NAME

Tk::FmtEntry - A Formatted Entry widget

SYNOPSIS

my $fe = $mw->FmtEntry(-fcmd => sub {...})->pack;   # The widget

DESCRIPTION

A normal Entry widget, but it may enforce some format upon the entered value. For example, telephone number, credit-card, or currency (money) formats. The programmer must supply the formatting function. See below for examples.

Otherwise, this acts a like a normal Entry widget, except for some restrictions on the use of -validate.

The following additional options/value pairs are supported:

-delay

Normally, the formatting function is called as soon as possible after any edit to the contents of the FmtEntry field. However, you may want to delay the formatting a bit, so multiple edits to the field can be "batched" and handled at once. This may also be useful when swipe-card reads are to be extracted from a field that's also a user-entry (type-in) field.

Use the -delay option to specify the number of milliseconds to wait before the formatting fuction is invoked. Edits to the field must be idle for this amount of time before the format function is called: if you set an 80ms delay, for example, and the edits come in with 65ms, 79ms, 55ms, then 81ms pauses, the call will occur after the last pause (not after each).

Note that users typically percieve delays of more than 80ms.

-formatcommand

Alias: -fcmd

The formatting function. This function is called to cleanup, correct, and/or [re]format the value contained in the entry. It is called with three arguments: the (possibly) unformatted string in the entry field, and the current insertion cursor index, and a reference to the FmtEntry widget.

The function should return two values: the cleaned-up string, and the revised insertion cursor point. If the revised insertion point is not returned (undef), then a simplistic correlation algorithm will be used to guess the new insertion cursor point. It's not a very good guess, so the cursor may not be where you expect it. Thus, your formatting function should calculate and return the revised position.

An example may help. Suppose we want to enforce all entries to be lowercase. In this simple example, the insertion curor point does not change, so we just return it un-altered, along with the lower-cased string:

my $fe = $mw->FmtEntry(-fcmd => \&fmt_lc)->pack;
  .
  .
  .
sub fmt_lc {
    my ($old, $i) = @_;
    return (lc $old, $i);
}

Interaction with -validate and -validatecommand

The only -validate options supported are 'none' and 'key'. Attempting to set any other -validate option will cause this widget to carp an error.

The -validatecommand (-vcmd) you specify is called out of Tk::FmtEntry's internal validator function, but it works just like it was called from Tk::Entry.

Insertion Cursor Positioning

The hardest thing about using this widget is that your custom formatting function needs to return a revised insert cursor position, in addition to the [re]formatted string. The 'insert cursor position' is where new characters are added when you type. If you don't correctly determine and return the revised position, things will go all screwy when the user types; their characters may get scrambled.

For example, the user wants to type 57 into a field that you're formatting as currency (money). The user first types "5", which leaves the cursor at position i=1 (just after the "5" character). Your formatting function changes this string to "$5.00". Because of the dollar sign prefix, the insertion cursor needs to change from position i=1 to position i=2. If this was not done, the cursor would remain at i=1 and typing another digit, say "7", would result in "$75.00" instead of the desired "$57.00".

How It Works

If the normal Entry widget allowed correction of the value during a -validatecommand callback, then this derived widget wouldn't be necessary. However, doing so can cause strange interactions with the -textvariable, especially if the -textvariable is changed outside the callback function. Hence, we have this FmtEntry widget.

This widget avoids the pitfalls of changing the Entry's value by deferring the change to some time after the Entry finishes its processing. It does this by using the -validatecommand callback on 'key' events, to set an afterIdle() "fixup" callback. This fixup callback is called moments after (unless -delay is used), and the formatting/correction/whatever is done there. If your Tk application is very busy, you may notice a delay before the (re)formatting occurs. I'm not thrilled with this approach, but it works pretty well.

METHODS

None.

ADVERTISED SUBWIDGETS

None.

EXAMPLES

Force Entry to be Uppercase

All letters typed in the FmtEntry field are uppercased.

use Tk;
use Tk::FmtEntry; 
my $mw = new MainWindow;
$mw->FmtEntry(-fcmd => \&fmt_uc)->pack;
MainLoop;

sub fmt_uc {
  my ($old, $i) = @_;
  return (uc $old, $i);
}

In this example, the entry is simply uppercased via the formatting function fmt_uc(). The cursor index $i does not change position, so it's just returned as-is.

