Spaces:
Sleeping
Sleeping
Update main.py
Browse files
main.py
CHANGED
|
@@ -11,7 +11,6 @@ from sqlalchemy.exc import IntegrityError
|
|
| 11 |
from thefuzz import process, fuzz
|
| 12 |
from werkzeug.utils import secure_filename
|
| 13 |
import tempfile
|
| 14 |
-
import stat
|
| 15 |
|
| 16 |
# ───────────────────────────────────────────────────────────────────────────────
|
| 17 |
# CONFIGURATION
|
|
@@ -23,95 +22,89 @@ log = logging.getLogger("product-pipeline-api")
|
|
| 23 |
app = Flask(__name__)
|
| 24 |
CORS(app)
|
| 25 |
|
| 26 |
-
# ---
|
| 27 |
-
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
|
| 31 |
-
|
| 32 |
-
|
| 33 |
-
# Standard data directory
|
| 34 |
-
os.path.join('data', 'products.db'),
|
| 35 |
-
# Temp directory (usually writable)
|
| 36 |
-
os.path.join(tempfile.gettempdir(), 'products.db'),
|
| 37 |
-
# Current working directory
|
| 38 |
-
'products.db',
|
| 39 |
-
# User home directory
|
| 40 |
-
os.path.expanduser('~/products.db'),
|
| 41 |
-
# In-memory database as last resort
|
| 42 |
-
':memory:'
|
| 43 |
-
]
|
| 44 |
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
|
| 51 |
-
|
| 52 |
-
|
| 53 |
-
|
| 54 |
-
|
| 55 |
-
|
| 56 |
-
# Test write permissions
|
| 57 |
-
test_file = db_path + '.test'
|
| 58 |
-
with open(test_file, 'w') as f:
|
| 59 |
-
f.write('test')
|
| 60 |
-
os.remove(test_file)
|
| 61 |
-
|
| 62 |
-
log.info(f"Selected database path: {os.path.abspath(db_path)}")
|
| 63 |
-
return db_path
|
| 64 |
-
|
| 65 |
-
except (OSError, PermissionError) as e:
|
| 66 |
-
log.warning(f"Cannot use database path {db_path}: {e}")
|
| 67 |
-
continue
|
| 68 |
|
| 69 |
-
#
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 70 |
return ':memory:'
|
| 71 |
|
| 72 |
-
|
| 73 |
-
DB_PATH = get_database_path()
|
| 74 |
app.config['SQLALCHEMY_DATABASE_URI'] = f'sqlite:///{DB_PATH}'
|
| 75 |
app.config['SQLALCHEMY_TRACK_MODIFICATIONS'] = False
|
| 76 |
|
| 77 |
-
# ---
|
| 78 |
-
def
|
| 79 |
-
"""
|
| 80 |
-
|
| 81 |
-
'uploads'
|
| 82 |
-
os.
|
| 83 |
-
|
| 84 |
-
|
| 85 |
-
|
| 86 |
-
|
| 87 |
-
|
| 88 |
-
|
| 89 |
-
|
| 90 |
-
|
| 91 |
-
with open(test_file, 'w') as f:
|
| 92 |
-
f.write('test')
|
| 93 |
-
os.remove(test_file)
|
| 94 |
-
log.info(f"Using upload folder: {os.path.abspath(folder)}")
|
| 95 |
-
return folder
|
| 96 |
-
except (OSError, PermissionError) as e:
|
| 97 |
-
log.warning(f"Cannot use upload folder {folder}: {e}")
|
| 98 |
-
continue
|
| 99 |
-
|
| 100 |
-
# Fallback to temp directory
|
| 101 |
-
return tempfile.gettempdir()
|
| 102 |
|
| 103 |
-
app.config['UPLOAD_FOLDER'] =
|
| 104 |
|
| 105 |
# --- File Upload Configuration ---
|
| 106 |
ALLOWED_EXTENSIONS = {'csv', 'xls', 'xlsx'}
|
| 107 |
|
| 108 |
-
# ---
|
|
|
|
| 109 |
try:
|
| 110 |
db = SQLAlchemy(app)
|
| 111 |
log.info("SQLAlchemy initialized successfully")
|
| 112 |
except Exception as e:
|
| 113 |
log.error(f"Failed to initialize SQLAlchemy: {e}")
|
| 114 |
-
|
|
|
|
|
|
|
|
|
|
| 115 |
|
| 116 |
# ───────────────────────────────────────────────────────────────────────────────
|
| 117 |
# DATABASE MODEL (Based on products-20.sql)
|
|
@@ -139,47 +132,6 @@ class Product(db.Model):
|
|
| 139 |
def __repr__(self):
|
| 140 |
return f'<Product {self.id}: {self.name}>'
|
| 141 |
|
| 142 |
-
# ───────────────────────────────────────────────────────────────────────────────
|
| 143 |
-
# DATABASE INITIALIZATION WITH ROBUST ERROR HANDLING
|
| 144 |
-
# ───────────────────────────────────────────────────────────────────────────────
|
| 145 |
-
|
| 146 |
-
def initialize_database():
|
| 147 |
-
"""Initialize database with comprehensive error handling."""
