In article <00e201c28426$051681e0$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (Patrick Salmon) writes:
>I have a piece of code that does repeated searches through an input =
>string and modifies it according to certain criteria. What I have works =
>great, but I'd like to know how to write it better/more efficiently.=20
>
>Given that I'm going over the same record until I get a hit (or not), =
>how would I re-write the following so that it's not a series of if =
>....if...if...if?
>
> if (/^\s[eE|aA]\d\d\d\d\d\D/){s/A/W95/;$w95++;print DFILE;next;}; # =
> (E or A + nnnnn) =3D Windows95
> if (/^\s[a-zA-Z]{5}dfs\d\d\d+/){s/A/W2K-S/;$w2ks++;print =
>DFILE;next;}; # W2K Server
> if (/^\s[a-zA-Z]{3}salnm+/){s/A/W2K-S/;$w2ks++;print DFILE;next;}; # =
> W2K Server
> if (/^\s[a-zA-Z]{3}scadc+/){s/A/W2K-DC/;$dc++;print DFILE;next;}; # =
> w2k Domain Controller =20
> if (/^\s\w\w\wnfris\d\d\d/){s/A/Snap/;$snaps++;print DFILE;next;}; # =
> Quantum Snap server
> if (/^\susu/){s/A/Non-MS/;$switch++;print DFILE;next;}; # Switch
> if (/^\sus/){s/A/Non-MS/;$legacy++;print DFILE;next;}; # Legacy =
>device (Print Server, etc)
> if (/^\ssap/){s/A/Non-MS/;$legacy++;print DFILE;next;}; # SAP =
>Host =20
> if (/^\srib/){s/A/Non-MS/;$legacy++;print DFILE;next;}; # SNMP =
>Host =20
> if (/^\s\wdcs+/){s/A/DCD/;$legacy++;print DFILE;next;}; # Data =
>Center Device
> if (/^\s[a-zA-Z]{3}de\d\d\d\d\d/){s/A/W2K/;$w2k++;$good++;print =
>DFILE;next;};# w2k desktop - obeying convention
> if (/^\s[a-zA-Z]{3}le\d\d\d\d\d/){s/A/W2K/;$w2k++;$good++;print =
>DFILE;next;};# w2k laptop - obeying convention
> if (/^\s[a-zA-Z]{3}d\d{1,6}/){s/A/W2K/;$w2k++;print DFILE;next;}; # =
> w2k desktop - Pre-Nextwave=20
> if (/^\s[a-zA-Z]{3}l\d{1,6}/){s/A/W2K/;$w2k++;print DFILE;next;}; # =
> w2k laptop - Pre-Nextwave
> if (/^\s[a-zA-Z]{3}l+/){s/A/W2K?/;$w2k++;print DFILE;next;}; # w2k =
>laptop - probable.
> if (/^\s[a-zA-Z]{3}d+/){s/A/W2K?/;$w2k++;print DFILE;next;}; # w2k =
>desktop - probable.
>
> {s/A/?/;} # Anything left over is interesting. Flag it for follow-up =
>with a "?".
Sure there's no print DFILE in that last case? Why the substitution then?
I'd do something like
my @rules = (
qr/^\s[eE|aA]\d\d\d\d\d\D/ => sub { s/A/W95/; $w95++ },
qr/^\s[a-zA-Z]{5}dfs\d\d\d+/ => sub { s/A/W2K-S/; $w2ks++;},
...
);
OUTER: while (<SOMEINPUTSTREAM>) {
for (my $i = 0; $i <= $#rules; $i += 2) {
/$rules[$i]/ and do { $rules[$i+1]->(); print DFILE; next OUTER };
}
s/A/?/;
}
If it weren't for the fact that sometimes you increment two variables I'd
have used a hash for the increments instead. Might still be better to have
a hash instead of all those little variables.
--
Peter Scott
http://www.perldebugged.com
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
