A week or so ago I came across (from RSS feeds? nah, I think it was in twitter? maybe a squirrel stepped o my keyboard?) Steph Gray’s Reading: keeping on top of stuff I save, some code he rolled. The way he describes the need rang a resonant bell with me:

Like a lot of people, I use services like Google Reader, Twitter, Delicious and Instapaper to help me find and store interesting links to articles, tools, apps or whatever. Personally, when I don’t have time to read it right now, I tend to star an item in Google Reader, ‘favourite’ it in Twitter, or mark it to ‘Read later’ in Instapaper – often five or ten things a day. I also save interesting stuff to Delicious, particularly where I think it may have longer term usefulness.

But I hardly ever actually go back and read those things.

Mainly it’s because it’s a bit of a faff finding the links, and because the time I’d get to do it (on the train, hanging around waiting for a toddler to wake up, etc) I’ve generally only got an iPhone to hand.

Hence Reading. This is a little sliver of PHP and jQuery…

Steph’s nifty script consists of a single PHP script plus the SimplePie RSS parser and a JQuery library to make the interface accordion work. My setup is at http://cogdogblog.com/reading:

If you open any of the “accordion” items, you get a full version of the site:

It’s something you cold implement without much code or PHP chops needed. In fact, all you need to edit is the line that lists the RSS feeds you want to use.

What it does is let you build your own list from all the RSS feeds for these sorts of things- your tags in delicious have a feed, your Google Reader shared (and starred) items have a feed, Instapaper has a feed- all you have to do is find them, and add them to the part of the code where you enter them in an array.

Delicious provides its RSS feed at the foot of most pages- you can do your entire stream, or create a special tag you want to use to designate items for this list. Instaper provides the feed somewhere in the account settings. You can get your Google Reader starred items via the Reader Settings – Folders and Tags (set the starred items to public, and you get a link for the RSS). Your shared reader items are set up in the Sharing Setup area.

Twitter does have a feed for your favorites, though it does not provide the link directly for you- I found the answer via twitterati, it is as simple as:

http://twitter.com/favorites/[insert_your_ID_here].rss

so my feed is http://twitter.com/favorites/cogdog.rss

What is also neat, or maybe neater, is that Step’s code has a mobile friendly version at the same URL- in this one, you get the same list, but the accordion opens to just a link to the site:

I have been doing a bit more of sharing in Google reader, especially when I am on the mobile version, and this is sure a better way of organizing then emailing links to myself (the same for twitter). I’ve not used Instapaper in eons, I keep forgetting about it.

Thanks Steph for sharing the code- again, you can get your own code from Steph’s site, and I see the plans for the next version are to make it so you can download all the items to a Google Spreadsheet.

What are you waiting for? Give it a try.

Well, let’s back up. If you are running online events where people might come from anywhere, there can always be challenges for getting the right time to people. For our NMC events, while our office is in Austin, we have staff in 3 other time zones, and people in our community from maybe another 14 or more zones on top of that.

If I say something starts at 2:00pm Central Time, then it puts it on everyone else to figure out when that is (and I’ve seen event sites where they do not even provide the reference time zone). If were all proper and versatile, we’d use Greenwich Mean Time, since everyone knows their offset, right? (I am -7 what are you?)

Most of the time , our event times are referenced to Pacific Time, since that is where our server is hosted, but even if we provide Pacific, Central, and Eastern Time, what about folks in Spain? New Zealand? Hong Kong?

This came to me a lot when our events picked up in Second Life back in 2006-2007. They used Pacific Time too as a reference time (actually for a long time they called it SLT for Second Life Time, and people still say that).

I started early using the World Time Clock fixed time. It allows you to enter a date time, and reference time zone, and generates a link that produces a big chart that converts that to other times around the world.

it’s silly to not think of the world as your audience, and most of the world is on a different time zone, like those weird half hour ones in South Australia.

