Homec4science
Diffusion invenio-infoscience (maint-2.0) Browse

R3600/invenio/legacy/webhelp/web/hacking/test-suite.webdoc
2ff09fdadaa3maint-2.0

Last Change
Raw File

test-suite.webdoc
View Options

## This file is part of Invenio.
## Copyright (C) 2007, 2008, 2010, 2011 CERN.
##
## Invenio is free software; you can redistribute it and/or
## modify it under the terms of the GNU General Public License as
## published by the Free Software Foundation; either version 2 of the
## License, or (at your option) any later version.
##
## Invenio is distributed in the hope that it will be useful, but
## WITHOUT ANY WARRANTY; without even the implied warranty of
## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
## General Public License for more details.
##
## You should have received a copy of the GNU General Public License
## along with Invenio; if not, write to the Free Software Foundation, Inc.,
## 59 Temple Place, Suite 330, Boston, MA 02111-1307, USA.
<!-- WebDoc-Page-Title: Test Suite Strategy -->
<!-- WebDoc-Page-Navtrail: <a class="navtrail" href="<CFG_SITE_URL>/help/hacking">Hacking Invenio</a> -->
<!-- WebDoc-Page-Revision: $Id$ -->
<h2>Contents</h2>
<strong>1. <a href="#1">General considerations</a></strong><br />
<strong>2. <a href="#2">Unit testing</a></strong><br />
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 2.1 <a href="#2.1">Unit testing philosophy</a><br />
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 2.2 <a href="#2.2">Writing unit tests</a><br />
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 2.3 <a href="#2.3">Running unit tests</a><br />
<strong>3. <a href="#3">Regression testing</a></strong><br />
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 3.1 <a href="#3.1">Regression testing philosophy</a><br />
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 3.2 <a href="#3.2">Writing regression tests</a><br />
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 3.3 <a href="#3.3">Running regression tests</a><br />
<strong>4. <a href="#4">Web testing (via real browser)</a></strong><br />
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 4.1 <a href="#4.1">Web testing philosophy</a><br />
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 4.2 <a href="#4.2">Writing web tests</a><br />
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 4.3 <a href="#4.3">Running web tests</a><br />
<strong>5. <a href="#5">Conclusions</a></strong><br />
<strong>6. <a href="#6">Additional information</a></strong><br />
<a name="1"></a><h2>1. General considerations</h2>
<p>This documents presents guidelines for unit testing and regression
testing homogenisation throughout all Invenio modules.
<p>Testing is an important coding activity. Most authors believe that
writing test cases should take between 10% and 30% of the project
time. But, even with such a large fraction, don't put too much belief
on such a testing. It cannot find bugs that aren't tested for. So,
while testing is an important activity inherent to safe software
development practices, it cannot become a substitute for pro-active
bug hunting, source code inspection, and bugfree-driven development
approach from the start.
<p>Testing should happen alongside with coding. If you write a
function, immediately load it into your toplevel, evaluate its
definition, and call it for a couple of arguments to make sure the
function works as expected. If not, then change the function
definition, re-evaluate it, re-call it, etc. Dynamic languages with
interactive toplevel such as Common Lisp or Python makes this easy for
you. Dynamic redefinition capabilities (full in Common Lisp, partial
in Python) are very programmer-friendly in this respect. If your test
cases are interesting to be kept, then keep them in a test file.
(It's almost all the time a good idea to store them in the test file,
since you cannot predict whether you won't want to change something in
the future.) We'll see below how to store your tests in a test file.
<p>When testing, it is nice to know some rules of thumb, like: check
your edge cases (e.g. null array), check atypical input values
(e.g. laaarge array instead of typically 5-6 elements only), check
your termination conditions, ask whether your arguments have already
been safe-proofed or whether it is in your mandate to check them,
write a test case for each `if-else' branch of the code to explore all
the possibilites, etc. Another interesting rule of thumb is the bug
frequency distribution. Experience has shown that the bugs tend to
cluster. If you discover a bug, there are chances that other bugs are
in the neighborhood. The famous 80/20 rule of thumb applies here too:
about 80% of bugs are located in about 20% of the code. Another rule
of thumb: if you find a bug caused by some coding practice pattern
thay may be used elsewhere too, look and fix other pattern instances.