|
| 148 |
-
max_retries = 3
|
| 149 |
-
retry_count = 0
|
| 150 |
-
|
| 151 |
-
while retry_count < max_retries:
|
| 152 |
-
try:
|
| 153 |
-
with app.app_context():
|
| 154 |
-
# Test database connection first
|
| 155 |
-
db.engine.execute('SELECT 1')
|
| 156 |
-
log.info("Database connection test successful")
|
| 157 |
-
|
| 158 |
-
# Create all tables
|
| 159 |
-
db.create_all()
|
| 160 |
-
log.info("Database tables created successfully")
|
| 161 |
-
return True
|
| 162 |
-
|
| 163 |
-
except Exception as e:
|
| 164 |
-
retry_count += 1
|
| 165 |
-
log.error(f"Database initialization attempt {retry_count} failed: {e}")
|
| 166 |
-
|
| 167 |
-
if retry_count < max_retries:
|
| 168 |
-
log.info(f"Retrying database initialization... (attempt {retry_count + 1}/{max_retries})")
|
| 169 |
-
|
| 170 |
-
# Try fallback to in-memory database
|
| 171 |
-
if retry_count == max_retries - 1:
|
| 172 |
-
log.warning("Falling back to in-memory database")
|
| 173 |
-
app.config['SQLALCHEMY_DATABASE_URI'] = 'sqlite:///:memory:'
|
| 174 |
-
# Recreate the database instance
|
| 175 |
-
global db
|
| 176 |
-
db = SQLAlchemy(app)
|
| 177 |
-
else:
|
| 178 |
-
log.error("All database initialization attempts failed")
|
| 179 |
-
return False
|
| 180 |
-
|
| 181 |
-
return False
|
| 182 |
-
|
| 183 |
# ───────────────────────────────────────────────────────────────────────────────
|
| 184 |
# DATA LOADING & PRE-PROCESSING
|
| 185 |
# ───────────────────────────────────────────────────────────────────────────────
|
|
@@ -190,86 +142,51 @@ EXISTING_PRODUCT_NAMES = []
|
|
| 190 |
HS_CODE_DESCRIPTIONS = {}
|
| 191 |
|
| 192 |
def parse_hs_codes_pdf(filepath='HS Codes for use under FDMS.pdf'):
|
| 193 |
-
"
|
| 194 |
-
log.info(f"Attempting to parse HS Codes from '{filepath}'...")
|
| 195 |
if not os.path.exists(filepath):
|
| 196 |
log.warning(f"HS Code PDF not found at '{filepath}'. Categorization will be limited.")
|
| 197 |
return []
|
| 198 |
-
|
| 199 |
codes = []
|
| 200 |
try:
|
| 201 |
with pdfplumber.open(filepath) as pdf:
|
| 202 |
-
for
|
| 203 |
-
|
| 204 |
-
|
| 205 |
-
if not text:
|
| 206 |
-
continue
|
| 207 |
-
|
| 208 |
-
# Improved regex to handle variations in PDF formatting
|
| 209 |
-
matches = re.findall(r'\"(\d{8})\"\s*,\s*\"(.*?)\"', text, re.DOTALL)
|
| 210 |
-
for code, desc in matches:
|
| 211 |
-
clean_desc = desc.replace('\n', ' ').strip()
|
| 212 |
-
if code and clean_desc:
|
| 213 |
-
codes.append({'code': code, 'description': clean_desc})
|
| 214 |
-
HS_CODE_DESCRIPTIONS[clean_desc] = code
|
| 215 |
-
|
| 216 |
-
except Exception as e:
|
| 217 |
-
log.warning(f"Error processing page {page_num + 1}: {e}")
|
| 218 |
continue
|
| 219 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 220 |
except Exception as e:
|
| 221 |
-
log.error(f"Failed to parse PDF
|
| 222 |
-
|
| 223 |
-
log.info(f"Successfully parsed {len(codes)} HS codes from PDF.")
|
| 224 |
return codes
|
| 225 |
|
| 226 |
def load_existing_products(filepath='Product List.csv'):
|
| 227 |
-
"
|
| 228 |
-
log.info(f"Attempting to load master product list from '{filepath}'...")
|
| 229 |
if not os.path.exists(filepath):
|
| 230 |
-
log.warning(f"Master product list not found at '{filepath}'.
|
| 231 |
return []
|
| 232 |
-
|
| 233 |
try:
|
| 234 |
-
# Try
|
| 235 |
-
|
| 236 |
-
df = None
|
| 237 |
-
|
| 238 |
-
for encoding in encodings:
|
| 239 |
try:
|
| 240 |
-
|
| 241 |
-
|
| 242 |
-
|
|
|
|
|
|
|
| 243 |
except UnicodeDecodeError:
|
| 244 |
continue
|
| 245 |
-
|
| 246 |
-
|
| 247 |
-
|
| 248 |
-
return []
|
| 249 |
-
|
| 250 |
-
# Flexible column detection - try different approaches
|
| 251 |
-
product_names = []
|
| 252 |
-
|
| 253 |
-
if len(df.columns) >= 2:
|
| 254 |
-
# Use second column if available
|
| 255 |
-
product_names = df.iloc[:, 1].dropna().astype(str).unique().tolist()
|
| 256 |
-
elif 'name' in df.columns:
|
| 257 |
-
product_names = df['name'].dropna().astype(str).unique().tolist()
|
| 258 |
-
elif 'product' in df.columns:
|
| 259 |
-
product_names = df['product'].dropna().astype(str).unique().tolist()
|
| 260 |
-
else:
|
| 261 |
-
# Use first column as fallback
|
| 262 |
-
product_names = df.iloc[:, 0].dropna().astype(str).unique().tolist()
|
| 263 |
-
|
| 264 |
-
# Clean product names
|
| 265 |
-
product_names = [name.strip() for name in product_names if name.strip()]
|
| 266 |
-
|
| 267 |
-
log.info(f"Loaded {len(product_names)} unique existing products from CSV.")
