in reply to using CGI, hash slices, and perldoc

I think the issue you're running into is that if CGI::Carp traps an error and attempts to output the message to the browser before your CGI script has had a chance to output proper CGI headers, you'll get that old familiar problem of attempting to generate output without proper headers.

The solution is for your CGI::Carp handle_errors() subref to handle the outputting of headers too. You may find it helpful to keep a flag that lets you know whether or not you've already output a header. The error handling has to be established in a BEGIN{} block so that compiletime errors will properly be handled too.

Regarding hash slices, consider the following examples:

@hash{ @keynames } = @valuelist; # Assign values to a bunch # of keynames at once. @valuelist = @hash{ @keynames }; # Grab a bunch of values given # a list of keynames. ( $address, $phone ) = @{$people{ $name }}{ 'address', 'phone' }; # Grab individual fields # from a record.

The usefulness of a language feature is just a matter of need and imagination.


Dave

Replies are listed 'Best First'.
Re^2: using CGI, hash slices, and perldoc
by greenFox (Vicar) on Feb 02, 2005 at 08:22 UTC