Spaces:
Runtime error
Runtime error
File size: 57,214 Bytes
a488104 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 | # app/main.py
import os
import uuid
import asyncio
import json
import pathlib
import ssl
import socket
import certifi
import requests
import urllib3
import time
from requests.adapters import HTTPAdapter
from urllib3.util.retry import Retry
from fastapi import FastAPI, UploadFile, File, BackgroundTasks, HTTPException
from fastapi.responses import JSONResponse, FileResponse
from fastapi.middleware.cors import CORSMiddleware
from pypdf import PdfReader
from pydantic import BaseModel
import motor.motor_asyncio
from bson.objectid import ObjectId
from pymongo.errors import ServerSelectionTimeoutError
import httpx
from bson import ObjectId
# Disable SSL warnings for requests
urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
# Load environment variables from .env file if it exists
env_path = pathlib.Path(__file__).parent.parent / '.env'
if env_path.exists():
print(f"Loading environment variables from {env_path}")
with open(env_path) as f:
for line in f:
line = line.strip()
if not line or line.startswith('#') or '=' not in line:
continue
key, value = line.split('=', 1)
os.environ[key] = value
# ----------------- Configuration -----------------
MONGO_URL = "mongodb://localhost:27017"
DB_NAME = "contracts_db"
COLLECTION_NAME = "contracts"
OPENROUTER_API_KEY = os.getenv("OPENROUTER_API_KEY") # set in .env
if not OPENROUTER_API_KEY:
print("WARNING: OPENROUTER_API_KEY environment variable is not set. API calls will fail.")
MISTRAL_MODEL = "mistralai/mistral-7b-instruct:free" # free model
UPLOAD_DIR = "uploads"
os.makedirs(UPLOAD_DIR, exist_ok=True)
# ----------------- Storage System -----------------
# Define a flag to track if MongoDB is available
USE_MONGO = True
# Create a directory for file-based storage as fallback
FILE_STORAGE_DIR = "file_storage"
os.makedirs(FILE_STORAGE_DIR, exist_ok=True)
# Try to connect to MongoDB
try:
# Set a short server selection timeout to fail fast if MongoDB is not available
client = motor.motor_asyncio.AsyncIOMotorClient(MONGO_URL, serverSelectionTimeoutMS=2000)
# Don't validate connection immediately - just set up the client
db = client[DB_NAME]
contracts_collection = db[COLLECTION_NAME]
print("✅ MongoDB client initialized - will test connection on first operation")
# We'll validate the connection on first use, not here
# This avoids blocking startup and allows for MongoDB to become available later
except Exception as e:
print(f"⚠️ MongoDB client initialization failed: {str(e)}")
print("⚠️ Falling back to file-based storage")
USE_MONGO = False
# Define dummy objects to prevent errors
client = None
db = None
contracts_collection = None
# ----------------- FastAPI App -----------------
app = FastAPI(
title="Contract Intelligence API",
# Increase maximum request size to handle large PDFs (100MB)
max_request_size=104857600
)
# Add CORS middleware
app.add_middleware(
CORSMiddleware,
allow_origins=["*"], # In production, replace with specific origins
allow_credentials=True,
allow_methods=["*"],
allow_headers=["*"],
)
# ----------------- Models -----------------
class ContractStatus(BaseModel):
status: str
progress: int
score: int | None = None
error: str | None = None
# ----------------- Helper Functions -----------------
async def resolve_hostname_to_ip(hostname: str) -> str:
"""Try to resolve hostname to IP address using different methods"""
try:
# Try standard resolution first
ip = socket.gethostbyname(hostname)
return ip
except socket.gaierror:
# If that fails, try some known IP addresses for common services
known_ips = {
"openrouter.ai": "104.18.6.192",
"api.openrouter.ai": "104.18.7.192",
"httpbin.org": "34.205.4.79"
}
return known_ips.get(hostname, hostname)
async def test_network_connectivity() -> bool:
"""Test basic network connectivity with multiple fallbacks"""
test_urls = [
"https://httpbin.org/get",
"https://8.8.8.8", # Google DNS
"https://1.1.1.1" # Cloudflare DNS
]
for url in test_urls:
try:
async with httpx.AsyncClient(timeout=10, verify=False) as client:
response = await client.get(url)
if response.status_code in [200, 301, 302]:
return True
except Exception:
continue
return False
async def extract_text_from_pdf(file_path: str) -> str:
"""Extract text from PDF with memory optimization"""
reader = PdfReader(file_path)
text = ""
# Process one page at a time to reduce memory usage
for page in reader.pages:
text += page.extract_text() or ""
# Free up memory by clearing page cache
page._objects = None
return text
def chunk_text(text: str, chunk_size: int = 8000, overlap: int = 500) -> list[str]:
"""Split text into chunks suitable for LLM processing"""
if len(text) <= chunk_size:
return [text]
chunks = []
start = 0
while start < len(text):
# Take a chunk of text
end = min(start + chunk_size, len(text))
# Try to find a good breaking point (newline or period)
if end < len(text):
# Look for a newline or period to break at
for break_char in ['\n\n', '\n', '. ', ', ', ' ']:
break_pos = text.rfind(break_char, start, end)
if break_pos > start + chunk_size // 2: # Ensure chunk isn't too small
end = break_pos + len(break_char)
break
chunks.append(text[start:end])
# Start next chunk with overlap for context
start = max(start, end - overlap)
return chunks
async def query_mistral_llm(contract_text: str) -> dict:
"""
Query the Mistral LLM via OpenRouter API with support for large documents"""
if not OPENROUTER_API_KEY:
raise ValueError("OPENROUTER_API_KEY environment variable is not set. Please set it before making API calls.")
