Best way to dust off an old project

[zorocke]

Today I was looking into dusting off an old project that I had never completed. Of course, all the dependencies and such have long since been moved or deleted on my system... So when i went to compile I was confronted with nothing but a list of errors.

After about 2 hours of trying to get the dependencies back and in their proper place such as Ogre and Newton, I am still confronted with a list of errors. Some of which to do newer versions of the libraries.

I'm wondering, should I spend more time trying to get my old project to build or might it be easier to create a new visual studios solution and start the project from scratch using my old code as a reference to speed the project along?

Currently i see more benefits in the second option such as getting to familiarize myself with my code, and an opportunity to possibly upgrade to better or more active dependencies.

[Klaim]

I'm wondering, should I spend more time trying to get my old project to build or might it be easier to create a new visual studios solution and start the project from scratch using my old code as a reference to speed the project along?

I've done this a long time ago. It was really easier to use the old code in a new VS project and re-setup the dependencies.

That said, it's totally dependant on the project. If the problem is just with the VS project/solution file, then just recreate it.

[stoneCold]

You can save yourself a lot of trouble by using environment variables, such as $(OGRE_HOME), etc. in your VS project settings (include/library directories).

[Klaim]

Personally I have projects and dependencies directly in my svn, making the project easy to get for other devs (just checkout and compile -- but still requires boost). That way i avoid any dependency problem.

[KungFooMasta]

I vote for re-writing, re-using old code/ideas where feasible. I have an old project as well, and that would be the approach I would take. My current framework is much better than the previous one, and I've learned a lot since then. (this is my first project btw )

[Fulmens]

Moreover, "rewriting" it will help you remember how it works. And be ashamed of the code you wrote back then
"Oh My Gawd, why would I do that in such a lame way ?!"

[zorocke]

I vote for re-writing, re-using old code/ideas where feasible. I have an old project as well, and that would be the approach I would take. My current framework is much better than the previous one, and I've learned a lot since then. (this is my first project btw )

I've started with this strategy. I'm documenting my progress on my blog.