Git-Mirrors/brozzler
1
0
Fork 0
mirror of https://github.com/internetarchive/brozzler.git synced 2025-04-20 23:56:34 -04:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

Merge pull request #195 from NGTmeaty/improve-login

Browse Source 
for qa test crawls
This commit is contained in:
Barbara Miller 2020-04-17 10:12:17 -07:00 committed by GitHub
commit ce2fe4754f
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
4 changed files with 81 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

38
brozzler/js-templates/try-login.js.j2
View File

@ -1,6 +1,12 @@
var __brzl_tryLoginState = 'trying';
var __brzl_tryLogin = function() {
var login_open = document.querySelectorAll("[class^='login-open']");
for(var x = 0; x < login_open.length; x++){
button = login_open[x];
button.click();
}
for (var i = 0; i < document.forms.length; i++) {
var form = document.forms[i];
if (form.method != 'post') {
@ -48,6 +54,38 @@ var __brzl_tryLogin = function() {
return;
}
}
if(__brzl_tryLoginState === 'trying'){
var submit = undefined;
var usernameInput = undefined;
var passwordField = undefined;
var login = document.querySelectorAll("[class^='login']");
for (var z = 0; z < login.length; z++){
var input = login[z];
if (input.type === "text" || input.type == "email") {
usernameField = input;
} else if (input.type === "password") {
passwordField = input;
} else if(input.onclick != null) {
submit = input;
} else if (input.type == "textarea") {
usernameField = undefined;
passwordField = undefined;
break;
}
}
if (submit && usernameField && passwordField){
usernameField.value = {{username|json}};
passwordField.value = {{password|json}};
submit.click();
__brzl_tryLoginState = 'maybe-submitted-form';
return;
}
}
__brzl_tryLoginState = 'login-form-not-found';
};

4
setup.py
View File

@ -2,7 +2,7 @@
'''
setup.py - brozzler setup script
Copyright (C) 2014-2019 Internet Archive
Copyright (C) 2014-2020 Internet Archive
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
@ -32,7 +32,7 @@ def find_package_data(package):
setuptools.setup(
name='brozzler',
version='1.5.19',
version='1.5.20',
description='Distributed web crawling with browsers',
url='https://github.com/internetarchive/brozzler',
author='Noah Levitt',

25
tests/htdocs/site11/form2.html Normal file
View File

@ -0,0 +1,25 @@
<html>
<head>
<title>brozzler login form test 2</title>
</head>
<body>
<div>
<a class="login-open-bt" onclick="document.getElementsByClassName('login-box')[0].style = '';">Sign In</li>
<div class="login-box" style="display: none;">
<div class="login-main">
<ul class="login-list">
<li>
<input type="text" class="login-ipt" placeholder="Email" id="loginEmail">
</li>
<li>
<input type="password" class="login-ipt" placeholder="Password" id="loginPsw">
</li>
<li>
<div class="login-bt" onclick="window.location.href = '/login-action';">Sign In</div>
</li>
<br>
</ul>
</div>
</div>
</body>
</html>

16
tests/test_brozzling.py
View File

@ -80,6 +80,8 @@ def httpd(request):
self.send_header('Content-Length', len(payload))
self.end_headers()
self.wfile.write(payload)
else:
super().do_POST()
# SimpleHTTPRequestHandler always uses CWD so we have to chdir
@ -265,6 +267,7 @@ def test_try_login(httpd):
response_urls.append(msg['params']['response']['url'])
chrome_exe = brozzler.suggest_default_chrome_exe()
form_url = 'http://localhost:%s/site11/form1.html' % httpd.server_port
form_url_other = 'http://localhost:%s/site11/form2.html' % httpd.server_port
favicon_url = 'http://localhost:%s/favicon.ico' % httpd.server_port
login_url = 'http://localhost:%s/login-action' % httpd.server_port
# When username and password are defined and initial page has login form,
@ -280,6 +283,19 @@ def test_try_login(httpd):
assert response_urls[2] == login_url
assert response_urls[3] == form_url
# We are now supporting a different type of form, we'll test that here.
response_urls = []
username = 'user1'
password = 'pass1'
with brozzler.Browser(chrome_exe=chrome_exe) as browser:
browser.browse_page(form_url_other, username=username, password=password,
on_response=on_response)
assert len(response_urls) == 4
assert response_urls[0] == form_url_other
assert response_urls[1] == favicon_url
assert response_urls[2] == login_url
assert response_urls[3] == form_url_other
# When username and password are not defined, just load the initial page.
response_urls = []
with brozzler.Browser(chrome_exe=chrome_exe) as browser: