content_types on the Content API, not the Streams API (#124)

Author: lovelydinosaur

docs/content-types.md

@@ -164,7 +164,8 @@ An interface for constructing HTTP content, along with relevant headers.
 
 The following method must be implemented...
 
-* `.encode()` - Returns a tuple of `(httx.Stream, str)`, representing the encoded data and the content type.
+* `.encode()` - Returns an `httx.Stream` representing the encoded data.
+* `.content_type()` - Returns a `str` indicating the content type.
 
 ---
 

docs/streams.md

@@ -14,7 +14,6 @@ The base `Stream` class. The core of the interface is a subset of Python's `io.I
 Additionally, the following properties are also defined...
 
 * `.size` - *(int or None)* Return an integer indicating the size of the stream, or `None` if the size is unknown. When working with HTTP this is used to either set a `Content-Length: <size>` header, or a `Content-Encoding: chunked` header.
-* `.content_type` - *(str or None)* Return a string indicating the content type of the data, or `None` if the content type is unknown. When working with HTTP this is used to optionally set a `Content-Type` header.
 
 The `Stream` interface and `ContentType` interface are related, with streams being used as the abstraction for the bytewise representation, and content types being used to encapsulate the parsed data structure.
 

src/ahttpx/_content.py

@@ -16,11 +16,26 @@
     "HTML",
 ]
 
+# https://github.com/nginx/nginx/blob/master/conf/mime.types
+_content_types = {
+    ".json": "application/json",
+    ".js": "application/javascript",
+    ".html": "text/html",
+    ".css": "text/css",
+    ".png": "image/png",
+    ".jpeg": "image/jpeg",
+    ".jpg": "image/jpeg",
+    ".gif": "image/gif",
+}
+
 
 class Content:
     def encode(self) -> Stream:
         raise NotImplementedError()
 
+    def content_type(self) -> str:
+        raise NotImplementedError()
+
 
 class Form(typing.Mapping[str, str], Content):
     """
@@ -63,8 +78,10 @@ def __init__(
 
     def encode(self) -> Stream:
         content = str(self).encode("ascii")
-        content_type = "application/x-www-form-urlencoded"
-        return ByteStream(content, content_type)
+        return ByteStream(content)
+
+    def content_type(self) -> str:
+        return "application/x-www-form-urlencoded"
 
     # Dict operations
 
@@ -163,6 +180,13 @@ def size(self) -> int:
     def encode(self) -> Stream:
         return FileStream(self._path)
 
+    def content_type(self) -> str:
+        _, ext = os.path.splitext(self._path)
+        ct = _content_types.get(ext, "application/octet-stream")
+        if ct.startswith('text/'):
+            ct += "; charset='utf-8'"
+        return ct
+
     def __lt__(self, other: typing.Any) -> bool:
         return isinstance(other, File) and other._path < self._path
 
@@ -185,6 +209,7 @@ def __init__(
             | typing.Sequence[tuple[str, File]]
             | None
         ) = None,
+        boundary: str = ''
     ) -> None:
         d: dict[str, list[File]] = {}
 
@@ -198,6 +223,7 @@ def __init__(
                 d.setdefault(k, []).append(v)
 
         self._dict = d
+        self._boundary = boundary or os.urandom(16).hex()
 
     # Standard dict interface
     def keys(self) -> typing.KeysView[str]:
@@ -231,6 +257,9 @@ def get_list(self, key: str) -> list[File]:
     def encode(self) -> Stream:
         return MultiPart(files=self).encode()
 
+    def content_type(self) -> str:
+        return f"multipart/form-data; boundary={self._boundary}"
+
     # Builtins
     def __getitem__(self, key: str) -> File:
         return self._dict[key][0]
@@ -268,8 +297,10 @@ def encode(self) -> Stream:
             separators=(",", ":"),
             allow_nan=False
         ).encode("utf-8")
-        content_type = "application/json"
-        return ByteStream(content, content_type)
+        return ByteStream(content)
+
+    def content_type(self) -> str:
+        return "application/json"
 
     def __repr__(self) -> str:
         return f"<JSON {self._data!r}>"
@@ -281,8 +312,10 @@ def __init__(self, text: str) -> None:
 
     def encode(self) -> Stream:
         content = self._text.encode("utf-8")
-        content_type = "text/plain; charset='utf-8'"
-        return ByteStream(content, content_type)
+        return ByteStream(content)
+
+    def content_type(self) -> str:
+        return "text/plain; charset='utf-8'"
 
     def __repr__(self) -> str:
         return f"<Text {self._text!r}>"
@@ -294,8 +327,10 @@ def __init__(self, text: str) -> None:
 
     def encode(self) -> Stream:
         content = self._text.encode("utf-8")
-        content_type = "text/html; charset='utf-8'"
-        return ByteStream(content, content_type)
+        return ByteStream(content)
+
+    def content_type(self) -> str:
+        return "text/html; charset='utf-8'"
 
     def __repr__(self) -> str:
         return f"<HTML {self._text!r}>"
@@ -336,5 +371,8 @@ def encode(self) -> Stream:
         files = [(key, file._path) for key, file in self._files.items()]
         return MultiPartStream(form, files, boundary=self._boundary)
 