For example, if there was an event, like tonight my dinner, at 6:00 PM, I’d enter the date and time in the form, and scroll all the way down to the bottom to select my time zone (U.S.A. – Arizona – Phoenix), and the form generates for me this URL

http://www.timeanddate.com/worldclock/fixedtime.html?month=10&day=23&year=2010&hour=18&min=0&sec=0&p1=197

which I can use to generate a URL that lets anyone else in the world know when I am having dinner, in their local time zone:

And in our event web pages I might say:

hey, tonight’s event is Alan’s Dinner! It is taking place October 22 at 6:00 PM PST (check for local time)

I was pretty adamant about getting our folks to add this “check for local time” link. If you inspect the URL, you can actually make new ones just by adjusting the values in the URL string (The site does provide a URL shortener if you dont want to display a monster long URL).

But here is the thing that struck me today- for our MIDEA site, I am using the WP custom post types for our events, so I already have built into the date, a value for the start date/time (it is the unix time version of the date/time, number of seconds since 1970). Since I already have the data, I thought it would be easy to write a function that auto generates the World Time Clock Link.

It was easy.

function get_worldtimeclock ($ts, $tz=197) {
	// generates a URL for a world time clock link for a given timestamp $ts
	// default timezine (for this site) is 197 or Phoenix, hey thats my time zone

	$when = getdate($ts);

	return 'http://www.timeanddate.com/worldclock/fixedtime.html?month=' . $when['mon']  . '&day=' . $when['mday']  . '&year=' . $when['year']  . '&hour=' . $when['hours']  . '&min=' . $when['minutes']  . '&sec=0&p1=' . $tz;
}

and you can see it in action in the top gray bars on pages like http://midea.nmc.org/event/horizonreportmuseumwebinar/.

It’s really bone simple PHP, just like to write it up. I set up a code demo page so you can see it in action:

Actually, it was just fun to dip my paws into the code. Even if it is bone head simple code.

As a first example, I am cleaning up an older WordPress site I use for logging my running/training; in the past, I kept a spreadsheet on my desktop for keeping a run log and then manually transferred the totals/averages/graphs to my web site by pasting into some text files (they are embedded with a PHP include). It worked, but it did have that tedious manual smell for something that should be more automated.

It seemed to make sense to transfer the spreadsheet to a Google Doc. I’m not going to detail all the bits in there, but essentially, for every run, I enter a distance and time on the appropriate date line, and the sheet calculates total/averages per week and on a final summary page.

This is just the data; I have another sheet that is just the table for the weekly totals of miles and time spent running; from here I can simply use Google’s built-in Publish Chart which provides a URL to embed this graphic:

This is subtle but powerful– this is an image, but not an image- as the data changes, so will the graphic. So I know longer have to manually make a graphic and upload it to my web site; I can simply use the HTML

<img src="http://spreadsheets.google.com/pub?key=tMIPDwydKpay-5Oq0_G769w&oid=4&output=image" width="100%"/>

The other piece uses Google’s ability to publish a worksheet as CSV- this is data you can then parse on your web site. So I have a sheet that simply does averages/totals for my running log:

To access this data, I click the Share button and select Publish As Web Page

The options in the top make it so the published data is made current whenever I request it; in the bottom I select the option to have it published as CSV (comma separated value), and I select just the spreadsheet I need.

Thus, this URL always gives me the data on the spreadsheet in a form I can use in my code…. this bit takes some PHP savvy to manage, but I use the same logic almost everywhere.

I use this on the sidebar of my blog to display an up to date listing of my totals. In my WordPress sidebar, I use a statement to bring in the code from an external file named totals.php that sits inside my theme directory:

<?php include (TEMPLATEPATH . "/totals.php"); ?>

I could have simply inserted the code into my sidebar template, but going modular like this makes it easier to separate code from format…

The basic logic is using the php file command to read int a remote file, which puts the contents into an array; where each array item is one line of data.

