Re: trying to use IO::Handle

Richard Lee wrote:
why does below fail?

----- where it's failing -----
open my $source, "<", "/tmp/server.log"
or die "Could NOT open /var/log/server.log: $!";

$source is an opened filehandle.

my $file = new IO::Handle;

$file is an IO::Handle object but it is not related to any file.

while ($_ = $file->getline($source) ) {

You are trying to getline() from an object that is not associated with a file and $source is ignored because getline() does not take an argument and as the example from the documentation below shows you are not testing if the returned value is defined which may exit out of the loop before the end-of-file.


Uncaught exception from user code:
usage: $io->getline() at ./ line 126
at /usr/lib/perl5/5.8.8/i386-linux-thread-multi/IO/ line 411
IO::Handle::getline('IO::Handle=GLOB(0x9af8c5c)', 'GLOB(0x9b12380)') called at ./ line 126
main::shove_it() called at ./ line 98

Does not take an argument.

This works like <$io> described in "I/O Operators" in perlop
<> except that
it's more readable and can be safely called in a list context but
still returns just one line. If used as the conditional +within a
|while| or C-style |for| loop, however, you will need to +emulate

the functionality of <$io> with |defined($_ = $io->getline)|.
defined() test required when used in while loop.

I was trying to speed up my program which has to go through huge server log (500Meg+) and when I was going over w/

while ($source) {

That won't work without the readline operator:

while ( <$source> ) {

Note that the built-in readline is short for:

while ( defined( $_ = <$source> ) ) {

Where the assignment to $_ and the defined test are applied by default.


I discovered(from Dprof) that it was using up too much CPU(it's on old system) and wanted to use a IO module to get down to lower level to gain some speed but I am not too familiar w/ using modules.. can someone help
on this?

IO::Handle is just an OO wrapper on top of Perl's built-in operators:

perldoc -m IO::Handle

[ SNIP ]

sub getline {
@_ == 1 or croak 'usage: $io->getline()';
my $this = shift;
return scalar <$this>;

So in the end you are just using Perl's <> readline operator but with the overhead of a method call.

Perl isn't a toolbox, but a small machine shop where you
can special-order certain sorts of tools at low cost and
in short order. -- Larry Wall

Relevant Pages

  • Re: getline
    ... the subsequent loop in that function depends on "str" but that loop doesn't so you could just populate "letter" once in the BEGIN section rather than every time the function gets called. ... maybe it's time to consider getline". ... To do that you need to loop on all the input records and call fooon each one. ... awk automatically does a loop on the input so the usual thought would be just to write: ...
  • Re: getline caveats misuse howto: final version
    ... but this is a state machine with a very limited number of states. ... Here's an example where I use getline. ... Fix PNG loading for non specialized types. ... Since the loop was never entered, the "sym" array will never be populated. ...
  • Re: get(char*, num, delim) question
    ... >> instructed to use the getfunction vice the getline() function. ... >Its still preferable to use an array instead of a pointer. ... >the end of the file so you end up going round the loop one too many times. ... instructor said we needed to use getbecause we need the practice ...
  • Re: Read everything from socket
    ... The problem is that getline is a blocking call, ... while when you know you got the last line, and the retr command uses a ... # Reading from RETR ... getline method doesn't seem to detect the end of the stream; the loop ...