Deitel and Deitel Book...
Jim Dennis
jimd at vega.starshine.org
Sun Mar 3 00:07:35 EST 2002
In article <899f842.0203010812.6151d817 at posting.google.com>,
Anthony_Barker wrote:
>"DeepBleu" <DeepBleu at DeepBleu.org> wrote in message
>> Deitel and Deitel book on Python? I personally would not touch it...
...
>> You want books on Python: Buy Beazley, Lundh, Holden, Hammond, Greyson
>> etc....
>> Leave Deitel alone!
>> DeepBleu
> I agree with you to a point. They don't seem passionate about
...
> presentations are excellent, as is his dead tree book. Eckel's, "The
> thinking in Python", is very good.
...
I thought Eckel's TiPy and his "Thinking in Patterns (with Python)"
had been merged. I didn't realize that either of them had been
released in dead tree editions. (I've read parts of them online
at his web site)
Bruce Eckel's MindView, Inc: Thinking in Patterns with Java
http://www.mindview.net/Books/TIPatterns/
Bruce Eckel's MindView, Inc: Thinking in Python
http://www.mindview.net/Books/TIPython
(Actually it seems, if I believe his web pages, that these are still
developmental versions, and have not yet been set to print).
> I would love to see books on Algorithms with Python (makes a perfect
> fit), and more advanced topics such as threads and extending python.
I would love to see a good book on "patterns" in Python. I'd like
it to be a simple booklet listing each of the GoF patterns, with a
couple of examples of each. It would be fine if it also covered some
additional patterns or some controversies or alternatives to the
classic GoF patterns (like the Alex Martelli's Borg vs. the GoF Singleton).
Don't even ask for my opinion on Thomas W. Christopher's
_Python_Programming_Patterns_ book (PTR/PH Prentice Hall, (c) 2002),
it has almost nothing to do with OODP (one chapter with superficial
coverage of 20 out of the GoF 23 patterns, and the example code is
convoluted, opaque and mostly irrelevant to Patterns). (As a saving
grace his chapters on threads and concurrency are somehwat better.
I think my best bet so far is to study online at:
Python Patterns
http://www.thinkware.se/cgi-bin/thinki.cgi/PythonPatterns
... and this seems to be the rule for Python. The best material
seems to be live and online. However, I still need something to
take to the coffee shop or into the can with me (no I don't have a
computer in *there* yet; and I haven't set up sitescooper to fetch
Python stuff to my Palm/Visor, yet).
I've been enjoying Steve Holden's book, though it is pretty basic
and a little slow in places. I thought his approach to introducing
OOP by starting with strawman procedural examples going through a
"objects as data structures" phase and iteratively refactoring that
into an OO design with constructors and other methods was interesting,
though a bit repetitive in the written form. (It would make alot of
sense in a lecture format, though). I also picked up Robert W. Bill's
new Jython book from New Riders. Based on that, Holden's work and
Beazeley's _Python_Essential_Reference_ I'd say that New Riders is
kicking @$$ in the Python arena.
Although I like most of the classic O'Reilly titles, I have to say that
I'm a little disappointed in Fredrik Lundh's _Python_Standard_Library_
--- it's just too short and doesn't offer comparisons among related
or similar modules. For example the coverage of the curses module
consists of one sentence of running prose and one 14-line code
example. That code example doesn't mention the curses.wrapper()
function, doesn't show his code in a try: finally: block (which could
leave the terminal in an unusable state if any exception is thrown)
and doesn't show or even mention ncurses color (which is a common
source of confusion in this NG). (Of course it might not be fair to
pick on the curses coverage, which seems to be relegated and accursed
in all Python documentation; but this is just one example).
Meanwhile I thought both the Lutz books were tedious (though I haven't
read the 2nd edition of Programming Python).
Meanwhile the O'Reilly book on _Python_&_XML_ by Christopher A. Jones
and Fred L. Drake, is very good.
More information about the Python-list
mailing list