Merge pull request #218 from satti-hari-krishna-reddy/master

feat: add GitHub Actions workflow to sync docs with OpenAI vector store

Author: frikky

.github/openai-file-map.json

@@ -0,0 +1,62 @@
+{
+  "docs/about.md": {
+    "file_id": "file-AzfzUW2xMs6yDbGjRkoCdf",
+    "hash": "6f1bddf26c36be9e3ee8f00ebfe7959c3db8f6b7629b5db22e97ee0abd9791c6"
+  },
+  "docs/AI.md": {
+    "file_id": "file-4gRa9ee9iST2qnnmu4S2La",
+    "hash": "db14b672ffed31d1a92a3bd29f74daefb7ebb44c734a72cd11417868f6deb2c0"
+  },
+  "docs/API.md": {
+    "file_id": "file-3ALFqypGXjfS3MsyZke1ca",
+    "hash": "77b4633187816433d92bbd833277aee26e0ade2fc3b992655259a3ed0c2b5336"
+  },
+  "docs/apps.md": {
+    "file_id": "file-NivELeCPkvDSNFSEf9CtvU",
+    "hash": "6e6b13a1499dd555fb81fc6c50ccc4d85ec7c6a809efc9c5f65a3f273dc40052"
+  },
+  "docs/architecture.md": {
+    "file_id": "file-1vJkH5FjEvf1k84XMJaKBr",
+    "hash": "f187e54d3692af5dc9ca78eb9dd5a99f685853fa77011115fa9249a8dfab662e"
+  },
+  "docs/configuration.md": {
+    "file_id": "file-MKKafNkugWBMVdpKcFuwe2",
+    "hash": "a05f696ba0cddb92aa1fb3a61269e44c7618d740bbf9d2e19c7dd05b59c3cafb"
+  },
+  "docs/creators.md": {
+    "file_id": "file-3nGiHFWH6oNcX3NCNNK6PV",
+    "hash": "8d754bdd7b1953b2cd05617bb3afa0cead051cd3d6e0e5ea469a2041e86bc7e4"
+  },
+  "docs/extensions.md": {
+    "file_id": "file-2C3XApcxBWSjzpTdJeeFoa",
+    "hash": "c557bb3edbb46fd40801acab6c9226d7ac815895fa4cb8b8c87d4e5ecce1650e"
+  },
+  "docs/features.md": {
+    "file_id": "file-Ueb9hgh2GkT4aZvdVqBT1R",
+    "hash": "0504f63a75efbafda83a637d0630b6a0f2db29bd984976cd985e2d8d82c1cef4"
+  },
+  "docs/getting_started.md": {
+    "file_id": "file-1WXn3ormjcadCchiJJpJHw",
+    "hash": "db77d9a74df99a7c4c757a58ab7e22f93b6b4a6dba65d1575e1af585551bc3bd"
+  },
+  "docs/liquid.md": {
+    "file_id": "file-KvBc6d8MWAWfi65y5xx6ZU",
+    "hash": "ee962e1870d1b7998314a69a0ed29d7d3b93a70ebd48c4b56d7e52f4c66fa9f0"
+  },
+  "docs/organizations.md": {
+    "file_id": "file-VVKqtp2vMBfLbp1XBzX4ad",
+    "hash": "a9c1135cbc4c6abaa6f2ca8204e1636ebf9bb014d642ab69acb3546eea4206fa"
+  },
+  "docs/triggers.md": {
+    "file_id": "file-Q1JYeWM3R8dgjjRBNQXBKv",
+    "hash": "447346bba4519fa44443cd14ad46be7743c45eb582cc9eb17ffc98c91c9e85be"
+  },
+  "docs/troubleshooting.md": {
+    "file_id": "file-BAzqnXh5TPqLvq6y9DCsFR",
+    "hash": "8e7c42ae6ac80fd4397f951596d1db9abb59db799b4b76720079f6149f6043cb"
+  },
+  "docs/workflows.md": {
+    "file_id": "file-5ZzHyGGVpGanyTVt6XPfAc",
+    "hash": "f6d0b5259a047c2a21610cadfff90b9a873f8b107f15ee480a261ec07d8e6e57"
+  }
+}

