taggedTextMast.php

<?php # THIS FILE MUST HAVE MAC LINE ENDINGS! That means \r aka carriage return
###############################################################################
# $Id: taggedTextMast.php 2322 2015-01-12 03:52:52Z quentin $
# This file will output a Tagged Text version of the masthead. The data MUST 
# be encoded in UTF-16 Big Endian with Carriage Returns for line endings for 
# everything to work. That means that some things may look really funky 
# programatically, but are needed due to the constraints of Tagged Text. Also, 
# this will case PHP to return the wrong line number for errors. Good thing I 
# don't have any :-)
#
# Note that PHP will slurp up a newline that immediately follows a closing php 
# tag. For more information, visit these (most likely outdated) links:
#
# http://www.adobe.com/go/learn_id_taggedtext
# http://help.adobe.com/en_US/InDesign/6.0/indesign_cs4_taggedtext.pdf
#
# Also, you could view information about the Daily Confusion which Ricky wrote 
# after reading the aformentioned documentation. Hopefully, I've brought that 
# knowledge over here appropriately.
###############################################################################

require_once dirname(__FILE__).'/common.php';

### Functions

# Takes position title and array of members, returns formatted text
function formatPosition($title, $members) {
  # Replace the middle space in <CAPITAL><PERIOD><SPACE><CAPITAL><PERIOD<SPACE>
  #   with a THIN SPACE (U+2009, &#8201).
  # This is for cases like "B. D. Colen" -> "B.<THIN SPACE>D. Colen",
  # "Bruno B. F. Faviero" -> "Bruno B.<THIN SPACE>F. Faviero", and
  # "Jonathon E. D. Richmond" -> "Jonathan E.<THIN SPACE>D. Richmond PhD"
  $members =  preg_replace(
     '/([A-Z]\.) ([A-Z]\. )/',
     '$1' . mb_convert_encoding('&#8201;', 'UTF-8', 'HTML-ENTITIES') . '$2',
     $members);

  $seperator = ", ";

  # Production positioning hack: replace "Associate " with "Assoc-iateN",
  # where the hyphen is discretionary (aka soft) and the N is a nonbreaking
  # space. This reduces the liklihood that a line will break as:
  # _DEPARTMENT_
  # Editor: .... Associate
  # Editor: ...
  # and cause confusion or bad copyfitting as a result.
  $title = preg_replace(
    '/Associate /',
    'Assoc' . mb_convert_encoding('&shy;', 'UTF-8', 'HTML-ENTITIES') .
    'iate'  . mb_convert_encoding('&nbsp;', 'UTF-8', 'HTML-ENTITIES'),
    $title);
  if ($title == '') return join($seperator, $members); // For AdBoard
  if (sizeof($members) > 1 and !preg_match("/Staff$/", $title)) {
    $title = $title . (substr($title, -1) == "s" ? "es" : "s");
  }
  return "<cTypeface:Bold>$title:<cTypeface:> " . join($seperator, $members);
}

# Takes department name and array of positions to array of members and returns 
# formatted text
function formatDepartment($name, $positions) {
  // Editors at Larger and AdBoard don't modify their dept names
  if(!($name == "Editors at Large" 
    or $name == "Advisory Board" 
    or $name == "Production Staff for This Issue")) {
    $name = $name . " Staff";
  }
  $folded = array();
  foreach ($positions as $title => $members) {
    array_push($folded, formatPosition($title, $members));
  }
  $folded = join('; ', $folded);
  return "<ParaStyle:PROD-Misc\\:MAST\\:PROD-MastDept>$name\r" .
    "<ParaStyle:PROD-Misc\\:MAST\\:PROD-MastPeople>$folded.";
}

$depts = getDepartments();
// Exec board gets special formatting
$exec = getDepartmentMembers(array_shift($depts));

header('Content-Type: text/plain; charset="UTF-16BE"');
header('Content-Disposition: attachment; filename=masthead.txt');
// XXX: Due to a bug in PHP - kinda http://bugs.php.net/34776 (although some 
// may also argue PHP itself is the bug) - the BOM must be output before we do 
// the rest of the conversion. Now, this will also cause headers to be sent, 
// which will mean that when we flush output at the bottom it'll try to send 
// headers again and emit a warning.
echo("\xfe\xff"); // BOM for proper UTF-16BE detection
mb_internal_encoding("UTF-8");
mb_http_output("UTF-16BE");
ob_start("mb_output_handler");
?>
<UNICODE-MAC>
<?php
foreach ($exec as $position => $members) {
  // Exec has Shift-returns (a.k.a just a newline) between titles and names
  $person = "\n".$members[0];
?>
<ParaStyle:PROD-Misc\:MAST\:PROD-MastTop><cTypeface:Bold><?=$position?><cTypeface:><?=$person?>

<?php
}
// Now deal with the rest of the departments
foreach ($depts as $dept) {
  $members = getDepartmentMembers($dept);
  if(sizeof($members) == 0) continue;
  echo formatDepartment($dept, $members) . "\r";
}

// Don't forget about Production staff for the issue!
echo formatDepartment('Production Staff for This Issue',
  array("Editor" => array("@@", "@@"),
        "Staff"  => array("@@"),
        "Copy Editors" => array("@@"))) . "\r";

// Time for the copyright with the current year
$currentyear=date("Y");
?>
<ParaStyle:PROD-Misc\:MAST\:PROD-MastBottom><cTypeface:Italic>The Tech<cTypeface:> (ISSN 0148-9607) is published on Thursdays during the academic year (except during MIT vacations) and monthly during the summer by The Tech, Room W20-483, 84 Massachusetts Avenue, Cambridge, Mass. 02139. Subscriptions are $50.00 per year (third class). <cTypeface:Bold><cCase:All Caps>Postmaster:<cCase:><cTypeface:> Please send all address changes to our mailing address: The Tech, P.O. Box 397029, Cambridge, Mass. 02139-7029. <cTypeface:Bold><cCase:All Caps>Telephone:<cCase:><cTypeface:> Editorial: (617) 253-1541. Business: (617) 258-8324. Facsimile: (617) 258-8226. <cTypeface:Italic>Advertising, subscription, and typesetting rates available.<cTypeface:> Entire contents <cTypeface:Bold Italic>© <?=$currentyear?> The Tech<cTypeface:>. <cTypeface:Italic>Printed by Turley Publications, Inc.<cTypeface:><?php
// Suppress the warning about headers. (See XXX at BOM above)
@ob_end_flush();