|
| 268 |
-
return product_names
|
| 269 |
-
|
| 270 |
except Exception as e:
|
| 271 |
log.error(f"Failed to load master product list: {e}")
|
| 272 |
-
|
| 273 |
|
| 274 |
# ───────────────────────────────────────────────────────────────────────────────
|
| 275 |
# CORE PROCESSING PIPELINE
|
|
@@ -282,127 +199,110 @@ def process_uploaded_file(filepath, filename):
|
|
| 282 |
"processed": 0, "added": 0, "updated": 0, "skipped_duplicates": 0,
|
| 283 |
"errors": [], "processed_data": []
|
| 284 |
}
|
| 285 |
-
|
|
|
|
| 286 |
try:
|
| 287 |
-
# Read file with robust error handling
|
| 288 |
-
df = None
|
| 289 |
file_ext = filename.rsplit('.', 1)[1].lower() if '.' in filename else ''
|
| 290 |
|
| 291 |
-
# Try
|
| 292 |
-
|
| 293 |
-
|
| 294 |
-
|
| 295 |
-
|
| 296 |
-
|
| 297 |
-
df = pd.read_csv(filepath, encoding=encoding, header=None)
|
| 298 |
-
elif file_ext in ['xls', 'xlsx']:
|
| 299 |
-
df = pd.read_excel(filepath, header=None, engine='openpyxl')
|
| 300 |
-
|
| 301 |
-
if df is not None:
|
| 302 |
-
log.info(f"Successfully read file with {encoding if file_ext == 'csv' else 'excel'} format")
|
| 303 |
break
|
| 304 |
-
|
| 305 |
-
except Exception as e:
|
| 306 |
-
log.warning(f"Failed to read with {encoding}: {e}")
|
| 307 |
-
continue
|
| 308 |
-
|
| 309 |
-
if df is None or df.empty:
|
| 310 |
-
results['errors'].append("Could not read the uploaded file or file is empty")
|
| 311 |
-
return results
|
| 312 |
-
|
| 313 |
-
# Flexible column detection
|
| 314 |
-
product_column_idx = None
|
| 315 |
-
if len(df.columns) >= 2:
|
| 316 |
-
product_column_idx = 1 # Second column
|
| 317 |
-
else:
|
| 318 |
-
product_column_idx = 0 # First column as fallback
|
| 319 |
-
|
| 320 |
-
log.info(f"Using column index {product_column_idx} for product names")
|
| 321 |
-
|
| 322 |
-
for index, row in df.iterrows():
|
| 323 |
-
try:
|
| 324 |
-
raw_name = row.iloc[product_column_idx] if len(row) > product_column_idx else None
|
| 325 |
-
results['processed'] += 1
|
| 326 |
-
|
| 327 |
-
if pd.isna(raw_name) or not str(raw_name).strip():
|
| 328 |
continue
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 329 |
|
| 330 |
-
|
| 331 |
-
|
| 332 |
-
|
| 333 |
-
|
| 334 |
-
|
| 335 |
-
|
| 336 |
-
|
| 337 |
-
|
| 338 |
-
|
| 339 |
-
|
| 340 |
-
|
| 341 |
-
|
| 342 |
-
|
| 343 |
-
|
| 344 |
-
|
| 345 |
-
|
| 346 |
-
|
| 347 |
-
|
| 348 |
-
|
| 349 |
-
|
| 350 |
-
|
| 351 |
-
|
| 352 |
-
|
| 353 |
-
|
| 354 |
-
|
| 355 |
-
|
| 356 |
-
processed_entry = {
|
| 357 |
-
"raw_name": str(raw_name),
|
| 358 |
-
"cleaned_name": validated_name,
|
| 359 |
-
"hs_code": hs_code,
|
| 360 |
-
"primary_category": best_hs_desc or "N/A",
|
| 361 |
-
"status": ""
|
| 362 |
-
}
|
| 363 |
-
|
| 364 |
-
# Database operations with error handling
|
| 365 |
try:
|
| 366 |
-
|
| 367 |
-
|
| 368 |
-
|
| 369 |
-
existing_product.hs_code = hs_code
|
| 370 |
-
existing_product.primary_category = best_hs_desc or "N/A"
|
| 371 |
-
db.session.commit()
|
| 372 |
-
results['updated'] += 1
|
| 373 |
-
processed_entry['status'] = 'Updated'
|
| 374 |
-
else:
|
| 375 |
-
results['skipped_duplicates'] += 1
|
| 376 |
-
processed_entry['status'] = 'Skipped (Duplicate)'
|
| 377 |
-
else:
|
| 378 |
-
new_product = Product(
|
| 379 |
-
name=validated_name,
|
| 380 |
-
hs_code=hs_code,
|
| 381 |
-
primary_category=best_hs_desc or 'N/A'
|
| 382 |
-
)
|
| 383 |
-
db.session.add(new_product)
|
| 384 |
-
db.session.commit()
|
| 385 |
-
results['added'] += 1
|
| 386 |
-
processed_entry['status'] = 'Added'
|
| 387 |
-
|
| 388 |
-
results['processed_data'].append(processed_entry)
|
| 389 |
-
|
| 390 |
except Exception as e:
|
| 391 |
-
|
| 392 |
-
|
| 393 |
-
|
| 394 |
-
|
| 395 |
-
|
| 396 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 397 |
except Exception as e:
|
| 398 |
-
|
| 399 |
-
|
| 400 |
-
|
| 401 |
-
|
| 402 |
-
|
| 403 |
-
|
| 404 |
-
|
| 405 |
-
|
|
|
|
|
|
|
| 406 |
return results
|
| 407 |
|
| 408 |
# ───────────────────────────────────────────────────────────────────────────────
|
|
@@ -417,35 +317,13 @@ def root():
|
|
| 417 |
return jsonify({
|
| 418 |
"ok": True,
|
| 419 |
"message": "The Product Validation server is running.",
|
| 420 |
-
"
|
| 421 |
-
"upload_folder": app.config['UPLOAD_FOLDER']
|
| 422 |
})
|
| 423 |
|
| 424 |
-
@app.get("/api/health")
|
| 425 |
-
def health_check():
|
| 426 |
-
"""Health check endpoint."""
|
| 427 |
-
try:
|
| 428 |
-
# Test database connection
|
| 429 |
-
with app.app_context():
|
| 430 |
-
db.engine.execute('SELECT 1')
|
| 431 |
-
|
| 432 |
-
return jsonify({
|
| 433 |
-
"ok": True,
|
| 434 |
-
"database": "connected",
|
| 435 |
-
"products_count": Product.query.count()
|
| 436 |
-
})
|
| 437 |
-
except Exception as e:
|
| 438 |
-
return jsonify({
|
| 439 |
-
"ok": False,
|
| 440 |
-
"database": "disconnected",
|
| 441 |
-
"error": str(e)
|
| 442 |
-
}), 500
|
| 443 |
-
|
| 444 |
@app.post("/api/upload")
|
| 445 |
def upload_products():
|
| 446 |
if 'file' not in request.files:
|
| 447 |
return jsonify({"ok": False, "error": "No file part in the request"}), 400
|
| 448 |
-
|
| 449 |
file = request.files['file']
|
| 450 |
if file.filename == '':
|
| 451 |
return jsonify({"ok": False, "error": "No file selected"}), 400
|
|
@@ -455,32 +333,20 @@ def upload_products():
|
|
| 455 |
filename = secure_filename(file.filename)
|
| 456 |
filepath = os.path.join(app.config['UPLOAD_FOLDER'], filename)
|
| 457 |
file.save(filepath)
|
| 458 |
-
|
| 459 |
results = process_uploaded_file(filepath, filename)
|
| 460 |
|
| 461 |
# Clean up uploaded file
|
| 462 |
try:
|
| 463 |
os.remove(filepath)
|
| 464 |
except:
|
| 465 |
-
pass # Non-critical
|
| 466 |
|
| 467 |
-
return jsonify({
|
| 468 |
-
"ok": True,
|
| 469 |
-
"message": "File processed successfully",
|
| 470 |
-
"results": results
|
| 471 |
-
})
|
| 472 |
-
|
| 473 |
except Exception as e:
|
| 474 |
-
log.error(f"Upload
|
| 475 |
-
return jsonify({
|
| 476 |
-
"ok": False,
|
| 477 |
-
"error": f"File processing failed: {str(e)}"
|
| 478 |
-
}), 500
|
| 479 |
|
| 480 |
-
return jsonify({
|
| 481 |
-
"ok": False,
|
| 482 |
-
"error": f"Invalid file type. Allowed types are: {', '.join(ALLOWED_EXTENSIONS)}"
|
| 483 |
-
}), 400
|
| 484 |
|
| 485 |
@app.get("/api/products")
|
| 486 |
def get_products():
|
|
@@ -489,41 +355,57 @@ def get_products():
|
|
| 489 |
all_products = Product.query.all()
|
| 490 |
products_list = [product.to_dict() for product in all_products]
|
| 491 |
log.info(f"Successfully retrieved {len(products_list)} products.")
|
| 492 |
-
return jsonify({
|
| 493 |
-
"ok": True,
|
| 494 |
-
"count": len(products_list),
|
| 495 |
-
"products": products_list
|
| 496 |
-
})
|
| 497 |
except Exception as e:
|
| 498 |
log.error(f"Could not retrieve products from database: {e}")
|
| 499 |
-
return jsonify({
|
| 500 |
-
"ok": False,
|
| 501 |
-
"error": f"Failed to retrieve products: {str(e)}"
|
| 502 |
-
}), 500
|
| 503 |
|
| 504 |
# ───────────────────────────────────────────────────────────────────────────────
|
| 505 |
# MAIN (Server Initialization)
|
| 506 |
# ───────────────────────────────────────────────────────────────────────────────
|
| 507 |
|
| 508 |
if __name__ == "__main__":
|
| 509 |
-
log.info("===== Application Startup
|
| 510 |
-
|
| 511 |
-
# Initialize database with error handling
|
| 512 |
-
if not initialize_database():
|
| 513 |
-
log.error("Failed to initialize database. Server may not function properly.")
|
| 514 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 515 |
# Load supporting data (non-critical)
|
| 516 |
try:
|
| 517 |
-
log.info("Loading supporting data...")