# Check if text needs to be chunked (over 15K characters is likely to exceed token limits)
if len(contract_text) > 15000:
print(f"Contract text is large ({len(contract_text)} chars), chunking for processing")
return await process_large_document(contract_text)
else:
# Process normally for smaller documents
return await query_mistral_llm_single(contract_text)
async def process_large_document(contract_text: str) -> dict:
"""Process a large document by chunking and combining results"""
# Split text into manageable chunks
chunks = chunk_text(contract_text)
print(f"Split document into {len(chunks)} chunks for processing")
# Process each chunk to extract information
chunk_results = []
for i, chunk in enumerate(chunks):
print(f"Processing chunk {i+1}/{len(chunks)}...")
try:
# Process each chunk with a simplified prompt focused on extraction
chunk_result = await query_mistral_llm_single(
chunk,
is_chunk=True,
chunk_num=i+1,
total_chunks=len(chunks)
)
chunk_results.append(chunk_result)
except Exception as e:
print(f"Error processing chunk {i+1}: {str(e)}")
# Combine results from all chunks
combined_result = combine_chunk_results(chunk_results)
return combined_result
def combine_chunk_results(chunk_results: list) -> dict:
"""Combine results from multiple chunks into a single coherent result"""
if not chunk_results:
return {}
# Initialize with empty structures
combined = {
"parties": [],
"financials": {},
"payment_terms": {},
"sla": {},
"contacts": [],
"additional_fields": {}
}
# Track seen items to avoid duplicates
seen_parties = set()
seen_contacts = set()
for result in chunk_results:
# Merge parties (avoiding duplicates)
if result.get("parties"):
for party in result["parties"]:
# Create a simple hash for deduplication
if isinstance(party, dict) and party.get("name"):
party_key = party["name"].lower()
if party_key not in seen_parties:
seen_parties.add(party_key)
combined["parties"].append(party)
elif isinstance(party, str) and party.lower() not in seen_parties:
seen_parties.add(party.lower())
combined["parties"].append(party)
# Merge financials (take most complete information)
if result.get("financials") and isinstance(result["financials"], dict):
combined["financials"].update(result["financials"])
# Merge payment terms
if result.get("payment_terms") and isinstance(result["payment_terms"], dict):
combined["payment_terms"].update(result["payment_terms"])
# Merge SLA information
if result.get("sla") and isinstance(result["sla"], dict):
combined["sla"].update(result["sla"])
# Merge contacts (avoiding duplicates)
if result.get("contacts"):
for contact in result["contacts"]:
# Create a simple hash for deduplication
if isinstance(contact, dict) and contact.get("name"):
contact_key = contact["name"].lower()
if contact_key not in seen_contacts:
seen_contacts.add(contact_key)
combined["contacts"].append(contact)
elif isinstance(contact, str) and contact.lower() not in seen_contacts:
seen_contacts.add(contact.lower())
combined["contacts"].append(contact)
# Merge additional fields
if result.get("additional_fields") and isinstance(result["additional_fields"], dict):
combined["additional_fields"].update(result["additional_fields"])
return combined
async def query_mistral_llm_single(contract_text: str, is_chunk=False, chunk_num=None, total_chunks=None) -> dict:
"""Query the Mistral LLM for a single chunk of text"""
system_prompt = "You are a specialized AI assistant for contract analysis. Your task is to extract structured information from contract documents and return it as valid JSON. Be precise and thorough in your extraction."
# Adjust prompt based on whether this is a chunk or full document
if is_chunk:
user_prompt = f"""You are analyzing chunk {chunk_num} of {total_chunks} from a large contract document.
Extract the following contract details as a valid JSON object with these keys:
1. parties: Array of entities involved in the contract (names, roles, addresses)
2. financials: Object containing monetary details (amounts, currencies, total value)
3. payment_terms: Object with payment schedule, methods, penalties
4. sla: Service level agreements, performance metrics, guarantees
5. contacts: Key personnel, points of contact, contact information
6. additional_fields: Object containing ANY other important information found in the contract
If any field is missing or cannot be determined, set it as null or an empty structure. Focus on extracting what's available in this chunk.
Your response MUST be a valid JSON object without any explanatory text before or after.
