If "¬" is the record seperator then use the record seperator variable
($/) to get the individual records:
{#create a new scope to prevent changes from effecting everthing else
local ($/) = "¬";
while (<FD>) {
chomp; #get rid of "¬" on the end;
do_something_with_a _record($_);
}
} # $/ goes out of scope and everyone is happy
On 05 Jul 2001 20:23:37 +0100, Mike Breeze wrote:
> Hi there,
>
> I'm writing some code that has to parse a text file containing a series
> of records. The records are a direct export from a third party DB and
> are delimited by the "¬" char (I hope that comes out on all your email
> clients but I guess it doesn't really matter.
>
> What bothers me is that the text file contains only one line which in
> turn contains many records. The test files I have are only 300 or 400K,
> but I'm a bit worried at the risk of receiving a huge file to parse
> which may leave me wanting for memory.
>
> I'm used to parsing files where each record is written to it's own line.
> I.e.
>
> while ( <FD> ) {
> my $job = Job->new();
> chomp;
> $job->FromStr( $_ );
> push ( @{ $self->{people} }, $job );
> }
>
> I could do:
>
> while ( <FD> ) {
> my @jobs = split /\¬/;
> foreach my $jobstr ( @jobs ) {
> my $job = Job->new();
> job->FromStr( $jobstr );
> push ( @{ $self->{people} }, $job );
> }
> }
>
> But this seems to me to be a little clunky given that I may receive a
> HUGE one lined file one day. Is this a valid risk, or am I being too
> careful?
>
> Cheers
>
> Breezy
>
>
--
Today is Sweetmorn, the 40th day of Confusion in the YOLD 3167
Or not.