|
| 518 |
HS_CODES_DATA = parse_hs_codes_pdf()
|
| 519 |
EXISTING_PRODUCT_NAMES = load_existing_products()
|
| 520 |
log.info("Supporting data loaded successfully")
|
| 521 |
except Exception as e:
|
| 522 |
log.warning(f"Failed to load supporting data: {e}")
|
| 523 |
log.info("Server will continue with limited functionality")
|
| 524 |
-
|
| 525 |
-
log.info(f"Server
|
| 526 |
-
|
| 527 |
-
|
| 528 |
port = int(os.environ.get("PORT", "7860"))
|
| 529 |
app.run(host="0.0.0.0", port=port, debug=False)
|
|
|
|
| 11 |
from thefuzz import process, fuzz
|
| 12 |
from werkzeug.utils import secure_filename
|
| 13 |
import tempfile
|
|
|
|
| 14 |
|
| 15 |
# ───────────────────────────────────────────────────────────────────────────────
|
| 16 |
# CONFIGURATION
|
|
|
|
| 22 |
app = Flask(__name__)
|
| 23 |
CORS(app)
|
| 24 |
|
| 25 |
+
# --- App Configuration ---
|
| 26 |
+
# --- ROBUST DATABASE PATH HANDLING ---
|
| 27 |
+
def setup_database_path():
|
| 28 |
+
"""Setup database path with fallbacks for Hugging Face Spaces."""
|
| 29 |
+
# Try original path first
|
| 30 |
+
DB_FOLDER = 'data'
|
| 31 |
+
DB_PATH = os.path.join(DB_FOLDER, 'products.db')
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
|
| 33 |
+
try:
|
| 34 |
+
os.makedirs(DB_FOLDER, exist_ok=True)
|
| 35 |
+
# Test if we can write to this directory
|
| 36 |
+
test_file = os.path.join(DB_FOLDER, 'test.tmp')
|
| 37 |
+
with open(test_file, 'w') as f:
|
| 38 |
+
f.write('test')
|
| 39 |
+
os.remove(test_file)
|
| 40 |
+
log.info(f"Using database path: {DB_PATH}")
|
| 41 |
+
return DB_PATH
|
| 42 |
+
except (OSError, PermissionError):
|
| 43 |
+
log.warning(f"Cannot write to {DB_FOLDER}, trying fallbacks...")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
|
| 45 |
+
# Fallback 1: Use temp directory
|
| 46 |
+
try:
|
| 47 |
+
temp_db_path = os.path.join(tempfile.gettempdir(), 'products.db')
|
| 48 |
+
test_file = os.path.join(tempfile.gettempdir(), 'test.tmp')
|
| 49 |
+
with open(test_file, 'w') as f:
|
| 50 |
+
f.write('test')
|
| 51 |
+
os.remove(test_file)
|
| 52 |
+
log.info(f"Using temp database path: {temp_db_path}")
|
| 53 |
+
return temp_db_path
|
| 54 |
+
except (OSError, PermissionError):
|
| 55 |
+
log.warning("Cannot write to temp directory, trying current directory...")
|
| 56 |
+
|
| 57 |
+
# Fallback 2: Current directory
|
| 58 |
+
try:
|
| 59 |
+
current_db_path = 'products.db'
|
| 60 |
+
test_file = 'test.tmp'
|
| 61 |
+
with open(test_file, 'w') as f:
|
| 62 |
+
f.write('test')
|
| 63 |
+
os.remove(test_file)
|
| 64 |
+
log.info(f"Using current directory database path: {current_db_path}")
|
| 65 |
+
return current_db_path
|
| 66 |
+
except (OSError, PermissionError):
|
| 67 |
+
log.warning("Cannot write to current directory, using in-memory database...")
|
| 68 |
+
|
| 69 |
+
# Final fallback: In-memory database
|
| 70 |
+
log.warning("Using in-memory database - data will not persist!")
|
| 71 |
return ':memory:'
|
| 72 |
|
| 73 |
+
DB_PATH = setup_database_path()
|
|
|
|
| 74 |
app.config['SQLALCHEMY_DATABASE_URI'] = f'sqlite:///{DB_PATH}'
|
| 75 |
app.config['SQLALCHEMY_TRACK_MODIFICATIONS'] = False
|
| 76 |
|
| 77 |
+
# --- ROBUST UPLOAD FOLDER HANDLING ---
|
| 78 |
+
def setup_upload_folder():
|
| 79 |
+
"""Setup upload folder with fallbacks."""
