105 lines
3.9 KiB
Python
105 lines
3.9 KiB
Python
from __future__ import annotations
|
|
|
|
import argparse
|
|
import base64
|
|
import json
|
|
import struct
|
|
import sys
|
|
import urllib.error
|
|
import urllib.request
|
|
import zlib
|
|
|
|
|
|
def png_chunk(kind: bytes, data: bytes) -> bytes:
|
|
return struct.pack(">I", len(data)) + kind + data + struct.pack(">I", zlib.crc32(kind + data) & 0xFFFFFFFF)
|
|
|
|
|
|
def rgb_png(width: int, height: int, pixels: list[tuple[int, int, int]]) -> bytes:
|
|
rows = []
|
|
for y in range(height):
|
|
start = y * width
|
|
row = b"".join(bytes(pixel) for pixel in pixels[start : start + width])
|
|
rows.append(b"\x00" + row)
|
|
header = struct.pack(">IIBBBBB", width, height, 8, 2, 0, 0, 0)
|
|
return b"\x89PNG\r\n\x1a\n" + png_chunk(b"IHDR", header) + png_chunk(b"IDAT", zlib.compress(b"".join(rows), 6)) + png_chunk(b"IEND", b"")
|
|
|
|
|
|
def make_image_data_url(width: int = 512, height: int = 512) -> str:
|
|
pixels = []
|
|
for y in range(height):
|
|
for x in range(width):
|
|
if x < width // 2 and y < height // 2:
|
|
pixels.append((230, 70, 70))
|
|
elif x >= width // 2 and y < height // 2:
|
|
pixels.append((70, 180, 95))
|
|
elif x < width // 2:
|
|
pixels.append((70, 100, 220))
|
|
else:
|
|
pixels.append((240, 210, 70))
|
|
return "data:image/png;base64," + base64.b64encode(rgb_png(width, height, pixels)).decode("ascii")
|
|
|
|
|
|
def post_json(url: str, payload: dict, timeout: int) -> dict:
|
|
request = urllib.request.Request(
|
|
url,
|
|
data=json.dumps(payload, ensure_ascii=False).encode("utf-8"),
|
|
headers={"Content-Type": "application/json"},
|
|
method="POST",
|
|
)
|
|
with urllib.request.urlopen(request, timeout=timeout) as response:
|
|
data = json.loads(response.read().decode("utf-8"))
|
|
if not isinstance(data, dict):
|
|
raise ValueError(f"JSON object expected from {url}")
|
|
return data
|
|
|
|
|
|
def run(args: argparse.Namespace) -> int:
|
|
payload = {
|
|
"base_url": args.video_base_url,
|
|
"image_base64": make_image_data_url(args.width, args.height),
|
|
"filename": "smoke-grid.png",
|
|
"prompt": args.prompt,
|
|
"model": args.model,
|
|
"max_tokens": args.max_tokens,
|
|
}
|
|
result = post_json(f"{args.chat_base_url.rstrip('/')}/api/video/analyze", payload, args.timeout)
|
|
print(json.dumps(result, ensure_ascii=False, indent=2))
|
|
if result.get("error"):
|
|
return 1
|
|
return 0 if result.get("text") and result.get("model") else 1
|
|
|
|
|
|
def format_url_error(exc: urllib.error.URLError) -> str:
|
|
if isinstance(exc, urllib.error.HTTPError):
|
|
try:
|
|
body = exc.read().decode("utf-8", errors="replace").strip()
|
|
except OSError:
|
|
body = ""
|
|
return f"HTTP {exc.code}: {body or exc.reason}"
|
|
return str(exc)
|
|
|
|
|
|
def main() -> int:
|
|
parser = argparse.ArgumentParser(description="Smoke test vision analysis through the model chat API.")
|
|
parser.add_argument("--chat-base-url", default="http://192.168.220.91:8765")
|
|
parser.add_argument("--video-base-url", default="http://docker-gpu.cin.su:8030")
|
|
parser.add_argument("--model", default="qwen2.5-vl-7b-instruct")
|
|
parser.add_argument("--prompt", default="Опиши изображение: какие крупные цветные области видны?")
|
|
parser.add_argument("--width", type=int, default=512)
|
|
parser.add_argument("--height", type=int, default=512)
|
|
parser.add_argument("--max-tokens", type=int, default=128)
|
|
parser.add_argument("--timeout", type=int, default=900)
|
|
parser.add_argument("--allow-unavailable", action="store_true")
|
|
args = parser.parse_args()
|
|
|
|
try:
|
|
return run(args)
|
|
except (OSError, ValueError, urllib.error.URLError) as exc:
|
|
message = format_url_error(exc) if isinstance(exc, urllib.error.URLError) else str(exc)
|
|
print(f"Smoke video analysis failed: {message}", file=sys.stderr)
|
|
return 0 if args.allow_unavailable else 1
|
|
|
|
|
|
if __name__ == "__main__":
|
|
raise SystemExit(main())
|