Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -1,5 +1,4 @@
|
|
| 1 |
-
from typing import
|
| 2 |
-
from urllib.parse import urlparse
|
| 3 |
import requests
|
| 4 |
import time
|
| 5 |
import json
|
|
@@ -98,21 +97,17 @@ def parse_huggingface_url(url: str) -> str:
|
|
| 98 |
|
| 99 |
return json.dumps(result, indent=2)
|
| 100 |
|
| 101 |
-
def parse_huggingface_url_with_summary(url: str
|
| 102 |
"""Parse URL and return both markdown summary and JSON."""
|
| 103 |
-
progress(0, desc="🔍 Starting URL parsing...")
|
| 104 |
|
| 105 |
if not url.strip():
|
| 106 |
return "# ❌ No URL Provided\n\nPlease enter a URL to parse.", "{}"
|
| 107 |
|
| 108 |
-
progress(0.3, desc="🔍 Analyzing URL format...")
|
| 109 |
json_result = parse_huggingface_url(url)
|
| 110 |
|
| 111 |
-
progress(0.7, desc="🔍 Generating summary...")
|
| 112 |
parsed_info = json.loads(json_result)
|
| 113 |
md_summary = format_url_summary(parsed_info)
|
| 114 |
|
| 115 |
-
progress(1.0, desc="✅ URL parsing complete!")
|
| 116 |
return md_summary, json_result
|
| 117 |
|
| 118 |
def format_url_summary(parsed_info: dict) -> str:
|
|
@@ -145,7 +140,7 @@ def format_url_summary(parsed_info: dict) -> str:
|
|
| 145 |
|
| 146 |
return md
|
| 147 |
|
| 148 |
-
def check_single_server_health(url: str
|
| 149 |
"""
|
| 150 |
Check health of a single MCP server from any URL format.
|
| 151 |
|
|
@@ -155,12 +150,10 @@ def check_single_server_health(url: str, progress=gr.Progress()) -> tuple:
|
|
| 155 |
Returns:
|
| 156 |
tuple: (markdown_summary, json_data)
|
| 157 |
"""
|
| 158 |
-
progress(0, desc="🏥 Starting health check...")
|
| 159 |
|
| 160 |
if not url.strip():
|
| 161 |
return "# ❌ No URL Provided\n\nPlease enter a URL to check.", "{}"
|
| 162 |
|
| 163 |
-
progress(0.1, desc="🏥 Parsing URL...")
|
| 164 |
parsed_info = json.loads(parse_huggingface_url(url))
|
| 165 |
|
| 166 |
if not parsed_info["is_valid"]:
|
|
@@ -182,7 +175,6 @@ def check_single_server_health(url: str, progress=gr.Progress()) -> tuple:
|
|
| 182 |
}
|
| 183 |
|
| 184 |
# Test 1: Check space URL health
|
| 185 |
-
progress(0.3, desc="🌐 Checking space accessibility...")
|
| 186 |
if parsed_info["space_url"] != "unknown":
|
| 187 |
start_time = time.time()
|
| 188 |
try:
|
|
@@ -206,7 +198,6 @@ def check_single_server_health(url: str, progress=gr.Progress()) -> tuple:
|
|
| 206 |
}
|
| 207 |
|
| 208 |
# Test 2: Check MCP endpoint health
|
| 209 |
-
progress(0.6, desc="🔧 Checking MCP endpoint...")
|
| 210 |
start_time = time.time()
|
| 211 |
try:
|
| 212 |
response = requests.get(parsed_info["mcp_endpoint"], timeout=8, stream=True)
|
|
@@ -228,8 +219,6 @@ def check_single_server_health(url: str, progress=gr.Progress()) -> tuple:
|
|
| 228 |
"error": str(e)
|
| 229 |
}
|
| 230 |
|
| 231 |
-
progress(0.9, desc="📊 Analyzing results...")