</p>
<p>In a nutshell, the best advice to write bug-free code is: <em>think
ahead</em>. Try to prepare in advance for unusual usage scenarios, to
foresee problems before they happen. Don't rely on typical input and
typical usage scenarios. Things have a tendency to become atypical
one day. Recall that testing is necessary, but not sufficient, to
write good code. Therefore, think ahead!
</p>
<a name="2"></a><h2>2. Unit testing</h2>
<a name="2.1"></a><h3>2.1 Unit testing philosophy</h3>
<p>Core functionality, such as the hit set intersection for the search
engine, or the text input manipulating functions of the BibConvert
language, should come with a couple of test cases to assure proper
behaviour of the core functionality. The test cases should cover
typical input (e.g. hit set corresponding to the query for ``ellis''),
as well as the edge cases (e.g. empty/full hit set) and other unusual
situations (e.g. non-UTF-8 accented input for BibConvert functions to
test a situation of different number of bytes per char).
</p>
<p>The test cases should be written for most important core
functionality. Not every function or class in the code is to be
throughly tested. Common sense will tell.
</p>
<p>Unit test cases are free of side-effects. Users should be able to
run them on production database without any harm to their data. This
is because the tests test ``units'' of the code, not the application
as such. If the behaviour of the function you would like to test
depends on the status of the database, or some other parameters that
cannot be passed to the function itself, the unit testing framework is
not suitable for this kind of situation and you should use the
regression testing framework instead (see below).
</p>
<p>For more information on Pythonic unit testing, see the
documentation to the unittest module at <a
href="http://docs.python.org/lib/module-unittest.html">http://docs.python.org/lib/module-unittest.html</a>.
For a tutorial, see for example <a
href="http://diveintopython.org/unit_testing/">http://diveintopython.org/unit_testing/</a>.
</p>
<a name="2.2"></a><h3>2.2 Writing unit tests</h3>
<p>Each core file that is located in the lib directory (such as the
<code>webbasketlib.py</code> in the example above) should come with a
testing file where the test cases are stored. The test file is to be
named identically as the lib file it tests, but with the suffix
<code>_unit_tests</code> (in our example,
<code>webbasketlib_unit_tests.py</code>).
</p>
<p>The test cases are written using Pythonic unittest TestCase class.
An example for testing search engine query parameter washing function:
<blockquote>
<pre>
$ cat /opt/invenio/lib/python/invenio.legacy.search_engine_unit_tests.py
[...]
import search_engine
import unittest
class TestWashQueryParameters(unittest.TestCase):
"""Test for washing of search query parameters."""
def test_wash_url_argument(self):
"""search engine washing of URL arguments"""
self.assertEqual(1, search_engine.wash_url_argument(['1'],'int'))
self.assertEqual("1", search_engine.wash_url_argument(['1'],'str'))
self.assertEqual(['1'], search_engine.wash_url_argument(['1'],'list'))
self.assertEqual(0, search_engine.wash_url_argument('ellis','int'))
self.assertEqual("ellis", search_engine.wash_url_argument('ellis','str'))
self.assertEqual(["ellis"], search_engine.wash_url_argument('ellis','list'))
self.assertEqual(0, search_engine.wash_url_argument(['ellis'],'int'))
self.assertEqual("ellis", search_engine.wash_url_argument(['ellis'],'str'))
self.assertEqual(["ellis"], search_engine.wash_url_argument(['ellis'],'list'))
[...]
</pre>
</blockquote>
</p>
<p>In addition, each test file is supposed to define a
<code>TEST_SUITE</code> variable that will return test suite with all
the tests available in this file:
<blockquote>
<pre>
$ cat /opt/invenio/lib/python/invenio.legacy.search_engine_unit_tests.py
[...]
TEST_SUITE = make_test_suite(TestWashQueryParameters,
TestStripAccents,
TestQueryParser,)
[...]
</pre>
</blockquote>
</p>
<p>In this way, all the test cases defined in the file
<code>search_engine_unit_tests.py</code> will be executed when the global
<code>testcase</code> executable is called.
<p>Note that it may be time-consuming to run all the tests in one go.
If you are interested in running tests only on a certain file (say
<code>search_engine_unit_tests.py</code>), then launch:
<blockquote>
<pre>
$ python /opt/invenio/lib/python/invenio.legacy.search_engine_unit_tests.py
</pre>
</blockquote>
</p>
<p>For full-scale examples, you may follow
<code>search_engine_unit_tests.py</code> and other <code>_unit_tests.py</code>
files in the source distribution.
