Wednesday, May 26, 2010

JUnit reporter for Unitest++

The C++ testing framework Unittest++ includes a XML reporter but it is not directly usable with most continuous integration (CI) servers such as Hudson and TeamCity. I wrote a XML reporter for Unittest++ that uses the same format as JUnit, the reporter will split test suites into different files and they can be directly used by for example the junitreport task in ant or in your CI build. Download the header and cpp files below and add them to your Unittest++ build.

This is how you normally use Unittest++'s built-in XML reporter:
std::ofstream f("test-report.xml");
UnitTest::XmlTestReporter reporter(f);
UnitTest::TestRunner runner(reporter);
return runner.RunTestsIf(UnitTest::Test::GetTestList(), NULL, UnitTest::True(), 0);
Simply replace it with the following,
std::string f = "TESTS-";
UnitTest::JUnitXmlTestReporter reporter(f);
UnitTest::TestRunner runner(reporter);
return runner.RunTestsIf(UnitTest::Test::GetTestList(), NULL, UnitTest::True(), 0);
Note that the JUnit reporter does not take a complete filename because the JUnit format requires that different test suites are split into different files. The output files are named TESTS-suitename.xml.

Download
JUnitXMLTestReporter.h
JUnitXMLTestReporter.cpp

Wednesday, February 17, 2010

Testing blogging from my iPhone, this seems to work ok. Yay.
Keep an eye on this blog if you are interested in continous integration methods for improving your software team. I am preparing a post on the subject....

Tuesday, February 16, 2010

Improving upon continuous integration

I am quite a big fan of continuous integration and other automatable "agile practices" for improving performance in software teams. Roughly half a year ago the team I am the scrum master of started improving upon our practices, one of the first tasks was to create a script that made it able to create a release with a single command. With this done we added a continuous integration server (CruiseControl) and started down the path of automated tests & tasks. A couple of months later this has had measurable impact on the time it takes to make releases and as we've added compile & test tasks that run for each commit to SVN, this makes us more confident about the quality of the software we make.

I also hold the believe that nothing is ever perfect and it's always possible to improve and tune your software process. The problem I've been seeing lately is that, yes the continuous server helps us easily find compilation errors, bugs and broken tests. But once the code is in our SVN the damage is already done, we are working on a product with multiple teams in multiple time zones. It's very easy for someone to make a commit that breaks the build and then leave for the day, national holidays, other work assignments etc. I'll shortly outline our current process:
  1. Write code & test
  2. Check in your changes to the version control system (VCS), such as SVN, GIT etc.
  3. Continuous integration server sees a change in the VCS and starts a build, tests and so on.
  4. If your code fails to compile or have broken some tests you will be notified and hopefully you are able to fix it before other developers update from the VCS.
  5. Other developers update from the VCS. They might update before you have had time to fix problems in the code you checked in.
  6. The bad code propagates to the whole team or the team might be sitting idle waiting for a fix for the bad code that is already in the VCS.
If you look at this process the crucial part is step 3, this where we get our benefits by finding bugs early but why not move this step to take place before the commit is made? Many teams I've heard about have policies that all developers must run the build and the tests prior to a commit but this breaks down every once in a while. Developers are stressed, they forget, they are lazy (yes!) and make mistakes. So, I've been on the look out for a more robust solution to this problem. I was long aware of SVN various commit hooks but didn't really think of it as being a viable option since it would involve a lot of scripting, configuration and time that we do not just have. Luckily one of the team members pointed out TeamCity to me. TeamCity's pre-tested commit features solves exactly this kind of problem. See there homepage for some nice diagrams.

TeamCity allows us to have the following process:
  1. Write code & test
  2. Send your code to TeamCity, which can run the same checks and builds as the continuous integration server but before the code is committed to the VCS.
  3. TeamCity automatically commits to the VCS if your code compile, pass all tests etc.
  4. Other developers update as normally from the VCS but they only get good code.
Yes! And it actually works. Well it took two days to set it up and get everything working. The trick was that the build agent must run as the same user as I am normally using since our build process is using cygwin and various cross compilation tools that for some reason didn't work otherwise. I've only been running this for a limited time but I am impressed with TeamCity and it will probably help us in the long run to make better software and more effectively.

