Michael J. Mathews


Markup::Perl - turn your CGI inside-out


  # don't write this...
  print "Content-type: text/html;\n\n";
  print "<html>\n<body>\n";
  print "<p>\nYour \"lucky number\" is\n";
  print "<i>", int rand 10, "</i>\n</p>\n";
  print "</body>\n</html>\n";
  # write this instead...
  use Markup::Perl;
  Your "lucky number" is
  <i><perl> print int rand 10 </perl></i>


For some problems, particularly in the presentation layer, thinking of the solution as a webpage that can run perl is more natural than thinking of it as a perl script that can print a webpage.

It's been done before, but this module is simple. The source code is compact: one file and less than 2k of code. Simply put: if you can do it in Perl, you can do it in Markup::Perl, only without all the print statements, heredocs and quotation marks.



It's a perl script when it starts. But as soon as the following line is encountered the rules all change.

  use Markup::Perl;

Every line after that follows this new rule: Anything inside <perl>...</perl> tags will be executed as perl. Anything not inside <perl>...</perl> tags will be printed as is.

So this...

  use Markup::Perl;
    Today's date is <perl> print scalar(localtime) </perl>

Is functionally equivalent to...

  print "<body>\n";
  print "Today's date is ";
  print scalar(localtime), "\n";
  print "</body>";

If you bear that in mind, you can see that this is also possible...

  use Markup::Perl;
    <perl> for (1..3) { </perl>
    <perl> } </perl>

Naturally, anything you can do in an ordinary perl script you can also do inside <perl></perl> tags. Use your favourite CPAN modules, define your own, whatever.


If you would like to have a some shared Markup::Perl code in a separate file, simply "include" it like so...

  use Markup::Perl;
    Today's date is <perl>src('inc/dateview.pml')</perl>

The included file can have the same mixture of literal text and <perl> tags allowed in the first file, and can even include other Markup::Perl files using its own src() calls. Lexical my variables defined in src files are independent of and inaccessible to code in the original file. Package variables are accessible across src files by using the variable's full package name.

Not all output happens in a stream-like way, but rather there is an attempt to be slightly intelligent by reordering certain things, such as printing of HTTP headers (including cookies). Thus you can use the header() call anywhere in your code, even conditionally, but the actual header, if you do print it, will always be at the very start of your document.



Adds the given name/value pair to the HTTP header. This can be called from anywhere in your Markup::Perl document.


Equivalent to CGI::param. Returns the GET or POST value with the given name.

Given a single string argument, returns the value of any cookie by that name, otherwise sets a cookie with the following values from @_: (name, value, expires, path, domain, secure).


Transforms the content of the given file to allow mixed literal text and executable <perl>...</perl> code, and evals that content.


For the sake of speed and simplicity, I've left some areas of the code less than bullet-proof. However, if you simply avoid the following bullets, this won't be a problem:

starting out

Keep the use Markup::Perl line simple. Its presence signals the beginning of new Markup::Perl syntax. The use line should be on a single line by itself.

tags that aren't tags

The parser is brutally simple. It just looks for <perl> and </perl> tags, regardless of whether or not you meant them to be treated as tags or not. For example printing a literal </perl> tag requires special treatment. You must write it in such a way that it doesn't look like </perl>. This is the same as printing a "</script>" tag from within a JavaScript block.

  print '<'.'/perl>';
including yourself

It is possible to include and run Markup::Perl code from other files using the src function. This will lead to a recursive loop if a file included in such a way also includes a file which then includes itself. This is the same as using the Perl do 'file.pl' function in such a way, and it's left to the programmer to avoid doing this.

use utf8

I've made every effort to write code that is UTF-8 friendly. So much so that you are likely to experience more problems for not using UTF-8. Saving your documents as UTF-8 (no BOM) is recommended; other settings may or may not work. Files included via the src function are always assumed to be UTF-8.


The author does not claim copyright on any part of this code; unless otherwise licensed, code in this work should be considered Public Domain.


Michael Mathews <micmath@gmail.com>, inspired by !WAHa.06x36 <paracelsus@gmail.com>.