|
| 232 |
-
|
| 233 |
# Determine overall status
|
| 234 |
space_ok = results["space_health"] is None or results["space_health"]["accessible"]
|
| 235 |
mcp_ok = results["mcp_health"]["accessible"]
|
|
@@ -246,7 +235,6 @@ def check_single_server_health(url: str, progress=gr.Progress()) -> tuple:
|
|
| 246 |
# Generate markdown summary
|
| 247 |
md = format_health_summary(results)
|
| 248 |
|
| 249 |
-
progress(1.0, desc="✅ Health check complete!")
|
| 250 |
return md, json.dumps(results, indent=2)
|
| 251 |
|
| 252 |
def format_health_summary(results: dict) -> str:
|
|
@@ -333,7 +321,7 @@ def extract_functions_from_source(source_code: str) -> List[Tuple[str, str, List
|
|
| 333 |
|
| 334 |
return functions
|
| 335 |
|
| 336 |
-
def discover_server_tools(url: str
|
| 337 |
"""
|
| 338 |
Discover available MCP tools from a server.
|
| 339 |
|
|
@@ -343,12 +331,10 @@ def discover_server_tools(url: str, progress=gr.Progress()) -> tuple:
|
|
| 343 |
Returns:
|
| 344 |
tuple: (markdown_summary, json_data)
|
| 345 |
"""
|
| 346 |
-
progress(0, desc="🛠️ Starting tools discovery...")
|
| 347 |
|
| 348 |
if not url.strip():
|
| 349 |
return "# ❌ No URL Provided\n\nPlease enter a URL to discover tools.", "{}"
|
| 350 |
|
| 351 |
-
progress(0.1, desc="🛠️ Parsing URL...")
|
| 352 |
parsed_info = json.loads(parse_huggingface_url(url))
|
| 353 |
|
| 354 |
if not parsed_info["is_valid"]:
|
|
@@ -364,15 +350,12 @@ def discover_server_tools(url: str, progress=gr.Progress()) -> tuple:
|
|
| 364 |
discovery_methods = []
|
| 365 |
|
| 366 |
# Method: Analyze app.py source code
|
| 367 |
-
progress(0.3, desc="🛠️ Fetching source code...")
|
| 368 |
try:
|
| 369 |
# Try to get app.py from HF spaces
|
| 370 |
if parsed_info["hf_spaces_url"] != "unknown":
|
| 371 |
app_url = f"{parsed_info['hf_spaces_url']}/raw/main/app.py"
|
| 372 |
-
progress(0.5, desc="🛠️ Analyzing app.py...")
|
| 373 |
response = requests.get(app_url, timeout=10)
|
| 374 |
if response.status_code == 200:
|
| 375 |
-
progress(0.7, desc="🛠️ Extracting functions...")
|
| 376 |
functions = extract_functions_from_source(response.text)
|
| 377 |
for func_name, docstring, params in functions:
|
| 378 |
tools.append({
|
|
@@ -385,8 +368,6 @@ def discover_server_tools(url: str, progress=gr.Progress()) -> tuple:
|
|
| 385 |
except Exception as e:
|
| 386 |
discovery_methods.append(f"Failed to analyze app.py: {str(e)}")
|
| 387 |
|
| 388 |
-
progress(0.9, desc="🛠️ Preparing results...")
|
| 389 |
-
|
| 390 |
# Prepare result
|
| 391 |
result = {
|
| 392 |
"original_url": url,
|
|
@@ -404,7 +385,6 @@ def discover_server_tools(url: str, progress=gr.Progress()) -> tuple:
|
|
| 404 |
# Generate markdown summary
|
| 405 |
md = format_tools_summary(result)
|
| 406 |
|
| 407 |
-
progress(1.0, desc="✅ Tools discovery complete!")
|
| 408 |
return md, json.dumps(result, indent=2)
|
| 409 |
|
| 410 |
def format_tools_summary(result: dict) -> str:
|
|
@@ -434,7 +414,7 @@ def format_tools_summary(result: dict) -> str:
|
|
| 434 |
|
| 435 |
return md
|
| 436 |
|
| 437 |
-
def monitor_multiple_servers(urls_text: str
|
| 438 |
"""
|
| 439 |
Monitor health and tools of multiple MCP servers simultaneously.
|
| 440 |
|
|
@@ -444,7 +424,6 @@ def monitor_multiple_servers(urls_text: str, progress=gr.Progress()) -> tuple:
|
|
| 444 |
Returns:
|
| 445 |
tuple: (markdown_summary, json_data)
|
| 446 |
"""
|
| 447 |
-
progress(0, desc="📊 Starting multi-server monitoring...")
|
| 448 |
|
| 449 |
if not urls_text.strip():
|
| 450 |
result = {
|
|
@@ -455,27 +434,22 @@ def monitor_multiple_servers(urls_text: str, progress=gr.Progress()) -> tuple:
|
|
| 455 |
md = "# ❌ No URLs Provided\n\nPlease enter URLs to monitor."
|
| 456 |
return md, json.dumps(result, indent=2)
|
| 457 |
|
| 458 |
-
progress(0.1, desc="📊 Parsing URL list...")
|
| 459 |
urls = [url.strip() for url in urls_text.strip().split('\n') if url.strip()]
|
| 460 |
|
| 461 |
if not urls:
|
| 462 |
result = {
|
| 463 |
"error": "No valid URLs found",
|
| 464 |
"servers": [],
|
| 465 |
-
"total_servers": 0
|
| 466 |
-
}
|
| 467 |
md = "# ❌ No Valid URLs\n\nPlease check the URL format."
|
| 468 |
return md, json.dumps(result, indent=2)
|
| 469 |
|
| 470 |
results = []
|
| 471 |
|
| 472 |
for i, url in enumerate(urls, 1):
|
| 473 |
-
|
| 474 |
-
progress(progress_pct, desc=f"📊 Checking server {i}/{len(urls)}: {url[:50]}...")
|
| 475 |
|
| 476 |
try:
|
| 477 |
-
print(f"🔍 Checking server {i}/{len(urls)}: {url}")
|
| 478 |
-
|
| 479 |
_, health_json = check_single_server_health(url)
|
| 480 |
health_data = json.loads(health_json)
|
| 481 |
|
|
@@ -499,8 +473,6 @@ def monitor_multiple_servers(urls_text: str, progress=gr.Progress()) -> tuple:
|
|
| 499 |
"combined_status": "error"
|
| 500 |
})
|
| 501 |
|
| 502 |
-
progress(0.95, desc="📊 Generating report...")
|
| 503 |
-
|
| 504 |
final_result = {
|
| 505 |
"servers": results,
|
| 506 |
"total_servers": len(urls),
|
|
@@ -510,7 +482,6 @@ def monitor_multiple_servers(urls_text: str, progress=gr.Progress()) -> tuple:
|
|
| 510 |
# Generate markdown summary
|
| 511 |
md = format_multiple_servers_summary(final_result)
|
| 512 |
|
| 513 |
-
progress(1.0, desc="✅ Multi-server monitoring complete!")
|
| 514 |
return md, json.dumps(final_result, indent=2)
|
| 515 |
|
| 516 |
def format_multiple_servers_summary(result: dict) -> str:
|
|
@@ -546,7 +517,7 @@ def format_multiple_servers_summary(result: dict) -> str:
|
|
| 546 |
|
| 547 |
return md
|
| 548 |
|
| 549 |
-
def validate_mcp_endpoint(url: str
|
| 550 |
"""
|
| 551 |
Validate that a URL is a working MCP endpoint by checking its schema.
|
| 552 |
|
|
@@ -556,12 +527,10 @@ def validate_mcp_endpoint(url: str, progress=gr.Progress()) -> tuple:
|
|
| 556 |
Returns:
|
| 557 |
tuple: (markdown_summary, json_data)
|
| 558 |
"""
|
| 559 |
-
progress(0, desc="✅ Starting MCP validation...")
|
| 560 |
|
| 561 |
if not url.strip():
|
| 562 |
return "# ❌ No URL Provided\n\nPlease enter a URL to validate.", "{}"
|
| 563 |
|
| 564 |
-
progress(0.1, desc="✅ Parsing URL...")
|
| 565 |
parsed_info = json.loads(parse_huggingface_url(url))
|
| 566 |
|
| 567 |
validation_result = {
|
|
@@ -581,8 +550,6 @@ def validate_mcp_endpoint(url: str, progress=gr.Progress()) -> tuple:
|
|
| 581 |
|
| 582 |
mcp_endpoint = validation_result["mcp_endpoint_url"]
|
| 583 |
|
| 584 |
-
progress(0.3, desc="✅ Building schema URL...")
|
| 585 |
-
|
| 586 |
# Construct schema URL from MCP SSE endpoint
|
| 587 |
# Example: https://user-space.hf.space/gradio_api/mcp/sse -> https://user-space.hf.space/gradio_api/mcp/schema
|
| 588 |
if mcp_endpoint.endswith("/sse"):
|
|
@@ -596,8 +563,6 @@ def validate_mcp_endpoint(url: str, progress=gr.Progress()) -> tuple:
|
|
| 596 |
|
| 597 |
print(f"ℹ️ Validating MCP: Original URL='{url}', Endpoint='{mcp_endpoint}', Schema='{mcp_schema_url}'")
|
| 598 |
|
| 599 |
-
progress(0.5, desc="✅ Testing schema endpoint...")
|
| 600 |
-
|
| 601 |
# Test MCP schema endpoint
|
| 602 |
try:
|
| 603 |
headers = {'User-Agent': 'MCP-Validator/1.0'}
|
|
@@ -605,8 +570,6 @@ def validate_mcp_endpoint(url: str, progress=gr.Progress()) -> tuple:
|
|
| 605 |
|
| 606 |
validation_result["schema_http_status"] = response.status_code
|
| 607 |
|
| 608 |
-
progress(0.7, desc="✅ Processing schema response...")
|
| 609 |
-
|
| 610 |
if response.status_code == 200:
|
| 611 |
try:
|
| 612 |
schema_data = response.json()
|
|
@@ -656,12 +619,9 @@ def validate_mcp_endpoint(url: str, progress=gr.Progress()) -> tuple:
|
|
| 656 |
validation_result["error"] = f"An unexpected error occurred during validation: {str(e_gen)}"
|
| 657 |
print(f"❌ Unexpected error during MCP validation for {mcp_schema_url}: {str(e_gen)}")
|
| 658 |
|
| 659 |
-
progress(0.9, desc="✅ Generating validation report...")
|
| 660 |
-
|
| 661 |
# Generate markdown summary
|
| 662 |
md = format_validation_summary(validation_result)
|
| 663 |
|
| 664 |
-
progress(1.0, desc="✅ MCP validation complete!")
|
| 665 |
return md, json.dumps(validation_result, indent=2)
|
| 666 |
|
| 667 |
def format_validation_summary(result: dict) -> str:
|
|
@@ -711,40 +671,18 @@ def format_validation_summary(result: dict) -> str:
|
|
| 711 |
|
| 712 |
return md
|
| 713 |
|
| 714 |
-
def scrape_hf_spaces_with_progress(max_pages: int, sort_by: str
|
| 715 |
-
"""Wrapper function for scraping
|
| 716 |
-
progress(0, desc="🕷️ Starting HuggingFace scraping...")
