Barry Gershenfeld wrote:
...I want to be selective about which sigs are selected...
I thought about making a script that...greps certain lines
>from the main sig file...
But my problem is that I need grep to cough up everything
between the two "%" delimiters, not just the matching line....
Learn awk, maybe?
As your ideas become fancier, the programming effort you need also
increases. This is the natural succession of things. While it may be
possible to do it all with a shell script, more advanced programs can make
it easier, more repeatable, and more modifiable. I see you've gotten sed
into the act, and that's the kind of next step I'm talking about.
When I climbed onto the Linux bandwagon, I figured I needed to progress
through 3 levels. I see there are really 4.
shell script
sed
awk
perl
And the reality was:
I didn't realize shell scripts even counted as a language. But growing up
with DOS Batch would make you think that. Shell scripts are good, and
sometimes, a script is enough. Often, there's a script behind the final
product anyway.
I don't know if I consider scripts (or batch) files to be a "language",
but they *are* a set of instructions, hence a program, especially if
branching occurs.
sed was easy enough, but every time I went to use it, it would break down as
soon as I needed to handle things that took more than one line. sed can
work across line endings, but it ain't pretty.
awk is like a real programming language. It has the guilty pleasures
associated with string handling, and easy (or no?) data typing, and of
course with regular expressions thrown in. awk solves most of my problems
that are like the one you propose.
perl, I never got to. I did read a "get started" article once, to sort of
look into it, but didn't have a pressing need for it. At the perl level,
you have python and php and as it happened I've also learned php.
So...awk is where I would go with this. And if you want to try it (have you
done any "programming language"s?), I'll help, as I suspect the onlookers
would, too.
Well, most don't esteem BASIC very highly, but that was my first
language on a Radio Shack Pocket PC with a whopping 1.3K RAM. Then
BASIC on Apple II with an incredible 64K RAM which eventually was too
small to run the paint program I was creating (so had to move to the
Apple IIe with 128K RAM).
I found out that goto was frowned upon, and so started avoiding it,
which really wasn't much of a problem after I got the hang of gosub.
Years later, frustrated with the glitches in the work order program that
we were using at work, I peeked into the code, and was surprised how
similar FoxBase (pre-FoxPlus) database programming code was to BASIC. I
pretty much picked up FoxBase on the fly (by the seat of my pants).
After I made several improvements to the code, I was loaned a book on
FoxBase programming. I occasionally referred to it (which was usually
spot on), but usually just figured things out from the context and from
its similarity to BASIC. The original FoxBase code for the work order
program ended up ballooning up to 3 times the original size. And I'll
tell you, it was *much* easier to use, and more feature rich.
Years later (several years ago), I dabbled in Perl, but didn't really
have a project that was driving me. Well, that's not exactly true. But
I kinda lost interest when I got waist deep in the mud when trying to do
something specific (namely, wanting the program to continue working
_until_ a keystroke was entered, temporarily jump into a subroutine to
process the keystroke, and then resume). I was told that I *might* be
able to accomplish what I wanted with things that employed :: in the
name. All in all, it went so far over my head so fast and not enough
help that I just finally lost interest. It is a project I *still* would
like to work on. And with this much passage of time, perl has no doubt
implemented a lot more into its code. I got a little excited when I
found out that perl no longer required a hash of hashes to get two
dimensional arrays (IIRC).
I seem to recall that with Perl, I could slurp the entire sig file into
a single variable. And I'm sure that I probably have forgotten *some*
perl, but I imagine that even that which I have forgotten would probably
come back pretty quickly.
I haven't messed with awk, not at all.
I would like to learn more. I don't feel like I got to a functional
level with perl. But I suppose I actually was "functional", tho barely
that. I never got to a level where I could call outside modules (those
things with :: in the names). But I remember getting to a point where I
realized I could use global variables and local variables, which could
both help me with the project I wanted to do.
I would love to become proficient in perl. I had a couple of books, one
very basic level (written for the total newb), and another that was
probably intermediate to advanced. I do not presently know the location
of those books except that they are somewhere in my possession.
--
Ralph
--------------------
There is no security without privacy. And liberty requires both security
and privacy.
--Bruce Schneier
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg
