* for the record, my steampunk goggles were bought off of Etsy.

Update 4: Plugin now hosted at wordpress.org.

Update 3: New version with support for viewport constants, and iPod Touch.

Update 2: James Craig points out that the ID attribute is not allowed in the META element, so I’ve updated the plugin.

Update: I should add what my aim is. Rather than creating a separate template for mobile, I want to have a theme that ‘does the right thing’ for iPhone and desktop browsers.

Some documentation:

1. Plugin tests user agent string.
   • If it contains ‘iPhone’ then it adds <meta content="width=480; maximum-scale=0.6667" name="viewport"/> to the document’s head.
   • It also adds iPhone specific CSS rules. For that reason, you should call wp_head() in your theme file after you declare your default CSS files.
   • These rules were written with Sandbox in mind. You will want to edit the CSS rules to suit your theme.
   • You can force the viewport and CSS rules by adding forceiphone=y to the query string portion of a URL.
2. Adds a boolean function: weh_is_iPhone().
   • Returns true if user agent string contains ‘iPhone’.
   • You can call this function from your theme to add or turn off generation of markup.
   • Most likely you’ll use it to turn off generation of sidebars.

Round trip.

ETA: there’s an updated version that runs on PHP 4.4.x.

ETA 2: And it works, too. I just tried it with Tim Bray’s APE.

And congratulations to the Torres’ on the occasion of the birth of their third child (Elias should confer with Tim and Lauren on the logistics of open source development while taking care of an infant.)

In order to use that with PHP 4, apply Niklas’ patch.

Finally, if you don’t want to syndicate your feed’s content as escaped HTML, apply my patch that combines my changes and Niklas’ to James’ code. However, you’re now responsible for making sure your entries are valid XHTML.

Note: after applying the patch, you’ll need to modify the format string passed to the_time to account for your offset from GMT. There’s probably another way to do this.

Photo credit: Scott Beale (Laughing Squid)

I showed my work to Bill Lazar, who told me that Scott Allan Wallick’s new theme, Sandbox, available for WordPress.com blogs, has hAtom baked in.

Well, that makes my work, Strangelove, v0.1, superfluous.

You can use Chris Casciano’s script for Net News Wire to read a Sandbox themed blog as Atom, or pipe it through Luke Arno’s hatom2atom proxy.

If you’re comfortable with editing PHP, open themes/sandbox/header.php, and add the following to the head section:

<link rel="alternate" type="application/atom+xml" title="<?php bloginfo('name'); ?> Atom Feed" href="http://www.lukearno.com/projects/hatom2atom/?ctype=application%2Fxml&tidy=yes&url=<?php echo urlencode(get_bloginfo('url')); ?>" />

This will use Luke Arno’s proxy to generate an Atom 1.0 feed by parsing the hAtom format in the page. Strangelove does that out of the box.

If you’re running PHP 5 with libxslt support, you can run Chris’ filter script instead of the remote proxy.

At the µformats session, where you see me doing my Nikita Khrushchev imitation, I talked about this. Thanks to Tantek for inviting Slashchick and me up to talk about our hacks.

Slashchick modified the Countdown plugin to publish events in the hCalendar format, so that you can extract them from a blog page into your calendar.

Additional Tags: 2006, wordcamp, microformats

It’s best to design your URLs without respect to how they will actually be implemented. Design them so they make sense to human clients and search engines. Then worry later about how you’ll actually implement the backend that serves representations of those URLs.

Sadly server support for such schemes is very lacking in 2006. Doing it on top of Apache requires serious mod_rewrite voodoo. (That’s how WordPress creates the very nice URLs you see on this site.) PHP doesn’t help out here since it’s still very tied to a one-URL, one .phtml file model.

Apache Cocoon and Popoon’s sitemap model does a lot of what ERH’s looking for. But those are front controller patterns that push the complexity of routing from the server and into your application. [ via Todd ]

I found Aidan Lister’s Duration.php. It defines a static method to convert a timestamp into a readable string.

// 2 hours, 20 minutes, 5 seconds
$time = 60*60*2 + 20*60 + 5;

// Convert to a string
echo Duration::toString($time);
echo "\n";

Useful. Thanks Aidan.

It turned out that this was Mark de Visser, Zend’s chief of marketing. He said he hadn’t read the post because he had spent the last two days at MashupCamp.

My take on PHP:

1. Gawd, it’s easy to jump in, and get running code.

2. Ugh, need native Unicode, badly.

   Don’t feel bad, Ruby has the same problem.

   PHP 6 is supposed to support UTF-8, and comes out later this year (says de Visser.)

3. PHP 5 has libxml2 and libxslt.

4. It’s still easy to do dumb things.

   Own up people, how many of you have written:

   
   $string = file_get_contents('http://example.com/restfulweb/thing?id=0001');
   if ($string != '') { // do stuff }
   

   when you were in a hurry? Then didn’t replace it with cURL or PEAR HTTP::Request?*

[*] Now, it turns out that you can check HTTP return codes when you open a URL using wrappers: $http_response_header is set after you call file, and you can look in the first element to get the HTTP response code.

I didn’t know that until I read a Yahoo! Developer Network article and looked at the example code.

So if you’re making unauthenticated GETs, that’s most of what you need.

So why doesn’t http://www.php.net/file_get_contents mention this? This useful nugget of info is buried in http://www.php.net/manual/wrappers.http.php.

WordPress advertises the URL /moreLikeThis/feed/atom/, however, that returns a 404 error.

More mod_rewrite hacking has fixed this.