feat(stats): include code blocks in word counting

Author: Reuben-Sun

api/ingestors/note_ingestor.py

@@ -43,8 +43,8 @@
 IMG_SRC_PATTERN = re.compile(r'<img\s+[^>]*src="([^"]+)"', re.IGNORECASE)
 HTML_IMG_TAG_PATTERN = re.compile(r"<img\s+[^>]*>", re.IGNORECASE)
 MARKDOWN_IMG_PATTERN = re.compile(r"!\[[^\]]*\]\(([^)]+)\)")
-CODE_BLOCK_PATTERN = re.compile(r"```[\s\S]*?```")
-INLINE_CODE_PATTERN = re.compile(r"`[^`\n]*`")
+CODE_BLOCK_PATTERN = re.compile(r"```[^\n]*\n([\s\S]*?)```")
+INLINE_CODE_PATTERN = re.compile(r"`([^`\n]*)`")
 LINK_PATTERN = re.compile(r"\[([^\]]+)\]\(([^)]+)\)")
 HTML_TAG_PATTERN = re.compile(r"<[^>]+>")
 MATH_BLOCK_PATTERN = re.compile(r"\$\$[\s\S]*?\$\$")
@@ -521,16 +521,16 @@ def _to_plain_text(self, markdown_body: str) -> str:
         if not base:
             return ""
 
-        text = CODE_BLOCK_PATTERN.sub(" ", base)
+        text = CODE_BLOCK_PATTERN.sub(lambda match: f" {match.group(1)} ", base)
         text = MATH_BLOCK_PATTERN.sub(" ", text)
         text = INLINE_MATH_PATTERN.sub(" ", text)
-        text = INLINE_CODE_PATTERN.sub(" ", text)
+        text = INLINE_CODE_PATTERN.sub(lambda match: f" {match.group(1)} ", text)
         text = MARKDOWN_IMG_PATTERN.sub(" ", text)
         text = HTML_IMG_TAG_PATTERN.sub(" ", text)
         text = LINK_PATTERN.sub(lambda match: f" {match.group(1)} ", text)
         text = HTML_TAG_PATTERN.sub(" ", text)
         text = re.sub(r"(?mi)^\s*#{1,6}\s*images\s*$", " ", text)
-        text = re.sub(r"[#>*_`~\-]+", " ", text)
+        text = re.sub(r"[#>*~]+", " ", text)
         text = re.sub(r"\s+", " ", text)
         return text.strip()
 

docs/project/entries/git-submodule-quick-notes.md

@@ -12,7 +12,7 @@ type: guide
 status: published
 related:
 - git
-word_count: 40
+word_count: 51
 image_count: 0
 ---
 

docs/project/entries/github-pages-openknowforge-web.md

@@ -14,7 +14,7 @@ type: guide
 status: published
 related:
 - openknowforge
-word_count: 355
+word_count: 428
 image_count: 0
 ---
 

docs/project/entries/kl-kl-divergence.md

@@ -11,7 +11,7 @@ date: '2026-03-26'
 type: concept
 status: published
 related: []
-word_count: 132
+word_count: 131
 image_count: 0
 ---
 

docs/project/entries/openknowforge.md

@@ -15,7 +15,7 @@ status: published
 related:
 - notes-explorer
 - search-index
-word_count: 153
+word_count: 249
 image_count: 0
 ---
 

docs/public/search-index.json

@@ -1,5 +1,5 @@
 {
-  "generatedAt": "2026-03-26T15:47:33+00:00",
+  "generatedAt": "2026-03-26T16:01:55+00:00",
   "notes": [
     {
       "slug": "note",
@@ -24,7 +24,7 @@
       "updated_at": "2026-03-26T13:02:27+00:00",
       "submitted_at": "2026-03-26T13:02:27+00:00",
       "date": "2026-03-26",
-      "word_count": 132,
+      "word_count": 131,
       "image_count": 0,
       "tags": [
         "math",
@@ -41,7 +41,7 @@
       "updated_at": "2026-03-26T12:29:03+00:00",
       "submitted_at": "2026-03-26T12:29:03+00:00",
       "date": "2026-03-26",
-      "word_count": 355,
+      "word_count": 428,
       "image_count": 0,
       "tags": [
         "github-pages",
@@ -60,7 +60,7 @@
       "updated_at": "2026-03-26T11:53:34+00:00",
       "submitted_at": "2026-03-26T11:53:34+00:00",
       "date": "2026-03-26",
-      "word_count": 40,
+      "word_count": 51,
       "image_count": 0,
       "tags": [
         "git",
@@ -77,7 +77,7 @@
       "updated_at": "2026-03-26T00:00:00+00:00",
       "submitted_at": "2026-03-26T00:00:00+00:00",
       "date": "2026-03-26",
-      "word_count": 153,
+      "word_count": 249,
       "image_count": 0,
       "tags": [
         "guide",

tests/test_api_note.py

@@ -382,6 +382,39 @@ def test_post_note_empty_title_fails_validation(client: TestClient) -> None:
     assert response.status_code == 422
 
 
+def test_word_count_includes_code_blocks_and_inline_code(client: TestClient) -> None:
+    create_resp = client.post(
+        '/note',
+        json={
+            'title': 'Code Count',
+            'content': '\n'.join(
+                [
+                    'before',
+                    '```bash',
+                    'git status',
+                    '```',
+                    '`inline_code`',
+                ]
+            ),
+            'tags': ['code'],
+            'images': [],
+            'type': 'note',
+            'status': 'published',
+            'related': [],
+            'submitted_at': '2026-03-26T16:00:00+00:00',
+        },
+    )
+    assert create_resp.status_code == 200
+    slug = create_resp.json()['result']['slug']
+    assert create_resp.json()['result']['word_count'] == 4
+
+    note_resp = client.get(f'/note/{slug}')
+    assert note_resp.status_code == 200
+    note = note_resp.json()['result']
+    assert note['word_count'] == 4
+    assert note['image_count'] == 0
+
+
 def test_existing_note_stats_are_backfilled_once(client: TestClient, tmp_path: Path) -> None:
     legacy_note_path = tmp_path / 'docs' / 'project' / 'entries' / 'legacy.md'
     legacy_note_path.parent.mkdir(parents=True, exist_ok=True)