feat: add option to allow custom extractors to skip default cleaners

src/extractors/custom/wikipedia.org/index.js

@@ -5,6 +5,8 @@ const WikipediaExtractor = {
       '#mw-content-text',
     ],
 
+    defaultCleaner: false,
+
     // transform top infobox to an image with caption
     transforms: {
       '.infobox img': ($node) => {

src/extractors/root-extractor.js

@@ -51,7 +51,7 @@ export function select(opts) {
   // contributors), return the string
   if (typeof extractionOpts === 'string') return extractionOpts;
 
-  const { selectors } = extractionOpts;
+  const { selectors, defaultCleaner = true } = extractionOpts;
 
   const matchingSelector = selectors.find(selector => $(selector).length === 1 && $(selector).text().trim() !== '');
 
@@ -88,9 +88,16 @@ export function select(opts) {
     // otherwise use the text of the node
     result = $(matchingSelector).text();
   }
+
+  // Allow custom extractor to skip default cleaner
+  // for this type; defaults to true
+  if (defaultCleaner) {
     return Cleaners[type](result, opts);
   }
 
+  return result;
+}
+
 function extractResult(opts) {
   const { type, extractor } = opts;