=== modified file 'import_typepad.module'
--- old/import_typepad.module	2009-03-27 17:34:20 +0000
+++ new/import_typepad.module	2009-03-30 15:16:15 +0000
@@ -153,16 +153,66 @@
       '#options' => $users);
   }
 
-  $import_form["taxonomy_mappings"] = array(
+  // Skip taxonomy fieldset if nothing to show
+  if ($cats) {
+    $import_form["taxonomy_mappings"] = array(
+      "#type" => "fieldset",
+      "#title" => t("Taxonomy mappings")
+    );
+
+    foreach ($cats as $cat=>$count){
+      $item = _import_typepad_taxonomy_list($cat);
+      $item['#title'] = "$cat ($count)";
+      $import_form["taxonomy_mappings"]['taxonomy_'.str_replace(' ','_',$cat)] = $item;
+    }
+  }
+
+  $import_form["tidy_content"] = array(
     "#type" => "fieldset",
-    "#title" => t("Taxonomy mappings"));
-
-  foreach ($cats as $cat=>$count){
-    $item = _import_typepad_taxonomy_list($cat);
-    $item['#title'] = "$cat ($count)";
-    $import_form["taxonomy_mappings"]['taxonomy_'.str_replace(' ','_',$cat)] = $item;
-  }
-
+    "#title" => t("Content Clean & Update"),
+  );
+  $import_form["tidy_content"]["import_images"] = array (
+    "#type" => "checkbox",
+    "#title" => "Copy images included in Blog posts to Drupal",
+    "#default_value" => false,
+    '#description' => t('If the IMCE module is installed, images will be copied to the IMCE image folder set for your role, otherwise to the Drupal <em>files</em> directory. Links are switched to the Drupal copy of the image if the copy succeeded.'),
+  );
+  // Check that php.ini setting
+  if (!ini_get('allow_url_fopen')) {
+    $import_form["tidy_content"]["import_images"]['#description'] = t('Image importing requires <em>allow_url_fopen</em> enabled in <em>php.ini</em>. Disabled until that\'s corrected.');
+    $import_form["tidy_content"]["import_images"]['#disabled'] = true;
+  }
+
+  $import_form["tidy_content"]["rm_ms_garbage"] = array (
+    "#type" => "checkbox",
+    "#title" => "Remove garbage metadata added if any posts were pasted from Microsoft Office",
+    "#default_value" => false,
+  );
+  $import_form["tidy_content"]["rm_empty_paragraphs"] = array (
+    "#type" => "checkbox",
+    "#title" => "Remove empty tag",
+    "#default_value" => false,
+    '#description' => t('Primarly for removing empty &lt;p> tags, but will also remove any empty tag and repeat &lt;br> tags.'),
+  );
+  $import_form["tidy_content"]['update_teasers'] = array (
+    "#type" => "checkbox",
+    "#title" => "Recreate teasers based on Drupal settings and cleaned up imported body",
+    "#default_value" => false,
+    '#description' => t('Only uses cleaned up version of body if appropriate checkboxes above are checked.'),
+  );
+  // Possible filter format user can use
+  $options[0] = 'Default';
+  foreach(filter_formats() as $format) {
+    $options[$format->format] = $format->name;
+  }
+
+  $import_form["tidy_content"]["applied_input_format"] = array (
+    "#type" => "select",
+    "#title" => "Set input format for imported nodes",
+    "#options" => $options,
+    '#description' => t('Note that the input format selections is based on your user account. If you are importing the blogs for another author, set in <em>Author mappings</em> above, they might not have permission to the same input formats. If you set the input format to one the author doesn\'t have permission for, they will <strong>not</strong> be able to edit the imported blog posts.'),
+  );
+  
   $import_form["import_warning"] = array (
     "#type" => "markup",
     "#value" => t('Importing may take a while, do not click \'Import\' more than once. To see progress, look at the <a href="?=admin/content/node">administer content</a> page in a new window.<br/><br/>'));
@@ -348,7 +398,7 @@
               foreach ($blogCats as $id=>$val){
                 $c[] = $taxonomy['taxonomy_'.str_replace(' ','_',$val)];
               }
-              _import_typepad_save($currentBlog, $c);
+              _import_typepad_save($currentBlog, $c, $authors['import_images'], $authors['applied_input_format'], $authors['rm_ms_garbage'], $authors['rm_empty_paragraphs'], $authors['update_teasers']);
               $success++;
       	    } else if ($previewCount < $preview){
     	      $output .= node_view($currentBlog);
@@ -421,7 +471,7 @@
       foreach ($blogCats as $id=>$val){
         $c[] = $taxonomy['taxonomy_'.str_replace(' ','_',$val)];
       }
-      _import_typepad_save($currentBlog, $c);
+      _import_typepad_save($currentBlog, $c, $authors['import_images'], $authors['applied_input_format'], $authors['rm_ms_garbage'], $authors['rm_empty_paragraphs'], $authors['update_teasers']);
       $success++;
     } else if ($previewCount < $preview){
       $output .= node_view($currentBlog);
@@ -446,9 +496,103 @@
 }
 
