Category: Technology

Google Maps API V3 geocoding with SharePoint 2010

This explains how to show addresses in a SharePoint 2010 list on a Google Map.

This article is based on Kyle Shaeffer’s Plotting Your SharePoint 2010 List Data on a Google Map. The main difference is I upgraded his method to Google Maps API V3.

Here’s how it’s done:

Create a list

Using SharePoint Designer 2010 (free download!), open your SharePoint site and:

  1. Under Site Objects, click Lists and Libraries.
  2. In the Lists and Libraries tab, click Custom List.
  3.  Enter MyData and press OK.
  4. Click on your new list, then select Edit list columns in the Customization section.
  5. Add a Multi Lines of Text field using the Add New Column button at top left:. Name it Address.

Create data view

Still in SharePoint Designer 2010:

  1. Click Site Pages under Site Objects on the left:
  2.  In the ribbon at top, select Page > ASPX. Name your new page map.aspx.
  3. Click on the new page, then click Edit file in the Customization section. If asked to open the page in advanced mode, click Yes.
  4. Make sure that Design or Split are selected at bottom. From the menu at top: Insert > Data View > MyData

Now you’ll see your data in a table in the page.

Add the JavaScripts

You’ll need to reference two script libraries: Google’s Maps API and jQuery.

Find the opening <form> element in the source code. Right after it, paste this code:

<script type="text/javascript" src="https://maps.googleapis.com/maps/api/js?key=API_KEY_GOES_HERE&sensor=false"></script>
<script type="text/javascript" src="https://ajax.googleapis.com/ajax/libs/jquery/1.7.1/jquery.min.js"></script>
 
<div id="map_canvas" style="width: 98%; height: 400px"></div>
 
<script type="text/javascript">
// set up the basic map object
var latlng = new google.maps.LatLng(34.603885, -4.626009);
var myOptions = {
	zoom: 2,
	center: latlng,
	mapTypeId: google.maps.MapTypeId.ROADMAP
};
var map = new google.maps.Map(document.getElementById("map_canvas"), myOptions);
 
// this is used for geocoding
var geocoder = new google.maps.Geocoder();
 
function codeAddress(address, theName) {
    geocoder.geocode( { 'address': address}, function(results, status) {
      if (status == google.maps.GeocoderStatus.OK) {
        var marker = new google.maps.Marker({
            map: map,
            position: results[0].geometry.location
        });
      } else {
        alert("Geocode was not successful for the following reason: " + status);
      }
    });
}
 
// this is called with body's onload to pull addresses after the page is done loading.
function initialize() {
	$("tr.ms-itmhover").each(function(i){
		codeAddress($(this).find('.ms-rtestate-field:eq(0)').text(), $(this).find('.ms-vb-title:eq(0) a').text());
	});
}
</script>

Replace API_KEY_GOES_HERE with your own Google API key. Don’t have one? Get it at https://code.google.com/apis/console/. Be sure to get a Google Maps API v3 key! v2 is deprecated.

Now you need to add an onload attribute to your body tag to fire off the initialize function:

<body onload="initialize()">

Viola! You have a map that dynamically pulls from the below list!

Caveats

The initialize method works by scraping from the rendered HTML. It’s finding every tr with class ms-itmhover–which is what is produced by the data view you inserted above. Then within each of those trs, it finds:

  • The first item with class ms-rtestate-field and gets its text contents. This will be your address.
  • The contents of the a tag within the first element with class ms-vb-title, which is the title field.

You’ll have to do JavaScript surgery if you change fields in a way that alters this. For example, if another multi line text field displays before your address, you’ll have to change .ms-rtestate-field:eq(0) to .ms-rtestate-field:eq(1). Or you’ll need to alter the XSLT behind the data view to produce some other classes or code.

Also, because this uses JavaScript to scrape the screen’s contents, you must deliver the data list in the page . It doesn’t have to be visible–you can make it invisible with CSS. If you don’t want the data to be on the page, you’ll need to use more sophisticated techniques, possibly some kind of AJAX.

Hiding Active Directory user IDs from WordPress author slugs

I recently set up a corporate WordPress blog system. With the Active Directory Integration plugin, users can sign in with their corporate ID and password.

But here’s a problem: each blog post has a link to the author’s profile. That profile’s URL includes the user ID. The corporation’s security standards say we can’t expose user IDs to the world, so the author profile URLs have to be sanitized.

It took a while to figure out a solution, but the end result is reasonable.