|
| 717 |
|
| 718 |
# Validate sort option
|
| 719 |
if sort_by not in SORT_OPTIONS:
|
| 720 |
sort_by = "relevance"
|
| 721 |
|
| 722 |
-
|
| 723 |
-
|
| 724 |
-
#
|
| 725 |
-
def progress_callback(message):
|
| 726 |
-
# Extract progress percentage from message if possible
|
| 727 |
-
if "page" in message.lower():
|
| 728 |
-
try:
|
| 729 |
-
# Try to extract current page from message
|
| 730 |
-
import re
|
| 731 |
-
match = re.search(r'page (\d+)(/(\d+))?', message.lower())
|
| 732 |
-
if match:
|
| 733 |
-
current = int(match.group(1))
|
| 734 |
-
total = int(match.group(3)) if match.group(3) else max_pages
|
| 735 |
-
progress_pct = min(0.9, (current / total) * 0.8 + 0.1) # Reserve 10% for final processing
|
| 736 |
-
progress(progress_pct, desc=f"🕷️ {message}")
|
| 737 |
-
else:
|
| 738 |
-
progress(None, desc=f"🕷️ {message}")
|
| 739 |
-
except:
|
| 740 |
-
progress(None, desc=f"🕷️ {message}")
|
| 741 |
-
else:
|
| 742 |
-
progress(None, desc=f"🕷️ {message}")
|
| 743 |
-
|
| 744 |
-
# Call the scraper with progress callback and sort option
|
| 745 |
md, json_data = scrape_and_monitor_hf_mcp_servers(max_pages, sort_by)
|
| 746 |
|
| 747 |
-
progress(1.0, desc="✅ HuggingFace scraping complete!")
|
| 748 |
return md, json_data
|
| 749 |
|
| 750 |
# Default URLs for testing
|
|
|
|
| 1 |
+
from typing import List, Tuple
|
|
|
|
| 2 |
import requests
|
| 3 |
import time
|
| 4 |
import json
|
|
|
|
| 97 |
|
| 98 |
return json.dumps(result, indent=2)
|
| 99 |
|
| 100 |
+
def parse_huggingface_url_with_summary(url: str) -> tuple:
|
| 101 |
"""Parse URL and return both markdown summary and JSON."""
|
|
|
|
| 102 |
|
| 103 |
if not url.strip():
|
| 104 |
return "# ❌ No URL Provided\n\nPlease enter a URL to parse.", "{}"
|
| 105 |
|
|
|
|
| 106 |
json_result = parse_huggingface_url(url)
|
| 107 |
|
|
|
|
| 108 |
parsed_info = json.loads(json_result)
|
| 109 |
md_summary = format_url_summary(parsed_info)
|
| 110 |
|
|
|
|
| 111 |
return md_summary, json_result
|
| 112 |
|
| 113 |
def format_url_summary(parsed_info: dict) -> str:
|
|
|
|
| 140 |
|
| 141 |
return md
|
| 142 |
|
| 143 |
+
def check_single_server_health(url: str) -> tuple:
|
| 144 |
"""
|
| 145 |
Check health of a single MCP server from any URL format.
|
| 146 |
|
|
|
|
| 150 |
Returns:
|
| 151 |
tuple: (markdown_summary, json_data)
|
| 152 |
"""
|
|
|
|
| 153 |
|
| 154 |
if not url.strip():
|
| 155 |
return "# ❌ No URL Provided\n\nPlease enter a URL to check.", "{}"
|
| 156 |
|
|
|
|
| 157 |
parsed_info = json.loads(parse_huggingface_url(url))
|
| 158 |
|
| 159 |
if not parsed_info["is_valid"]:
|
|
|
|
| 175 |
}
|
| 176 |
|
| 177 |
# Test 1: Check space URL health
|
|
|
|
| 178 |
if parsed_info["space_url"] != "unknown":
|
| 179 |
start_time = time.time()
|
| 180 |
try:
|
|
|
|
| 198 |
}
|
| 199 |
|
| 200 |
# Test 2: Check MCP endpoint health
|
|
|
|
| 201 |
start_time = time.time()
|
| 202 |
try:
|
| 203 |
response = requests.get(parsed_info["mcp_endpoint"], timeout=8, stream=True)
|
|
|
|
| 219 |
"error": str(e)
|
| 220 |
}
|
| 221 |
|
|
|
|
|
|
|
| 222 |
# Determine overall status
|
| 223 |
space_ok = results["space_health"] is None or results["space_health"]["accessible"]
|
| 224 |
mcp_ok = results["mcp_health"]["accessible"]
|
|
|
|
| 235 |
# Generate markdown summary
|
| 236 |
md = format_health_summary(results)
|
| 237 |
|
|
|
|
| 238 |
return md, json.dumps(results, indent=2)
|
| 239 |
|
| 240 |
def format_health_summary(results: dict) -> str:
|
|
|
|
| 321 |
|
| 322 |
return functions
|
| 323 |
|
| 324 |
+
def discover_server_tools(url: str) -> tuple:
|
| 325 |
"""
|
| 326 |
Discover available MCP tools from a server.
