From a640652a9bf9f0d10848a812bd2b3ada8562dcdc Mon Sep 17 00:00:00 2001
From: "Timothy J. Baek" <timothyjrbeck@gmail.com>
Date: Sat, 23 Dec 2023 23:05:52 -0800
Subject: [PATCH] fix: large file upload issue

---
 backend/apps/web/routers/utils.py | 90 ++++++++++++++++++++++++-------
 1 file changed, 71 insertions(+), 19 deletions(-)

diff --git a/backend/apps/web/routers/utils.py b/backend/apps/web/routers/utils.py
index 8c822f02..ee125975 100644
--- a/backend/apps/web/routers/utils.py
+++ b/backend/apps/web/routers/utils.py
@@ -21,19 +21,74 @@ class UploadBlobForm(BaseModel):
     filename: str
 
 
+from urllib.parse import urlparse
+
+
+def parse_huggingface_url(hf_url):
+    # Parse the URL
+    parsed_url = urlparse(hf_url)
+
+    # Get the path and split it into components
+    path_components = parsed_url.path.split("/")
+
+    # Extract the desired output
+    user_repo = "/".join(path_components[1:3])
+    model_file = path_components[-1]
+
+    return [user_repo, model_file]
+
+
+def download_file_stream(url, file_path, chunk_size=1024 * 1024):
+    done = False
+
+    if os.path.exists(file_path):
+        current_size = os.path.getsize(file_path)
+    else:
+        current_size = 0
+
+    headers = {"Range": f"bytes={current_size}-"} if current_size > 0 else {}
+
+    with requests.get(url, headers=headers, stream=True) as response:
+        total_size = int(response.headers.get("content-length", 0)) + current_size
+
+        with open(file_path, "ab") as file:
+            for data in response.iter_content(chunk_size=chunk_size):
+                current_size += len(data)
+                file.write(data)
+
+                done = current_size == total_size
+
+                progress = round((current_size / total_size) * 100, 2)
+                yield f'data: {{"progress": {progress}, "current": {current_size}, "total": {total_size}}}\n\n'
+
+
+@router.get("/download")
+async def download(
+    url: str = "https://huggingface.co/TheBloke/stablelm-zephyr-3b-GGUF/resolve/main/stablelm-zephyr-3b.Q2_K.gguf",
+):
+    user_repo, model_file = parse_huggingface_url(url)
+
+    os.makedirs("./uploads", exist_ok=True)
+    file_path = os.path.join("./uploads", f"{model_file}")
+
+    return StreamingResponse(
+        download_file_stream(url, file_path), media_type="text/event-stream"
+    )
+
+
 @router.post("/upload")
 async def upload(file: UploadFile = File(...)):
     os.makedirs("./uploads", exist_ok=True)
     file_path = os.path.join("./uploads", file.filename)
 
-    def file_write_stream():
+    async def file_write_stream():
         total = 0
         total_size = file.size
         chunk_size = 1024 * 1024
 
         done = False
         try:
-            with open(file_path, "wb") as f:
+            with open(file_path, "wb+") as f:
                 while True:
                     chunk = file.file.read(chunk_size)
                     if not chunk:
@@ -50,26 +105,23 @@ async def upload(file: UploadFile = File(...)):
                     yield f"data: {json.dumps(res)}\n\n"
 
                 if done:
-                    with open(file_path, "rb") as f:
-                        hashed = calculate_sha256(f)
+                    f.seek(0)
+                    hashed = calculate_sha256(f)
+                    f.seek(0)
 
-                        f.seek(0)
-                        file_data = f.read()
+                    url = f"{OLLAMA_API_BASE_URL}/blobs/sha256:{hashed}"
+                    response = requests.post(url, data=f)
 
-                        url = f"{OLLAMA_API_BASE_URL}/blobs/sha256:{hashed}"
+                    if response.ok:
+                        res = {
+                            "done": done,
+                            "blob": f"sha256:{hashed}",
+                        }
+                        os.remove(file_path)
 
-                        response = requests.post(url, data=file_data)
-
-                        if response.ok:
-                            res = {
-                                "done": done,
-                                "blob": f"sha256:{hashed}",
-                            }
-                            os.remove(file_path)
-
-                            yield f"data: {json.dumps(res)}\n\n"
-                        else:
-                            raise "Ollama: Could not create blob, Please try again."
+                        yield f"data: {json.dumps(res)}\n\n"
+                    else:
+                        raise "Ollama: Could not create blob, Please try again."
 
         except Exception as e:
             res = {"error": str(e)}