Re: Simple typo in numerical sort, major error

really? maybe you should get out more...

it is a bit unfortunate that that particular typo still produces valid code... maybe someone will be able to add a catcher for this to the warnings pragma, but I bet theres heaps more permutations than the one you found and that ikegami mentioned.

Comment on Re: Simple typo in numerical sort, major error

Replies are listed 'Best First'.
Re^2: Simple typo in numerical sort, major error by JavaFan (Canon) on Dec 30, 2009 at 23:44 UTC
`open my $fh, ">", "important-file" or die;` [download] Very costly if the `>` was a mistyped `<`. IMO, the `warnings` pragma is the wrong type of tool for detecting typos like the OP made. After all, such a check only needs to be done once (or once after each change). There's absolutely no point in paying the penalty for each and every run. Check like this (if once could actually make a check that doesn't trigger too many false positives) belong in a linter. Perl::Critic is a well used linter in the Perl world. And of course one has their test suite, which should be able to catch mistakes like that.	[reply] [d/l] [select]

Replies are listed 'Best First'.

Re^2: Simple typo in numerical sort, major error
by JavaFan (Canon) on Dec 30, 2009 at 23:44 UTC

open my $fh, ">", "important-file" or die;
[download]

>

<

IMO, the warnings pragma is the wrong type of tool for detecting typos like the OP made. After all, such a check only needs to be done once (or once after each change). There's absolutely no point in paying the penalty for each and every run. Check like this (if once could actually make a check that doesn't trigger too many false positives) belong in a linter. Perl::Critic is a well used linter in the Perl world. And of course one has their test suite, which should be able to catch mistakes like that.

[reply]
[d/l]
[select]