Linux Programing Hints

Perl is often considered a scripting language for systems administrators. Jim demonstrates that it is useful to applications and scientific programmers as well—as a prototyping tool.

You have seen two examples of how Perl can be used for prototyping. I hope that from these examples you have gained a feel for Perl's syntax. More importantly, I hope that you have seen how using Perl can free you from concentrating on programming specific details, like memory allocation. Instead, you can direct your efforts toward getting your algorithm up and running. I have discovered that, in many cases, the Perl prototype was sufficient for my purposes, saving me the time of coding the program in C or C++ at all!

When I do recode a prototyped algorithm from Perl to another language, I have found that it is easy to change gears. The logic is behind me, freeing me to concentrate on C specifics, memory allocation/deallocation, input/output, error reporting, etc.

My suggestion to the reader is to program a simple application in Perl and see for yourself how this very elegant and powerful language works. You may not save any time with the first program or two, but it will not be long before the benefits of Perl appear. If you feel ambitious, try writing a routine to replace my point_on_line. I mentioned earlier that my algorithm for testing whether a point is on a line is not very efficient. Another, more efficient scheme, is to first check whether the point's x coordinate is within the x range of the line and, if so, whether the point's y coordinate satisfies the equation of the line. Vertical lines are special cases.

Among the many algorithms I have prototyped in Perl are LZW data compression (the same as used in the UNIX compress utility), RSA encryption, many matrix operations including eigenvalue/eigenvector determination and a code generator that outputs C code from a database. I even have a little program called “perls” that reads a database of perl programming tips and prints a random tip to the screen. [I can provide this program to The Linux Journal and/or its audience via Internet. Let me know if you are interested.]

[Yes, we are. We would like to put it on our web site, perhaps even in a cgi script.]

Jim Shapiro is a consultant specializing in programming mathematical algorithms. He is presently developing a GIS system for a telecommunications company. When he isn't on his Linux system hacking away in C or Perl he can often be found on the squash courts. Jim is a founding member of LUGOR, the Linux User's Group Of the Rockies.

Perl Resources

Programming Perl by Larry Wall and Randal L. Schwartz, O'Reilly & Associates, Inc., 1992. If you are serious about learning Perl, this is the book to read. It is all here, including some very sophisticated examples. Not recommended for beginners, however.

Learning Perl by Randal L. Schwartz, O'Reilly & Associates, Inc., 1993. A tutorial divided into lesson sized chapters.

Teach Yourself Perl in 21 Days by David Till, SAMS publishing, 1995. My personal favorite. Looks more daunting (841 pages) than it is. I got so excited I read it in seven days. Read this one, then “Programming perl”, and you will soon be an expert.

The “man” pages. Not bad if you want to get the flavor of the language, but mine seem dated.