Update keepEmptyParagraphs option to handle multimedia content in par…

…agraphs

src/util/saveArticles.ts

@@ -941,19 +941,16 @@ export function applyOtherTreatments(parsoidDoc: DominoElement, dump: Dump) {
   })
 
   /* Remove empty paragraphs */
-  // TODO: This option should be applied for page/html or/and page/mobile-html endpoints
   if (!dump.opts.keepEmptyParagraphs) {
-    // Mobile view === details
-    // Desktop view === section
-    const sections: DominoElement[] = Array.from(parsoidDoc.querySelectorAll('details, section'))
-    for (const section of sections) {
-      if (
-        section.children.length ===
-        Array.from(section.children).filter((child: DominoElement) => {
-          return child.matches('summary')
-        }).length
-      ) {
-        DU.deleteNode(section)
+    const paragraphs: DominoElement[] = Array.from(parsoidDoc.querySelectorAll('p'))
+    const mediaTags = ['img', 'video', 'audio', 'embed', 'object', 'iframe', 'canvas', 'svg', 'picture', 'track', 'source']
+    for (const paragraph of paragraphs) {
+      const hasNoMediaContent = !mediaTags.some((tag) => paragraph.querySelector(tag))
+      // Check if no media content inside first
+      if (hasNoMediaContent) {
+        if (!paragraph.textContent || (paragraph.textContent && paragraph.textContent.trim().length === 0)) {
+          DU.deleteNode(paragraph)
+        }
       }
     }
   }

test/unit/saveArticles.test.ts

@@ -66,8 +66,6 @@ describe('saveArticles', () => {
     let dump2: Dump
     let articleHtml: string
 
-    // TODO: This test will cause issues with intergational tests of multimedia content
-    /*
     beforeEach(async () => {
       const classes = await setupScrapeClasses({ mwUrl: 'https://en.wikivoyage.org' }) // en wikipedia
       dump = classes.dump
@@ -96,7 +94,6 @@ describe('saveArticles', () => {
       const paragraphs = Array.from(doc.querySelectorAll('p'))
       expect(paragraphs.length).toEqual(4)
     })
-    */
 
     /*
       TODO: Investigate empty section behavior for other endpoints such as page/html and page/mobile html