any) -> list[tuple[str, Any]]:
 
         if isinstance(value, dict):
             for k, v in value.items():
-                key = k.decode("latin-1") if isinstance(k, bytes) else k
-                value = v.decode("latin-1") if isinstance(v, bytes) else v
+                key = k.decode("utf-8") if isinstance(k, bytes) else k
+                value = v.decode("utf-8") if isinstance(v, bytes) else v
                 headers.append((key, value))
         if isinstance(value, list):
             for k, v in value:
-                key = k.decode("latin-1") if isinstance(k, bytes) else k
-                value = v.decode("latin-1") if isinstance(v, bytes) else v
+                key = k.decode("utf-8") if isinstance(k, bytes) else k
+                value = v.decode("utf-8") if isinstance(v, bytes) else v
                 headers.append((key, value))
 
         return headers
@@ -347,7 +347,7 @@ def build_from_scope(cls, scope: Scope) -> URL:
         host_header = None
         for key, value in scope["headers"]:
             if key == b"host":
-                host_header = value.decode("latin-1")
+                host_header = value.decode("utf-8")
                 break
 
         if host_header is not None:
@@ -625,7 +625,7 @@ def __init__(
         if isinstance(value, str):  # type: ignore
             super().__init__(parse_qsl(value, keep_blank_values=True))  # type: ignore
         elif isinstance(value, bytes):  # type: ignore
-            super().__init__(parse_qsl(value.decode("latin-1"), keep_blank_values=True))  # type: ignore
+            super().__init__(parse_qsl(value.decode("utf-8"), keep_blank_values=True))  # type: ignore
         else:
             super().__init__(*args)
 

lilya/requests.py

@@ -327,7 +327,7 @@ async def send_push_promise(self, path: str) -> None:
             raw_headers: list[tuple[bytes, bytes]] = []
             for name in SERVER_PUSH_HEADERS:
                 for value in self.headers.getlist(name):
-                    raw_headers.append((name.encode("latin-1"), value.encode("latin-1")))
+                    raw_headers.append((name.encode("utf-8"), value.encode("utf-8")))
             await self._send({"type": "http.response.push", "path": path, "headers": raw_headers})
 
     def __repr__(self) -> str:

lilya/responses.py

@@ -107,7 +107,7 @@ def make_headers(
                 headers.setdefault("content-type", content_type)
 
         raw_headers = [
-            (name.encode("latin-1"), f"{value}".encode(errors="surrogateescape"))
+            (name.encode("utf-8"), f"{value}".encode(errors="surrogateescape"))
             for name, value in headers.items()
         ]
         self.raw_headers = raw_headers
@@ -173,7 +173,7 @@ def set_cookie(
             ], "samesite must be either 'strict', 'lax' or 'none'"
             cookie[key]["samesite"] = samesite
         cookie_val = cookie.output(header="").strip()
-        self.headers.add("set-cookie", cookie_val.encode("latin-1"))
+        self.headers.add("set-cookie", cookie_val.encode("utf-8"))
 
     def delete_cookie(
         self,
@@ -212,11 +212,9 @@ def encoded_headers(self) -> list[Any]:
 
         for name, value in list(self.headers.multi_items()):
             if not isinstance(value, bytes):
-                headers.append(
-                    (name.encode("latin-1"), f"{value}".encode(errors="surrogateescape"))
-                )
+                headers.append((name.encode("utf-8"), f"{value}".encode(errors="surrogateescape")))
             else:
-                headers.append((name.encode("latin-1"), value))
+                headers.append((name.encode("utf-8"), value))
         return headers
 
     def message(self, prefix: str) -> dict[str, Any]: