The Perl Toolchain Summit needs more sponsors. If your company depends on Perl, please support this very important event.

NAME

HTML::QuickTable - Quickly create fairly complex HTML tables

SYNOPSIS

    use HTML::QuickTable;

    my $qt = HTML::QuickTable->new(
                   table_width  => '95%',       # opt method 1
                   td => {bgcolor => 'gray'},   # opt method 2
                   font_face => 'arial',        # set font
                   font => {face => 'arial'},   # same thing
                   labels => 1,                 # make top <th>?
                   stylesheet => 1,             # use stylesheet?
                   styleclass => 'mytable',     # class to use
                   useid  => 'results',         # id="results_r1c2" etc
                   header => 0,                 # print header?
             );

    my $table1 = $qt->render(\@array_of_data);

    my $table2 = $qt->render(\%hash_of_keys_and_values);

    my $table3 = $qt->render($object_with_param_method);

DESCRIPTION

This modules lets you easily create HTML tables. Like CGI::FormBuilder, this module does a lot of thinking for you. For a comprehensive module that gives you the ability to tweak every aspect of table building, see HTML::Table or Data::Table. This one gives you a lot of control, but is really designed as an easy way to expand arbitrary data structures.

The simplest table can be created with nothing more than:

    my $qt = HTML::QuickTable->new;
    print $qt->render(\@data);

Where @data would be an array holding your data structure. For example, the data structure:

    @data = (
        [ 'nwiger', 'Nathan Wiger', 'x43264', 'nate@wiger.org' ],
        [ 'jbobson', 'Jim Bobson', 'x92811', 'jim@bobson.com' ]
    );

Would be rendered as something like:

    <table>
    <tr><td>nwiger</td><td>Nathan Wiger</td><td>x43264</td><td>nate@wiger.org</td></tr>
    <tr><td>jbobson</td><td>Jim Bobson</td><td>x92811</td><td>jim@bobson.com</td></tr>
    </table>

Of course, the best use for this module is on dynamic data, say something like this:

    use DBI;
    use HTML::QuickTable;

    my $qt = HTML::QuickTable->new(header => 1);    # print header
    my $dbh = DBI->connect( ... );

    my $all_arrayref = $dbh->selectall_arrayref("select * from billing");
    
    print $qt->render($all_arrayref);

With header => 1, you will get a brief CGI header as well as some basic HTML to prettify things. As such, the above will print out all the rows that your query selected in an HTML table.

FUNCTIONS

new(opt => val, opt => val)

The new() function takes a list of options and returns a $qt object, which can then be used to render() different data. The new() function has a flexible options-parsing mechanism that allows you to specify settings to pretty much any element of the table.

Options include:

header => 1 | 0

If set to 1, a basic CGI header and leading HTML is printed out. Useful if you're really looking for quick and dirty. Defaults to 0.

htmlize => 1 | 0

If set to 1, then all values will be run through a simple filter that creates links for things that look like email addresses or websites. Also, *word* will be changed to <b>word</b>, and _word_ will be changed to <i>word</i>.

labels => 1 | 0 | LTRB

If set to 1, then the first row of the data is used as the labels of the data columns, and is placed in <th> tags. For example, if we assume our above data structure, and said:

    my $qt = HTML::QuickTable->new(... labels => 1);

    unshift @data, ['User', 'Name', 'Ext', 'Email'];

    print $qt->render(\@data);

You would get something like this:

    <table>
    <tr><th>User</th><th>Name</th><th>Ext</th><th>Email</th></tr>
    <tr><td>nwiger</td><td>Nathan Wiger</td><td>x43264</td><td>nate@wiger.org</td></tr>
    <tr><td>jbobson</td><td>Jim Bobson</td><td>x92811</td><td>jim@bobson.com</td></tr>
    </table>

Since the labels are placed in <th> tags, you can then use the extra HTML options described below to alter the way that the labels look.

You can also set this to a string that includes the characters L, T, R, and B, to specify that <th> tags should be created for the Left, Top, Right, and Bottom rows and columns. So for example:

    labels => 'LT'

Would alter the table so that both the first row AND first column had <th> instead of <td> elements. This is useful for creating tables that have two axes, such as calendars.

null => $string

If set, then null (undef) fields will be set to that string instead. This is useful if pulling a bunch of records out of a database and not wanting to get blank table spaces everywhere there's a null field. For example:

    my $qt = HTML::QuickTable->new(null => '-');
    my $all_arrayref = $sth->fetchall_arrayref;
    print $qt->render($all_arrayref);

By default null table elements are left blank.

nulltags => \%hash

In addition to just changing the string used to represent null data, you may want to change the look of it as well. These tags will become attributes to the <td> element holding the null field. So, settings like this:

    null => 'N/A',
    nulltags => {bgcolor => 'gray'},

Would result in an element like the following for null fields:

    <td bgcolor="gray">N/A<td>

Make sense?

stylesheet => 1 | '/path/to/style.css'

If set, then any font settings are ignored and instead all table elements are wrapped with a class= attribute. The class name is whatever styleclass is set to (see below). See also the useid option to generate id tags in an intelligent way.

styleclass => $string | \@array

This used as a style class to use if the above setting is used. If set to a string, it is passed directly to the class tag. If set to an arrayref, then those styles are alternated between on a row-by-row (tr) basis. For example:

    styleclass => [qw(one two)]

Would yield XHTML similar to:

    <table class="one">
      <tr class="one">
        <td class="one">a</td>
        <td class="one">b</td>
        <td class="one">c</td>
        <td class="one">d</td>
      </tr>
      <tr class="two">
        <td class="two">e</td>
        <td class="two">f</td>
        <td class="two">g</td>
        <td class="two">h</td>
      </tr>
    </table>

Notice that the table gets the style of the first array element.

text => $string

Just like FormBuilder, this text is printed out for you to easily annotate your table.

title => $string

If you set header => 1, then you can also specify the title to be prefixed to the document. Otherwise this option is ignored.

useid => $baseid

If set, then unique id tags are automatically generated for each and every table element, allowing you to address the entire table on a per-element basis via Javascript or CSS. These tags take the format:

    $baseid[_rX[cY]]

Where X is the row number and Y is the column number. So this setting:

    useid => 'results'

Would yield XHTML like:

    <table id="results">
      <tr id="results_r1">
        <th id="results_r1c1">n1</th>
        <th id="results_r1c2">n2</th>
        <th id="results_r1c3">n3</th>
        <th id="results_r1c4">n4</th>
      </tr>
      <tr id="results_r2">
        <td id="results_r2c1">1</td>
        <td id="results_r2c2">2</td>
        <td id="results_r2c3">3</td>
        <td id="results_r2c4">4</td>
      </tr>
    </table>

Notice that the table gets the baseid verbatim.

vertical => 1 | 0

If you set this to 1, then it fundamentally changes the way in which data is expanded. Instead of walking the data structure and building rows horizontally, each element of data will become a column. This option is described more below under render().

body => {opt => val, opt => val}
font => {opt => val, opt => val}
table => {opt => val, opt => val}
td => {opt => val, opt => val}
th => {opt => val, opt => val}
tr => {opt => val, opt => val}

These options can be used to set attributes to be used on the applicable tag. For example, if you wanted the table width to be 95% and the border to be 1, you would say:

    my $qt = HTML::QuickTable->new(table => {width => '95%', border => 1});

Of course, you can specify as many different options as you want:

    my $qt = HTML::QuickTable->new(table => {width => '95%', border => 1},
                                   td    => {class => 'td_el'},
                                   font  => {face => 'arial,helvetica'} );

As an alternative form, you can also use:

body_opt => val
font_opt => val
table_opt => val
td_opt => val
th_opt => val
tr_opt => val

Instead of having to specify a hashref, you can use this option form to specify HTML tags. For example, if you want to set the font face, either of these will do the exact same thing:

    my $qt = HTML::QuickTable->new(font => {face => 'verdana'});
    my $qt = HTML::QuickTable->new(font_face => 'verdana');

Again, you can specify any HTML tag you want and it will get included. Anything after the underscore is taken as the tag name and placed into the output HTML verbatim.

render(\@data | \%data | $object)

The render() function can accept either an arrayref, hashref, or object. It then recursively expands the data per the options you specified to new(). Each data structure is rendered differently:

arrayref (\@array)

An arrayref should expand intuitively; each row in the array becomes another row in the table. If you specify the labels option, then the first row is taken as the column labels and is placed within <th> elements.

object ($object)

An object also expands quite simply. First, the object's param() method is called to get a list of keys. Then, for each key the value is placed in the array. The key is taken as the label for that column, and is placed within a <th>. As an example, you can dump a nice table of your CGI query with:

    use CGI;
    use HTML::QuickTable;

    my $cgi = CGI->new;
    my $qt  = HTML::QuickTable->new(header => 1);

    print $qt->render($cgi); 
hashref (\%hash)

A hashref is first sorted by key. Then, each data element becomes a data element for that column. For example:

    %user = (
        'nwiger'  => ['Nathan Wiger', 'nate@wiger.org'],
        'jbobson' => ['Jim Bobson', 'jim@bobson.com']
    );

    print $qt->render(\%user);

Would be rendered as:

    <table>
    <tr><td>jbobson</td><td>Jim Bobson</td><td>jim@bobson.com</td></tr>
    <tr><td>nwiger</td><td>Nathan Wiger</td><td>nate@wiger.org</td></tr>
    </table>

Note that it's very similar to the way arrays are handled. The benefit here is that this allows you to expand arbitrary data structures.

If it's a hashref of hashrefs, for example:

    %user = (
        'nwiger'  => { name => 'Nathan Wiger', email => 'nate@wiger.org' },
        'jbobson' => { name => 'Jim Bobson', email => 'jim@bobson.com'}
    );

    print $qt->render(\%user);

Then some Major Magic (tm) happens and you'll get something like this:

    <table>
    <tr><th></th><th>email</th><th>name</th></tr>
    <tr><td>jbobson</td><td>jim@bobson.com</td><td>Jim Bobson</td></tr>
    <tr><td>nwiger</td><td>nate@wiger.org</td><td>Nathan Wiger</td></tr>
    </table>

Notice that the keys were sorted alphabetically and output in order. But, note that the top-level key is not labeled in the <th>. To change this, you must specify the keylabel option to new():

    my $qt = HTML::QuickTable->new(keylabel => 'user');
    # ...
    print $qt->render(\%user);

That would create the same HTML as above, except the first column label would be "user".

NOTES

The 'B' option to 'labels' is currently broken, due to the fact that render() recursively calls itself and thus loses track of where it is. But who the heck puts labels at the bottom of an HTML table??

If you run into a bug, please DO NOT submit it via rt.cpan.org - that just causes me alot of extra work. Email me at the below address, and include the version string your eyes are about to pass over.

SEE ALSO

HTML::Table, Data::Table, SQL::Abstract, CGI::FormBuilder

VERSION

$Id: QuickTable.pm,v 1.12 2005/05/10 21:10:52 nwiger Exp $

AUTHOR

Copyright (c) 2001-2005 Nathan Wiger <nate@wiger.org>. All Rights Reserved.

This module is free software; you may copy this under the terms of the GNU General Public License, or the Artistic License, copies of which should have accompanied your Perl kit.