A couple of minor caveats:
  • TeamCity has no command line tool, so there is no way to script commits. Everything is done with IDE plugins.
  • Java centric, be ready for a lot of scripting & configuration if your build is complex using lots of different build systems (makefile, vcproj/sln, cygwin etc).

Tuesday, January 5, 2010

Increasing java heap size when using maven

If you get out of memory exceptions while using maven you can tweak the amount of heap space available to the JVM with the following environment variable,

export MAVEN_OPTS=-Xmx512m

I didn't test any other but i presume you can set other Java options also with the same variable.

Wednesday, December 30, 2009

Adding custom Javascript bindings to WebKIT

This post will show you how to add custom objects / functions which are implemented in C but used in a Javascript that's part of a HTML document that WebKIT renders. We will add a class "myclass" to the Javascript engine instance and it will have one static function named "mymethod()" that returns a single string.

First you need to install all the needed dependencies, assuming you are using Ubuntu or other Debian based system run the following command as root:
  • apt-get install libwebkit-1.0.1 libwebkit-dev
The following very simple HTML page will be used to test that the Javascript binding to the native C function works properly.

simple.html
<html>
<body>
<h1>String in html</h1>

<script type="text/javascript">
document.write("<h1>String from JS:");
document.write(myclass.mymethod());
document.write("</h1>");
</script>

</body>
</html>

C implementation of myclass.mymethod()

mymethod() will be implemented as a static function on the myclass object. The JavaScript framework used by WebKIT can be used with an API that's documented on Apple's pages. The important thing for this simple class is the JSClassDefinition and JSStaticFunction struct both have to be filled out with information about our class, including its static function(s) (mymethod), initialize/constructor callback, finalize/destructor callback and so on. To add the class to the Javascript engine we need something called the JSGlobalContextRef, how you get it depends on the flavor of WebKIT you are using. In this tutorial I will show how to do it with the GTK port of WebKIT. Connect a handler to the window-object-cleared signal which is sent when a new page is loaded (found this out after a lot of googling, reference). In the callback you connect you can call webkit_web_frame_get_global_context(), it will return the JSGlobalContextRef you need.

Enough talking, here is the actual source code for the example,


#include <gtk/gtk.h>
#include <webkit/webkit.h>
#include <JavaScriptCore/JavaScript.h>

static void myclass_init_cb(JSContextRef ctx, JSObjectRef object)
{
// ...
}

static void myclass_finalize_cb(JSObjectRef object)
{
// ...
}

static JSValueRef myclass_mymethod(JSContextRef context,
JSObjectRef function,
JSObjectRef thisObject,
size_t argumentCount,
const JSValueRef arguments[],
JSValueRef *exception)
{
JSStringRef string = JSStringCreateWithUTF8CString("mystring");
return JSValueMakeString(context, string);
}

static const JSStaticFunction class_staticfuncs[] =
{
{ "mymethod", myclass_mymethod, kJSPropertyAttributeReadOnly },
{ NULL, NULL, 0 }
};

static const JSClassDefinition class_def =
{
0,
kJSClassAttributeNone,
"TestClass",
NULL,

NULL,
class_staticfuncs,

myclass_init_cb,
myclass_finalize_cb,

NULL,
NULL,
NULL,
NULL,
NULL,
NULL,
NULL,
NULL,
NULL
};

static void addJSClasses(JSGlobalContextRef context)
{
JSClassRef classDef = JSClassCreate(&class_def);
JSObjectRef classObj = JSObjectMake(context, classDef, context);
JSObjectRef globalObj = JSContextGetGlobalObject(context);
JSStringRef str = JSStringCreateWithUTF8CString("myclass");
JSObjectSetProperty(context, globalObj, str, classObj,
kJSPropertyAttributeNone, NULL);
}

static void window_object_cleared_cb(WebKitWebView *web_view,
WebKitWebFrame *frame,
gpointer context,
gpointer arg3,
gpointer user_data)