+    def content_type(self) -> str:
+        return f"multipart/form-data; boundary={self._boundary}"
+
     def __repr__(self) -> str:
         return f"<MultiPart form={self._form.multi_items()!r}, files={self._files.multi_items()!r}>"

src/ahttpx/_request.py

@@ -35,7 +35,9 @@ def __init__(
             elif isinstance(content, Stream):
                 self.stream = content
             elif isinstance(content, Content):
+                ct = content.content_type()
                 self.stream = content.encode()
+                self.headers = self.headers.copy_set("Content-Type", ct)
             else:
                 raise TypeError(f'Expected `Content | Stream | bytes | None` got {type(content)}')
 
@@ -51,9 +53,6 @@ def __init__(
             elif content_length > 0:
                 self.headers = self.headers.copy_set("Content-Length", str(content_length))
 
-            if self.stream.content_type is not None:
-                self.headers = self.headers.copy_set("Content-Type", self.stream.content_type)
-
     @property
     def body(self) -> bytes:
         if not hasattr(self, '_body'):

src/ahttpx/_response.py

@@ -94,7 +94,9 @@ def __init__(
             elif isinstance(content, Stream):
                 self.stream = content
             elif isinstance(content, Content):
+                ct = content.content_type()
                 self.stream = content.encode()
+                self.headers = self.headers.copy_set("Content-Type", ct)
             else:
                 raise TypeError(f'Expected `Content | Stream | bytes | None` got {type(content)}')
 
@@ -111,9 +113,6 @@ def __init__(
             else:
                 self.headers = self.headers.copy_set("Content-Length", str(content_length))
 
-            if self.stream.content_type is not None:
-                self.headers = self.headers.copy_set("Content-Type", self.stream.content_type)
-
     @property
     def reason_phrase(self):
         return _codes.get(self.status_code, "Unknown Status Code")
@@ -129,9 +128,11 @@ def text(self) -> str:
         if not hasattr(self, '_body'):
             raise RuntimeError("'.text' cannot be accessed without calling '.read()'")
         if not hasattr(self, '_text'):
-            content_type = self.headers.get('Content-Type', '')
-            media, opts = parse_opts_header(content_type)
-            charset = opts.get('charset', 'utf-8')
+            ct = self.headers.get('Content-Type', '')
+            media, opts = parse_opts_header(ct)
+            charset = 'utf-8'
+            if media.startswith('text/'):
+                charset = opts.get('charset', 'utf-8')
             self._text = self._body.decode(charset)
         return self._text
 

src/ahttpx/_streams.py

@@ -3,19 +3,6 @@
 import os
 
 
-# https://github.com/nginx/nginx/blob/master/conf/mime.types
-_content_types = {
-    ".json": "application/json",
-    ".js": "application/javascript",
-    ".html": "text/html",
-    ".css": "text/css",
-    ".png": "image/png",
-    ".jpeg": "image/jpeg",
-    ".jpg": "image/jpeg",
-    ".gif": "image/gif",
-}
-
-
 class Stream:
     async def read(self, size: int=-1) -> bytes:
         raise NotImplementedError()
@@ -30,10 +17,6 @@ async def close(self) -> None:
     def size(self) -> int | None:
         return None
 
-    @property
-    def content_type(self) -> str | None:
-        return None
-
     async def __aenter__(self):
         return self
 
@@ -47,10 +30,9 @@ async def __aexit__(
 
 
 class ByteStream(Stream):
-    def __init__(self, data: bytes = b'', content_type: str | None = None, mode='r'):
+    def __init__(self, data: bytes = b'', mode='r'):
         self._buffer = io.BytesIO(data)
         self._size = len(data)
-        self._content_type = content_type
         self._allow_write = {'r': False, 'w': True}[mode]
 
     async def read(self, size: int=-1) -> bytes:
@@ -71,10 +53,6 @@ def getvalue(self) -> bytes:
     def size(self) -> int | None:
         return self._size
 
-    @property
-    def content_type(self) -> str | None:
-        return self._content_type
-
 
 class FileStream(Stream):
     def __init__(self, path):
@@ -100,14 +78,6 @@ async def close(self) -> None:
     def size(self) -> int | None:
         return self._size
 
-    @property
-    def content_type(self) -> str | None:
-        _, ext = os.path.splitext(self._path)
-        ct = _content_types.get(ext, "application/octet-stream")
-        if ct is not None and ct.startswith('text/'):
-            ct += "; charset='utf-8'"
-        return ct
-
     async def __aenter__(self):
         await self.open()
         return self
@@ -243,7 +213,3 @@ async def close(self) -> None:
     @property
     def size(self) -> int | None:
         return None
-
-    @property
-    def content_type(self) -> str | None:
-        return f"multipart/form-data; boundary={self._boundary}"