Tag: C++

Orthodox canonical class form

This post is more about the basics of C++ and the last things C++11 brings to the orthodox canonical class form (OCCF). Some times the basics of a programming language are assumed by everybody but there is always a the time when someone asks why. Implement the complete OCCF is something that not always is required. In fact when you work with Qt framework and specially with widgets or inheritance from QObject is something you usually don’t care. And that’s an error.

The orthodox canonical class form before C++11

In C++98 and C++03 the OCCF had four different methods that the C++ compiler is willing to generate:

  • Default constructor
  • Copy constructor
  • Destructor
  • Copy assignment operator

A standard class should look like the following code:

As you may see, this is simple and it should be nothing to worry about. In general, when we work with objects (not pointers to an object) we can assume that there is no need to overwrite all methods. When we copy an object, all its data is copied to the other using the copy constructors of their data members. That’s fine and most of the time it’s also enough. The problem arises when we want to work with pointers to objects, members that are pointers to objects, o more complex data structures (smart pointers, etc).

A problematic example

Imagine the following situation:

In the class B we can assume that we don’t need to deallocate memory for the integer so we discard the destructor. The copy is really cheap so we also discard copy constructor or copy assignment operator. It means that the compiler will generate those methods for us copying data members.

Now, let’s evaluate the class A. We have a member that is a pointer to a class B. In the constructor we allocate the memory and then we deallocate it in the destructor, that’s fine. If we are not copying or assigning objects that’s perfect but we all now that we like to copy, assign and delete objects all the time. In fact, we like to create pointers to objects and deallocate them whenever it’s possible! Well, I’m joking. Maybe we don’t like to do it all the time but we do need to so many times…

The problem implementation

Taking the code written above. We start writing and the code flows and we do:

We assume that all the methods we call are defined and don’t do any weird stuff like deallocating memory or so. Even in that case, we have the problem in lines 7 and 13. I written both different approaches because sometimes we forget that going out of scope implies a call to the destructor of the class (line 7). Other times we explicitly allocate and deallocate memory (so we do with the destructor) like in the line 13.

In these cases, the copy constructor and the copy assignment operator generated by the compiler makes the member of the class A to share the address between all the objects. The reason is because C++ copies or assigns the values and in a pointer is the address it points to. So, when we are deleting the object in lines 7 or 13 we are in fact deallocating the memory of the member field of object a!

The solution: implement the orthodox canonical class form

In order to avoid such a nasty behavior we must define the copy methods in the class A. It means that we have to implement the full orthodox canonical class form: 4 methods before C++11.

Doing as the code above we avoid pointers that points to the same address in multiple objects. I think it is necessary to point that in the copy assignment operator we need to check that the address we’re giving by parameter is not our own address. If we don’t do it, we’d be leaking memory or destroying our own object.

The orthodox canonical class form in C++11 and later

In C++11 the orthodox canonical class form has changed by including two new member functions:

  • Move constructor
  • Move assignment operator

Since it implies a lot of new things I prefer to explain it in its own post.

Bicing Barcelona with Qt

When I was living in Barcelona I saw that there is a OpenData service to get the information of the Bicing (bike public service). They provide the data in several formats such as CSV, XML and JSON. I wanted to write an app for Windows and Linux that retrieves the information from the stations and shows their current status. And for that, I chose JSON since I think the format is easier to read than XML.

After thinking a while I realised that it could be useful not just have every Bicing station data on demand but also a historic. To achieve that I split the app in two separated parts: “backend” that works as a deamon or service, and “frontend” that is the actual GUI app.


Customization of a QCompleter

Time ago I had the task to create a QComboBox and a QLineEdit that open a completer so the user can select between all the suggested options. So far it seems a quite easy task by using the QCompleter that Qt provides.

The interesting part starts when I had to add customized styles for both the QComboBox and the QLineEdit, and make them match with the styles that I also had to add to the QCompleter. In addition I had a language constraint that initially I didn’t take into account but it impacted the performance of the widget. That’s the default filter implemented in QCompleter.


Integrating Qt Test in Travis CI

For a long time I wanted to create a full Continuous Integration environment that uses Qt for both building apps and testing. In fact, one of the most important things for me was to find the right way to integrate Qt Test. For CI I’ve started to use Travis since it is one of the most widely used environments in GitHub. I took that decision after seeing other Qt projects (Musescore and Stellarium) using it.

The main difference between those projects and the one I wanted to set up, is that they’re using CTest as a testing tool. With that tool they can report when the tests are broken easily. But, how the hell I was going to do it with Qt?


Simple bugs that crashed my mind

This post is more about a memorial of my errors or those I found during the years than a tutorial or research entry. Even though, I’ll try to give an explanation about why it happened and what we try to take into account for the future. They’re usually simple errors but hard to find when you don’t know what you’re looking for. Some times it crashes but other times it just causes unexpected behavior.

I’ve some for C++ and other about Qt. I think it’s quite interesting to see both of them because their types and reasons are quite different. While in C++ we have full control about memory allocation and deallocation in Qt we have the re-parenting issue.


My C++11 favorite features

Where I came from

I started developing C++ at university where I had my first contact with programming languages. I always had a computer at home since MS-DOS but I never knew or asked myself about how applications run until I started the university. When I started I found it difficult to think as a computer so my efforts where more oriented to memorize routines. Totally wrong, as you may see.

I’ve always worked with C++ during my university studies but it doesn’t mean that I knew how it works or how to use it properly. In fact, the first year I didn’t learn OOP at all, it was just structural programming to learn about loops, arrays and basic data structures. The main problem: not a single explanation of why we were using it in that way. Time after that I started with OOP, some templates and finally the conjunction of OOP with GUI applications. A total┬ádisaster as you will find out.