Contract text (chunk {chunk_num}/{total_chunks}):
{contract_text}
"""
else:
user_prompt = f"""Extract the following contract details as a valid JSON object with these keys:
1. parties: Array of entities involved in the contract (names, roles, addresses)
2. financials: Object containing monetary details (amounts, currencies, total value)
3. payment_terms: Object with payment schedule, methods, penalties
4. sla: Service level agreements, performance metrics, guarantees
5. contacts: Key personnel, points of contact, contact information
6. additional_fields: Object containing ANY other important information found in the contract
If any field is missing or cannot be determined, set it as null. For additional_fields, include any important information that doesn't fit the standard categories.
Your response MUST be a valid JSON object without any explanatory text before or after. Format:
{{"parties": [...], "financials": {{...}}, "payment_terms": {{...}}, "sla": {{...}}, "contacts": {{...}}, "additional_fields": {{...}}}}
Contract text:
{contract_text}
"""
try:
# First try with httpx
async with httpx.AsyncClient(timeout=60, verify=False) as client:
headers = {
"Authorization": f"Bearer {OPENROUTER_API_KEY}",
"Content-Type": "application/json"
}
payload = {
"model": "mistralai/mistral-7b-instruct",
"messages": [
{"role": "system", "content": system_prompt},
{"role": "user", "content": user_prompt}
]
}
response = await client.post(
"https://openrouter.ai/api/v1/chat/completions",
json=payload,
headers=headers
)
if response.status_code == 200:
data = response.json()
if "choices" in data and len(data["choices"]) > 0:
text_output = data["choices"][0]["message"]["content"]
try:
# First try direct JSON parsing
extracted = json.loads(text_output)
return extracted
except json.JSONDecodeError:
print("JSON parsing failed, attempting to extract JSON from text")
# Try to extract JSON from text (sometimes model wraps JSON in markdown or explanations)
import re
json_match = re.search(r'\{[\s\S]*\}', text_output)
if json_match:
try:
json_str = json_match.group(0)
# Try to parse as is
try:
extracted = json.loads(json_str)
return extracted
except json.JSONDecodeError:
# Try to fix truncated JSON by adding missing closing braces
print("Attempting to fix truncated JSON")
# Count opening and closing braces
open_braces = json_str.count('{')
close_braces = json_str.count('}')
if open_braces > close_braces:
# Add missing closing braces
json_str += '}' * (open_braces - close_braces)
try:
extracted = json.loads(json_str)
return extracted
except json.JSONDecodeError:
pass
except Exception as e:
print(f"Error extracting JSON: {str(e)}")
# If we get here, either the API call failed or JSON parsing failed
print(f"OpenRouter API call failed with status {response.status_code}")
print(f"Response: {response.text}")
# Try the fallback method
return await query_mistral_llm_requests_fallback(contract_text)
except Exception as e:
print(f"Error in query_mistral_llm: {str(e)}")
# Try the fallback method
return await query_mistral_llm_requests_fallback(contract_text)
async def query_mistral_llm_requests_fallback(contract_text: str) -> dict:
"""
Fallback function using requests library with aggressive SSL bypass
"""
if not OPENROUTER_API_KEY:
raise ValueError("OPENROUTER_API_KEY environment variable is not set.")
system_prompt = "You are a specialized AI assistant for contract analysis. Your task is to extract structured information from contract documents and return it as valid JSON. Be precise and thorough in your extraction."
user_prompt = f"""Extract the following contract details as a valid JSON object with these keys:
1. parties: Array of entities involved in the contract (names, roles, addresses)
2. financials: Object containing monetary details (amounts, currencies, total value)
3. payment_terms: Object with payment schedule, methods, penalties
4. sla: Service level agreements, performance metrics, guarantees
5. contacts: Key personnel, points of contact, contact information
6. additional_fields: Object containing ANY other important information found in the contract
If any field is missing or cannot be determined, set it as null. For additional_fields, include any important information that doesn't fit the standard categories.
Your response MUST be a valid JSON object without any explanatory text before or after. Format:
{{"parties": [...], "financials": {{...}}, "payment_terms": {{...}}, "sla": {{...}}, "contacts": {{...}}, "additional_fields": {{...}}}}
Contract text:
{contract_text}
"""
headers = {
"Authorization": f"Bearer {OPENROUTER_API_KEY}",
"Content-Type": "application/json",
"User-Agent": "ContractIntelligence/1.0"
}
payload = {
"model": MISTRAL_MODEL,
"messages": [
{"role": "system", "content": system_prompt},
{"role": "user", "content": user_prompt}
],
"temperature": 0.3,
"max_tokens": 1000
}
# Configure requests session with SSL bypass
session = requests.Session()
session.verify = False # Disable SSL verification completely
# Add retry strategy
retry_strategy = Retry(
total=3,
backoff_factor=1,
status_forcelist=[429, 500, 502, 503, 504],
)
adapter = HTTPAdapter(max_retries=retry_strategy)
session.mount("http://", adapter)
session.mount("https://", adapter)
# Create custom SSL context that's more permissive
import ssl
ssl_context = ssl.create_default_context()
ssl_context.check_hostname = False
ssl_context.verify_mode = ssl.CERT_NONE
endpoints = [
"https://openrouter.ai/api/v1/chat/completions",
"https://api.openrouter.ai/api/v1/chat/completions"
]
for endpoint in endpoints:
try:
print(f"Requests fallback: trying {endpoint}")
response = session.post(
endpoint,
json=payload,
headers=headers,
timeout=60,
verify=False
)
print(f"Requests response status: {response.status_code}")
if response.status_code == 200:
data = response.json()
if "choices" in data and len(data["choices"]) > 0:
text_output = data["choices"][0]["message"]["content"]
print("✅ Requests fallback successful!")
