http://qs321.pair.com?node_id=543962


in reply to Re^5: Un "tie"ing a "tie"
in thread Un "tie"ing a "tie"

I spent time addressing all these concerns in my post. I find it weird that you chose to disreguard them when suggesting an improvement.

Support for all the special variables needs a little thought, because they are dynamically scoped. I'd probably want to capture their values in the "open" function and store those values in the object.

Actually, it's much simpler than that. Inherit PRINT and PRINTF from Tie::Handle, and override WRITE like I did.

I'm not sure that binmode is required for this filehandle. All it does is call a user-defined perl function for each line (implicitly text). If that code prints to a "real" filehandle, then it is that filehandle that needs to support binmode.

It should have a dummy in case you want to use the handle somewhere that calls binmode. As is, it can't be used as a replacement for a normal file handle, which is the whole reason to use tie.

I'd be interested to know how to rewrite the "open" function to avoid needing 5.8.x -- the requirement exists because of bugs in 5.6.x, so the simplest (and safest) thing it to require a version of perl that works reliably with tied filehandles.

See wrap in my post. It works fine in 5.6.1.

Replies are listed 'Best First'.
Re^7: Un "tie"ing a "tie"
by dpuu (Chaplain) on Apr 18, 2006 at 01:32 UTC
    I spent time addressing all these concerns in my post. I find it weird that you chose to disreguard them when suggesting an improvement.

    Sorry. I cut&paste from one of my existing scripts (which didn't support them), and didn't spend the time to merge it with yours. Yes, I probably should think about modifying my code to take some of the improvements you suggest

    Your "wrap" works because you are willing to require the user to pass the handle-container in as a typeglob (returning localised filehandles starts to hit the bugs). I didn't want to do that because the natural interface is a signature of (&) so that its single arg is a codeblock. To mimic the "open" builtin it'd need a signature of (*&) which would clutter user-code with the "sub" keyword.

    --Dave
    Opinions my own; statements of fact may be in error.