Using BIRT to Print Using DOM

There is an issue which I have in developing with Dojo's and related libraries: Without true HTML source printing a page is not functional. The standard JavaScript print method produces print-screen functionality not print-page. The page is created from the DOM not HTML. So, here is a thought: What if we used the DOM-XML to push information to the BIRT Engine to print a report of the page or pages in question? I have been perusing this thought and wanted to start with an existing code base rather than attempting to create all which would be needed on my own. I have found the following link which may be helpful from an existing code basis and a sound starting point: <a class='bbc_url' href='http://www.mozilla.org/newlayout/doc/layout-2006-12-14/master.xhtml'>Mozilla's Layout Engine</a><br />
<br />
Am I missing something simple? Is there an easy way to simply print the full page when starting with the DOM and very little true HTML?<br />
<br />
Any comments or help would be greatly appreciated. THANKS ALL

Find more posts tagged with

Comments

JasonW

Can you give some more detail on what you are trying to do? Is this a print of a BIRT report? Is the PDF print not acceptable?

Jason

dewittm

No this is not printing a normal BIRT report. I am developing a web based application using a package called Wavemaker (WM) for the GUI. WM is heavily Dojo, Dojox and AJAX based. The actual html look like the following:<br />
<br />
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><br />
<html xmlns="http://www.w3.org/1999/xhtml"><br />
<head><br />
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><br />
<meta http-equiv="cache-control" content="must-revalidate" /><br />
<meta name="generator" content="WaveMaker Visual Ajax Studio" /><br />
<title>Project</title><br />
<link rel="shortcut icon" href="/wavemaker/lib/boot/images/favicon.ico" /><br />
<style type="text/css"><br />
@import "/wavemaker/lib/boot/boot.css";<br />
@import "app.css";<br />
<br />
/* styling for the wavemaker application element */<br />
#wavemakerNode {<br />
height: 100%;<br />
overflow: hidden;<br />
position: relative;<br />
}<br />
</style><br />
<br />
<br />
<script type="text/javascript" src="config.js"></script><br />
<script type="text/javascript" src="/wavemaker/lib/boot/boot.js"></script><br />
<br />
</head><br />
<body class="tundra"><br />
<br />
<div id="wavemakerNode"><br />
<div id="_wm_loading" style="z-index: 100;"><br />
<table><tr><td>/wavemaker/lib/boot/images/loader.gif Loading...</td></tr></table><br />
</div>
</div>
<script type="text/javascript" src="/wavemaker/lib/runtimeLoader.js"></script><br />
<br />
<script type="text/javascript" src="lib_project.js"></script><br />
<script type="text/javascript" src="types.js"></script><br />
<script type="text/javascript" src="Project.js"></script><br />
<script type="text/javascript"><br />
new Project({domNode: "wavemakerNode"}).run();<br />
</script><br />
<br />
</body><br />
</html><br />
<br />
Where the actual work is hidden in the javascript and DOM. The affect is that if one calls the standard js print method, only the viewable screen is printed not the page as desired. One can zoom out to see the complete screen; however, on a lengthy screen the font size is then too small to read.<br />
The point to the original post was to ask if perhaps with the XML from the DOM and the use of XSLT, maybe, and the BIRT engine, maybe, one could create an input for the BIRT engine which would resolve the DOM print issue. By printing a BIRT "report" of the page which was displayed. This would alleviate the need to manage a DOM generated screen and an associated form (only for printing) with the same content which overtime would likely become unwieldy for a large project such as the one I am working on. Any help would be greatly appreciated as I am quickly approaching a deadline and have not solution for this issue to date.

JasonW

You may be able to write a report that uses BIRT XML data source to parse and print what you need. The question on this approach is how much different would the xml be for every print page. Obviously you could always use BIRT's DEAPI to create the reports on the fly. So you would write a servlet that uses the DE API and RE API. The servlet would take the XML parse it and use the DE API to create tables etc and then call the RE API to run it.

dewittm

Okay. DE-API & RE-API answer the high level question of interfacing with BIRT; however, do you have any ideas on parsing through a DOM to define the input XML for DE-API? Realizing this is not exactly what you are addressing in the forum, but this is truly where I am lost. I ask this not just for the assistance to me but there is a large number of developers out here using the DOJO & DOJOX approach which need this type of solution for printing pages. I believe it would be very beneficial to Actuate to be the solution for this and positions BIRT as a defacto tool for the project. At that point one would need a really strong argument to walk away from BIRT or BIRT Enterprise where it answers other questions, such as, reporting and plotting. Just my thoughts and sales pitch.

JasonW

Are you talking about parsing the dom looking for objects like tables?

dewittm

Tables are part of the equation; however, what I am attempting to explain is I want to print whatever is on the page. In the world of DOJO's below is all the available html for the attached page. With DOJO's if you call the "Print" method you get the screen print not the page print.<br />
<br />
The attached page, as one can see by the right hand slide bar" is only part of the page, due to the users resolution. If one attempts to zoom out far enough to get the page on the screen it becomes too tiny to read.<br />
<br />
I am looking, perhaps fruitlessly, to use BIRT as a way to overcome this issue. The DOM is the only place in the application that the information on the page exists in total. BIRT creates using XML. The DOM is XML. The thought was perhaps there is a way to feed the DOM to the BIRT engine where a report could construct the page for the purpose of printing it.<br />
<br />
AVAILABLE HTML FROM THE ATTACHED PAGE:<br />
===============================================<br />
<br />
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><br />
<html xmlns="http://www.w3.org/1999/xhtml"><br />
<head><br />
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><br />
<meta http-equiv="cache-control" content="must-revalidate" /><br />
<meta name="generator" content="WaveMaker Visual Ajax Studio" /><br />
<title>SANG</title><br />
<link rel="shortcut icon" href="lib/boot/images/favicon.ico" /><br />
<style type="text/css"><br />
@import "lib/boot/boot.css";<br />
@import "app.css";<br />
<br />
/* styling for the wavemaker application element */<br />
#wavemakerNode {<br />
height: 100%;<br />
overflow: hidden;<br />
position: relative;<br />
}<br />
</style><br />
<br />
<br />
<script type="text/javascript" src="config.js"></script><br />
<script type="text/javascript" src="lib/boot/boot.js"></script><br />
<br />
</head><br />
<body class="tundra"><br />
<br />
<div id="wavemakerNode"><br />
<div id="_wm_loading" style="z-index: 100;"><br />
<table><tr><td>lib/boot/images/loader.gif Loading...</td></tr></table><br />
</div>
</div>
<script type="text/javascript" src="lib/runtimeLoader.js"></script><br />
<br />
<script type="text/javascript" src="lib_project.js"></script><br />
<script type="text/javascript" src="types.js"></script><br />
<script type="text/javascript" src="rtlgenerator.js"></script><br />
<script type="text/javascript" src="SANG.js"></script><br />
<script type="text/javascript"><br />
new SANG({domNode: "wavemakerNode"}).run();<br />
</script><br />
<br />
<br />
<script type="text/javascript" src="localization.js"></script><br />
<script type="text/javascript" src="constants.js"></script><br />
<br />
</body><br />
</html><br />
======================================<br />
<br />
See attachment for the DOM's version of this rendered image.

JasonW

BIRT can use xml as a datasource, which can be set on the fly or using a scripted dataset any xml parser can be called to parse the xml. The issue is that columns have to be known before hand. A different approach would be to store the dom in some type of session variable and in the beforeFactory script of the birt report you could access that variable and then parse it and create the report items on the fly.