paddleocr-skills 1.0.0 → 1.1.0

package/templates/ppocrv5/scripts/ppocrv5/requirements.txt

@@ -1,4 +1,4 @@
-# Runtime dependencies for PP-OCRv5 API Skill
-
-httpx>=0.24.0
-python-dotenv>=0.19.0
+# Runtime dependencies for PP-OCRv5 API Skill
+
+httpx>=0.24.0
+python-dotenv>=0.19.0

package/templates/ppocrv5/scripts/ppocrv5/smoke_test.py

@@ -1,139 +1,139 @@
-#!/usr/bin/env python3
-"""
-Smoke Test for PP-OCRv5 API Skill
-Verifies that AISTUDIO_HOST and PADDLE_OCR_TOKEN are correctly configured
-and that the provider API is accessible.
-"""
-
-import json
-import os
-import subprocess
-import sys
-from pathlib import Path
-
-
-def main():
-    print("=" * 60)
-    print("PP-OCRv5 API Skill - Smoke Test")
-    print("=" * 60)
-
-    # Check configuration (all sources)
-    print("\n[1/3] Checking configuration...")
-
-    # Add scripts dir to path for imports
-    script_dir = Path(__file__).parent
-    sys.path.insert(0, str(script_dir))
-
-    from _lib import Config
-
-    try:
-        # Try to get config from .env file
-        api_url = Config.get_api_url()
-        token = Config.get_token()
-    except ValueError as e:
-        print(f"\nConfiguration error: {e}")
-        sys.exit(1)
-
-    test_file_url = os.getenv("TEST_FILE_URL", "").strip()
-
-    if not test_file_url:
-        print("WARNING: TEST_FILE_URL is not set, using default test image")
-        # Use a default public test image (Chinese text)
-        test_file_url = "https://raw.githubusercontent.com/PaddlePaddle/PaddleOCR/release/2.7/doc/imgs/11.jpg"
-
-    print(f"  API_URL: {api_url}")
-    print(f"  PADDLE_OCR_TOKEN: {'*' * 8}{token[-4:] if len(token) > 4 else '****'}")
-    print(f"  TEST_FILE_URL: {test_file_url}")
-
-    # Run ocr_caller.py
-    print("\n[2/3] Running OCR on test file...")
-
-    script_dir = Path(__file__).parent
-    ocr_caller = script_dir / "ocr_caller.py"
-
-    cmd = [
-        sys.executable,
-        str(ocr_caller),
-        "--mode", "auto",
-        "--file-url", test_file_url,
-        "--max-attempts", "2",
-        "--budget-ms", "20000"
-    ]
-
-    try:
-        result = subprocess.run(
-            cmd,
-            capture_output=True,
-            text=True,
-            timeout=30
-        )
-    except subprocess.TimeoutExpired:
-        print("ERROR: OCR call timed out after 30 seconds")
-        sys.exit(4)
-    except Exception as e:
-        print(f"ERROR: Failed to run ocr_caller.py: {e}")
-        sys.exit(4)
-
-    # Parse output
-    print("\n[3/3] Validating response...")
-
-    if result.returncode != 0:
-        print(f"ERROR: ocr_caller.py exited with code {result.returncode}")
-        print("\nStderr:")
-        print(result.stderr)
-        print("\nStdout:")
-        print(result.stdout)
-        sys.exit(result.returncode)
-
-    try:
-        response = json.loads(result.stdout)
-    except json.JSONDecodeError as e:
-        print(f"ERROR: Failed to parse JSON response: {e}")
-        print("\nStdout:")
-        print(result.stdout)
-        sys.exit(4)
-
-    # Validate response structure
-    if not response.get("ok"):
-        error = response.get("error", {})
-        print(f"ERROR: OCR failed with error code: {error.get('code')}")
-        print(f"Message: {error.get('message')}")
-        print(f"\nFull response:\n{json.dumps(response, indent=2, ensure_ascii=False)}")
-        sys.exit(3)
-
-    # Check for content
-    result_data = response.get("result", {})
-    full_text = result_data.get("full_text", "")
-    pages = result_data.get("pages", [])
-
-    total_items = sum(len(page.get("items", [])) for page in pages)
-
-    if not full_text and total_items == 0:
-        print("WARNING: OCR succeeded but returned no text. This may indicate:")
-        print("  - The test image is blank or unreadable")
-        print("  - Provider API is working but returned empty results")
-        print(f"\nFull response:\n{json.dumps(response, indent=2, ensure_ascii=False)}")
-        # Still pass, as API is working
-    else:
-        print(f"SUCCESS: OCR completed")
-        print(f"  - Total text items: {total_items}")
-        print(f"  - Quality score: {response.get('quality', {}).get('quality_score', 0):.4f}")
-        print(f"  - Avg confidence: {response.get('quality', {}).get('avg_rec_score', 0):.4f}")
-        print(f"  - Mode: {response.get('agent_trace', {}).get('mode')}")
-        print(f"  - Attempts: {len(response.get('agent_trace', {}).get('attempts', []))}")
-
-        # Print first 200 chars of text
-        if full_text:
-            preview = full_text[:200].replace("\n", " ")
-            if len(full_text) > 200:
-                preview += "..."
