Add workflow and script for docs JSON export

Introduces a GitHub Actions workflow to build Sphinx HTML documentation and export structured documentation data to docs.json using a new Python script. The script parses generated HTML files to extract classes, methods, attributes, and functions, then outputs them in JSON format for use as an artifact.

Author: Lulalaby

.github/workflows/docs-json-export.yml

@@ -0,0 +1,55 @@
+name: Docs JSON Export
+
+on:
+  push:
+    branches:
+      - master
+  workflow_dispatch:
+
+jobs:
+  export-docs-json:
+    name: Export docs.json
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout repository
+        id: checkout
+        uses: actions/checkout@v5
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        id: setup-python
+        with:
+          python-version: "3.13"
+          cache: "pip"
+          cache-dependency-path: "requirements/docs.txt"
+          check-latest: true
+
+      - name: Install dependencies
+        id: install-deps
+        run: pip install beautifulsoup4
+
+      - name: Build Sphinx HTML docs
+        id: build-sphinx
+        run: sphinx-build -b html docs docs/_build/html
+
+      - name: Export docs.json
+        id: generate-json
+        run: python scripts/docs_json_exporter.py
+
+      - name: Upload docs.json as artifact
+        uses: actions/[email protected]
+        id: artifact-upload
+        with:
+          name: Pycord Docs JSON
+          path: docs.json
+          retention-days: 1
+
+      - name: Show docs.json summary
+        run: |
+          head -n 40 docs.json || tail -n 40 docs.json
+
+      - name: Output artifact ID
+        run:  |
+          echo "artifact-id=${{ steps.artifact-upload.outputs.artifact-id }}" >> $GITHUB_OUTPUT
+          echo "artifact-url=${{ steps.artifact-upload.outputs.artifact-url }}" >> $GITHUB_OUTPUT
+          echo "::notice::Artifact uploaded: ${{ steps.artifact-upload.outputs.artifact-url }}"

scripts/docs_json_exporter.py

@@ -0,0 +1,58 @@
+
+import os
+import json
+from bs4 import BeautifulSoup
+
+def log(msg):
+    print(f"::notice::{msg}")
+
+log("Starting docs JSON export...")
+folders = [
+    "docs/_build/html/api",
+    "docs/_build/html/ext",
+]
+result = {}
+try:
+    for folder in folders:
+        if not os.path.isdir(folder):
+            log(f"Skipping missing folder: {folder}")
+            continue
+        base_html = os.path.normpath("docs/_build/html")
+        for root, _, files in os.walk(folder):
+            rel_dir = os.path.relpath(root, base_html).replace("\\", "/") + "/"
+            if rel_dir not in result:
+                result[rel_dir] = {}
+            for html_file in files:
+                if not html_file.endswith(".html"):
+                    continue
+                file_path = os.path.join(root, html_file)
+                with open(file_path, encoding="utf-8") as f:
+                    soup = BeautifulSoup(f, "html.parser")
+                    page_index = {}
+                    for class_dl in soup.find_all("dl", class_="class"):
+                        dt = class_dl.find("dt")
+                        class_name = dt.get("id") if dt else None
+                        if not class_name:
+                            class_name = dt.text.split(":")[-1].strip() if dt else None
+                        members = []
+                        for member_dl in class_dl.find_all("dl", class_=["attribute", "method"]):
+                            for member_dt in member_dl.find_all("dt"):
+                                member_id = member_dt.get("id")
+                                member_name = member_id.split(".")[-1] if member_id else member_dt.text.split(":")[-1].strip()
+                                if member_name:
+                                    members.append(member_name)
+                        page_index[class_name] = members
+                    for func_dl in soup.find_all("dl", class_="function"):
+                        dt = func_dl.find("dt")
+                        func_name = dt.get("id") if dt else None
+                        if not func_name:
+                            func_name = dt.text.split(":")[-1].strip() if dt else None
+                        page_index[func_name] = []
+                    result[rel_dir][html_file] = page_index
+    cleaned_result = {k: v for k, v in result.items() if v}
+    with open("docs.json", "w", encoding="utf-8") as out:
+        json.dump(cleaned_result, out, indent=2, ensure_ascii=False)
+    log("Exported docs to docs.json")
+    log("To upload as artifact: docs.json")
+except Exception as e:
+    print(f"::error::Docs JSON export failed: {e}")