|
| 327 |
|
|
|
|
| 331 |
Returns:
|
| 332 |
tuple: (markdown_summary, json_data)
|
| 333 |
"""
|
|
|
|
| 334 |
|
| 335 |
if not url.strip():
|
| 336 |
return "# ❌ No URL Provided\n\nPlease enter a URL to discover tools.", "{}"
|
| 337 |
|
|
|
|
| 338 |
parsed_info = json.loads(parse_huggingface_url(url))
|
| 339 |
|
| 340 |
if not parsed_info["is_valid"]:
|
|
|
|
| 350 |
discovery_methods = []
|
| 351 |
|
| 352 |
# Method: Analyze app.py source code
|
|
|
|
| 353 |
try:
|
| 354 |
# Try to get app.py from HF spaces
|
| 355 |
if parsed_info["hf_spaces_url"] != "unknown":
|
| 356 |
app_url = f"{parsed_info['hf_spaces_url']}/raw/main/app.py"
|
|
|
|
| 357 |
response = requests.get(app_url, timeout=10)
|
| 358 |
if response.status_code == 200:
|
|
|
|
| 359 |
functions = extract_functions_from_source(response.text)
|
| 360 |
for func_name, docstring, params in functions:
|
| 361 |
tools.append({
|
|
|
|
| 368 |
except Exception as e:
|
| 369 |
discovery_methods.append(f"Failed to analyze app.py: {str(e)}")
|
| 370 |
|
|
|
|
|
|
|
| 371 |
# Prepare result
|
| 372 |
result = {
|
| 373 |
"original_url": url,
|
|
|
|
| 385 |
# Generate markdown summary
|
| 386 |
md = format_tools_summary(result)
|
| 387 |
|
|
|
|
| 388 |
return md, json.dumps(result, indent=2)
|
| 389 |
|
| 390 |
def format_tools_summary(result: dict) -> str:
|
|
|
|
| 414 |
|
| 415 |
return md
|
| 416 |
|
| 417 |
+
def monitor_multiple_servers(urls_text: str) -> tuple:
|
| 418 |
"""
|
| 419 |
Monitor health and tools of multiple MCP servers simultaneously.
|
| 420 |
|
|
|
|
| 424 |
Returns:
|
| 425 |
tuple: (markdown_summary, json_data)
|
| 426 |
"""
|
|
|
|
| 427 |
|
| 428 |
if not urls_text.strip():
|
| 429 |
result = {
|
|
|
|
| 434 |
md = "# ❌ No URLs Provided\n\nPlease enter URLs to monitor."
|
| 435 |
return md, json.dumps(result, indent=2)
|
| 436 |
|
|
|
|
| 437 |
urls = [url.strip() for url in urls_text.strip().split('\n') if url.strip()]
|
| 438 |
|
| 439 |
if not urls:
|
| 440 |
result = {
|
| 441 |
"error": "No valid URLs found",
|
| 442 |
"servers": [],
|
| 443 |
+
"total_servers": 0 }
|
|
|
|
| 444 |
md = "# ❌ No Valid URLs\n\nPlease check the URL format."
