Anjuta versus Eclipse: Using an existing code base

I’ve recently been using Anjuta 2. It’s far better than Anjuta 1, and less annoying than Eclipse’s arbitrarily eccentric UI, though Eclipse is obviously more feature-packed. Here is an example of how Anjuta makes a very commonly-used feature simple and explicit, while Eclipse hides the feature and leaves you in doubt. Nobody will make money selling Anjuta training courses.

Using an existing code base with Anjuta:

  • Select the File/”New Project From Existing Sources” menu item.
  • In the wizard, click Next through the title page.
  • In the wizard, enter the project title and browse to the path to your existing code and click Next.
  • In the wizard, click Apply on the confirmation page.

Using an existing code base with Eclipse:

  • Select the File/New/Project… menu item.
  • Choose C++/”Standard Make C++ Project”
  • In the New Project wizard, enter a project name. Uncheck “Use default location” and browse for the path to your existing code.
  • In the New Project wizard, ignore the (very complex) “Referenced C/C++ Projects” page and click Finish.
  • Wonder if it adds or changes files in your existing directory. Actually, the “Standard Make C++ Project” doesn’t, but others do. Presumably it always does if you did not uncheck “Use default location”.
  • (There’s also a way to create a new Eclipse project by checking out from svn or your favorite source control system, if you have an appropriate Eclipse plugin. I find that annoying as a way to create a project for an existing code base.

3 thoughts on “Anjuta versus Eclipse: Using an existing code base

  1. Hi Murray,

    A point of clarification: you don’t mean “Eclipse”, you mean “the Eclipse CDT”.

    While I agree the CDT’s project creation wizards could be improved, I disagree about checking out from a VCS. I think that’s the best way to do it. You get the full advantages of your VCS right from within the IDE. Existing projects can be “linked” projects, too, but I’m not super-familiar with that.

    Anyway, I’ll pass along your comments to some CDT peeps. Your continued attempts to use it are appreciated :)

    Take care,


  2. We do appreciate the feedback. I notice that you mention Standard Make projects with CDT. In the most recent release of CDT, 4.0, this concept is gone. Now you create a C or a C++ project and this brings up the new project dialog. Select Makefile as the project type and select the toolchain you are using, i.e. Linux GCC. The rest is the same, the project name and changing the location.

    The Eclipse UI is probably the hardest thing to get used to. It isn’t arbitrary though as Eclipse is used for many more things than C/C++ development. As the CDT matures we’ll continue to focus on cleanup up the eccentricities. :)

  3. Doug, yes, I am using Eclipse 3.2 on Ubuntu Hardy. You mention some changes, but you don’t mention that it will become explicitly obvious when I am using existing files from the location and when I am creating files in the location. There’s no reason for the behaviour to be hidden. I hope it can be improved in future versions of the CDT.

    I also think it’s obvious that project plugins need to share concepts/UI such as this – at the moment they all seem to completely implement their own UI and logic for their wizards.

Comments are closed.