try:
# First try direct JSON parsing
extracted = json.loads(text_output)
return extracted
except json.JSONDecodeError:
print("JSON parsing failed, attempting to extract JSON from text")
# Try to extract JSON from text (sometimes model wraps JSON in markdown or explanations)
import re
json_match = re.search(r'\{[\s\S]*\}', text_output)
if json_match:
try:
json_str = json_match.group(0)
# Try to parse as is
try:
extracted = json.loads(json_str)
return extracted
except json.JSONDecodeError:
# Try to fix truncated JSON by adding missing closing braces
print("Attempting to fix truncated JSON")
# Count opening and closing braces
open_braces = json_str.count('{')
close_braces = json_str.count('}')
open_brackets = json_str.count('[')
close_brackets = json_str.count(']')
# Add missing closing braces and brackets
if open_braces > close_braces:
json_str += '}' * (open_braces - close_braces)
if open_brackets > close_brackets:
json_str += ']' * (open_brackets - close_brackets)
try:
extracted = json.loads(json_str)
print("Successfully fixed truncated JSON")
return extracted
except json.JSONDecodeError:
print("Could not fix truncated JSON")
except Exception as e:
print(f"Error processing JSON match: {str(e)}")
# Try to extract JSON from raw text using a more aggressive approach
try:
# Look for JSON-like structure and try to parse it
# First, try to find the raw JSON structure
raw_json = text_output
# Try to extract the JSON content from the raw text
# This is a more aggressive approach to find valid JSON
import re
# Find all JSON-like structures
json_candidates = re.findall(r'\{[^{}]*(?:\{[^{}]*\}[^{}]*)*\}', raw_json)
for candidate in json_candidates:
try:
# Try to parse each candidate
parsed = json.loads(candidate)
if isinstance(parsed, dict) and any(key in parsed for key in ['parties', 'financials', 'payment_terms', 'sla', 'contacts']):
print("Found valid JSON structure in raw text")
return parsed
except:
continue
except Exception as e:
print(f"Advanced JSON extraction failed: {str(e)}")
# If all parsing fails, try to extract structured data from the raw text
try:
# Try to parse the raw text as JSON with missing closing braces
raw_text = text_output
# Find the start of a JSON object
json_start = raw_text.find('{')
if json_start >= 0:
partial_json = raw_text[json_start:]
# Count opening and closing braces
open_braces = partial_json.count('{')
close_braces = partial_json.count('}')
open_brackets = partial_json.count('[')
close_brackets = partial_json.count(']')
# Add missing closing braces and brackets
if open_braces > close_braces:
partial_json += '}' * (open_braces - close_braces)
if open_brackets > close_brackets:
partial_json += ']' * (open_brackets - close_brackets)
try:
fixed_json = json.loads(partial_json)
print("Successfully parsed fixed JSON")
return fixed_json
except:
pass
except Exception as e:
print(f"Failed to fix JSON: {str(e)}")
# If all parsing fails, return structured data with the raw text
return {
"parties": None,
"financials": None,
"payment_terms": None,
"sla": None,
"contacts": None,
"additional_fields": {
"raw_text": text_output
}
}
else:
print(f"Requests error: {response.status_code} - {response.text}")
except Exception as e:
print(f"Requests exception for {endpoint}: {str(e)}")
continue
raise ValueError("All requests fallback attempts failed")
async def query_mistral_llm(contract_text: str) -> dict:
"""
Send contract text to Mistral model via OpenRouter API with comprehensive fallbacks.
Returns structured JSON fields.
"""
# Check if API key is available
if not OPENROUTER_API_KEY:
raise ValueError("OPENROUTER_API_KEY environment variable is not set. Please set it before making API calls.")
system_prompt = "You are a specialized AI assistant for contract analysis. Your task is to extract structured information from contract documents and return it as valid JSON. Be precise and thorough in your extraction."
user_prompt = f"""Extract the following contract details as a valid JSON object with these keys:
1. parties: Array of entities involved in the contract (names, roles, addresses)
2. financials: Object containing monetary details (amounts, currencies, total value)
3. payment_terms: Object with payment schedule, methods, penalties
4. sla: Service level agreements, performance metrics, guarantees
5. contacts: Key personnel, points of contact, contact information
6. additional_fields: Object containing ANY other important information found in the contract
If any field is missing or cannot be determined, set it as null. For additional_fields, include any important information that doesn't fit the standard categories.