Credit Card Format

This example formats the entry as a 16-digit credit card number, like XXXX-XXXX-XXXX-XXXX . Only digits are accepted.

!/usr/bin/perl -w
use strict;
use Tk;
use Tk::FmtEntry;

my $mw = new MainWindow;
$mw->Label(-text => 'Enter a Credit Card number')->pack;
$mw->FmtEntry(-fcmd => \&fmt_cc)->pack;
MainLoop;

sub fmt_cc {
    my ($old, $i) = @_;

    # To figure the new insert cursor position, 
    #  format just the left half and see where it lands.
    my $lf = substr($old, 0, $i);
    $lf =~ s/[^\d]//g;    			# remove all but digits
    $lf = substr($lf, 0, 16);                  # max 16 digits
    while ($lf =~ s/(\d{4})(\d)/$1-$2/) { };   # group to fours
    my $j = length($lf);                       # get new position

    # Now format again the whole thing
    my $new = $old;
    $new =~ s/[^\d]//g;                        # nuke all but digits
    $new = substr($new, 0, 16);                # max 16 digits
    while ($new =~ s/(\d{4})(\d)/$1-$2/) { };  # group to fours
    return ($new, $j);
}

Note how the revised insert cursor position is determined. Although there's likely more efficient methods, a simple approach is to split the old string at the old cursor postion (call this the 'left' part), then format this left part and see how big it is. The length is the new cursor position. Then repeat, formatting the whole old string and return this as the new.

Money Format

Example formatting function for currency. This formatter uses the trick of placing a marker character (an asterix) into a second "marked" string, then after formatting it correlates the marked string to the new string, to determine the cursor position. It's not perfect but it works reasonably well (it has a problem with leading zeros - see if you can fix it!)

#!/usr/bin/perl -w
# Example Tk::FmtEntry with cash (money) style formatting
use strict;
use Tk;
use Tk::FmtEntry;

my $mw = new MainWindow;
$mw->Label(-text => 'Enter Money Amount')->pack;
$mw->FmtEntry(-fcmd => \&fmt_cash)->pack;
MainLoop;

sub fmt_cash {
    my ($old, $i) = @_;

    # Make the new string
    my $new = $old;
    $new =~ s/[^\d\.]//g;             # remove all but digits and decimal
    if ($new eq q{.}) {$old = $new = "0."; ++$i;}   # special for dp-only
    $new =~ s/(\.\d{0,2}).*$/$1/;                   # max two past dp
    $new = sprintf '$%4.2f', $new if $new ne q{};   # if blank, leave blank

    # Add commas
    $new = reverse $new;
    $new =~ s/(\d{3})(?=\d)(?!\d*\.)/$1,/g;
    $new = reverse $new;

    # Make a marked string
    my $mrk
        = substr($old, 0, $i) . q{*} . substr($old, $i, length($old) - $i);
    $mrk =~ s/[^\d\.\*]//g;    # remove all but digits, decimal, and marker

    # Find new insert point
    my $j = 0;
    my $k = 0;
    foreach my $c (split //, $new) {
        if ($c eq q{$} || $c eq q{,}) {
            $j++;
            next;
        }
        last if substr($mrk, $k++, 1) eq q{*};    # found the marker
        $j++;
    }
    return ($new, $j);
}

AUTHOR

Steve Roscio <roscio@cpan.org>

Copyright (c) 2010, Steve Roscio <roscio@cpan.org>. All rights reserved.

This module is free software; you can redistribute it and/or modify it under the same terms as Perl itself. See perlartistic.

DISCLAIMER OF WARRANTY

Because this software is licensed free of charge, there is no warranty for the software, to the extent permitted by applicable law. Except when otherwise stated in writing the copyright holders and/or other parties provide the software "as is" without warranty of any kind, either expressed or implied, including, but not limited to, the implied warranties of merchantability and fitness for a particular purpose. The entire risk as to the quality and performance of the software is with you. Should the software prove defective, you assume the cost of all necessary servicing, repair, or correction.

In no event unless required by applicable law or agreed to in writing will any copyright holder, or any other party who may modify and/or redistribute the software as permitted by the above licence, be liable to you for damages, including any general, special, incidental, or consequential damages arising out of the use or inability to use the software (including but not limited to loss of data or data being rendered inaccurate or losses sustained by you or third parties or a failure of the software to operate with any other software), even if such holder or other party has been advised of the possibility of such damages.

KEYWORDS

FmtEntry