Spaces:
Sleeping
Sleeping
Update main.py
Browse files
main.py
CHANGED
|
@@ -8,6 +8,17 @@ import json
|
|
| 8 |
import asyncio
|
| 9 |
from crawl4ai import AsyncWebCrawler
|
| 10 |
from crawl4ai.extraction_strategy import JsonCssExtractionStrategy
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
|
| 12 |
# Định nghĩa model request body
|
| 13 |
class URLProcessRequest(BaseModel):
|
|
|
|
| 8 |
import asyncio
|
| 9 |
from crawl4ai import AsyncWebCrawler
|
| 10 |
from crawl4ai.extraction_strategy import JsonCssExtractionStrategy
|
| 11 |
+
import modal
|
| 12 |
+
|
| 13 |
+
crawler = modal.Image.debian_slim(python_version="3.10").pip_install_from_requirements("requirements.txt").run_commands(
|
| 14 |
+
"apt-get update",
|
| 15 |
+
"apt-get install -y software-properties-common",
|
| 16 |
+
"apt-add-repository non-free",
|
| 17 |
+
"apt-add-repository contrib",
|
| 18 |
+
"playwright install-deps chromium",
|
| 19 |
+
"playwright install chromium",
|
| 20 |
+
"playwright install",
|
| 21 |
+
)
|
| 22 |
|
| 23 |
# Định nghĩa model request body
|
| 24 |
class URLProcessRequest(BaseModel):
|