</p>
<a name="2.3"></a><h3>2.3 Running unit tests</h3>
<p>Invenio test suite can be run at any time via:
<blockquote>
<pre>
$ /opt/invenio/bin/inveniocfg --run-unit-tests
</pre>
</blockquote>
This command will run all available unit tests provided with CDS
Invenio.
</p>
<p>The informative output is of the form:
<blockquote>
<pre>
Invenio v0.3.2.20040519 test suite results:
===========================================
search engine washing of query patterns ... ok
search engine washing of URL arguments ... ok
search engine stripping of accented letters ... ok
bibindex engine list union ... ok
----------------------------------------------------------------------
Ran 4 tests in 0.121s
OK
</pre>
</blockquote>
In case of problems you will see failures like:
<blockquote>
<pre>
Invenio v0.3.2.20040519 test suite results:
===========================================
search engine washing of query patterns ... FAIL
search engine washing of URL arguments ... ok
search engine stripping of accented letters ... ok
bibindex engine list union ... ok
======================================================================
FAIL: search engine washing of query patterns
----------------------------------------------------------------------
Traceback (most recent call last):
File "/opt/invenio/lib/python/invenio.legacy.search_engine_unit_tests.py", line 25, in test_wash_pattern
self.assertEqual("ell*", search_engine.wash_pattern('ell*'))
File "/usr/lib/python2.3/unittest.py", line 302, in failUnlessEqual
raise self.failureException, \
AssertionError: 'ell*' != 'ell'
----------------------------------------------------------------------
Ran 4 tests in 0.091s
FAILED (failures=1)
</pre>
</blockquote>
</p>
<p>The test suite compliance should be checked before each CVS commit.
(And, obviously, double-checked before each Invenio release.)
</p>
<a name="3"></a><h2>3. Regression testing</h2>
<a name="3.1"></a><h3>3.1 Regression testing philosophy</h3>
<p>In addition to the above-mentioned unit testing of important
functions, a regression testing should ensure that the overall
application functionality is behaving well and is not altered by code
changes. This is especially important if a bug had been previously
found. Then a regression test case should be written to assure that
it will never reappear. (It also helps to scan the neighborhood of
the bug, or the whole codebase for occurrences of the same kind of
bug, see the 80/20 thumb rule cited above.)
</p>
<p>Moreover, the regression test suite should be used when the
functionality of the item we would like to test depends on
extra-parametrical status, such as the database content. Also, the
regression framework is suitable for testing the overall behaviour of
web pages. (In extreme programming, the regression testing is called
<em>acceptance testing</em>, the name that evolved from previous
<em>functionality testing</em>.)
</p>
<p>Within the framework of the regression test suite, we have liberty
to alter database content, unlike that of the unit testing framework.
We can also simulate the web browser in order to test web
applications. (Note that for real browser tests, we have yet
another <a href="#4">web test suite</a> described below.)
</p>
<p>As an example of a regression test, we can test whether the web
pages are alive; whether searching for Ellis in the demo site produces
indeed 12 records; whether searching for aoeuidhtns produces no hits
but the box of nearest terms, and with which content; whether
accessing the Theses collection page search prompts an Apache password
prompt; whether the admin interface is really accessible only to
admins or also to guests, etc.
</p>
<p>For more information on regression testing, see for example <a
href="http://c2.com/cgi/wiki?RegressionTesting">http://c2.com/cgi/wiki?RegressionTesting</a>.
</p>
<a name="3.2"></a><h3>3.2 Writing regression tests</h3>
<p>Regression tests are written per application (or sub-module) in
files named like <code>websearch_regression_unit_tests.py</code> or
<code>websubmitadmin_regression_tests.py</code>.
</p>
<p>When writing regression tests, you can assume that the site is in
the fresh demo mode (Atlantis Institute of Fictive Science). You can
also safely write not only database-read-only tests, but you can also
safely insert/update/delete into/from the database whatever values you
need for testing. Users are warned prior to running the regression
test suite about its possibly destructive side-effects. (See below.)
Therefore you can create users, create user groups, attach users to
groups to test the group joining process etc, as needed.