{
JSGlobalContextRef jsContext = webkit_web_frame_get_global_context(frame);
addJSClasses(jsContext);
}

static GtkWidget* main_window;
static WebKitWebView* web_view;

static void destroy_cb(GtkWidget* widget, gpointer data)
{
gtk_main_quit ();
}

static GtkWidget* create_browser()
{
GtkWidget* scrolled_window = gtk_scrolled_window_new (NULL, NULL);
gtk_scrolled_window_set_policy (GTK_SCROLLED_WINDOW (scrolled_window), GTK_POLICY_AUTOMATIC, GTK_POLICY_AUTOMATIC);

web_view = WEBKIT_WEB_VIEW (webkit_web_view_new ());
gtk_container_add (GTK_CONTAINER (scrolled_window), GTK_WIDGET (web_view));

g_signal_connect (G_OBJECT (web_view), "window-object-cleared", G_CALLBACK(window_object_cleared_cb), web_view);

return scrolled_window;
}

static GtkWidget* create_window()
{
GtkWidget* window = gtk_window_new (GTK_WINDOW_TOPLEVEL);
gtk_window_set_default_size (GTK_WINDOW (window), 500, 500);
g_signal_connect (G_OBJECT (window), "destroy", G_CALLBACK (destroy_cb), NULL);
return window;
}

int main (int argc, char* argv[])
{
gtk_init (&argc, &argv);
if (!g_thread_supported())
g_thread_init (NULL);

GtkWidget* vbox = gtk_vbox_new(FALSE, 0);
gtk_box_pack_start(GTK_BOX(vbox), create_browser (), TRUE, TRUE, 0);

main_window = create_window();
gtk_container_add(GTK_CONTAINER (main_window), vbox);

gchar* uri = (gchar*) "file://simple.html";
webkit_web_view_open(web_view, uri);

gtk_widget_grab_focus (GTK_WIDGET (web_view));
gtk_widget_show_all (main_window);
gtk_main ();

return 0;
}

Compile & Test

Finally, to compile and test the example code run the following commands:
  • gcc test.c -o test `pkg-config --cflags --libs webkit-1.0`
  • ./test

Wednesday, December 16, 2009

Clutter tutorials

For some time I've been planning to look into Clutter more seriously, it looks like a really nice framework to build a UI engine on top of. Today I found the following two articles that gives a decent introduction to how the API works,

This post goes through the basics of Clutter's C API. It covers how to create an initial empty window, how to render simple rectangles (with textures) and finally how to animate and scale them.
http://tuxradar.com/content/clutter-beginners-tutorial

This one mentions on how it's possible to use gjs to create JavaScript bindings which gives you an easier to use development environment (no memory management and other C "features").
http://townx.org/blog/elliot/introduction-sorts-javascript-desktop-application-development-gjs-and-clutter

Monday, December 14, 2009

Creating a web service client using gsoap.

In this post I'll continue my posts related to creating web services (Part 1, Part 2), as have been shown it's easy to setup and create the server part of the service using Java, Axis2 and Tomcat. However you do not always have access to a very hungry and memory intensive Java stack so this post will cover how you can create the client in native C code. For this I'll be using gSOAP. If you have the web service up and running as describe previously you should be able to access it's WSDL document at http://localhost:8080/axis2/services/ExampleService?wsdl. The WSDL document describes the service in detail and can be used to generate the needed code to access the service. First you start off by installing the package:
  • apt-get install gsoap
Now you can use the wsdl2h tool to generate C bindings (it can also be used to generate C++ bindings, just drop the -c argument)
  • wsdl2h -c -o mywebservice.h http://localhost:8080/axis2/services/ExampleService?wsdl
  • soapcpp2 -C -c mywebservice.h -I/usr/include/gsoap
The following very simple code segment shows how to call the remote service, both the getSomeValue() and setSomeValue() function. It can be compiled by doing:
  • gcc -I/usr/include/gsoap myclient.c -o myclient.o soapC.c soapClient.c /usr/include/gsoap/stdsoap2.c
myclient.c
#include "soapH.h"
#include "stdio.h"
#include "MyWebService.nsmap"

int main(int argc, char *argv[])
{
struct soap *soap = soap_new();
struct _ns1__getSomeValueResponse response;
struct _ns1__setSomeValue value;

*value.args0 = 1.0f;

if (soap_send___ns2__setSomeValue(soap, NULL, NULL, &value) == SOAP_OK)
printf("okay");
else
soap_print_fault(soap, stderr); // display the SOAP fault on the stderr stream

if (soap_call___ns2__getSomeValue(soap, NULL, NULL, &response) == SOAP_OK)
printf("value: %f\n", *response.return_);
else // an error occurred
soap_print_fault(soap, stderr); // display the SOAP fault on the stderr stream
}