Your response MUST be a valid JSON object without any explanatory text before or after. Format:
{{"parties": [...], "financials": {{...}}, "payment_terms": {{...}}, "sla": {{...}}, "contacts": {{...}}, "additional_fields": {{...}}}}
Contract text:
{contract_text}
"""
headers = {
"Authorization": f"Bearer {OPENROUTER_API_KEY}",
"Content-Type": "application/json",
"HTTP-Referer": "http://localhost:8000",
"X-Title": "Contract Intelligence API",
"User-Agent": "ContractIntelligence/1.0"
}
payload = {
"model": MISTRAL_MODEL,
"messages": [
{"role": "system", "content": system_prompt},
{"role": "user", "content": user_prompt}
],
"temperature": 0.3,
"max_tokens": 1000
}
# Multiple endpoint strategies with DNS resolution fallbacks
endpoint_strategies = [
# Strategy 1: Standard hostnames
{
"name": "Standard OpenRouter endpoint",
"url": "https://openrouter.ai/api/v1/chat/completions"
},
{
"name": "Alternative OpenRouter endpoint",
"url": "https://api.openrouter.ai/api/v1/chat/completions"
},
# Strategy 2: Direct IP addresses (bypassing DNS)
{
"name": "Direct IP for openrouter.ai",
"url": "https://104.18.6.192/api/v1/chat/completions",
"headers": {**headers, "Host": "openrouter.ai"}
},
{
"name": "Direct IP for api.openrouter.ai",
"url": "https://104.18.7.192/api/v1/chat/completions",
"headers": {**headers, "Host": "api.openrouter.ai"}
}
]
# Different client configurations to try
client_configs = [
{
"name": "Standard secure",
"timeout": httpx.Timeout(60.0, connect=30.0),
"verify": True,
"follow_redirects": True
},
{
"name": "Extended timeout",
"timeout": httpx.Timeout(120.0, connect=60.0),
"verify": True,
"follow_redirects": True
},
{
"name": "No SSL verification",
"timeout": httpx.Timeout(60.0, connect=30.0),
"verify": False,
"follow_redirects": True
},
{
"name": "Basic connection",
"timeout": httpx.Timeout(30.0, connect=15.0),
"verify": False,
"follow_redirects": False
}
]
last_error = None
for strategy in endpoint_strategies:
endpoint = strategy["url"]
endpoint_headers = strategy.get("headers", headers)
print(f"Trying strategy: {strategy['name']} - {endpoint}")
for config in client_configs:
try:
print(f" Using config: {config['name']}")
async with httpx.AsyncClient(**{k: v for k, v in config.items() if k != 'name'}) as client:
try:
response = await client.post(
endpoint,
json=payload,
headers=endpoint_headers
)
print(f" Response status: {response.status_code}")
response.raise_for_status()
data = response.json()
# Parse response according to the OpenRouter API format
if "choices" in data and len(data["choices"]) > 0:
text_output = data["choices"][0]["message"]["content"]
print(f"✅ Success with {strategy['name']}")
# Try to parse JSON from response
try:
# First try direct JSON parsing
extracted = json.loads(text_output)
return extracted
except json.JSONDecodeError:
print("Warning: Could not parse LLM response as JSON, attempting to extract JSON from text")
# Try to extract JSON from text (sometimes model wraps JSON in markdown or explanations)
import re
json_match = re.search(r'\{[\s\S]*\}', text_output)
if json_match:
try:
json_str = json_match.group(0)
# Try to parse as is
try:
extracted = json.loads(json_str)
return extracted
except json.JSONDecodeError:
# Try to fix truncated JSON by adding missing closing braces
print("Attempting to fix truncated JSON")
# Count opening and closing braces
open_braces = json_str.count('{')
close_braces = json_str.count('}')
open_brackets = json_str.count('[')
close_brackets = json_str.count(']')
# Add missing closing braces and brackets
if open_braces > close_braces:
json_str += '}' * (open_braces - close_braces)
if open_brackets > close_brackets:
json_str += ']' * (open_brackets - close_brackets)
try:
extracted = json.loads(json_str)
print("Successfully fixed truncated JSON")
return extracted
except json.JSONDecodeError:
print("Could not fix truncated JSON")
except Exception as e:
print(f"Error processing JSON match: {str(e)}")
# Try to extract JSON from raw text using a more aggressive approach
try:
# Look for JSON-like structure and try to parse it
# First, try to find the raw JSON structure
raw_json = text_output
# Try to extract the JSON content from the raw text
# This is a more aggressive approach to find valid JSON
import re
# Find all JSON-like structures
json_candidates = re.findall(r'\{[^{}]*(?:\{[^{}]*\}[^{}]*)*\}', raw_json)
for candidate in json_candidates:
try:
# Try to parse each candidate
parsed = json.loads(candidate)
if isinstance(parsed, dict) and any(key in parsed for key in ['parties', 'financials', 'payment_terms', 'sla', 'contacts']):
print("Found valid JSON structure in raw text")
return parsed
except:
continue
except Exception as e:
print(f"Advanced JSON extraction failed: {str(e)}")
# If all parsing fails, try to extract structured data from the raw text
try:
# Try to parse the raw text as JSON with missing closing braces
raw_text = text_output
# Find the start of a JSON object
json_start = raw_text.find('{')
if json_start >= 0:
partial_json = raw_text[json_start:]
# Count opening and closing braces
open_braces = partial_json.count('{')
close_braces = partial_json.count('}')
open_brackets = partial_json.count('[')
close_brackets = partial_json.count(']')
# Add missing closing braces and brackets
if open_braces > close_braces:
partial_json += '}' * (open_braces - close_braces)
if open_brackets > close_brackets:
partial_json += ']' * (open_brackets - close_brackets)
try:
fixed_json = json.loads(partial_json)
print("Successfully parsed fixed JSON")
return fixed_json
except:
pass
except Exception as e:
print(f"Failed to fix JSON: {str(e)}")
# If all parsing fails, return structured data with the raw text
return {
"parties": None,
"financials": None,
"payment_terms": None,
"sla": None,
"contacts": None,
"additional_fields": {
"raw_text": text_output
}
}
else:
last_error = ValueError(f"Unexpected API response format: {data}")
except httpx.HTTPStatusError as e:
last_error = ValueError(f"HTTP {e.response.status_code}: {e.response.text}")
print(f" HTTP Error: {e.response.status_code}")
except httpx.ConnectError as e:
last_error = ValueError(f"Connection error: {str(e)}")
print(f" Connection Error: {str(e)}")
except httpx.TimeoutException as e:
last_error = ValueError(f"Timeout error: {str(e)}")
print(f" Timeout Error: {str(e)}")
except Exception as e:
last_error = ValueError(f"Request error: {str(e)}")
print(f" Request Error: {str(e)}")
except Exception as e:
last_error = ValueError(f"Client creation error: {str(e)}")
print(f" Client Error: {str(e)}")
# If all httpx attempts failed, try requests fallback
if last_error:
print("🔄 Trying requests library fallback...")
try:
return await query_mistral_llm_requests_fallback(contract_text)
except Exception as fallback_error:
print(f"Requests fallback also failed: {str(fallback_error)}")
# Provide a helpful error message based on the original httpx errors
error_msg = str(last_error)
if "getaddrinfo failed" in error_msg or "Name or service not known" in error_msg:
raise ValueError("DNS resolution failed. This could be due to: 1) No internet connection, 2) DNS server issues, 3) Firewall blocking requests. Please check your network settings.")
elif "SSL" in error_msg.upper() or "certificate" in error_msg.lower() or "handshake" in error_msg.lower():
raise ValueError("SSL/TLS connection failed. This might be due to corporate firewall or security settings. All SSL bypass attempts failed.")
elif "timeout" in error_msg.lower():
raise ValueError("Connection timeout. Please check your internet speed and try again.")
else:
raise ValueError(f"All connection attempts failed. HTTPX error: {error_msg}. Requests error: {str(fallback_error)}")
# This should never be reached
raise ValueError("All API connection attempts failed without specific error")
# Mock data generation removed as per requirement
async def process_contract(contract_id: str, file_path: str):
"""Process a contract file and extract information with memory optimization"""
global USE_MONGO
try:
# Update status to processing
await update_contract_status(contract_id, status="processing", progress=5)
# Get file size for progress tracking
file_size = os.path.getsize(file_path)
is_large_file = file_size > 10 * 1024 * 1024 # 10MB threshold
if is_large_file:
print(f"Processing large PDF ({file_size / (1024 * 1024):.2f} MB) with optimized memory usage")
# Extract text from PDF with memory optimization
contract_text = await extract_text_from_pdf(file_path)
# Free up memory by clearing references
import gc
gc.collect()
await update_contract_status(contract_id, status="processing", progress=30)
# Check internet connectivity before proceeding
has_connectivity = await test_network_connectivity()
if not has_connectivity:
print(f"No internet connectivity detected for contract {contract_id}")
await update_contract_status(contract_id, status="error", progress=100,
error="No internet connectivity detected. Please check your network connection.")
return
await update_contract_status(contract_id, status="processing", progress=50)
# Query LLM for extraction with chunking for large documents
extracted_data = await query_mistral_llm(contract_text)
# Free up memory again
contract_text = None
gc.collect()
await update_contract_status(contract_id, status="processing", progress=90)
# Update contract with extracted data
await update_contract_status(contract_id, status="completed", progress=100)
await update_contract(contract_id, {
"extracted_data": extracted_data
})
except Exception as e:
error_message = str(e)
# Provide more user-friendly error messages
if "getaddrinfo failed" in error_message:
error_message = "DNS resolution failed. Please check your internet connection and DNS settings."
elif "SSL" in error_message.upper():
error_message = "SSL connection failed. This might be due to firewall or network security settings."
elif "timeout" in error_message.lower():
error_message = "Connection timeout. Please check your internet connection."