-            print(f"\n  Preview: {preview}")
-
-    print("\n" + "=" * 60)
-    print("Smoke test PASSED")
-    print("=" * 60)
-    sys.exit(0)
-
-
-if __name__ == "__main__":
-    main()
+#!/usr/bin/env python3
+"""
+Smoke Test for PP-OCRv5 API Skill
+Verifies that AISTUDIO_HOST and PADDLE_OCR_TOKEN are correctly configured
+and that the provider API is accessible.
+"""
+
+import json
+import os
+import subprocess
+import sys
+from pathlib import Path
+
+
+def main():
+    print("=" * 60)
+    print("PP-OCRv5 API Skill - Smoke Test")
+    print("=" * 60)
+
+    # Check configuration (all sources)
+    print("\n[1/3] Checking configuration...")
+
+    # Add scripts dir to path for imports
+    script_dir = Path(__file__).parent
+    sys.path.insert(0, str(script_dir))
+
+    from _lib import Config
+
+    try:
+        # Try to get config from .env file
+        api_url = Config.get_api_url()
+        token = Config.get_token()
+    except ValueError as e:
+        print(f"\nConfiguration error: {e}")
+        sys.exit(1)
+
+    test_file_url = os.getenv("TEST_FILE_URL", "").strip()
+
+    if not test_file_url:
+        print("WARNING: TEST_FILE_URL is not set, using default test image")
+        # Use a default public test image (Chinese text)
+        test_file_url = "https://raw.githubusercontent.com/PaddlePaddle/PaddleOCR/release/2.7/doc/imgs/11.jpg"
+
+    print(f"  API_URL: {api_url}")
+    print(f"  PADDLE_OCR_TOKEN: {'*' * 8}{token[-4:] if len(token) > 4 else '****'}")
+    print(f"  TEST_FILE_URL: {test_file_url}")
+
+    # Run ocr_caller.py
+    print("\n[2/3] Running OCR on test file...")
+
+    script_dir = Path(__file__).parent
+    ocr_caller = script_dir / "ocr_caller.py"
+
+    cmd = [
+        sys.executable,
+        str(ocr_caller),
+        "--mode", "auto",
+        "--file-url", test_file_url,
+        "--max-attempts", "2",
+        "--budget-ms", "20000"
+    ]
+
+    try:
+        result = subprocess.run(
+            cmd,
+            capture_output=True,
+            text=True,
+            timeout=30
+        )
+    except subprocess.TimeoutExpired:
+        print("ERROR: OCR call timed out after 30 seconds")
+        sys.exit(4)
+    except Exception as e:
+        print(f"ERROR: Failed to run ocr_caller.py: {e}")
+        sys.exit(4)
+
+    # Parse output
+    print("\n[3/3] Validating response...")
+
+    if result.returncode != 0:
+        print(f"ERROR: ocr_caller.py exited with code {result.returncode}")
+        print("\nStderr:")
+        print(result.stderr)
+        print("\nStdout:")
+        print(result.stdout)
+        sys.exit(result.returncode)
+
+    try:
+        response = json.loads(result.stdout)
+    except json.JSONDecodeError as e:
+        print(f"ERROR: Failed to parse JSON response: {e}")
+        print("\nStdout:")
+        print(result.stdout)
+        sys.exit(4)
+
+    # Validate response structure
+    if not response.get("ok"):
+        error = response.get("error", {})
+        print(f"ERROR: OCR failed with error code: {error.get('code')}")
+        print(f"Message: {error.get('message')}")
+        print(f"\nFull response:\n{json.dumps(response, indent=2, ensure_ascii=False)}")
+        sys.exit(3)
+
+    # Check for content
+    result_data = response.get("result", {})
+    full_text = result_data.get("full_text", "")
+    pages = result_data.get("pages", [])
+
+    total_items = sum(len(page.get("items", [])) for page in pages)
+
+    if not full_text and total_items == 0:
+        print("WARNING: OCR succeeded but returned no text. This may indicate:")
+        print("  - The test image is blank or unreadable")
+        print("  - Provider API is working but returned empty results")
+        print(f"\nFull response:\n{json.dumps(response, indent=2, ensure_ascii=False)}")
+        # Still pass, as API is working
+    else:
+        print(f"SUCCESS: OCR completed")
+        print(f"  - Total text items: {total_items}")
+        print(f"  - Quality score: {response.get('quality', {}).get('quality_score', 0):.4f}")
+        print(f"  - Avg confidence: {response.get('quality', {}).get('avg_rec_score', 0):.4f}")
+        print(f"  - Mode: {response.get('agent_trace', {}).get('mode')}")
+        print(f"  - Attempts: {len(response.get('agent_trace', {}).get('attempts', []))}")
+
+        # Print first 200 chars of text
+        if full_text:
+            preview = full_text[:200].replace("\n", " ")
+            if len(full_text) > 200:
+                preview += "..."
+            print(f"\n  Preview: {preview}")
+
+    print("\n" + "=" * 60)
+    print("Smoke test PASSED")
+    print("=" * 60)
+    sys.exit(0)
+
+
+if __name__ == "__main__":
+    main()