Font Awesome File Icons: Mapping MIME Types to correct icon classes

font-awesome-mime-type-icons.php

<?php

/**
 * Get font awesome file icon class for specific MIME Type
 * @see https://gist.github.com/guedressel/0daa170c0fde65ce5551
 *
 */
function ($mime_type) {
  // List of official MIME Types: http://www.iana.org/assignments/media-types/media-types.xhtml
  $icon_classes = array(
    // Media
    'image' => 'fa-file-image-o',
    'audio' => 'fa-file-audio-o',
    'video' => 'fa-file-video-o',
    // Documents
    'application/pdf' => 'fa-file-pdf-o',
    'application/msword' => 'fa-file-word-o',
    'application/vnd.ms-word' => 'fa-file-word-o',
    'application/vnd.oasis.opendocument.text' => 'fa-file-word-o',
    'application/vnd.openxmlformats-officedocument.wordprocessingml' => 'fa-file-word-o',
    'application/vnd.ms-excel' => 'fa-file-excel-o',
    'application/vnd.openxmlformats-officedocument.spreadsheetml' => 'fa-file-excel-o',
    'application/vnd.oasis.opendocument.spreadsheet' => 'fa-file-excel-o',
    'application/vnd.ms-powerpoint' => 'fa-file-powerpoint-o',
    'application/vnd.openxmlformats-officedocument.presentationml' => 'fa-file-powerpoint-o',
    'application/vnd.oasis.opendocument.presentation' => 'fa-file-powerpoint-o',
    'text/plain' => 'fa-file-text-o',
    'text/html' => 'fa-file-code-o',
    'application/json' => 'fa-file-code-o',
    // Archives
    'application/gzip' => 'fa-file-archive-o',
    'application/zip' => 'fa-file-archive-o',
  );
  foreach ($icon_classes as $text => $icon) {
    if (strpos($mime_type, $text) === 0) {
      return $icon;
    }
  }
  return 'fa-file-o';
}

Instead of all those weird loops when you have a dictionary-like structure it's more efficient to do sth like this:
return iconClasses[this.mimeType.split('/')[0]] || iconClasses[this.mimeType] || 'fa-file-o';

I agree that the loop is not very efficient, @7sedam7's solution is incomplete, since the loops are designed to find a prefix of the mimetype. For example, the actual prefix for a DOCX document is application/vnd.openxmlformats-officedocument.wordprocessingml.document. Given icon class only contain the prefix. An optimal implementation will probably require a variant of tries. Note that the problem here is to find the longest prefix of a given string, which is not the standard use case for tries. This will be overkill for such a small data set.

I propose the following TypeScript code to simplify the loop, however (using the Fontawesome React package) :

const iconClasses: { [key: string]: IconDefinition } = {
  // Media
  "image": faFileImage,
  "audio": faFileAudio,
  "video": faFileVideo,
  // Documents
  "application/pdf": faFilePdf,
  "application/msword": faFileWord,
  "application/vnd.ms-word": faFileWord,
  "application/vnd.oasis.opendocument.text": faFileWord,
  "application/vnd.openxmlformats-officedocument.wordprocessingml":
  faFileWord,
  "application/vnd.ms-excel": faFileExcel,
  "application/vnd.openxmlformats-officedocument.spreadsheetml":
  faFileExcel,
  "application/vnd.oasis.opendocument.spreadsheet": faFileExcel,
  "application/vnd.ms-powerpoint": faFilePowerpoint,
  "application/vnd.openxmlformats-officedocument.presentationml":
  faFilePowerpoint,
  "application/vnd.oasis.opendocument.presentation": faFilePowerpoint,
  "text/plain": faFileAlt,
  "text/html": faFileCode,
  "application/json": faFileCode,
  // Archives
  "application/gzip": faFileArchive,
  "application/zip": faFileArchive
};

export function getFontAwesomeIconFromMIME(mimeType: string): IconDefinition {
  // List of official MIME Types: http://www.iana.org/assignments/media-types/media-types.xhtml

  const candidate = Object.entries(iconClasses).find(([k]) =>
    mimeType.startsWith(k)
  )

  if (candidate === undefined) {
    return faFile;
  } else {
    return candidate[1];
  }

}