await update_contract_status(contract_id, status="error", progress=100, error=error_message)
# ----------------- Routes -----------------
@app.post("/contracts/upload")
async def upload_contract(file: UploadFile = File(...), background_tasks: BackgroundTasks = None):
if not file.filename.endswith(".pdf"):
raise HTTPException(status_code=400, detail="Only PDF files are supported")
contract_id = str(uuid.uuid4())
file_path = os.path.join(UPLOAD_DIR, f"{contract_id}.pdf")
# Stream the file to disk in chunks to handle large files
file_size = 0
chunk_size = 1024 * 1024 # 1MB chunks
try:
with open(file_path, "wb") as f:
while True:
chunk = await file.read(chunk_size)
if not chunk:
break
file_size += len(chunk)
f.write(chunk)
print(f"Uploaded file size: {file_size / (1024 * 1024):.2f} MB")
except Exception as e:
print(f"Error during file upload: {str(e)}")
if os.path.exists(file_path):
os.remove(file_path) # Clean up partial file
raise HTTPException(status_code=500, detail=f"File upload failed: {str(e)}")
# Create contract document
contract_doc = {
"_id": contract_id,
"filename": file.filename,
"status": "uploaded",
"progress": 0,
"file_size_bytes": file_size,
"extracted_data": None,
"error": None,
"created_at": int(time.time()) # Unix timestamp for sorting
}
# Store contract metadata based on available storage
global USE_MONGO
if USE_MONGO:
try:
# Set a short timeout for this operation
await asyncio.wait_for(
contracts_collection.insert_one(contract_doc),
timeout=2.0
)
except (asyncio.TimeoutError, ServerSelectionTimeoutError) as e:
print(f"MongoDB connection timed out: {str(e)}")
print("Switching to file-based storage for future operations")
USE_MONGO = False
# Fall back to file storage
save_contract_to_file(contract_id, contract_doc)
except Exception as e:
print(f"MongoDB insert failed: {str(e)}")
# Fall back to file storage if MongoDB insert fails
save_contract_to_file(contract_id, contract_doc)
else:
# Use file-based storage
save_contract_to_file(contract_id, contract_doc)
# Launch background processing
background_tasks.add_task(process_contract, contract_id, file_path)
return {"contract_id": contract_id}
@app.get("/contracts/{contract_id}/status", response_model=ContractStatus)
async def get_contract_status(contract_id: str):
global USE_MONGO
contract = None
# Try to get contract from available storage
if USE_MONGO:
try:
# Set a short timeout for this operation
contract = await asyncio.wait_for(
contracts_collection.find_one({"_id": contract_id}),
timeout=2.0
)
except (asyncio.TimeoutError, ServerSelectionTimeoutError) as e:
print(f"MongoDB connection timed out: {str(e)}")
print("Switching to file-based storage for future operations")
USE_MONGO = False
# Fall back to file storage
contract = load_contract_from_file(contract_id)
except Exception as e:
print(f"MongoDB find_one failed: {str(e)}")
# Fall back to file storage
contract = load_contract_from_file(contract_id)
else:
# Use file-based storage
contract = load_contract_from_file(contract_id)
if not contract:
raise HTTPException(status_code=404, detail="Contract not found")
return ContractStatus(
status=contract.get("status"),
progress=contract.get("progress"),
score=None,
error=contract.get("error")
)
@app.get("/contracts/{contract_id}")
async def get_contract_data(contract_id: str):
global USE_MONGO
contract = None
# Try to get contract from available storage
if USE_MONGO:
try:
# Set a short timeout for this operation
contract = await asyncio.wait_for(
contracts_collection.find_one({"_id": contract_id}),
timeout=2.0
)
except (asyncio.TimeoutError, ServerSelectionTimeoutError) as e:
print(f"MongoDB connection timed out: {str(e)}")
print("Switching to file-based storage for future operations")
USE_MONGO = False
# Fall back to file storage
contract = load_contract_from_file(contract_id)
except Exception as e:
print(f"MongoDB find_one failed: {str(e)}")
# Fall back to file storage
contract = load_contract_from_file(contract_id)
else:
# Use file-based storage
contract = load_contract_from_file(contract_id)
if not contract:
raise HTTPException(status_code=404, detail="Contract not found")
# Return the full contract object instead of just extracted_data
# This allows the frontend to display contract info for any status
return contract
@app.get("/contracts/{contract_id}/download")
async def download_contract_pdf(contract_id: str):
"""Download the original PDF file for a contract"""
# Check if contract exists
global USE_MONGO
contract = None
# Try to get contract from available storage to verify it exists
if USE_MONGO:
try:
contract = await asyncio.wait_for(
contracts_collection.find_one({"_id": contract_id}),
timeout=2.0
)
except Exception:
# Fall back to file storage
contract = load_contract_from_file(contract_id)
else:
# Use file-based storage
contract = load_contract_from_file(contract_id)
if not contract:
raise HTTPException(status_code=404, detail="Contract not found")
# Construct the path to the PDF file
file_path = os.path.join(UPLOAD_DIR, f"{contract_id}.pdf")
# Check if file exists
if not os.path.exists(file_path):
raise HTTPException(status_code=404, detail="PDF file not found")
