Gauging difficulty of a project: Converting an application to a library

[Banana]

In another thread, I mentioned my incredulity over ridiculous pricing of a SSH library.

Seeing that my search has been unfruitful, I came up with a idea of making one myself. There's plenty of source codes out there for me to take and modify (already checked the licensing). This also happen to be perfect answer to my other thing- I want to learn a bit more about programming, but needed a project, so this may be it.

But I want to understand what I am getting myself into. Googling on converting an application to a library was kind of skeptical- some says it's a simple recompile; while other says major restructuring of code is require to make it a library.

Maybe somoene a bit more experienced can give me a good idea of whether this project is do-able and perhaps give some advice or two?

TIA. :)

[AsLan^]

libssh?

[Banana]

I think I originally skipped this package because firstly, there wasn't any mention of supporting tunneling, which would be the primary use and secondly, didn't mention Windows as supported as I need this as well. (Though, recompiling the source code on a different platform would be far more easier than modifying a source code for an application to a library).

[AsLan^]

What language are you thinking of using?

For example, Perl has the Net::SSH::Perl module to create a SSH connection.

[Banana]

I was sorta of assuming C or C++, as it's the language the original SSH packages were written, but it's cool to know there's Ruby or Perl implementation.

Ideally, I want to compile them into a standalone .DLL without any dependencies (ala PuTTY) or installing Cygwin (a great shell, but definitely not for an average user!) on the end machine.