It has been ten years since David Beazley wrote the first edition of Python Essential Reference. The book has proven itself as a valuable resource to Python developers and has been kept current over those ten years, with the fourth edition coming at an interesting time for Python. Python 3 was a major release that broke backwards compatibility. Python 3 has been around for a year now. That said, the current download page at the official Python site states, “If you don’t know which version to use, start with Python 2.6.4; more existing third party software is compatible with Python 2 than Python 3 right now.” Beazley in keeping with the pragmatic roots of a reference that sticks to what is ‘essential’ has removed the coverage on features from 2 that were removed from 3. At the same time, the primary focus for new features that came with 3 is limited to those that have been back-ported to 2. This approach, born out of a desire to keep the reference relevant, provides a blended approach that is above all else practical.
The end result of that choice is a reference document consisting of those parts of Python that are shared between versions 2 and 3. This is a significant portion of the language and I think this approach is really what will give this reference more traction than many of the other guides that focus purely on 3. I think that those are valuable and over time the balance will shift but as of right now, for a little while to come, this book takes the most realistic approach. That feels very fuzzy, but I have no idea how long it will be until Python 3 truly is the dominant version and Python 2 is truly put to bed.
If I had to guess how Beazley’s Python Essential Reference has held in there over the years, the key would be that there is a lot of what a developer needs and very little of what she doesn’t need. There is a twenty-four page tutorial introduction, but this is not a guide on how to program or how to use Python for beginners. An experienced programmer could probably use this reference to shift to Python as a new language, but someone completely new to writing code would probably not want to start here. A quick look at the table of contents shows that an explanation of the language itself is covered in under 200 pages. Extending and embedding Python also get their own section, but close to 400 pages is given to the Python library.
An inevitable question is what one will gain with this reference over the online documentation. A good example to see how things vary is to look at chapter nineteen, Operating System Services and the online documentation for Generic Operating System Services. The online documentation is very thorough, and covers each piece of the library starting with os and io, building from there. While every facet is documented much of it is rather brief. For example section 16.2.3. Raw File I/O is a very straightforward listing of the very low level functionality available via io.FileIO. In contrast, looking at the 3.1.1 Docs for Raw IO shows that parameters for FileIO changed with that version. Looking to the documentation for 2.7a1 Raw File I/O shows that these changes are being back-ported to Python 2.
In Python Essential Reference none of this hunting down changes and checking to see if they are coming to 2 are necessary. Beazley shows them in his documentation. This is the strength of his choice on how to handle these types of situations. On top of that, Beazley provides more than the online documents by including four paragraphs of additional information on Raw I/O and when its use is appropriate. This added content is probably available googling around for it, but then I have to take the time to check dates on posts to see if things are still current and in general just hope that things are accurate. I have never read a technical book that was completely error free, and there are probably at the very least some typos in Python Essential Reference, though I haven’t caught any of them on my read through or use of the book yet. But the important thing is that I don’t expect the book to be perfect, rather I value it for being a known quantity. I am aware of just when the material was compiled, who put it together and I have it all in one place.
The index is solid. It would seem that one should be able to take this for granted with a technical reference but I’ve seen some sad exceptions. Between the thorough index and the detailed table of contents I’ve never had to spend more than a few seconds looking for what I need. This is the result of those tools as well as the fact that this is not an exhaustive reference. After initially reading through the book for this review, I’ve taken some time just to use it day to day, as I doubt many will be reading it from front to back. I don’t use Python professionally. I’m purely a hobbyist when it comes to programming, but I’ve found that if I want to get the most out of the time I do have to play with personal projects, I want this book close. I’m not cranking out code that fast to begin with and so I need all the help I can get. I’ve found that Beazley seems to have hit that sweet spot where he gives enough information to get me where I need to be without bogging down in too many details or the things that I just don’t need to know. I imagine this proper balance of information is due to Beazley’s extensive experience with Python and that of Noah Gift the technical editor for the book.
I’ve mentioned repeatedly that I approve of how the shift between Python 2 and 3 has been handled. Beazley hasn’t completely integrated everything and left some of the unique new features of 3 out in the cold. There is an appendix that deals specifically with Python 3. It is short but does have some value. New features, common pitfalls for those making the move from 2 to 3 and how to run both at the same time in a single environment are covered. This is helpful and keeps my desk a little neater, though I think if I were going to be spending extensive time working with Python 3 then I would probably want to have another reference on hand.
If you are a week-end hacker like me, or someone that is writing Python on the clock, I think that this compact reference is very useful. I don’t have any trouble running across huge technical books that do come in handy for any project that requires something heavy. I also see a lot of little books that seem to be quickly produced summaries of what is already out there, spending most of their short content on fluff. Every so often though, someone hits that sweet spot of concise usefulness. Beazley did this with Python Essential Reference and this new edition continues that history in strong fashion.
Title: Python Essential Reference 4th ed.
Author: David M. Beazley
Publisher: Addison-Wesley Professional