I found this post at StackExchange’s WordPress site. Adding the first two code snippets (below) to the end of wp-config.php tells WordPress to use the user’s nickname metadata to construct the profile URLs:

add_filter( 'request', 'wpse5742_request' );function wpse5742_request( $query_vars )
 
{
 if ( array_key_exists( 'author_name', $query_vars ) ) {
 global $wpdb;
 $author_id = $wpdb-&gt;get_var( $wpdb-&gt;prepare( "SELECT user_id FROM {$wpdb-&gt;usermeta} WHERE meta_key='nickname' AND meta_value = %s", $query_vars['author_name'] ) );
 if ( $author_id ) {
 $query_vars['author'] = $author_id;
 unset( $query_vars['author_name'] );
 }
 }
 return $query_vars;
}
 
add_filter( 'author_link', 'wpse5742_author_link', 10, 3 );
function wpse5742_author_link( $link, $author_id, $author_nicename)
{
 $author_nickname = get_user_meta( $author_id, 'nickname', true );
 if ( $author_nickname ) {
 $link = str_replace( $author_nicename, $author_nickname, $link );
 }
 return $link;
}

But wait, there’s more!

Now you have to get a proper value into the nickname field. Active Directory Integration makes this easy. In this plugin’s settings, go to the User Meta tab and enter this into the Additional User Attributes field: mailnickname:string:nickname. You’ll may need to replace mailnickname with your own Active Directory user attribute if it isn’t appropriate for you.

That’s it. The next time a user logs in, the nickname field is updated, and all future profile URLs for that user will not have a user ID.

Educause: Outsource the Transactional, Keep the Transformative

Educause’s recent Outsource the Transactional, Keep the Transformative complements my recent article questioning the value of IT projects.

In my article, I said that work  easily expressed as a classical/waterfall project probably has less intrinsic business value. This includes stuff like routine infrastructure and commodity services.

The Educause piece has a graph showing how Pepperdine University rated the value of its in-house IT services:

(Click the image to see the full size.)

Sampling of services with lowest value or transformative potential, mostly the stuff of process or where classical/waterfall projects are normative:

  • Windows administration
  • Security
  • DBAs
  • Help Desk (outsourcing this is the theme of the Educause article)

And services with the highest value or transformative potential, the stuff of agility:

  • EIS functional (basically business analysts)
  • Portal
  • Technology & Learning Team
  • University Planning (basically enterprise architecture)

I chuckled at security’s inclusion in the least value segment. Security has always lived in a halo of insecurity; their value is in what didn’t happen. How do you express that? Regardless, you’d be crazy not to have a good security team.

How about the rest? Does this mean Windows administrators, DBAs, help desk staff, etc. should quake in their boots?

Depends. In the near term, I don’t see a sea change, but I also don’t see growing opportunities for work on the bottom left of the graph. Long-term, the outsourcing question is “when”, not “if”. And until then, the value of work on the bottom left isn’t intrinsic; it’s measured by what it enables the people on the top right to do. If the bottom lefters aren’t helping the top righters, they are nailing their own coffin.

Once the outsourcing begins, I’d hope that it results in either:

  1. Equivalent position at the outsourcer. I saw a university do this when it outsourced its trades department (HVAC, plumbing, landscaping, etc.). If you enjoy the trade, this can be good; you’ll probably have more growth opportunities versus pigeonholing with an employer where your trade is ephemeral. This is already happening to pure-play programmers, too, but that’s a subject for a different day.
  2. You get reassigned within the original employer. This is what Pepperdine did (see bottom of page 1 of this). But–and this is a big but!–it’s only going to work if you’re versatile and able. If you’re a one shot wonder, have little skills depth, or find it difficult to adapt, this could be the beginning of a downward spiral.

So what to make of this? Two things.

First, if your job is routine, process-driven, or involves a lot of waterfall projects, you may not be high on the business value ladder. These positions will be scrutinized. Make sure you are versatile enough to merit and survive reassignment.

Second, watch what’s going on around you. Where are you, your immediate coworkers, and your department headed? Is the cliched “writing on the wall”? Sometimes, you need to be proactive and transform yourself up the business value ladder.

Transformation, agility, and business value are the future of IT. Process and waterfall’s share of that future is declining. Be prepared.

Do projects matter for IT?

Visit http://ericbrown.com/do-projects-matter-for-it.htm for a recent article I wrote on projects in IT.

Two emerging trends:

  1. Projects don’t well-explain IT’s value to the business, and it’s getting worse.
  2. Work that is best organized as a classical (waterfall) project may not be valuable to your business. If you can’t express the project as an agile project, you’re probably doing something either quite complex or really vanilla. Either way, it’s not where IT is going.

Provoked? Read the article!