C and C++ together?

Hi again,

I don't understand the rationale behind clubbing C and C++ together throughout the website? They are different languages and should be treated as such. The statistics are not so helpful when they say "C/C++ is(are) the most popular language(s) to be used by open source projects". For instance, it doesn't answer the question: "Is Java more popular than C++?"

How difficult would it be to parse the files with the aim of differentiating between C and C++? If you're going by file extensions, that would be easy; even otherwise I don't think it would be too tough.

I'm probably just nitpicking, but it would be cool to see updated stats treating C and C++ seperately :)

Keep up the great work! Anant

I'd just like to echo this sentiment. C and C++ have very different usage patterns and distinct user communities. C++ projects are not, in general, of interest to C users. And while C++ users may be interested in C projects, certainly C++ users can be expected to have a particular interest in C++ projects.

Header files are indeed tough. Keyword detection is problematic in that C++ keywords that are not C keywords can be used as identifiers in C. Matching keywords with proximate syntax might be workable. Emacs mode settings are a handy thing to look for; but certainly not something that can be relied upon generally.

I understand that getting this right is not exactly trivial. Thanks, though, for acknowledging the issue.

Header files could be handled by allocating the lines of code proportionally to the lines of C vs C++ in the same directory.

I guess in most cases it would be 100% C or 100% C++ anyway.