String handling, money, time, dates. Email, network sockets, cgi, xml. The staples of the cookbook, and Python Cookbook certainly has these. However, interspersed throughout are chapters that seem to have come from at least one other completely different book, a more discursive rumination on Python programming in general. Each chapter begins with a mini essay from a Python luminary, and the discussion of each recipe is fairly extensive.
If you do any scientific or engineering work, you'll know that Python is everywhere on the scientific desktop, providing bindings, scripting and GUI front ends for ancient Fortran/C monstrosities. Reflecting this interest, there is a strong emphasis on performance, with chapters devoted to algorithms and searching and sorting.
Elsewhere, those who have graduated from the plethora of beginner's books, but have been bemused by the complete lack of any intermediate texts, will be pleased to find chapters on Python shortcuts (getting the most out of sequences for the most part) and one on generators and iterators. Futher, there is a chapter on OOP the Python way (including examples of dynamic delegation and design patterns implementation), and one on metaclasses.
This is an extremely useful book, particularly the chapters on using Python's basic collections, which will furnish the reader with some essential idioms for efficient use. However, this, and the OOP chapters would have been better as a separate book. But in lieu of a Thinking in Python or Effective Python, you need this book if you want to do any serious development in Python.
As a cookbook, it has everything you will be expecting as a springboard for exploring the standard library, except for regular expressions. But these are so well covered in introductory books, that you won't need enormous coverage here. On the other hand, the material is presented in a fairly wordy manner, which makes for interesting reading, but for dipping in and out, makes finding things more difficult than it might be.
The other notable thing about Python Cookbook is that it has rather a large number of errors in it. You will want to check the O'Reilly website for the errata, especially if you don't have the most recent printing, rather than scratching your head over why the Singleton implementation doesn't work at all.
Nonetheless, this is a vital resource for Python 2.4 users; even if you don't think you need a traditional cookbook, there is an enormous amount of material here you can benefit from.
I can't help but compare this book to Tom Christiansen's Perl Cookbook. It's not bad, but if you've ever used the Perl book, you're likely to be disappointed by this one.
One problem is that the "recipes" in this book are often for very specific tasks that are unlikely to correspond exactly to what the reader wants to do ("Looking up Holidays Automatically", "Enriching the Dictionary Type with Ratings Functionality", "Converting Among Temperature Scales", etc.).
What do headings like these really tell the reader about the technical problems that the recipes solve? In the Perl Cookbook, you can flip through the pages and within seconds find a heading that corresponds exactly to what you want to do. Here, you have to go through a process of interpretation, and as often as not read a couple of paragraphs of each recipe to see if it's any use to you.
This lack of informativeness about content applies also to some whole chapters, notably "Time and Money" and "Python Shortcuts". What have time and money got to do with the structure of Python? How do I know in advance if my task falls into the category of a "shortcut"? Programming guides should structure themselves around tasks, not concepts.
The book also suffers somewhat from verbosity, especially in the chapter introductions. You don't buy a book like this to be reminded how great Python is ("Python shines in this task", etc.). This may sound picky, but in a cookbook, succinctness is of the essence: you don't want to have to study every paragraph on the offchance that there's useful information in it.
Maybe I'm being harsh, but knowing from the Perl book just how good a programming cookbook CAN be, I have to express my disappointment with this one. I am finding "Python in a Nutshell" a much more helpful, more succinct, and no less comprehensive, guide to the language.