Archive of UserLand's first discussion group, started October 5, 1998.

Re: Open Source Projects and quality...

Author:A.M. Kuchling
Posted:8/26/1999; 10:19:04 AM
Topic:Opening Up Linux Journal and O'Reilly
Msg #:9960 (In response to 9941)
Prev/Next:9959 / 9961

Perl's source code is also quite difficult to read. Ever wonder why no one else uses Perl's regex engine, instead always writing a brand-new engine with the same syntax? (For example, there's a Perl-like regex engine inside Mozilla, there's a C library called PCRE, and Tcl 8.x uses a new regex engine from Henry Spencer.) A while back I pored over at the code in Perl's regexec.c and regcomp.c, and while it's presumably comprehensible to someone, comments are rare and not very helpful; much easier just to write a new engine...

This property even made it into the Jargon File's definition of "joe code": http://www.tuxedo.org/~esr/jargon/html/entry/joe-code.html Recent versions of Perl don't seem to be improving matters.


There are responses to this message:


This page was archived on 6/13/2001; 4:52:08 PM.

© Copyright 1998-2001 UserLand Software, Inc.