.github/scripts/sync_vector_store.py

@@ -0,0 +1,104 @@
+import os
+import json
+import hashlib
+from pathlib import Path
+from openai import OpenAI
+
+DOCS_DIR = Path("docs")
+MAP_FILE = Path(".github/openai-file-map.json")
+
+OPENAI_API_KEY = os.environ.get("OPENAI_API_KEY")
+VECTOR_STORE_ID = os.environ.get("OPENAI_DOCS_VS_ID")
+
+if not OPENAI_API_KEY or not VECTOR_STORE_ID:
+    raise RuntimeError("Missing OPENAI_API_KEY or OPENAI_DOCS_VS_ID")
+
+client = OpenAI(api_key=OPENAI_API_KEY)
+
+def compute_hash(path: Path) -> str:
+    h = hashlib.sha256()
+    with open(path, "rb") as f:
+        for chunk in iter(lambda: f.read(8192), b""):
+            h.update(chunk)
+    return h.hexdigest()
+
+def load_map() -> dict:
+    if MAP_FILE.exists():
+        with open(MAP_FILE, "r") as f:
+            return json.load(f)
+    return {}
+
+def save_map(m: dict):
+    MAP_FILE.parent.mkdir(parents=True, exist_ok=True)
+    with open(MAP_FILE, "w") as f:
+        json.dump(m, f, indent=2)
+
+def upload_and_attach(path: Path) -> str:
+    print(f"Uploading: {path}")
+
+    uploaded = client.files.create(
+        file=open(path, "rb"),
+        purpose="assistants"
+    )
+
+    client.vector_stores.files.create(
+        vector_store_id=VECTOR_STORE_ID,
+        file_id=uploaded.id
+    )
+
+    return uploaded.id
+
+def delete_from_vector_store(file_id: str, path: str):
+    try:
+        client.vector_stores.files.delete(
+            vector_store_id=VECTOR_STORE_ID,
+            file_id=file_id
+        )
+        print(f"Deleted from vector store: {path}")
+    except Exception as e:
+        print(f"Warning: failed to delete {path}: {e}")
+
+file_map = load_map()
+
+current_files = {}
+for f in DOCS_DIR.rglob("*"):
+    if f.is_file():
+        rel_path = f.as_posix()
+        current_files[rel_path] = compute_hash(f)
+
+indexed = 0
+deleted = 0
+
+# Handle new + updated files
+for path, current_hash in current_files.items():
+    if path not in file_map:
+        file_id = upload_and_attach(Path(path))
+        file_map[path] = {
+            "file_id": file_id,
+            "hash": current_hash
+        }
+        indexed += 1
+
+    else:
+        # Existing file, check content
+        if file_map[path]["hash"] != current_hash:
+            delete_from_vector_store(file_map[path]["file_id"], path)
+            file_id = upload_and_attach(Path(path))
+            file_map[path] = {
+                "file_id": file_id,
+                "hash": current_hash
+            }
+            indexed += 1
+
+# Handle deleted files
+for path in list(file_map.keys()):
+    if path not in current_files:
+        delete_from_vector_store(file_map[path]["file_id"], path)
+        del file_map[path]
+        deleted += 1
+
+save_map(file_map)
+
+print("Sync complete.")
+print(f"Uploaded/updated: {indexed}")
+print(f"Deleted: {deleted}")

.github/workflows/sync-vector-store.yml

@@ -0,0 +1,43 @@
+name: Sync Vector Store
+
+on:
+  push:
+    branches:
+      - master
+    paths-ignore:
+      - ".github/openai-file-map.json"
+
+permissions:
+  contents: write
+
+jobs:
+  sync:
+    runs-on: ubuntu-latest
+
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.11"
+
+      - run: pip install openai
+
+      - name: Sync docs to vector store
+        env:
+          OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
+          OPENAI_DOCS_VS_ID: ${{ secrets.OPENAI_DOCS_VS_ID }}
+        run: |
+          python .github/scripts/sync_vector_store.py
+
+      - name: Commit updated map
+        run: |
+          if git status --porcelain | grep ".github/openai-file-map.json"; then
+            git config user.name "github-actions[bot]"
+            git config user.email "github-actions[bot]@users.noreply.github.com"
+            git add .github/openai-file-map.json
+            git commit -m "chore: sync vector store"
+            git push
+          else
+            echo "No changes"
+          fi

docs/workflows.md

@@ -765,5 +765,4 @@ Azure DevOps integration enables workflow backup with your organization’s Azur
 4. **Username**
    - Use your **Azure DevOps email address** as the username.
 
-Note: Azure devops support is available from version 2.1.1 and above
-   
+Note: Azure devops support is available from version 2.1.1 and above