</p>
<p>For testing web pages using GET arguments, you can take advantage
of the following helper function:
<blockquote>
<pre>
$ cat /opt/invenio/lib/python/invenio.testsuite.py
[...]
def test_web_page_content(url, username="guest", expected_text="</html>"):
"""Test whether web page URL as seen by user USERNAME contains
text EXPECTED_TEXT. Before doing the tests, login as USERNAME.
(E.g. interesting values are "guest" or "admin".)
Return empty list in case of problems, otherwise list of error
messages that may have been encountered during processing of
page.
"""
</pre>
</blockquote>
For example you can test whether admins can access WebSearch Admin
interface but guests cannot:
<blockquote>
<pre>
test_web_page_content(CFG_SITE_URL + '/admin/websearch/websearchadmin.py',
username='admin')
test_web_page_content(CFG_SITE_URL + '/admin/websearch/websearchadmin.py',
username='guest',
expected_text='Authorization failure')
</pre>
</blockquote>
or you can test whether searching for aoeuidhtns produces nearest
terms box:
<blockquote>
<pre>
test_web_page_content(CFG_SITE_URL + '/search?p=aoeuidhtns',
expected_text='Nearest terms in any collection are')
</pre>
</blockquote>
</p>
<p>For testing web pages using POST argumens or for other more
advanced testing you should use directly <code>mechanize</code> Python
module that simulates the browser. It can post forms, follow links,
go back to previous pages, etc. An example of how to test the login
page functionality:
<blockquote>
<pre>
browser = mechanize.Browser()
browser.open(CFG_SITE_SECURE_URL + "/youraccount/login")
browser.select_form(nr=0)
browser['p_un'] = 'userfoo'
browser['p_pw'] = 'passbar'
browser.submit()
username_account_page_body = browser.response().read()
try:
string.index(username_account_page_body,
"You are logged in as userfoo.")
except ValueError:
self.fail('ERROR: Cannot login as userfoo.')
</pre>
</blockquote>
<p>For full-scale examples, you may follow
<code>websearch_regression_tests.py</code> and other
<code>_regression_tests.py</code> files in the source distribution.
</p>
<p>It is to be noted that regression tests using the simple mechanize
browser emulation (i) cannot test advanced web technologies such as
JavaScript and AJAX. Hence, for example, we cannot use them to test
submission pages that are currently heavily replying on JavaScript.
Moreover, with the mechanise browser emulation, it may be (ii)
cumbersome to write tests for operations that require going through
several pages etc. For these purposes we use the <a href="#4">web
test suite</a> that can run in a real browser. We shall describe this
test suite below. Nevertheless, for quick testing of HTTP GET pages
or other single-page web operations, the regression tests are very
easy to write and more practical then using the below-described web
test suite.</p>
<a name="3.3"></a><h3>3.3 Running regression test suite</h3>
<p>The regression test suite can be run by invoking:
<blockquote>
<pre>
$ /opt/invenio/bin/inveniocfg --run-regression-tests
</pre>
</blockquote>
similarly to the unit test suite cited above. The notable exception
when compared to running the unit test suite is:
</p>
<ul>
<li>we assume the site to be in demo mode (Atlantis Institute of
Fictive Science)
<li>we shall pollute the database with test data as it seems fit for
the regression testing purposes.
</ul>
<p>
Therefore beware, <strong>running regression test suite requires clean
demo site and may destroy your data forever</strong>. The user is
warned about this prior to running the suite and is given a chance to
abort the process:
<blockquote>
<pre>
$ /opt/invenio/bin/inveniocfg --run-regression-tests
**********************************************************************
** **
** *** I M P O R T A N T W A R N I N G *** **
** **
** The regression test suite needs to be run on a clean demo site **
** that you can obtain by doing: **
** **
** $ inveniocfg --drop-demo-site \ **
** --create-demo-site \ **
** --load-demo-records **
** **
** Note that DOING THE ABOVE WILL ERASE YOUR ENTIRE DATABASE. **
** **
** (In addition, due to the write nature of some of the tests, **
** the demo database may be altered with junk data, so that **
** it is recommended to rebuild the demo site anew afterwards.) **
** **
**********************************************************************
Please confirm by typing "Yes, I know!": NO
Aborted.
</pre>
</blockquote>
</p>
<p>If you choose to continue, the regression test suite will produce
the output similar to the unit test suite that was discussed
previously.