# Get original filename from contract metadata
filename = contract.get("filename", f"contract_{contract_id}.pdf")
# Return the file as a downloadable response
return FileResponse(
path=file_path,
filename=filename,
media_type="application/pdf"
)
@app.get("/contracts")
async def list_contracts():
"""Get a list of all contracts with their status information"""
global USE_MONGO
contracts = []
if USE_MONGO:
try:
# Set a short timeout for this operation
cursor = contracts_collection.find({}, {
"_id": 1,
"filename": 1,
"status": 1,
"progress": 1,
"error": 1,
"created_at": 1
})
# Use a timeout for the cursor iteration
async def get_docs_with_timeout():
docs = []
async for doc in cursor:
# Convert MongoDB _id to contract_id string for frontend
# Handle ObjectId serialization
if isinstance(doc["_id"], ObjectId):
doc["contract_id"] = str(doc["_id"])
else:
doc["contract_id"] = doc["_id"]
# Remove the _id field to avoid serialization issues
doc.pop("_id")
docs.append(doc)
return docs
# Execute with timeout
contracts = await asyncio.wait_for(get_docs_with_timeout(), timeout=3.0)
except (asyncio.TimeoutError, ServerSelectionTimeoutError) as e:
print(f"MongoDB connection timed out: {str(e)}")
print("Switching to file-based storage for future operations")
USE_MONGO = False
# Fall back to file storage
contracts = load_all_contracts_from_files()
except Exception as e:
print(f"MongoDB find failed: {str(e)}")
# Fall back to file storage
contracts = load_all_contracts_from_files()
else:
# Use file-based storage
contracts = load_all_contracts_from_files()
# Sort by most recently created first (if created_at exists)
contracts.sort(key=lambda x: x.get("created_at", 0), reverse=True)
return contracts
# ----------------- File-based Storage Functions -----------------
def save_contract_to_file(contract_id: str, contract_data: dict):
"""Save contract data to a JSON file"""
try:
file_path = os.path.join(FILE_STORAGE_DIR, f"{contract_id}.json")
with open(file_path, 'w') as f:
json.dump(contract_data, f)
print(f"✅ Saved contract {contract_id} to file storage")
return True
except Exception as e:
print(f"❌ Error saving contract to file: {str(e)}")
return False
def load_contract_from_file(contract_id: str) -> dict:
"""Load contract data from a JSON file"""
try:
file_path = os.path.join(FILE_STORAGE_DIR, f"{contract_id}.json")
if not os.path.exists(file_path):
return None
with open(file_path, 'r') as f:
return json.load(f)
except Exception as e:
print(f"❌ Error loading contract from file: {str(e)}")
return None
def load_all_contracts_from_files() -> list:
"""Load all contracts from JSON files"""
contracts = []
try:
for filename in os.listdir(FILE_STORAGE_DIR):
if filename.endswith('.json'):
contract_id = filename.replace('.json', '')
contract = load_contract_from_file(contract_id)
if contract:
# Add contract_id field for frontend compatibility
contract['contract_id'] = contract['_id']
# Remove _id to avoid serialization issues
if '_id' in contract:
contract.pop('_id')
contracts.append(contract)
return contracts
except Exception as e:
print(f"❌ Error loading contracts from files: {str(e)}")
return []
async def update_contract_status(contract_id: str, status=None, progress=None, error=None):
"""Update contract status in the appropriate storage
Can be called with either individual parameters or as a dictionary
"""
# Create update data dictionary from parameters
update_data = {}
if isinstance(status, str):
update_data["status"] = status
if progress is not None:
update_data["progress"] = progress
if error is not None:
update_data["error"] = error
# Call the update function with the dictionary
await update_contract(contract_id, update_data)
async def update_contract(contract_id: str, update_data: dict):
"""Update contract in the appropriate storage"""
if USE_MONGO:
try:
# Try MongoDB first
await _update_mongo_contract(contract_id, update_data)
except Exception as e:
print(f"MongoDB update failed: {str(e)}")
# Fall back to file storage
_update_file_contract(contract_id, update_data)
else:
# Use file-based storage
_update_file_contract(contract_id, update_data)
async def _update_mongo_contract(contract_id: str, update_data: dict):
"""Update contract in MongoDB"""
global USE_MONGO
try:
# Set a short timeout for this operation
await asyncio.wait_for(
contracts_collection.update_one(
{"_id": contract_id},
{"$set": update_data}
),
timeout=2.0
)
except (asyncio.TimeoutError, ServerSelectionTimeoutError) as e:
print(f"MongoDB connection timed out: {str(e)}")
print("Switching to file-based storage for future operations")
USE_MONGO = False
# Fall back to file storage
_update_file_contract(contract_id, update_data)
except Exception as e:
print(f"MongoDB update failed: {str(e)}")
# Fall back to file storage
_update_file_contract(contract_id, update_data)
def _update_file_contract(contract_id: str, update_data: dict):
"""Update contract in file storage"""
try:
contract = load_contract_from_file(contract_id)
if contract:
contract.update(update_data)
save_contract_to_file(contract_id, contract)
except Exception as e:
print(f"File storage update failed: {str(e)}")
|