Re: What's the right way to write a method which returns one line at a time from a file?

while(my $line = $reader->get_next_line()){

I just wanted to point out that this suffers from the same issue that AnomalousMonk correctly pointed out deeper in the thread: if the file ends on a line containing just "0" with no newline, this loop won't catch that. You would have to say while( defined( my $line = $reader->get_next_line() ) ) instead. Alternatively, note it's possible to overload the <> operator (note that overloaded <> in list context wasn't implemented until 5.18). See for example my use in Algorithm::Odometer::Tiny; you'd just have to change $self->() to your method call, and then you could write while( my $line = <$reader> ) and Perl will automatically add the defined call.

By the way, in your post here, I don't understand the point of the two $self->{file} = shift; lines, especially the second one? Why change the filename while reading the file?

Comment on Re: What's the right way to write a method which returns one line at a time from a file? Select or Download Code

Replies are listed 'Best First'.
Re^2: What's the right way to write a method which returns one line at a time from a file? by Cody Fendant (Hermit) on Nov 28, 2020 at 03:26 UTC
By the way, in your post here, I don't understand the point of the two $self->{file} = shift; lines, especially the second one? Why change the filename while reading the file? Looking at it, you're right, but I think the second one, if you mean the one further down the page, is the one that needs to exist and the first one is the one which doesn't. I instantiate the module without naming the file, then call `get_lines` with the file name. It doesn't need to be passed as an argument to `get_filehandle` because it's already there. My brain was having a very bad day as you can probably tell.	[reply] [d/l] [select]
Re^3: What's the right way to write a method which returns one line at a time from a file? by haukex (Archbishop) on Nov 29, 2020 at 12:16 UTC
I instantiate the module without naming the file, then call `get_lines` with the file name. It doesn't need to be passed as an argument to `get_filehandle` because it's already there. Yes, you're right, because you call `get_filehandle` inside of `get_lines`, it actually could make sense to pass the filename to the `get_lines` call. However, I think it could potentially still be confusing because with the code you showed, if all the user uses is `get_lines`, it'll only ever open one file - say I call `my $x = $obj->get_lines("foo.txt")`, and then `my $y = $obj->get_lines("bar.txt")`, now `$y` contains the second line of `foo.txt`. So that's why it might be better to separate the two actions - opening the file and reading from it - into two methods. Update: The reason I questioned whether it makes sense to pass a filename to `get_filehandle` is that I was imagining `$self->{file}` to be an object property that might deserve its own setter, but that's not as important. BTW, you might want to consider renaming `get_filehandle` to something like `open_file` to make it more clear what the method is doing.	[reply] [d/l] [select]


"be consistent"
	PerlMonks