|
| 80 |
+
try:
|
| 81 |
+
upload_folder = 'uploads'
|
| 82 |
+
os.makedirs(upload_folder, exist_ok=True)
|
| 83 |
+
test_file = os.path.join(upload_folder, 'test.tmp')
|
| 84 |
+
with open(test_file, 'w') as f:
|
| 85 |
+
f.write('test')
|
| 86 |
+
os.remove(test_file)
|
| 87 |
+
return upload_folder
|
| 88 |
+
except (OSError, PermissionError):
|
| 89 |
+
log.warning("Cannot create uploads folder, using temp directory")
|
| 90 |
+
return tempfile.gettempdir()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 91 |
|
| 92 |
+
app.config['UPLOAD_FOLDER'] = setup_upload_folder()
|
| 93 |
|
| 94 |
# --- File Upload Configuration ---
|
| 95 |
ALLOWED_EXTENSIONS = {'csv', 'xls', 'xlsx'}
|
| 96 |
|
| 97 |
+
# --- ROBUST DATABASE INITIALIZATION ---
|
| 98 |
+
db = None
|
| 99 |
try:
|
| 100 |
db = SQLAlchemy(app)
|
| 101 |
log.info("SQLAlchemy initialized successfully")
|
| 102 |
except Exception as e:
|
| 103 |
log.error(f"Failed to initialize SQLAlchemy: {e}")
|
| 104 |
+
# Try in-memory fallback
|
| 105 |
+
app.config['SQLALCHEMY_DATABASE_URI'] = 'sqlite:///:memory:'
|
| 106 |
+
db = SQLAlchemy(app)
|
| 107 |
+
log.warning("Fell back to in-memory database")
|
| 108 |
|
| 109 |
# ───────────────────────────────────────────────────────────────────────────────
|
| 110 |
# DATABASE MODEL (Based on products-20.sql)
|
|
|
|
| 132 |
def __repr__(self):
|
| 133 |
return f'<Product {self.id}: {self.name}>'
|
| 134 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 135 |
# ───────────────────────────────────────────────────────────────────────────────
|
| 136 |
# DATA LOADING & PRE-PROCESSING
|
| 137 |
# ───────────────────────────────────────────────────────────────────────────────
|
|
|
|
| 142 |
HS_CODE_DESCRIPTIONS = {}
|
| 143 |
|
| 144 |
def parse_hs_codes_pdf(filepath='HS Codes for use under FDMS.pdf'):
|
| 145 |
+
log.info(f"Parsing HS Codes from '{filepath}'...")
|
|
|
|
| 146 |
if not os.path.exists(filepath):
|
| 147 |
log.warning(f"HS Code PDF not found at '{filepath}'. Categorization will be limited.")
|
| 148 |
return []
|
|
|
|
| 149 |
codes = []
|
| 150 |
try:
|
| 151 |
with pdfplumber.open(filepath) as pdf:
|
| 152 |
+
for page in pdf.pages:
|
| 153 |
+
text = page.extract_text()
|
| 154 |
+
if not text:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 155 |
continue
|
| 156 |
+
# Improved regex to handle variations in PDF formatting
|
| 157 |
+
matches = re.findall(r'\"(\d{8})\"\s*,\s*\"(.*?)\"', text, re.DOTALL)
|
| 158 |
+
for code, desc in matches:
|
| 159 |
+
clean_desc = desc.replace('\n', ' ').strip()
|
| 160 |
+
if code and clean_desc:
|
| 161 |
+
codes.append({'code': code, 'description': clean_desc})
|
| 162 |
+
HS_CODE_DESCRIPTIONS[clean_desc] = code
|
| 163 |
except Exception as e:
|
| 164 |
+
log.error(f"Failed to parse PDF: {e}")
|
| 165 |
+
log.info(f"Successfully parsed {len(codes)} HS codes.")
|
|
|
|
| 166 |
return codes
|
| 167 |
|
| 168 |
def load_existing_products(filepath='Product List.csv'):
|
| 169 |
+
log.info(f"Loading master product list from '{filepath}'...")
|
|
|
|
| 170 |
if not os.path.exists(filepath):
|
| 171 |
+
log.warning(f"Master product list not found at '{filepath}'. Validation may be limited.")
|
| 172 |
return []
|
|
|
|
| 173 |
try:
|
| 174 |
+
# Try multiple encodings
|
| 175 |
+
for encoding in ['utf-8', 'latin-1', 'cp1252']:
|
|
|
|
|
|
|
|
|
|
| 176 |
try:
|
| 177 |
+
# Based on the CSV structure, the 'name' is in the second column.
|
| 178 |
+
df = pd.read_csv(filepath, usecols=[1], names=['name'], header=0, encoding=encoding)
|
| 179 |
+
product_names = df['name'].dropna().astype(str).unique().tolist()
|
| 180 |
+
log.info(f"Loaded {len(product_names)} unique existing products with {encoding} encoding.")
|
| 181 |
+
return product_names
|
| 182 |
except UnicodeDecodeError:
|
| 183 |
continue
|
| 184 |
+
except Exception as e:
|
| 185 |
+
log.warning(f"Error with {encoding} encoding: {e}")
|
| 186 |
+
continue
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 187 |
except Exception as e:
|
| 188 |
log.error(f"Failed to load master product list: {e}")
|
| 189 |
+
return []
|
| 190 |
|
| 191 |
# ───────────────────────────────────────────────────────────────────────────────
|
| 192 |
# CORE PROCESSING PIPELINE
|
|
|
|
| 199 |
"processed": 0, "added": 0, "updated": 0, "skipped_duplicates": 0,
|
| 200 |
"errors": [], "processed_data": []
|
| 201 |
}
|
| 202 |
+
df = None
|
| 203 |
+
|
| 204 |
try:
|
|
|
|
|
|
|
| 205 |
file_ext = filename.rsplit('.', 1)[1].lower() if '.' in filename else ''
|
| 206 |
|
| 207 |
+
# Try multiple encodings for CSV files
|
| 208 |
+
if file_ext == 'csv':
|
| 209 |
+
for encoding in ['utf-8', 'latin-1', 'cp1252']:
|
| 210 |
+
try:
|
| 211 |
+
df = pd.read_csv(filepath, header=None, usecols=[1], names=['product_name'], encoding=encoding)
|
| 212 |
+
log.info(f"Successfully read CSV with {encoding} encoding")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 213 |
break
|
| 214 |
+
except (UnicodeDecodeError, ValueError):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 215 |
continue
|
| 216 |
+
except Exception as e:
|
| 217 |
+
log.warning(f"Error reading CSV with {encoding}: {e}")
|
| 218 |
+
continue
|
| 219 |
+
elif file_ext in ['xls', 'xlsx']:
|
| 220 |
+
try:
|
| 221 |
+
df = pd.read_excel(filepath, header=None, usecols=[1], names=['product_name'], engine='openpyxl')
|
| 222 |
+
except Exception as e:
|
| 223 |
+
log.error(f"Error reading Excel file: {e}")
|
| 224 |
|
| 225 |
+
if df is None:
|
| 226 |
+
results['errors'].append("Could not read the uploaded file with any supported encoding.")
