note 88374 deleted from function.date by tularis

Note Submitter: uramihsayibok, gmail, com ---- In a refreshing change, this code snippet is *not* about calculating date differences or anything like that. <background-story> I've always preferred date() over strftime() because of what each offers. For example, date has an st/nd/rd for the day number (S) while strftime does not. It also has an unpadded day number (j) that strftime doesn't (%e pads it with a space). On the other hand, each character in the format string for date() is translated unless you prepend a backslash... Recently I've wanted to create links with date information in it. Like with "January 21st 2009" each part would be a link (January => /2009/01, 21st => /2009/01/21, 2009 => /2009). date() makes this difficult because the HTML markup gets the treatment as well as the Y/m/d characters. Escaping all of those is ugly, not to mention annoying. </background-story> Here's my combination of date's placeholders and strftime's % markers. It's the best solution I could think of (at 5am mind you) but I'm open to suggestions. Lots of comments because the code is rather tricky. <?php /** * Combines placeholders from date() with the % marker from strftime() * * Like strftime, use %% for a literal %. * * @see date, strftime * @param string $format The format of the outputted date string * @param int $timestamp An integer Unix timestamp that defaults to the current local time * @return A formatted date string */ function strfdate($format, $timestamp = null) { // look for tokens if (preg_match_all('/(?<!%)(%%)*%(.)/', $format, $matches)) { // passing false or null as the timestamp doesn't work so we // have to generate the default ourselves if ($timestamp === null) $timestamp = time(); // run each token through date - all at once // combines them into a "a!b!c!d" list, runs it through date, // and splits it apart again $parts = explode("!", date(implode("!", $matches[2]), $timestamp)); // (! should never show up in date() output so this works) // a second function is used sequentially: // the Xth time this function is called it will replace the Xth token // with the corresponding element in $parts (which is passed through as $a) // // @param array $a Meant to be $parts as defined above // @return $a[X] where X is how many times this function has been called $replace = create_function('$a', 'static $i = 0; return $a[$i++];'); // find each token and get the replacement data from the $replace function // note how the text being replaced isn't used anywhere: we already know // what it is as it was used to create the $parts array $result = preg_replace('/(%*)%./e', '"$1" . $replace($parts)', $format); // finally "unescape" any %s $result = str_replace("%%", "%", $result); return $result; } return $format; } ?>