Add support for multipart/form-data to FormRequest

docs/topics/request-response.rst

@@ -799,7 +799,7 @@ fields with form data from :class:`Response` objects.
 
 .. class:: scrapy.http.request.form.FormRequest
 .. class:: scrapy.http.FormRequest
-.. class:: scrapy.FormRequest(url, [formdata, ...])
+.. class:: scrapy.FormRequest(url, [formdata, is_multipart, ...])
 
     The :class:`FormRequest` class adds a new keyword parameter to the ``__init__`` method. The
     remaining arguments are the same as for the :class:`Request` class and are
@@ -810,10 +810,14 @@ fields with form data from :class:`Response` objects.
        body of the request.
     :type formdata: dict or collections.abc.Iterable
 
+    :param is_multipart: is a boolean flag to determine if the ``Content-Type`` should be sent
+        as ``multipart/form-data``.
+    :type is_multipart: bool
+
     The :class:`FormRequest` objects support the following class method in
     addition to the standard :class:`Request` methods:
 
-    .. classmethod:: FormRequest.from_response(response, [formname=None, formid=None, formnumber=0, formdata=None, formxpath=None, formcss=None, clickdata=None, dont_click=False, ...])
+    .. classmethod:: FormRequest.from_response(response, [formname=None, formid=None, formnumber=0, formdata=None, formxpath=None, formcss=None, is_multipart=False, clickdata=None, dont_click=False, ...])
 
        Returns a new :class:`FormRequest` object with its form field values
        pre-populated with those found in the HTML ``<form>`` element contained
@@ -862,6 +866,10 @@ fields with form data from :class:`Response` objects.
           request, even if it was present in the response ``<form>`` element.
        :type formdata: dict
 
+       :param is_multipart: if True, the form will be sent using the header
+        ``Content-Type: multipart/form-data``.
+        :type is_multipart: bool
+
        :param clickdata: attributes to lookup the control clicked. If it's not
          given, the form data will be submitted simulating a click on the
          first clickable element. In addition to html attributes, the control

scrapy/http/request/form.py

@@ -34,7 +34,11 @@ class FormRequest(Request):
     valid_form_methods = ["GET", "POST"]
 
     def __init__(
-        self, *args: Any, formdata: FormdataType = None, **kwargs: Any
+        self,
+        *args: Any,
+        formdata: FormdataType = None,
+        is_multipart: bool = False,
+        **kwargs: Any,
     ) -> None:
         if formdata and kwargs.get("method") is None:
             kwargs["method"] = "POST"
@@ -46,7 +50,12 @@ def __init__(
             form_query_str = _urlencode(items, self.encoding)
             if self.method == "POST":
                 self.headers.setdefault(
-                    b"Content-Type", b"application/x-www-form-urlencoded"
+                    b"Content-Type",
+                    (
+                        b"application/x-www-form-urlencoded"
+                        if not is_multipart
+                        else b"multipart/form-data"
+                    ),
                 )
                 self._set_body(form_query_str)
             else:

tests/test_http_request.py

@@ -514,6 +514,37 @@ def test_multi_key_values(self):
             r3.body, b"colours=red&colours=blue&colours=green&price=%C2%A3+100"
         )
 
+    def test_multipart_formdata(self):
+        data = {"foo": "bar"}
+        r1 = self.request_class(
+            "http://www.example.com", formdata=data, is_multipart=True
+        )
+        self.assertEqual(r1.headers[b"Content-Type"], b"multipart/form-data")
+
+    def test_from_response_multipart_formdata(self):
+        response = _buildresponse(
+            b"""<form action="post.php" method="POST" enctype="multipart/form-data">
+            <input type="hidden" name="foo" value="bar">
+            <input type="hidden" name="foo" value="baz">
+            <input type="hidden" name="bar" value="baz">
+            </form>""",
+            url="http://www.example.com/this/list.html",
+        )
+        req = self.request_class.from_response(
+            response,
+            formdata={"fullname": ["john", "doe"], "age": "30"},
+            is_multipart=True,
+        )
+
+        self.assertEqual(req.method, "POST")
+        self.assertEqual(req.headers[b"Content-type"], b"multipart/form-data")
+        self.assertEqual(req.url, "http://www.example.com/this/post.php")
+        fs = _qs(req)
+        self.assertEqual(set(fs[b"foo"]), {b"bar", b"baz"})
+        self.assertEqual(set(fs[b"fullname"]), {b"john", b"doe"})
+        self.assertEqual(fs[b"bar"], [b"baz"])
+        self.assertEqual(fs[b"age"], [b"30"])
+
     def test_from_response_post(self):
         response = _buildresponse(
             b"""<form action="post.php" method="POST">