// get the CSV data as an array from the remote URL
$lines = file('http://spreadsheets.google.com/pub?key=tMIPDwydKpay-5Oq0_G769w&single=true&gid=0&output=csv');

// get rid of header row
$headers = array_shift($lines);

// Loop through data- therer is only one line hear
foreach ($lines as $line) {
	$ldata =  explode(',', trim($line)); // split row to its own array of elements

	if ($ldata[0] == '') break; // an empty line means we are done, so exit the foreach loop

      // now we can just output the information as an HTML list, referencing the appropriate array items
       echo '<li>Total Miles Biking  <strong>' . $ldata[0] . '</strong></li><li>Total Miles Running  <strong>' . $ldata[1] . '</strong></li><li>Total Hours Running  <strong>' . $ldata[2] . '</strong></li><li>Ave Pace Running  <strong>' . $ldata[3] . '</strong></li>';
}

And what is cool is there does not seem to be much of an impact of hitting Google each time. It must be ‘a caching. Good stuff.

This is pretty simple example. I am running some more complex examples on the NMC Virtual Worlds site where the reason to do this was to have a larger amount of tabular data be displayed in a more readable fashion on our web site– and to make it so someone other than me can be in charge of keeping the data up to date bu simply editing a Google doc (note that a spreadsheet used for this purpose must be public viewable, so don’t mix in any company secrets in there).

So for example, our Clients listing is coming from a google spreadsheet, and the parsing tests the existence of certain items to format the output– e.g. if the organization URL column has a value, we echo the name as a hyperlink; if we have the coordinates of their land in Second Life, we add a teleport link. The code that generates this page also sorts it on output.

This one technically is not hitting the spreadsheet every time…. I set this up so a unix cron (a timed script) calls every hour a PHP script which generates the formatted content as a text file on the web server, so the web page actually just reads it in via an include statement… this is a simple form of caching.

Another page has actually two chunks of data that are generated from different spreadsheets. Out current land availability page lists plots of Second Life land available from NMC for rental.

The main body content is a list of land with the output formatted around data in a spreadsheet. And the current land prices on the right side come form another sheet- this way we can make adjustments to the web page at any time simply by editing a spreadsheet, and never touch the web page- again it works like this:

1. Every hour a cron script on the server calls a PHP script.
2. The PHP script reads in CSV data, sucks it into an array (one line as an item), marches through items, and parses each item as an array representing the cells, and outputs the formatted as HTML to a small text file.
3. The WordPress page simply uses an include statement to display the content (I use the Exec-PHP plugin to be able to run statements in a page).

I am thinking more and more how I can use this process….

Of course, this is all baby stuff compared to the masterful data wrangling by the Jedi Master Tony Hirst….

The main navigation of most WordPress sites is driven by cycling through all of the content that are WordPress pages– which is nice, but sometimes you have content that you don;t want cluttering the navbar. Or maybe you want to insert something that is not a Page into the navbar. I’ve been amazed that so few people seem to know how to subvert it, and below is a little trick I use when I want to wedge some other kind of WordPress content into the navbar by Pages process.

Mostly this is in your HEAD… I mean the header.php file of your templates that generates it as an Unordered List:

<div id="nav">
<ul>
  <li class="<?php if (is_home()) {echo 'page_item current_page_item'; } ?>">
    <a href="<?php echo get_settings('home'); ?>/">Home</a></li>
  <?php wp_list_pages('title_li' ); ?>
</ul>
</div>

The first li item creates the Home link, and in thes case applies some CSS logic if the home page is in view. The rest of it happens in the PHP code statement that uses the WordPress fucntion to return a list of all Pages and putting each one in an

<?php wp_list_pages('exclude=5,11&title_li' ); ?>

<?php wp_list_pages('include=3,5,12,23,34&title_li' ); ?>

<?php wp_list_pages('include=3,5,12,23,34
         &title_li&sort_column=menu_order' ) ?>

<div id="nav">
<ul>
  <li class="<?php if (is_home()) {echo 'page_item current_page_item'; } ?>">
    <a href="<?php echo get_settings('home'); ?>/">Home</a></li>
  <li><a href="http://cogdogblog.com/">Good Stuff</a></li>
  <?php wp_list_pages('title_li' ); ?>
  <li><a href="http://flickr.com/photos/cogdog">Photos</a></li>
</ul>
</div>

RedirectPermanent /tips http://www.mycoolstuff.org/my-cool-tips

<gcse>url|width|height|title</gsce>

include("extensions/googleCSE.php");