Test Cases

This kind of string should match our pattern:

• a string of six (6) and only six (6) digits
• a string of ten (10) and only ten (10) digits

All other strings should not match our pattern.

The Regular Expression Pattern

^\d{6}(?:\d{4})?$

The explanation

1. ^: matches the beginning of the string
2. \d{6} matches exactly six digits.
3. (?:\d{4})? this non-capturing (non-capturing because of ?:) group matches exactly four digits. This group is optional.
4. $: matches the end of the string

So, this regular expression could be read as:
Match a string of exactly six (6) digits which could be optionally followed by a string of exactly four (4) digits.
In conclusion, this regex will match a six (6) digits string or a ten (10) digits string (6 required digits + 4 optional digits).

At least once in his career, a SharePoint developer is going to be faced with a security problem.

When a user executes an action in SharePoint, i.e. modifies an item in a list, the request on the server executes using the security context of the user. So, if the user does not have the right privileges to accomplish specific actions, the request could fail with a security exception.

A way to overcome this behaviour is to use the RunWithElevatedPrivileges method from the SPUtility class.

An important thing to keep in mind when using this method is the initial context of the object on which you will execute specific actions.

In the example below, even if you use the RunWithElevatedPrivileges method, the security context of the SPWeb object (web variable) is inherited from the user permissions.
This behaviour is normal because the SPSite site variable is a reference to the properties variable which was created with the user’s security context.

public override void ItemUpdated(SPItemEventProperties properties)
{
	SPSecurity.RunWithElevatedPrivileges(delegate(){
		using (SPSite site = properties.ListItem.Web.Site)
		{
			using (SPWeb web = site.RootWeb)
			{
				web.AllowUnsafeUpdates = true;
                                // Do some actions
			}
		}
	});
}

To run in a context with full control, all objects should be instantiated inside the RunWithElevatedPrivileges delegate method.
By doing this, objects will inherit their permission from the RunWithElevatedPrivileges context which is full control:

public override void ItemUpdated(SPItemEventProperties properties)
{
	SPSecurity.RunWithElevatedPrivileges(delegate(){
		using (SPSite site = new SPSite(properties.ListItem.Web.Site.ID))
		{
			using (SPWeb web = site.RootWeb)
			{
				web.AllowUnsafeUpdates = true;
                                // Do some actions
			}
		}
	});
}

The semantic of the <abbr> HTML tag

W3C HTML specification defines ABBR tag as

ABBR:
Indicates an abbreviated form (e.g., WWW, HTTP, URI, Mass., etc.).

Example:

1.5 <abbr title="Liter">l.</abbr>

This example will render (move your mouse over the string l.) as: 1.5 l.

How Jaws can correctly read <abbr> title?

By default, Jaws does not read the title of the abbr tag but only the abbreviation itself. I think it is a nonsense because this option should be turned on by default.

Activating this option is quite simple. Just follow the steps below:

1. Open Jaws
2. Open the Configuration Manager in the Utilities menu
3. Choose HTML Options in the Set Options menu
4. Check the Expand Abbreviations box
5. Et voilà!

Here is the exception I encountered:

Unexpected end of file while parsing Name has occurred. Line 1, position 708. System.Xml.XmlException: Unexpected end of file while parsing Name has occurred. Line 1, position 708.[...]

After a lot of searches on the Internet, I’d found a very helpful post on the InfoPathDev forums.

To summarise, if the server on which you’re developing is running with Windows SharePoint Services SP1 (Service Pack 1), your InfoPath form must not have a useless secondary Data Connection (declared but not used at all on your form).

Very simple… if you know it!

So, my co-worker wanted to know how to retrieve all digits (and only digits: 0 to 9) following a non-digit character and preceding a pipe “|” character ?

Test Cases

This kind of string should match our pattern:

• test123|another456|more789| (should match three strings: “123″, “456″ and “789″)
• justone1|

All other strings should not match our pattern.

The Regular Expression Pattern

(?<!\d)(\d+)(?=\|)

The explanation

1. (?<!\d): negative lookbehind expression tells to the regex engine that the character preceding our matching pattern must not be a digit. Because I’m using a lookaround which is a “zero-width assertion”, this first expression will not be a part of the matching string
2. (\d+) matches one or more digit character. This group is the only capturing group (our matching string) of the regex.
3. (?=|): positive lookahead expression tells to the regex engine that the character following our matching pattern must not be a pipe “|”. Because I’m using a lookaround which is a “zero-width assertion”, this third expression will not be  a part of the matching string

So, this regular expression could be read as:
Match a string of one or more digit character which is preceded by any character except a digit and which is followed by a pipe “|”.

Test Cases

These strings should not match our pattern:

• /
• an empty string

All other strings should match our pattern.

The Regular Expression Pattern

^/.+|[^/].+$

The explanation

1. ^ matches the beginning of the string
2. /.+ matches a string beginning by a “/” and following by at least one character
3. | OR operator
4. [^/].+ matches a string beginning by any character except “/” and following by at least one character
5. $ matches the end of the string

So, this regular expression could be read as:
Match a string beginning by a “/” and following by at least one character or match a string beginning by any character except “/” and following by at least one character.

There are lots of articles and resources on the Internet which talk about the theory of the accessibility matter. However, there are not many technical articles which explain how the standards are really implemented or how a web developer has to implement them.

This first article focuses on how to give the… focus on a non-focusable HTML element with jQuery.

It extends (and I hope, completes) the original article written by C. Blouch and published on the Donald F. Evans’s website.

The theory

Giving the focus to a non-focusable HTML element seems to be quite simple. Many articles on the Web were written about it.

To summarize them, you only have to :

1. Set the tabindex attribute of the selected element to -1
2. Give the focus to the DOM node with the JavaScript method .focus()

Quite simple, isn’t it?

In fact, after doing few tests with Jaws, it demonstrates that it does not really work that way…

The reality

By default, only a few HTML elements are focusable like anchors or form inputs.
So, in order to give the focus to a non-focusable HTML element correctly, don’t forget to add an important third step:

1. Set the tabindex attribute of the selected element to -1
2. Give the focus to the DOM node with the JavaScript method focus()
3. Use the setTimeout JavaScript function to delay the execution of the focus() function

Without the use of the setTimeout function, Jaws will simply ignore the focus and will not redirect the user to the element where you want to give the focus.

The JavaScript code

$(document).ready(function () {

	$("#addNMoveFocus").click(function(e){

		AddNMoveFocus("ul#nl", "<li><b>A simple b tag inside a li tag</b></li>", "ul li:last b");
		e.preventDefault();

	});

});

function AddNMoveFocus (o1, content, o2){
	AddContent(o1, content);
	MoveFocus(o2);
}

function AddContent(o, content) {
	$(o).append(content);
}

var f; // must be declared as global
function MoveFocus(o){
	$(o).attr("tabindex", "-1");
	f = $(o).get(0);
	setTimeout("f.focus();", 0);
}

The examples

Examples listed below were tested with IE7, Jaws 7.10, Jaws 8, Jaws 9 and Jaws 10

• View the working example (with the use of setTimeout function).
• View the not working example (without the use of setTimeout function).