You originally said "easier to build and *understand*" (emphasis mine).
Machine code is not mnemonic, but assembly is. From where I sit, going
from APL to COBOL is a closer analogy. Speaking only for myself, I don't
see any "conceptual increment" (more power for less effort) in this
case. I can actually get more done with less effort with the RE syntax,
but perhaps that's because I've been using REs for decades. I use a few
high-level RE comments to make them easier to understand and I don't
mind looking up the less-frequently-used modifiers when I need to.
To expand upon John Gamble's comment, I think a good name, description,
synopsis, and clear benefits are more likely to get me to install and
try out a module. I'm not sure the characteristics you've chosen to
emphasize would be the ones to "sell" your module. Dynamic RE
construction and targeting programmers less familiar/comfortable with RE
syntax might be more appropriate, for example.
- Brian
On 2013-08-04 14:28, Shawn H Corey wrote:
On Sun, 04 Aug 2013 14:13:59 -0500
Brian Katzung <bri...@kappacs.com> wrote:
That's an interesting idea, but I'm not sure you're accomplishing
your stated goals. Consider that the RE /^\s*test\s*$/ is 14
characters and you've coded it in 91 non-white-space characters (650%
of the original size).
The difference is between writing machine code and writing assembler.
The goal is not to create readable REs but to make it easier to create
them.
--
Brian Katzung, Kappa Computer Solutions, LLC
Software development and mixed operating system support
for business, education, and science
Phone: 847.412.0713 http://www.kappacs.com
