Remove the heading numbers

in all the source docs.

docs/utils/build_markdown.py

@@ -10,6 +10,7 @@ sys.path.append(osp.dirname(osp.dirname(osp.dirname(osp.abspath(__file__)))))
 import re
 import glob
 import shutil
+from docs.utils.number_headings import number_headings
 from docs.utils.extract_code_python import ExtractCodeBlocksPython
 from docs.utils.extract_code_java import ExtractCodeBlocksJava
 from docs.utils.extract_code_cpp import ExtractCodeBlocksCpp
@@ -20,7 +21,7 @@ from docs.utils.extract_code_go import ExtractCodeBlocksGo
 from docs.utils.extract_code_zig import ExtractCodeBlocksZig
 
 
-def build_markdown(md_path):
+def build_code_blocks(md_path):
     with open(md_path, "r") as f:
         lines = f.readlines()
     
@@ -114,6 +115,9 @@ if __name__ == "__main__":
     shutil.copytree("docs", "build", dirs_exist_ok=True)
     shutil.rmtree("build/utils")
     
-    # Build docs
+    # Build code blocks
     for md_path in glob.glob("docs/chapter_*/*.md"):
-        build_markdown(md_path)
+        build_code_blocks(md_path)
+
+    # Build headings
+    number_headings("mkdocs.yml", "build")

docs/utils/number_headings.py

@@ -0,0 +1,81 @@
+import re
+
+def get_heading_info_from_nav(mkdocs_path):
+    """
+    Get heading info from mkdocs navigation
+    """
+    
+    with open(mkdocs_path) as f:
+        lines = f.readlines()[125:]
+    # Get nav lines
+    for i, line in enumerate(lines):
+        if "nav:" in line:
+            break
+    lines = lines[i:]
+
+    # Search articles
+    articles = []
+    for line in lines:
+        level = 0
+        level_re = None
+        while level_re is None and level < 3:
+            level += 1
+            level_pat = level * "  "
+            level_pat = f"^{level_pat}- \d"
+            level_re = re.search(level_pat, line)
+        # Only add articles with heading level 2
+        if level != 2:
+            continue
+
+        number_pat = level * "\d+."
+        number_re = re.search(number_pat, line)
+        number = re.search(number_pat, line).group(0) if number_re else None
+        file_path = re.search("\S+\/\S+\.md", line).group(0)
+        
+        article = {
+            "level": level,
+            "number": number,
+            "file_path": file_path
+        }
+        articles.append(article)
+        
+        print(f"{file_path}, heading number is {number}")
+
+    return articles
+
+def number_article(article, base_dir="build"):
+    """
+    Number a doc
+    """
+    
+    with open(f"{base_dir}/{article['file_path']}", "r") as f:
+        lines = f.readlines()
+
+    # Add h1, h2 heading numbers
+    h2_count = 1
+    for i, line in enumerate(lines):
+        h1_re = re.search("^(#)\s+\S+", line)
+        if h1_re is not None:
+            h1 = h1_re.group(1)
+            lines[i] = line.replace(h1, f"# {article['number']}")
+            continue
+
+        h2_re = re.search("^(##)\s+\S+", line)
+        if h2_re is not None:
+            h2 = h2_re.group(1)
+            lines[i] = line.replace(h2, f"## {article['number']}{h2_count}.")
+            h2_count += 1
+
+    with open(f"{base_dir}/{article['file_path']}", "w") as f:
+        f.writelines(lines)
+
+
+def number_headings(mkdocs_path, build_dir):
+    """
+    Build headings
+    """
+    
+    articles = get_heading_info_from_nav(mkdocs_path)
+
+    for article in articles:
+        number_article(article, base_dir=build_dir)