Re: Mastering Perl 5 by Eric C. Hermann

[davorg]

Since this is my first Perl book, I can't rate it based upon my knowledge of Perl. But as far as programming books go, I'd say it's possibly the best programming book I've read (so far) because it simply made sense, and It wasn't so boring.

Reading stuff like this in a review worries me rather a lot. Surely the most important thing about a Perl book is whether or not it teaches good Perl. I've ranted on this theme at various other times, but this is the reason why I ignore all reader reviews at sites like Amazon. Most people who review technical books are unqualified to do so. This renders their review worthless.

My theory is that a good book needs two things:

• It needs to present good quality information
• It needs to present it in a way that its target audience finds easy to follow

The target audience for this book is obviously people with little or no Perl knowledge. Those people are therefore qualified to comment on the second of these criteria, did the book speak a language that you understood, but they are exactly the worse set of people to judge a book on the first criteria. As far as they know, the book could be teaching them the skills to be the next Matt Wright (this, by the way, would be a bad idea!) They simply have no way of knowing.

He's a simple checklist that I use when flicking thru books in a bookshop to see how good a beginners book might be:

• Does the book encourage the use of -w and use strict?
• Does it explain the return values of localtime correctly?
• Does it use -T and CGI.pm for CGI programs?
• Does it explain finding, installing and using modules from CPAN?

Fulfilling these criteria doesn't necessarily mean that a book is any good, but not fulfilling them guarantees that the book will be a waste of money.

How does Mastering Perl 5 rate on these criteria?

--
<http://www.dave.org.uk>

"The first rule of Perl club is you don't talk about Perl club."

[staeryatz]

Herrmann DOES encourage using strict and -w. It's one of the first things he mentions near the beginning of the book, before teaching how to code. The whole second chapter of the book is dedicated to debugging, making your code easier to debug, how to follow "Good Coding Practices", and "Avoiding Common Perl 5 Mistakes". It also shows how to debug in Emacs, in ActiveState Windows Debugger, and Debugging CGI programs.

Yes, it does explain all the return values of localtime (page 862). An array that contains the following in order:


• The seconds from 0 to 60


• The minutes from 0 to 60


• The hour of the day from 1 to 12


• The day of the month from 1 to 31


• The month of the year from 0 to 11


• The year in four-digit format


• The day of the week from 0 to 6


• The day of the year from 1 to 365


• Whether or not it is daylight saving time (1 indicates true)



Sorry no CGI.pm that I've seen. Herrmann shows how to roll your own CGI scripts, and how to decode the HTTP headers yourself (and what it actually is doing). Later on, it shows how to protect your CGI scripts from possibly being exploited by hackers by filtering meta-characters from text entries, which can be used to escape the CGI code and start shell commands on the server.

CPAN? Yes indeed. At the time this book was written, Herrmann referred to it as ftp://ftp.cdrom.com/pub/perl/CPAN/ in the introduction of the book (unfortunately this address has been depreciated since then). Other websites and Perl newsgroups are also listed.

The book also shows some pretty nifty advanced Windows things, like how to load DLL's and call their functions, and calling the Win32 API functions. Please don't tell me that this is beginner stuff...

[davorg]

Actually, localtime returns a list not an array. Either the author got that wrong or he didn't differentiate between the two well enough for you to pick understand that there is a difference.

Also, the year is not returned as four digits. It is the number of years since 1900 (currently 101).

The CGI information sounds a bit flakey to me. Does he mention taint mode at all?

I don't understand why he references CPAN thru a third-party URL that is liable to change or disappear. Why not use one of the Perl community's own URLs of http://www.cpan.org or http://search.cpan.org.

On the basis of this test the book looks like something that I couldn't recommend, but I should probably get hold of a copy and send a review to the publishers.

Thanks for bringing it to my attention.

--
<http://www.dave.org.uk>

"The first rule of Perl club is you do not talk about Perl club."
-- Chip Salzenberg