→ Digital humanities

Maintained by: David J. Birnbaum (djbpitt@gmail.com) [Creative Commons BY-NC-SA 3.0 Unported License] Last modified: 2021-12-27T22:03:49+0000

JavaScript overview

Why JavaScript?

JavaScript has become a general-purpose programming language used in a number of contexts, supporting a range of initiatives in software development. In this course, we use JavaScript code on the web, known as client-side JavaScript, to create Dynamic HTML, or DHTML. The difference between plain ol’ HTML and HTML embedded with JavaScript (DHTML) is that the latter is interactive, which is to say that it enables the user to interact with and change the page in the browser.

Client and server

When a web user types a URL into the browser address bar and a website loads, the user’s computer is operating as a client, and the website the user sees is a server (because it is serving web pages to the client). Right now, as you read this page, your computer is a client that downloaded this webpage from the Obdurodon server. How servers know what information to send and where to send it is not our concern in this course. What you should know, however, is that the client downloads information from the server, usually via some sort of web browser. The browser knows how to present the information it downloads because it comes in a standard format, such as HTML or XML.

How JavaScript works with HTML

When the client (your personal computer, for example) downloads a web page, it may also download a separate CSS document automatically, without asking for it explicitly, and that CSS stylesheet will be used to style the HTML or XML in the browser. Whether a separate CSS stylesheet is delivered alongside a web page depends, as you know, on whether the author of the web page has incorporated a link in the HTML <head> that points to a separate stylesheet. JavaScript operates similarly; the developer of a web page can include a link to a separate JavaScript file, which will then be delivered automatically along with the web page when the user requests the latter. This is called client-side JavaScript because the JavaScript code is run locally, that is, it is executed and processed entirely by the client computer, and does not have to return to the server for more information. This is the preferred way to let the user modify the browser display; the slowest part of loading and styling a web page is typically sending a request to the server and waiting for the information one needs to be returned, so that process is avoided when possible by using JavaScript. If all of the desired information can already be accessed by the client, modifying the rendering client-side is much faster than sending a new request to the server.

Web document basics

JavaScript, like all of the languages we have been working with in this class, understands how to deal with hierarchies. All of our XML and HTML documents have been in hierarchical formats, and JavaScript is just as capable as XPath in being able to traverse and manipulate document hierarchies, and to interact with elements according to their properties. There are some differences, of course, but conceptually, we can use the same types of operations we have been using in all of our X-languages in JavaScript.

The main difference between client-side JavaScript’s understanding of document hierarchies and our XML-based understanding is that the main entry point to a hierarchy in JavaScript is no longer the document node (which we address by beginning a path expression with a slash [/] in XPath), but what is called a window object. The reason for this is that JavaScript is capable of working with the entire browser window, and not just the XML or HTML document it is rendering. The window object represents a frame or window within your web browser, and can be referred to with the keyword window.

Think of window as a reference to the very top of your document, much like / resets the context of our XPath expressions (that is, starts a new path at the very top of the document hierarchy). Note that object is a common term within computer science, and generally refers to a named variable with any number of values or properties. We can think of XML elements as objects, because not only can they have a text value, but they can also have a number of properties; for any XML element, its properties can be said to include both its attributes and its relation to the document: any ancestor, sibling, or child elements, that is, all of the axes we have become familiar with. To learn JavaScript, we must adopt this new nomenclature of objects and properties.

The window object has a property called document, which is a Document object that can be referred to with the keyword document (yes, objects can have properties that are also objects). A Document object represents a Node that is the start of an HTML or XML document tree. In the case of the window, its document property contains all of the Element objects that compose our website, in either HTML or XML form. We will use the document keyword frequently to access various elements in our web document. To put this in plain language: a window in your browser has a document (a Document object that is its document property), and the document has elements (the Element objects it contains).

Associating JavaScript actions with HTML

Most of the projects in this class will use JavaScript to change the appearance of a web page in a user’s browser in response to user interactions, so we will focus on how to accomplish that type of task. User interactions with a document like click, mouseover, or button press are called events, and we can tell the browser to respond to specific events by creating a JavaScript event handler. An event handler has a target, which is the object in the document where the event occurs. A target might be a button the user can click or a menu tab that will open up into a list when moused over. Elements in the document can have event handler properties; for example, a menu tab in the document might have the property of being associated with an event handler that knows what to do when a user mouses over the tab. The components of an event handler, then, are the target (the element in the document where the event occurs), the event (what has to happen to the target to cause the event handler to fire), and the event handler code, which describes what happens when the event occurs at the target.

All event handlers have names that are the concatenation of the prefix on and the name of some event, like click or mouseover (that is, onclick or onmouseover). We can assign these properties to a specific action that we want to occur when the event happens, that is, on the occurrence of the event. There are three practical ways of doing this, all of which have analogs in CSS.

1. Assign a value to the target element’s attribute whose name corresponds to the desired event handler

Description
Event handlers can be referenced explicitly in HTML by using attributes that correspond to their names. For example, the onclick event handler can be deployed for a particular element by assigning a value to the @onclick attribute of that element.
Example
<p onclick="stuff();"> will cause a JavaScript function called stuff() (which we write ourselves) to fire whenever the user clicks on that particular paragraph.
CSS analog
<p style="color: red;"> uses inline CSS to color a paragraph red.

2. Write code to assign the event handlers in a separate document and link to it from the <head> of the HTML document

Description
Event handlers can be configured in an external JavaScript document. In this case nothing has to be written into attributes within the HTML; both the code that fires when events occur and the information about which events on which elements cause which code to run all resides in the external file.
Example
<script type="text/javascript" src="stuff.js">//</script> causes the JavaScript code in an external file called stuff.js to load and run. To enhance compatibility with older browsers, it is common to write separate start and end tags with two slashes between them, instead of a regular XML empty element.
CSS analog
<link rel="stylesheet" type="text/css" href="style.css"/> causes an external CSS stylesheet to be associated with the HTML document. The stylesheet contains the information that associates styles with specific elements, just as an external JavaScript file may contain the information to associate event handlers with their targets, that is, with the elements in the HTML document where the events occur.