 /**
+ * Parse for image urls & copy to Drupal
+ */
+function _import_typepad_import_images($body) {
+  if (module_exists('imce')) {
+    $conf = imce_settings_user();
+    $conf['allowedExt'] = $GLOBALS['imce_ext']; 
+  } else {
+    $conf['dir'] = file_directory_path();
+    $conf['allowedExt'] = array('.jpg', '.gif', '.png');
+  }
+
+  // finds any <a><img></a> sets, copies both the href & src; only works for absolute
+  // paths, either starting with http or /; this prevents any bad images from causing
+  // an post import failure
+  preg_match_all("%href=\"((http:|/)[^\"]+)[^>]+><img.*?src=\"((http:|/)[^\"]+)%", $hay, $matches);
+  // If you don't want the absolute path requirement uncomment this line & comment out the previous one
+  //preg_match_all("%href=\"([^\"]+)[^>]+><img.*?src=\"([^\"]+)%", $hay, $matches);
+
+  // Combine href & src files into one array
+  $combined = array_merge($matches[1], $matches[2]);
+
+  // finds all <img> regardless of link tag to ensure any unlinkded images are also imported, copies
+  // src only; only works for absolute paths, either starting with http or /; this prevents any bad
+  // images from causing an post import failure
+  preg_match_all("%<img.*?src=\"((http:|/)[^\"]+)%", $body, $matches);
+  // If you don't want the absolute path requirement uncomment this line & comment out the previous one
+  //preg_match_all("|<img.*?src=\"([^\"]+)|", $body, $matches);
+  
+  $copied = 0;
+  // combine then skip dups 
+  $combined = array_merge($combined, $matches[1]);
+  foreach(array_unique($combined) as $url) {
+    $file_name = substr($url, strrpos($url, '/') + 1);
+    if ($dot = strrpos($file_name, '.')) {
+      $ext = substr($file_name, $dot);
+      $file_name = substr($file_name, 0, $dot);
+      if (!in_array($ext, $conf['allowedExt'])) {
+        drupal_set_message(t('Invalid file extension - %ext.', array('%ext' => $ext)), 'error');
+        return FALSE;
+      }
+    } else {
+      // Really just a guess; proper way to handle it would be to copy file to temp directory, check
+      // file type of temp file to pick intelligent extension, then move to permenant location
+      $ext = '.jpg';
+    }
+
+    //clear filename and copy file
+    $cleared = preg_replace("/[^\w\-\_]/", '_', substr($file_name, 0)) . $ext;
+    $newpath = file_create_filename($cleared, $_SERVER['DOCUMENT_ROOT'] . base_path() . $conf['dir']);
+    // Need doc root here, but not in array below
+    if (!@copy($url, $newpath)) {
+      drupal_set_message(t('Error copying %file.', array('%file' => $url)), 'error');
+    } else {
+      @chmod($newpath, 0664);
+      $copied++;
+      $source_files[] = $url;
+      // Strip the doc root so that URL references are absolute from wwwroot. Doc root path 
+      // was needed to copy & chmod above, as well as check for collisions in file names
+      $dest_files[] = substr($newpath, strlen($_SERVER['DOCUMENT_ROOT']));
+    }
+  }
+
+  $body = str_replace($source_files, $dest_files, $body);
+  drupal_set_message(t('%num images copied succesfully', array('%num' => $copied)));
+  return $body;
+}
+
+/**
+ * Clean up MS garbage & empty tags - repeat BR tags too, but there's a comment on
+ * how to disable that if you prefer
+ */
+function _import_typepad_rm_garbage($body, $rmGarbage, $rmEmptyP) {
+  $garbage = array();
+
+  if ($rmGarbage) {
+    $garbage[] = "|<meta.*?</style>|s";
+    $garbage[] = "| class=\"MsoNormal[^\"]{0,}\"|s";
+  }
+  if ($rmEmptyP) {
+    // This one would remove all tags containing just one encoded char, probably
+    // OK, but real culprit is nbsp
+    //$garbage[] = "%<[^/>]+>(&[a-z0-9]{2,4};|)</[^>]+>%";
+    $garbage[] = "%<[^/>]+>(&nbsp;|)</[^>]+>%";
+
+    // Comment out this next line if you don't want to include repeat <br> tags
+    $garbage[] = "%(<br[ ]?/>\s?){2,}%";
+  }
+  if ($garbage) {
+    $body = preg_replace($garbage, '', $body);
+  }
+  return $body;
+}
+
+/**
  * Save the content into the database, first the blogs, then the comments
  */
-function _import_typepad_save($currentBlog, $terms){
+function _import_typepad_save($currentBlog, $terms, $cpImages, $format, $rmGarbage, $rmEmptyP, $updateTeaser){
   if ($currentBlog != null){
   // Apply all the substitutions
   $subReplace = $_REQUEST['import_typepad_substitution_replace'];
@@ -460,6 +604,24 @@
       $currentBlog->body = str_replace($val, $subWith[$id], $currentBlog->body);
     }
   
+  if ($rmGarbage || $rmEmptyP) {
+    $currentBlog->body = _import_typepad_rm_garbage($currentBlog->body, $rmGarbage, $rmEmptyP);
+    $currentBlog->teaser = _import_typepad_rm_garbage($currentBlog->teaser, $rmGarbage, $rmEmptyP);
+  }
+
+  if ($cpImages) {
+    $currentBlog->body = _import_typepad_import_images($currentBlog->body);
+  }
+
+  // If specified, use format_format
+  if ($format > 0) {
+    $currentBlog->format = $format;
+  }
+
+  if ($drupalTeaser) {
+    $currentBlog->teaser = node_teaser($currentBlog->body, $currentBlog->format);
+  }
+
   // Save the entry
   node_save($currentBlog);