Fix auto-generate-examples.py for hosted images

Author: alvarobartt

docs/scripts/auto-generate-examples.py

@@ -2,9 +2,17 @@
 import re
 
 
+DOCS_EXAMPLES_DIR = "docs/source/examples"
+DOCS_EXAMPLES_RELATIVE_PREFIX = "../../../"
+DOCUMENTATION_IMAGES_BASE_URL = (
+    "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/google-cloud"
+)
+IMAGE_EXTENSIONS = (".png", ".jpg", ".jpeg", ".gif", ".webp", ".svg")
+
+
 def process_readme_files():
     print("Processing README.md files from examples/gke and examples/cloud-run...")
-    os.makedirs("docs/source/examples", exist_ok=True)
+    os.makedirs(DOCS_EXAMPLES_DIR, exist_ok=True)
 
     for dir in ["gke", "cloud-run", "vertex-ai/notebooks"]:
         for root, _, files in os.walk(f"examples/{dir}"):
@@ -13,6 +21,64 @@ def process_readme_files():
                     process_file(root, file, dir)
 
 
+def is_relative_image_path(path):
+    normalized_path = path.split("#", 1)[0].split("?", 1)[0]
+    return normalized_path.startswith("./") and normalized_path.lower().endswith(IMAGE_EXTENSIONS)
+
+
+def build_local_image_path(root, image_path):
+    return os.path.normpath(os.path.join(root, image_path))
+
+
+def build_docs_image_path(local_image_path):
+    return f"{DOCS_EXAMPLES_RELATIVE_PREFIX}{local_image_path}".replace(os.sep, "/")
+
+
+def build_hub_image_url(local_image_path):
+    return (
+        f"{DOCUMENTATION_IMAGES_BASE_URL}/{local_image_path}".replace(os.sep, "/")
+    )
+
+
+def resolve_relative_image_path(root, image_path):
+    local_image_path = build_local_image_path(root, image_path)
+
+    if os.path.exists(local_image_path):
+        return build_docs_image_path(local_image_path)
+
+    return build_hub_image_url(local_image_path)
+
+
+def replace_relative_image_references(content, root):
+    def replace_markdown_image(match):
+        image_path = match.group("path")
+        if not is_relative_image_path(image_path):
+            return match.group(0)
+
+        resolved_path = resolve_relative_image_path(root, image_path)
+        return f"![{match.group('alt')}]({resolved_path})"
+
+    def replace_html_image(match):
+        image_path = match.group("path")
+        if not is_relative_image_path(image_path):
+            return match.group(0)
+
+        resolved_path = resolve_relative_image_path(root, image_path)
+        return match.group(0).replace(image_path, resolved_path, 1)
+
+    content = re.sub(
+        r"!\[(?P<alt>[^\]]*)\]\((?P<path>[^)]+)\)",
+        replace_markdown_image,
+        content,
+    )
+    content = re.sub(
+        r'(<img\b[^>]*\bsrc=["\'])(?P<path>[^"\']+)(["\'][^>]*>)',
+        replace_html_image,
+        content,
+    )
+    return content
+
+
 def process_file(root, file, dir):
     dir_name = dir if not dir.__contains__("/") else dir.replace("/", "-")
 
@@ -21,9 +87,9 @@ def process_file(root, file, dir):
     base = os.path.basename(subdir)
 
     if file_path == f"examples/{dir}/README.md":
-        target = f"docs/source/examples/{dir_name}-index.mdx"
+        target = f"{DOCS_EXAMPLES_DIR}/{dir_name}-index.mdx"
     else:
-        target = f"docs/source/examples/{dir_name}-{base}.mdx"
+        target = f"{DOCS_EXAMPLES_DIR}/{dir_name}-{base}.mdx"
 
     print(f"Processing {file_path} to {target}")
     with open(file_path, "r") as f:
@@ -33,23 +99,18 @@ def process_file(root, file, dir):
     # We only uncomment the metadata block to avoid uncommenting other HTML comments
     content = re.sub(r"<!--\s*(---.*?---)\s*-->", r"\1", content, flags=re.DOTALL)
 
+    content = replace_relative_image_references(content, root)
+
     # Replace image and link paths
     content = re.sub(
-        r"\(\./(imgs|assets)/([^)]*\.png)\)",
-        r"(https://raw.githubusercontent.com/huggingface/Google-Cloud-Containers/main/"
-        + root
-        + r"/\1/\2)",
-        content,
-    )
-    content = re.sub(
-        r"\(\.\./([^)]+)\)",
+        r"(?<!!)\(\.\./([^)]+)\)",
         r"(https://github.com/huggingface/Google-Cloud-Containers/tree/main/examples/"
         + dir
         + r"/\1)",
         content,
     )
     content = re.sub(
-        r"\(\.\/([^)]+)\)",
+        r"(?<!!)\(\.\/([^)]+)\)",
         r"(https://github.com/huggingface/Google-Cloud-Containers/tree/main/"
         + root
         + r"/\1)",