Skip to content

Scrapy, a fast high-level web crawling & scraping framework for Python.

Fixed bug FormRequest.from_response() clickdata ignores input[type=image]

Viral Mehtad5b7ebcfdcfd
diff --git a/scrapy/http/request/form.py b/scrapy/http/request/form.py
index d9d178a3e..184ee2599 100644
--- a/scrapy/http/request/form.py
+++ b/scrapy/http/request/form.py
@@ -170,10 +170,12 @@ def _get_clickable(clickdata, form):
"""
clickables = [
el for el in form.xpath(
- 'descendant::*[(self::input or self::button)'
- ' and re:test(@type, "^submit$", "i")]'
- '|descendant::button[not(@type)]',
- namespaces={"re": "http://exslt.org/regular-expressions"})
+ 'descendant::*[(self::input or self::button)'
+ ' and re:test(@type, "^submit$", "i")]'
+ '|descendant::*[(self::input or self::button)'
+ ' and re:test(@type, "^image$", "i")]'
+ '|descendant::button[not(@type)]',
+ namespaces={"re": "http://exslt.org/regular-expressions"})
]
if not clickables:
return
diff --git a/tests/test_http_request.py b/tests/test_http_request.py
index fca8ff411..73a74cd5d 100644
--- a/tests/test_http_request.py
+++ b/tests/test_http_request.py
@@ -532,6 +532,16 @@ class FormRequestTest(RequestTest):
req = self.request_class.from_response(response, dont_click=True)
fs = _qs(req)
self.assertEqual(fs, {b'i1': [b'i1v']})
+
+ def test_from_response_clickdata_does_not_ignore_image(self):
+ response = _buildresponse(
+ """<form>
+ <input type="text" name="i1" value="i1v">
+ <input id="image" name="i2" type="image" value="i2v" alt="Login" src="http://my.image.org/1.jpg">
+ </form>""")
+ req = self.request_class.from_response(response, dont_click=True)
+ fs = _qs(req)
+ self.assertEqual(fs, {b'i1': [b'i1v'], b'i2': [b'i2v']})
def test_from_response_dont_submit_reset_as_input(self):
response = _buildresponse(