|
| 445 |
return md, json.dumps(result, indent=2)
|
| 446 |
|
| 447 |
results = []
|
| 448 |
|
| 449 |
for i, url in enumerate(urls, 1):
|
| 450 |
+
print(f"🔍 Checking server {i}/{len(urls)}: {url}")
|
|
|
|
| 451 |
|
| 452 |
try:
|
|
|
|
|
|
|
| 453 |
_, health_json = check_single_server_health(url)
|
| 454 |
health_data = json.loads(health_json)
|
| 455 |
|
|
|
|
| 473 |
"combined_status": "error"
|
| 474 |
})
|
| 475 |
|
|
|
|
|
|
|
| 476 |
final_result = {
|
| 477 |
"servers": results,
|
| 478 |
"total_servers": len(urls),
|
|
|
|
| 482 |
# Generate markdown summary
|
| 483 |
md = format_multiple_servers_summary(final_result)
|
| 484 |
|
|
|
|
| 485 |
return md, json.dumps(final_result, indent=2)
|
| 486 |
|
| 487 |
def format_multiple_servers_summary(result: dict) -> str:
|
|
|
|
| 517 |
|
| 518 |
return md
|
| 519 |
|
| 520 |
+
def validate_mcp_endpoint(url: str) -> tuple:
|
| 521 |
"""
|
| 522 |
Validate that a URL is a working MCP endpoint by checking its schema.
|
| 523 |
|
|
|
|
| 527 |
Returns:
|
| 528 |
tuple: (markdown_summary, json_data)
|
| 529 |
"""
|
|
|
|
| 530 |
|
| 531 |
if not url.strip():
|
| 532 |
return "# ❌ No URL Provided\n\nPlease enter a URL to validate.", "{}"
|
| 533 |
|
|
|
|
| 534 |
parsed_info = json.loads(parse_huggingface_url(url))
|
| 535 |
|
| 536 |
validation_result = {
|
|
|
|
| 550 |
|
| 551 |
mcp_endpoint = validation_result["mcp_endpoint_url"]
|
| 552 |
|
|
|
|
|
|
|
| 553 |
# Construct schema URL from MCP SSE endpoint
|
| 554 |
# Example: https://user-space.hf.space/gradio_api/mcp/sse -> https://user-space.hf.space/gradio_api/mcp/schema
|
| 555 |
if mcp_endpoint.endswith("/sse"):
|
|
|
|
| 563 |
|
| 564 |
print(f"ℹ️ Validating MCP: Original URL='{url}', Endpoint='{mcp_endpoint}', Schema='{mcp_schema_url}'")
|
| 565 |
|
|
|
|
|
|
|
| 566 |
# Test MCP schema endpoint
|
| 567 |
try:
|
| 568 |
headers = {'User-Agent': 'MCP-Validator/1.0'}
|
|
|
|
| 570 |
|
| 571 |
validation_result["schema_http_status"] = response.status_code
|
| 572 |
|
|
|
|
|
|
|
| 573 |
if response.status_code == 200:
|
| 574 |
try:
|
| 575 |
schema_data = response.json()
|
|
|
|
| 619 |
validation_result["error"] = f"An unexpected error occurred during validation: {str(e_gen)}"
|
| 620 |
print(f"❌ Unexpected error during MCP validation for {mcp_schema_url}: {str(e_gen)}")
|
| 621 |
|
|
|
|
|
|
|
| 622 |
# Generate markdown summary
|
| 623 |
md = format_validation_summary(validation_result)
|
| 624 |
|
|
|
|
| 625 |
return md, json.dumps(validation_result, indent=2)
|
| 626 |
|
| 627 |
def format_validation_summary(result: dict) -> str:
|
|
|
|
| 671 |
|
| 672 |
return md
|
| 673 |
|
| 674 |
+
def scrape_hf_spaces_with_progress(max_pages: int, sort_by: str) -> tuple:
|
| 675 |
+
"""Wrapper function for scraping."""
|
|
|
|
| 676 |
|
| 677 |
# Validate sort option
|
| 678 |
if sort_by not in SORT_OPTIONS:
|
| 679 |
sort_by = "relevance"
|
| 680 |
|
| 681 |
+
# Call the scraper with sort option
|
| 682 |
+
# The imported scrape_and_monitor_hf_mcp_servers function
|
| 683 |
+
# will print its own progress to the console.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 684 |
md, json_data = scrape_and_monitor_hf_mcp_servers(max_pages, sort_by)
|
| 685 |
|
|
|
|
| 686 |
return md, json_data
|
| 687 |
|
| 688 |
# Default URLs for testing
|