3. Write code to assign the event handlers inside a <script> element in the <head> of the HTML document

Description
As with #2, above, nothing has to be written into attributes on the target elements within the HTML. Instead, the handlers are assigned within the JavaScript script. The difference between #2 and #3 is that in #2 the <script> element links to an external file that contains JavaScript code, but it does not contain any JavaScript code itself. With #3 there is no external file, and the raw JavaScript code is written between the start and end <script> tags.
Example
<script type="text/javascript"> (raw JavaScript code goes here) </script>
CSS analog
This is analogous to writing raw CSS rules inside a <style> element in the <head> of the HTML document.

Recall that the preferred way to cause CSS styling to be applied to HTML is to create an external CSS document and link to it from the HTML. It is possible to write CSS rules directly inside the HTML, either inside a <style> element in the <head> (analogously to our #3, above) or in @style attributes on individual elements (analogously to our #1, above), but for a variety of reasons that is rarely the best approach. Similarly, the preferred way to associate JavaScript with HTML is to use the second of the methods above, that is, to write all of the JavaScript event handlers outside the HTML, in a separate JavaScript file, and associate that file with the HTML. Nonetheless, because the first approach is easier to understand and to code and the third avoids the step of managing an additional document, we may use them for some assignments in this course. You can combine these three methods within in a single document, just as you can combine the different CSS methods.

External JavaScript scripts are, by convention, stored with a .js file extension (just as CSS files by convention end in .css). You can create and edit JavaScript in <oXygen/>.

How JavaScript does its work

How do we describe actions in JavaScript? That is, how to do we tell JavaScript what to do when an event occurs on a target? Functions in JavaScript, as in XPath, perform actions using a set of input parameters. A simple function in JavaScript called alert() takes a single string as its input, and alerts the user by creating a tiny popup window with the string as a message inside it. For example, if we wanted a click event for some target to alert the message, Catz r kewl!, we would assign the event handler like this: onclick="alert('Catz r kewl!')", which we have done for this <span> element. We’ve used strategy #1, inline JavaScript attached to an @onclick attribute, to make this happen; you can see our underlying code by viewing the source of this page.

The special load event

The window has its own load event, which is triggered when the content of the document inside it is finished loading and is ready to be presented in the browser. This lets it function as a bootstrapping event; because it fires automatically whenever the document loads, it can be used to initialize other JavaScript functions. In conformity with the general convention, the event handler for the load event is called onload. We can assign this event handler to a specific action that we would like to be performed upon the triggering of the load event.

Event handlers normally listen on the element to which they’re attached. Recall that the target of the event handler is the element where the event occurs. For example, the onclick event listener in the example above listens for a click on the <span> that carries the @onclick attribute, and when the user clicks on that <span>, it activates the alert() function. Clicking elsewhere doesn’t trigger the alert; the event handler is listening for clicks only on the <span> with which it’s associated.

That’s fine for elements like <span>, but the window isn’t an element and doesn’t have attributes, so if we want an event to fire when the window loads, we need a different strategy. The way to associate an onload event handler with the window is to use dot notation; writing window.onload="alert('Catz r kewl!');" would mean fire this event (pop up the alert) when the window onload event occurs (that is, when the window finishes loading the document).

Okay, time to write your first JavaScript

Try creating an HTML document, with a paragraph (<p>) containing some text, and a span (<span>) containing part of the text within that paragraph. Give the span an @onclick attribute, the value of which is an alert (alert()) that has a specific message associated with it (inside quotation marks between the parentheses). In Real Life you’ll want to style the span, so that the text is visibly different from the rest of the paragraph and your users will know that it’s clickable. You can do that now, or, for this exercise, you can just remember which portion of the text you assigned an event handler to.

After you’ve tested that, modify the HTML document by including a <script> element within the <head> that contains this line of code: window.onload="alert('Insert message here');". Note that the XHTML rules require that you include a @type attribute on your script element with a value text/javascript.

We ended that single line of code with a semicolon. The rules for JavaScript, as for CSS, are that you can include multiple statements in a set of rules, and each must end with a semicolon. The semicolon is optional on the last (or only) statement in both CSS and JavaScript, but we strongly advise including it, so that you won’t forget about it should you go back and add another line. When you’re done, refresh your page, and you should get an alert with the message you wrote in your script before your paragraph’s content is even drawn to the screen.

The resulting HTML document should look something like this:

<!DOCTYPE html SYSTEM "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
    <head>
        <title>Your first JavaScript script</title>
        <script type="text/javascript">
            window.onload = alert('This message came from loading the window!');
        </script>
    </head>
    <body>
        <p>This is a sample paragraph of text meant to <span
                onclick="alert('This message came from clicking the span!');"
                style="color:blue; text-decoration:underline; cursor:pointer">exemplify the
                circumstance</span> of possibly having an HTML paragraph, perchance.</p>
    </body>
</html>

What next?

The use of JavaScript to modify CSS in memory, changing the appearance of a page that has already been loaded in the browser, is called DHTML (dynamic HTML). You can see it in action at http://zatochnik.obdurodon.org. Clicking on numbers in the SVG at the top or on individual paragraphs in either of the columns on the sides causes paragraphs to appear and disappear (toggling the CSS display property) and change color and other font properties (toggling the CSS color and font-decoration properties). Those events also change the color and opacity of components of the SVG. Some tasks you might want to consider for your projects might include: