I finished reading this a few weeks ago, but I still haven't finished it properly. I've been working through the exercises, and I told myself I wouldn't let myself get too far ahead, but I'm afraid I've still got a few chapters of them to go. Still, I better get on with the review!
Ian Stewart's non-pop-sci books are apparently just as good as his more popularist ones. The presentation is great. Everything is introduced in a welcomely incremental manner, and the proofs nice and intuitive. It's all in a fairly topological framework. At 280 pages with a fairly tight table of contents, it's very thorough in what it covers, and the exercises ensure understanding.
When I mentioned to a friend that I was interested in complex analysis, he wholeheartedly recommended this book, as the modern standard introduction. I must admit I can fault neither his advice, nor this book (modulo a few minor typos in the exercises... surprisingly frustrating to eventually discover!). Having said that, it's left me wanting more... Riemann surfaces, here I come!