Date: Tue, 9 Sep 1997 14:31:46 -0400 (EDT) From: Temcguire@aol.com To: hackers@freebsd.org Subject: Interpreter compilers Message-ID: <970909142903_-131376035@emout04.mail.aol.com>
next in thread | raw e-mail | index | archive | help
Experts, For several years now (well since about 1987) I have been using awkcc (from AT&T Toolchest) to compile awk programs. I use awk and the compiled executables to do pattern analysis research with medical datasets. The compiled programs run approximately 10x faster than the interpretive versions. Recently, I installed FreeBSD (2.2.2) on a Pentium Pro in an attempt to upgrade from SPARC. When I attempted to build awkcc (from AT&T Toolchest) I discovered that flex (GNU) was not happy with awk.lx.l in the awkcc source. Seems, that flex -l option also would not accommodate the lack of POSIX compliance in awk.lx.l. I have tried blex (Berkeley port?) and had similar results. Question#1: Any suggestions on how to get around differences between AT&T lex and GNU flex differences within FreeBSD? In the realm of rapid prototyping with interpretive languages, there appear to be several options in FreeBSD (1997). I noticed tcl/tk, JAVA and PERL in particular and observed scheme and python. It appears that a compiler exists for scheme (scheme-to-C). Question#2: Can PERL scripts be translated into C and compiled into executables? Do regular expressions and associative arrays (awk) exist in any other languages that can be translated and compiled? Thank you for any possible assistance! Dr. Thomas E. McGuire 411 East 9th Avenue Salt Lake City, UT 84103 P.S., The Pentium Pro I purchased (Micron Inc.) has the option of a second processor. I have 64Mb of memory and wonder if I installed another processor, would FreeBSD operate much faster?
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?970909142903_-131376035>