That is far from the only problem with libtcod. While I applaud Jice for making it available years ago, and I assume it is/was undoubtedly a good solution to the needs of Doryen, as a general purpose lib today it has a huge number of shortcomings. The least of which is that the last stable release is dated August, 29 2012. A much bigger problem that has existed since its initial release is that it attempts to do way too much and, as a result, has a ton of cruft that is only usable in specific, narrow, use case scenarios. Add in the toolchain issues we have fought with for the past twenty years in the pursuit of portable C++ code and heh... much of this thread is illustrative.
I do believe we need new tutorials using narrowly scoped, modern, solutions, but... I'm not volunteering. There are way too many bad practices being encouraged these days and, frankly, the definition of roguelike has long past veered into the land of the lost and meaningless. As a point, personally I consider the entire 7DRL concept to be encouraging of bad practices in programming and design, so anything I would write would have zero appeal to those who disagree.
That's a nice thing bubbling in your sig.
Infra Arcana is a great game.
Whilst reserving your methods in order to make them current at some point, there is a lively thing going on regarding those few notches down the timescale. There is a fact, to be honest, that says how devs (and programmers in general) find their pioneering efforts most admirable at some point. Kinda loosenes 'em up, gives em' a big grin - if you know what I mean.
Now, I know how getting tough on all those malpractices certainly brings things up into perspective or at least makes you less self-aware, being that humans
make mistakes.. and altough, for my part, I never liked things I couldn't finish - it all seemed quite simple upon reminescing. Even though a mere make-believe fail-attempt makes you frown once or twice, it actually brings you closer to the sense of achievement where perfection is more hard-earned, and less stumbled upon, vaguely embraced...
Be it somehow you agreed or disagreed, even flaws in foundations have their positive perks. And the simple things such as coding style, pseudo-programming guidelines or naming conventions - they all bring one's wisdom to our grasp, and even better if the practice is condoned by many. IMO, struggling with styles and specific workarounds when using a library of choice for developing a RL game for a challenge such as 7DRL is a sign of inexperience, nothing more. Remembering how making a RL is all about the attempt to create your own game brings up the simple truth which repeats itself through our programming endeavors, which is -
errors are bound to happen.
Seeing that the year 2015 has started with some really good efforts from the community which I've been a fan of since, yeah, as long as I knew what OOP was made me enthusiastic on many levels. Upon seeing how players who've never asked what a RL game is venture into a procgen dungeon I found most assuring the thought of RL community getting better by the day. The errors, seen as mere obstacles that had to be crossed made me believe once again that coding RL games isn't a skillset that's either not there or it's witheld, but more of an art-form - a choice of deliberance to make a computer game as you imagined it.
