grep for text from a postscript file

[J M]

How can I grep for each record in array of text filter against postscript
files?

ArrayF
aaaa
bbbb
abcd
ABcd

I needs to run each record against postscript file and see if there is any
filter matches?

TIA!

 

[Walter Roberson]

:How can I grep for each record in array of text filter against postscript
:files?

:ArrayF
:aaaa
:bbbb
:abcd
:ABcd

:I needs to run each record against postscript file and see if there is any
:filter matches?

Tricky. Postscript is a complete programming language in itself.
There is no possible [finite] general mechanism to determine whether
arbitrary Postscript programs wil produce particular outputs.

Consider for example that a Postscript driver might choose to use
absolute positioning to place the AB and then relative positioning
for the following cd. Absolute positioning of an uppercase letter
introducing a word happens quite a bit in Postscript drivers, in
my experience.


What happens if the driver choses to define producing cd as a function,

/frotz1733 (cd) def
[.... thousands of lines of postscript here]
(AB) 12 timesnewroman
frotz1733

?


If you know that the target might appear *as code* then you just
have a traditional matching problem

use strict;
use warnings;
my @ArrayF = qw(aaaa bbbb abcd ABcd);
my $matchstring = join '|', @ArrayF;
print grep { m/$matchstring/o } (<>)


Note 1: as @ArrayF gets big, the backtracking involved in the matching
can really slow you down.

Note 2: Watch out for metacharacters in @ArrayF

Note 3: Your target list suggests you'd be better off doing a case-
insensitive search such as m/aaaa|bbbb|abcd/i .

 

[Darren Dunham]

have a traditional matching problem

use strict;
use warnings;
my @ArrayF = qw(aaaa bbbb abcd ABcd);
my $matchstring = join '|', @ArrayF;
print grep { m/$matchstring/o } (<>)

Note 1: as @ArrayF gets big, the backtracking involved in the matching
can really slow you down.

If you can guarantee surrounding context, then this can be much faster..

for instance..
m/^$matchstring$/o or
m/\($matchstring\)/o

If you can't, *and* you care which one it matched, you might want to
sort the list by length first (so the longest matches are tried first).

join '|', sort {length $b <=> length $a} @ArrayF;