|
| 227 |
+
return results
|
| 228 |
+
|
| 229 |
+
except Exception as e:
|
| 230 |
+
log.error(f"Could not read the uploaded file: {e}")
|
| 231 |
+
results['errors'].append(f"Invalid file format or corrupt file: {e}")
|
| 232 |
+
return results
|
| 233 |
+
|
| 234 |
+
if df.empty:
|
| 235 |
+
results['errors'].append("The uploaded file is empty.")
|
| 236 |
+
return results
|
| 237 |
+
|
| 238 |
+
for index, row in df.iterrows():
|
| 239 |
+
try:
|
| 240 |
+
raw_name = row['product_name']
|
| 241 |
+
results['processed'] += 1
|
| 242 |
+
|
| 243 |
+
if pd.isna(raw_name) or not str(raw_name).strip():
|
| 244 |
+
continue
|
| 245 |
+
|
| 246 |
+
cleaned_name = str(raw_name).strip()
|
| 247 |
+
|
| 248 |
+
# Fuzzy matching with error handling
|
| 249 |
+
best_match, score = (cleaned_name, 100)
|
| 250 |
+
if EXISTING_PRODUCT_NAMES:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 251 |
try:
|
| 252 |
+
best_match, score = process.extractOne(
|
| 253 |
+
cleaned_name, EXISTING_PRODUCT_NAMES, scorer=fuzz.token_sort_ratio
|
| 254 |
+
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 255 |
except Exception as e:
|
| 256 |
+
log.warning(f"Fuzzy matching failed for '{cleaned_name}': {e}")
|
| 257 |
+
|
| 258 |
+
validated_name = best_match if score >= FUZZY_MATCH_THRESHOLD else cleaned_name
|
| 259 |
+
|
| 260 |
+
# HS Code matching with error handling
|
| 261 |
+
best_hs_desc, hs_code = None, None
|
| 262 |
+
if HS_CODE_DESCRIPTIONS:
|
| 263 |
+
try:
|
| 264 |
+
best_hs_desc, _ = process.extractOne(
|
| 265 |
+
validated_name, list(HS_CODE_DESCRIPTIONS.keys())
|
| 266 |
+
)
|
| 267 |
+
hs_code = HS_CODE_DESCRIPTIONS.get(best_hs_desc)
|
| 268 |
+
except Exception as e:
|
| 269 |
+
log.warning(f"HS code matching failed for '{validated_name}': {e}")
|
| 270 |
+
|
| 271 |
+
processed_entry = {
|
| 272 |
+
"raw_name": str(raw_name), "cleaned_name": validated_name, "hs_code": hs_code,
|
| 273 |
+
"primary_category": best_hs_desc or "N/A", "status": ""
|
| 274 |
+
}
|
| 275 |
+
|
| 276 |
+
try:
|
| 277 |
+
existing_product = Product.query.filter_by(name=validated_name).first()
|
| 278 |
+
if existing_product:
|
| 279 |
+
if hs_code and existing_product.hs_code != hs_code:
|
| 280 |
+
existing_product.hs_code = hs_code
|
| 281 |
+
existing_product.primary_category = best_hs_desc or "N/A"
|
| 282 |
+
db.session.commit()
|
| 283 |
+
results['updated'] += 1
|
| 284 |
+
processed_entry['status'] = 'Updated'
|
| 285 |
+
else:
|
| 286 |
+
results['skipped_duplicates'] += 1
|
| 287 |
+
processed_entry['status'] = 'Skipped (Duplicate)'
|
| 288 |
+
else:
|
| 289 |
+
new_product = Product(name=validated_name, hs_code=hs_code, primary_category=best_hs_desc or 'N/A')
|
| 290 |
+
db.session.add(new_product)
|
| 291 |
+
db.session.commit()
|
| 292 |
+
results['added'] += 1
|
| 293 |
+
processed_entry['status'] = 'Added'
|
| 294 |
+
results['processed_data'].append(processed_entry)
|
| 295 |
except Exception as e:
|
| 296 |
+
db.session.rollback()
|
| 297 |
+
log.error(f"Database error for '{validated_name}': {e}")
|
| 298 |
+
results['errors'].append(f"DB Error on '{validated_name}': {e}")
|
| 299 |
+
processed_entry['status'] = 'Error'
|
| 300 |
+
results['processed_data'].append(processed_entry)
|
| 301 |
+
except Exception as e:
|
| 302 |
+
log.error(f"Error processing row {index}: {e}")
|
| 303 |
+
results['errors'].append(f"Row {index + 1} error: {str(e)}")
|
| 304 |
+
continue
|
| 305 |
+
|
| 306 |
return results
|
| 307 |
|
| 308 |
# ───────────────────────────────────────────────────────────────────────────────
|
|
|
|
| 317 |
return jsonify({
|
| 318 |
"ok": True,
|
| 319 |
"message": "The Product Validation server is running.",
|
| 320 |
+
"database": "in-memory" if DB_PATH == ':memory:' else "persistent"
|
|
|
|
| 321 |
})
|
| 322 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 323 |
@app.post("/api/upload")
|
| 324 |
def upload_products():
|
| 325 |
if 'file' not in request.files:
