async def test_async_try_login_basic_digest_ntlm_wrong_credentials():
target_url = "http://perdu.com/"
respx.get(target_url).mock(return_value=httpx.Response(
200,
text=
"<html><head><title>Vous Etes Perdu ?</title></head><body><h1>Perdu sur l'Internet ?</h1> \
<h2>Pas de panique, on va vous aider</h2> \
<strong><pre> * <----- vous êtes ici</pre></strong><a href='http://perdu.com/foobar/'></a> \
<a href='http://perdu.com/foobar/signout'></a> \
<div><a href='http://perdu.com/a/b/signout'></a></div></body></html>"
))
auth_urls = [["http://perdu.com/login1", 401],
["http://perdu.com/login2", 403],
["http://perdu.com/login3", 404]]
crawler = AsyncCrawler(Request(target_url), timeout=1)
crawler._auth_credentials = ["username", "password"]
for auth_url, status_code in auth_urls:
respx.get(auth_url).mock(
return_value=httpx.Response(status_code, text="KO"))
is_logged_in, form, disconnect_urls = await crawler._async_try_login_basic_digest_ntlm(
auth_url)
assert is_logged_in is False
assert len(form) == 0
assert len(disconnect_urls) == 0
async def test_async_try_login_post_form_not_detected():
target_url = "http://perdu.com/"
respx.get(target_url).mock(return_value=httpx.Response(
200,
text=
"<html><head><title>Vous Etes Perdu ?</title></head><body><h1>Perdu sur l'Internet ?</h1> \
<h2>Pas de panique, on va vous aider</h2> \
<strong><pre> * <----- vous êtes ici</pre></strong><a href='http://perdu.com/foobar/'></a> \
<a href='http://perdu.com/foobar/signout'></a> \
<div><a href='http://perdu.com/a/b/signout'></a></div></body></html>"
))
crawler = AsyncCrawler(Request(target_url), timeout=1)
crawler._auth_credentials = ["username", "password"]
is_logged_in, form, disconnect_urls = await crawler._async_try_login_post(
"username", "password", target_url)
assert form == {}
assert len(disconnect_urls) == 0
assert is_logged_in is False
async def test_async_try_login_post_good_credentials():
target_url = "http://perdu.com/"
respx.get(target_url).mock(return_value=httpx.Response(
200,
text=
"<html><head><title>Vous Etes Perdu ?</title></head><body><h1>Perdu sur l'Internet ?</h1> \
<h2>Pas de panique, on va vous aider</h2> \
<strong><pre> * <----- vous êtes ici</pre></strong><a href='http://perdu.com/foobar/'></a> \
<a href='http://perdu.com/foobar/signout'></a> \
<form name='loginform' method='post' action='userinfo.php'> \
<table cellpadding='4' cellspacing='1'> \
<tbody><tr><td>Username : </td><td><input name='uname' type='text' size='20'></td></tr> \
<tr><td>Password : </td><td><input name='pass' type='password' size='20'></td></tr> \
<tr><td colspan='2' align='right'><input type='submit' value='login' style='width:75px;'></td></tr> \
</tbody></table> \
</form> \
<div><a href='http://perdu.com/a/b/signout'></a></div></body></html>"
))
respx.post(target_url + 'userinfo.php').mock(return_value=httpx.Response(
200,
text=
"<html><head><title>Vous Etes Perdu ?</title></head><body><h1>Perdu sur l'Internet ?</h1> \
<h2>Pas de panique, on va vous aider</h2> \
<strong><pre> * <----- vous êtes ici</pre></strong><a href='http://perdu.com/foobar/'></a> \
<a href='http://perdu.com/foobar/signout'>disconnect</a> \
<div><a href='http://perdu.com/a/b/signout'></a></div></body></html>"
))
crawler = AsyncCrawler(Request(target_url), timeout=1)
crawler._auth_credentials = ["username", "password"]
is_logged_in, form, disconnect_urls = await crawler._async_try_login_post(
"username", "password", target_url)
assert form == {'login_field': 'uname', 'password_field': 'pass'}
assert len(disconnect_urls) == 2
assert "http://perdu.com/foobar/signout" in disconnect_urls
assert "http://perdu.com/a/b/signout" in disconnect_urls
assert is_logged_in is True
