Alright, it's time for a full report...
I jokinly said that Mass Effect 3 would ruin the 7DRL Challenge for me after I realized the game was going to be released just before the 7DRL week. Evidently, it didn't, but my starting time got pushed to Sunday and I still hadn't finished the game when I began working on Drakefire Chasm.
I got off to a good start, though, and I had dungeon generation already in place that day. But on Monday I returned to ME3 and finally finished it. However, I then spent Tuesday being completely baffled by the ending, and thus lost another day. After that, it was 7DRL all the way.
On Saturday, my success looked uncertain, and to be sure, I pulled an all-nighter. And it was a good thing, because I just made it. The last hour I spent searching for a bug that was preventing highscores being saved. Finally I located it in if(!TCODSystem::isDirectory("data")) TCODSystem::createDirectory("data");. From some reason it was causing a crash. I had the saving happening upon game exit and didn't notice the crash before moving the score saving elsewhere. In Rings of Valor, a similar line never caused a problem; maybe the new version of libtcod changed something, I don't know.
Another thing I re-used was most of Untitled Roguelike Project's target.cpp (~2k lines). But there was no way I could have done all the targeting and animation logic from scratch (especially since it was Saturday already). It was still a lot of work, to go through all that code to make stuff compatible with DfC. It all might have been easier if I had just started coding the game upon URLP.
I'm happy I decided to ditch much of the plans I had last year. It was way out of the scope of a 7DRL. I think I got in almost everything I wanted this time. Fire Elemental Swarm ability I had to change into Blast Wave, though -- it would have required me to include logic for allied entities, and I didn't really have the time to do that.
I also wish I had had more time to try out the game. During the short balance tuning period, the game went from easy as goblin pie to nigh impossible and vice versa. Also, for some reason, I had designed each of the abilities for a maxed stat at the end of the game. Because of this, many of them aren't very good the moment you can get them. (Some never get good, to be honest.)
What I ultimately wanted to achieve with Drakefire Chasm, was the feeling of being a raid boss; multiple adventurers trying to kill you and your millions of health. I think I strayed from that, but maybe it's for the best. Forever lasting battles to kill all the healers probably wouldn't be that fun.
I will likely release a new version at some point in the future. I've noticed quite a few minor bugs that need to be fixed. The game balance could use some work as well, and of course general polishing wouldn't hurt at all.
I updated the zip to include a readme file, which may explain some stuff in the game. It's downloadable at the same place: https://sites.google.com/site/dwcroguelikes/.