|
| 326 |
return jsonify({"ok": False, "error": "No file part in the request"}), 400
|
|
|
|
| 327 |
file = request.files['file']
|
| 328 |
if file.filename == '':
|
| 329 |
return jsonify({"ok": False, "error": "No file selected"}), 400
|
|
|
|
| 333 |
filename = secure_filename(file.filename)
|
| 334 |
filepath = os.path.join(app.config['UPLOAD_FOLDER'], filename)
|
| 335 |
file.save(filepath)
|
|
|
|
| 336 |
results = process_uploaded_file(filepath, filename)
|
| 337 |
|
| 338 |
# Clean up uploaded file
|
| 339 |
try:
|
| 340 |
os.remove(filepath)
|
| 341 |
except:
|
| 342 |
+
pass # Non-critical cleanup
|
| 343 |
|
| 344 |
+
return jsonify({"ok": True, "message": "File processed successfully", "results": results})
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 345 |
except Exception as e:
|
| 346 |
+
log.error(f"Upload error: {e}")
|
| 347 |
+
return jsonify({"ok": False, "error": f"Upload failed: {str(e)}"}, 500)
|
|
|
|
|
|
|
|
|
|
| 348 |
|
| 349 |
+
return jsonify({"ok": False, "error": f"Invalid file type. Allowed types are: {', '.join(ALLOWED_EXTENSIONS)}"}), 400
|
|
|
|
|
|
|
|
|
|
| 350 |
|
| 351 |
@app.get("/api/products")
|
| 352 |
def get_products():
|
|
|
|
| 355 |
all_products = Product.query.all()
|
| 356 |
products_list = [product.to_dict() for product in all_products]
|
| 357 |
log.info(f"Successfully retrieved {len(products_list)} products.")
|
| 358 |
+
return jsonify({"ok": True, "count": len(products_list), "products": products_list})
|
|
|
|
|
|
|
|
|
|
|
|
|
| 359 |
except Exception as e:
|
| 360 |
log.error(f"Could not retrieve products from database: {e}")
|
| 361 |
+
return jsonify({"ok": False, "error": f"Database error: {str(e)}"}, 500)
|
|
|
|
|
|
|
|
|
|
| 362 |
|
| 363 |
# ───────────────────────────────────────────────────────────────────────────────
|
| 364 |
# MAIN (Server Initialization)
|
| 365 |
# ───────────────────────────────────────────────────────────────────────────────
|
| 366 |
|
| 367 |
if __name__ == "__main__":
|
| 368 |
+
log.info("===== Application Startup =====")
|
|
|
|
|
|
|
|
|
|
|
|
|
| 369 |
|
| 370 |
+
# Robust database initialization
|
| 371 |
+
max_retries = 3
|
| 372 |
+
for attempt in range(max_retries):
|
| 373 |
+
try:
|
| 374 |
+
with app.app_context():
|
| 375 |
+
log.info(f"Initializing server (attempt {attempt + 1})...")
|
| 376 |
+
db.create_all()
|
| 377 |
+
log.info("Database tables created successfully")
|
| 378 |
+
break
|
| 379 |
+
except Exception as e:
|
| 380 |
+
log.error(f"Database initialization failed (attempt {attempt + 1}): {e}")
|
| 381 |
+
if attempt == max_retries - 1:
|
| 382 |
+
log.error("All database initialization attempts failed, trying in-memory fallback")
|
| 383 |
+
try:
|
| 384 |
+
app.config['SQLALCHEMY_DATABASE_URI'] = 'sqlite:///:memory:'
|
| 385 |
+
# Reinitialize the database connection
|
| 386 |
+
from flask_sqlalchemy import SQLAlchemy
|
| 387 |
+
db.init_app(app)
|
| 388 |
+
with app.app_context():
|
| 389 |
+
db.create_all()
|
| 390 |
+
log.warning("Using in-memory database - data will not persist")
|
| 391 |
+
break
|
| 392 |
+
except Exception as fallback_error:
|
| 393 |
+
log.error(f"Even in-memory database failed: {fallback_error}")
|
| 394 |
+
raise
|
| 395 |
+
else:
|
| 396 |
+
import time
|
| 397 |
+
time.sleep(1) # Brief pause between retries
|
| 398 |
+
|
| 399 |
# Load supporting data (non-critical)
|
| 400 |
try:
|
|
|
|
| 401 |
HS_CODES_DATA = parse_hs_codes_pdf()
|
| 402 |
EXISTING_PRODUCT_NAMES = load_existing_products()
|
| 403 |
log.info("Supporting data loaded successfully")
|
| 404 |
except Exception as e:
|
| 405 |
log.warning(f"Failed to load supporting data: {e}")
|
| 406 |
log.info("Server will continue with limited functionality")
|
| 407 |
+
|
| 408 |
+
log.info(f"Server is ready. Database: {DB_PATH}")
|
| 409 |
+
|
|
|
|
| 410 |
port = int(os.environ.get("PORT", "7860"))
|
| 411 |
app.run(host="0.0.0.0", port=port, debug=False)
|