</p>
<a name="4"></a><h2>4. Web testing (via real browser)</h2>
<a name="4.1"></a><h3>4.1 Web testing philosophy</h3>
<p>The web test suite bears many similarities with the black-box
functional acceptance testing that the regression test suite also
provides. We have seen that the regression test suite (using
mechanize) can very well take care of testing of simple single-page
web scenarios. Nevertheless, it is useful to distinguish between
testing of web pages via an emulated browser (mechanize) and testing
of web pages via a real browser. The latter tests run slower, but are
closer to the end user environment, can use real JavaScript engine,
and can generally more easily test various complex multi-page
scenarios. It is the latter type of acceptance tests that we are
concerned with here.</p>
<p>Invenio currently uses the
<a href="http://www.mozilla.com/en-US/firefox/">Firefox</a> browser
with the <a href="http://selenium-ide.openqa.org/">Selenium IDE</a>
extension to automatize its real-browser web tests.</p>
<a name="4.2"></a><h3>4.2 Writing web tests</h3>
<p>Selenium IDE uses "Selenise HTML" to represent test cases via HTML
tables with rows representing actions and expected results. You can
take a look at <code>test_search_ellis.html</code> example and emulate
it.</p>
<p>However, the most practical way to write web test cases might be to
use Selenium IDE test recording capabilities: you surf in Firefox as a
normal user would, and you add simple tests for the presence of
certain text on the pages as you go. Selenium IDE can record and
store tests for further replaying. (You can watch an explanatory
screencast on <a href="http://seleniumhq.org/movies/intro.mov">how
to record a test</a>.)
<p>(Note that one inconvenience with recording test cases is that this
is usually done <em>a posteriori</em> when the web pages are written
already, which does not go well with with test-driven development
practices. But some of TDD goals can be met by using regression test
suite with mechanize instead.)</p>
<p>When the test is recorded and the Selenese HTML test case is
written, edit its source and pay attention to using default URLs
as <code>http://localhost</code>, and install it into <code>web</code>
directory following e.g. <code>websearch/web</code> example.
<a name="4.3"></a><h3>4.3 Running web tests</h3>
<p>Run web tests via:
<blockquote>
<pre>
$ /opt/invenio/bin/inveniocfg --run-web-tests
</pre>
</blockquote>
<p>This will build complete web test suite out of
module-specific <code>test_foo_bar.html</code> files, and will open a
Firefox window with Selenium IDE running the built complete test
suite. Check the results there.
<p>Note that if you want to web-test a remote Invenio demo site (say
<code>foo.site.com</code>) using your local Firefox with Selenium IDE,
you can install a local Invenio site as usual, and then do:</p>
<blockquote>
<pre>
$ sudo -u apache perl -pi -e 's,localhost,foo.site.com,g' /opt/invenio/lib/webtest/invenio/test*html
$ /opt/invenio/bin/inveniocfg --run-web-tests --yes-i-know
</pre>
</blockquote>
<a name="5"></a><h2>5. Conclusions</h2>
<p>A uniform testing technique and three test suites (unit test suite,
regression test suite, web test suite) were discussed. Each
programmer should plan to write the test code alongside the core code
development to test the building blocks of his/her code (unit tests)
as well as the overall application behaviour (regression tests, web
tests). The guidelines were given how to do so.
</p>
<a name="6"></a><h2>6. Additional information</h2>
<dl>
<dt>More information can be found on the URLs mentioned above:
<dd>
<pre>
<a href="http://c2.com/cgi/wiki?UnitTest">http://c2.com/cgi/wiki?UnitTest</a>
<a href="http://c2.com/cgi/wiki?RegressionTesting">http://c2.com/cgi/wiki?RegressionTesting</a>
<a href="http://docs.python.org/lib/module-unittest.html">http://docs.python.org/lib/module-unittest.html</a>
<a href="http://diveintopython.org/unit_testing/">http://diveintopython.org/unit_testing/</a>
<a href="http://wwwsearch.sourceforge.net/mechanize/">http://wwwsearch.sourceforge.net/mechanize/</a>
<a href="http://selenium.openqa.org/">http://selenium.openqa.org/</a>
</pre>
</dl>
<dl>
<dt>and elsewhere:
<dd>
<pre>
Steve McConnell: "Code Complete"
FIXME: list of other interesting references, like Kent Beck papers, etc
</pre>
</dl>
c4science · Help