Merge pull request #202 from adobe/wrong-paragraph

Section and subitems type is wrong for paragraph with a link

README.md

@@ -212,7 +212,7 @@ The `types` property is an array of string values that describes the type of the
 
 - `has-<type>`: for each type of content that occurs at least once in the section, e.g. has-heading
 - `has-only-<type>`: for sections that only have content of a single type, e.g. has-only-image
-- `is-<type-1>-<type-2>-<type3>`, `is-<type-1>-<type-2>`, and `is-<type-1>` for the top 3 most frequent types of children in the section. For instance a gallery with a heading and description would be `is-image-paragraph-heading`. You can infer additional types using [`utils.types`](#infer-content-types-with-utilstypes).
+- `is-<type-1>-<type-2>-<type3>`, `is-<type-1>-<type-2>`, and `is-<type-1>` for the top 3 most frequent types of children in the section. For instance a gallery with a heading and description would be `is-image-text-heading`. You can infer additional types using [`utils.types`](#infer-content-types-with-utilstypes).
 - `nb-<type>-<occurences>`: number of occurences of each type in the section
 
 Each section has additional content-derived metadata properties, in particular:
@@ -579,8 +579,8 @@ Step 5 (diff only):
 +        ],
 +        "title": "Hello World",
 +        "types": [
-+          "has-paragraph",
-+          "has-only-paragraph"
++          "has-text",
++          "has-only-text"
 +        ],
 +        "intro": "Hello World",
 +        "meta": {}

src/html/get-metadata.js

@@ -87,13 +87,20 @@ function sectiontype(section) {
     if (type === 'paragraph' && pChildren && pChildren.length > 0) {
       // if child is a paragraph, check its children, it might contain an image or a list
       // which are always wrapped by default.
-      pChildren.forEach(({ type: subType }) => {
-        // exclude text nodes which are default paragraph content
-        if (subType !== 'text') {
-          const mycount = mycounter[subType] || 0;
-          mycounter[subType] = mycount + 1;
-          node.data.types.push(`is-${subType}`);
+      pChildren.forEach((p) => {
+        let prefix = 'has';
+        if (p.type === 'text') {
+          // do not count "empty" paragraphs
+          if (p.value === '\n' || p.value === '') return;
+
+          // paragraph with type text "is" a text
+          prefix = 'is';
         }
+        if (!node.data.types.includes(`${prefix}-${p.type}`)) {
+          node.data.types.push(`${prefix}-${p.type}`);
+        }
+        const mycount = mycounter[p.type] || 0;
+        mycounter[p.type] = mycount + 1;
       });
     }
 

test/fixtures/sections/2images.json

@@ -0,0 +1,42 @@
+[
+  {
+    "title": "helix-logo\nhelix-logo",
+    "types": [
+      "has-image",
+      "nb-image-2",
+      "has-only-image"
+    ],
+    "image": "./helix_logo.png",
+    "intro": "helix-logo\nhelix-logo",
+    "meta": {},
+    "type": "root",
+    "children": [
+      {
+        "type": "paragraph",
+        "children": [
+          {
+            "type": "image",
+            "title": null,
+            "url": "./helix_logo.png",
+            "alt": "helix-logo"
+          },
+          {
+            "type": "text",
+            "value": "\n"
+          },
+          {
+            "type": "image",
+            "title": null,
+            "url": "./helix_logo.png",
+            "alt": "helix-logo"
+          }
+        ],
+        "data": {
+          "types": [
+            "has-image"
+          ]
+        }
+      }
+    ]
+  }
+]

test/fixtures/sections/2images.md

@@ -0,0 +1,2 @@
+![helix-logo](./helix_logo.png)
+![helix-logo](./helix_logo.png)