mousom's picture
Upload folder using huggingface_hub
e6c137b verified
Raw History Blame Contribute Delete
51.9 kB
from fastapi import FastAPI, HTTPException from pydantic import BaseModel, Field from typing import Dict, List, Optional, Any import requests from bs4 import BeautifulSoup import re import time import logging from urllib.parse import urlparse from datetime import datetime import uvicorn import os # Configure logging logging.basicConfig(level=logging.INFO) logger = logging.getLogger(__name__) # FastAPI app initialization app = FastAPI( title="Caltrans Payment Tracking System", description="API for retrieving contract payment history and detailed monthly estimates", version="1.0.0" ) # Pydantic Models class PaymentRecord(BaseModel): estimate_number: str = Field(..., description="Estimate number") detail_link: Optional[str] = Field(None, description="Link to payment details") voucher_link: Optional[str] = Field(None, description="Link to voucher") type: str = Field(..., description="Payment type") payment_type: str = Field(..., description="Payment classification") release_date: str = Field(..., description="Payment release date") held_amount: float = Field(..., description="Amount held") disbursed_amount: float = Field(..., description="Amount disbursed") class ContractDetails(BaseModel): contract_number: str = Field(..., description="Contract number") contractor_name: Optional[str] = Field(None, description="Contractor name") last_updated: Optional[str] = Field(None, description="Last update date") job_description: Optional[str] = Field(None, description="Job description") contract_accepted: Optional[str] = Field(None, description="Contract acceptance date") stop_notice_deadline: Optional[str] = Field(None, description="Stop notice deadline") suit_filing_expiration: Optional[str] = Field(None, description="Suit filing expiration") estimated_completion: Optional[str] = Field(None, description="Estimated completion date") current_withhold: Optional[float] = Field(None, description="Current withhold amount") currently_held: Optional[float] = Field(None, description="Currently held amount") need_to_hold: Optional[float] = Field(None, description="Amount needed to hold") class ContactInfo(BaseModel): email: Optional[str] = Field(None, description="Contact email") phone: Optional[str] = Field(None, description="Contact phone") class PaymentSummary(BaseModel): total_payments: int = Field(..., description="Total number of payments") total_disbursed: float = Field(..., description="Total amount disbursed") total_held: float = Field(..., description="Total amount held") latest_payment_date: str = Field(..., description="Date of latest payment") class Quantities(BaseModel): original: float = Field(..., description="Original quantity") this_period: float = Field(..., description="This period quantity") percent_this_period: float = Field(..., description="Percentage this period") job_to_date: float = Field(..., description="Job to date quantity") percent_jtd: float = Field(..., description="Percentage job to date") class Amounts(BaseModel): original: float = Field(..., description="Original amount") this_period: float = Field(..., description="This period amount") job_to_date: float = Field(..., description="Job to date amount") moh: float = Field(..., description="Materials on hand") class LineItem(BaseModel): item_number: str = Field(..., description="Item number") revenue_code: str = Field(..., description="Revenue code") description: str = Field(..., description="Work description") sub_item: str = Field(..., description="Y=Subcontracted, N=Self-performed") quantities: Quantities units: str = Field(..., description="Unit of measurement") unit_price: float = Field(..., description="Unit price") amounts: Amounts class DetailedPaymentRecord(BaseModel): estimate_number: str = Field(..., description="Estimate number") detail_link: Optional[str] = Field(None, description="Link to payment details") voucher_link: Optional[str] = Field(None, description="Link to voucher") type: str = Field(..., description="Payment type") payment_type: str = Field(..., description="Payment classification") release_date: str = Field(..., description="Payment release date") held_amount: float = Field(..., description="Amount held") disbursed_amount: float = Field(..., description="Amount disbursed") detailed_estimate: Optional[Dict[str, Any]] = Field(None, description="Detailed CAS145 estimate data") class MonthlyPaymentSummary(BaseModel): month: str = Field(..., description="Month in YYYY-MM format") payment_count: int = Field(..., description="Number of payments in month") total_disbursed: float = Field(..., description="Total disbursed in month") total_held: float = Field(..., description="Total held in month") payments: List[DetailedPaymentRecord] = Field(..., description="Individual payments with detailed estimates") class ContractResponse(BaseModel): contract_details: ContractDetails payment_history: List[PaymentRecord] contact_info: ContactInfo summary: PaymentSummary monthly_breakdown: Dict[str, float] = Field(..., description="Monthly disbursement breakdown") scraped_at: str = Field(..., description="Timestamp when data was scraped") class CaltransPETSScraper: """ Powerful web scraper for California Department of Transportation (Caltrans) Payment Estimation Tracking System (PETS) """ def __init__(self, base_url: str = "https://misc-external.dot.ca.gov/pets/", delay: float = 1.0, timeout: int = 30): """ Initialize the scraper Args: base_url: Base URL for the PETS system delay: Delay between requests (seconds) timeout: Request timeout (seconds) """ self.base_url = base_url self.delay = delay self.timeout = timeout self.session = requests.Session() # Set up headers to mimic a real browser self.session.headers.update({ 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36', 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8', 'Accept-Language': 'en-US,en;q=0.5', 'Accept-Encoding': 'gzip, deflate', 'Connection': 'keep-alive', 'Upgrade-Insecure-Requests': '1', }) # Set up logging self.logger = logging.getLogger(__name__) def _make_request(self, url: str, method: str = 'GET', **kwargs) -> Optional[requests.Response]: """Make a safe HTTP request with error handling""" try: self.logger.info(f"Making {method} request to: {url}") if method.upper() == 'GET': response = self.session.get(url, timeout=self.timeout, **kwargs) else: response = self.session.post(url, timeout=self.timeout, **kwargs) response.raise_for_status() time.sleep(self.delay) # Respect server return response except requests.exceptions.RequestException as e: self.logger.error(f"Request failed for {url}: {str(e)}") return None def extract_contract_details(self, soup: BeautifulSoup) -> Dict[str, any]: """Extract contract details from a contract page""" contract_info = {} try: # Extract contract number and contractor name from header header = soup.find('h1', class_='rpt77tx') if header: header_text = header.get_text(strip=True) # Extract contract number (format: 033F0704) contract_match = re.search(r'Contract:\s*(\w+)', header_text) if contract_match: contract_info['contract_number'] = contract_match.group(1) # Extract contractor name lines = header_text.split('\n') if len(lines) > 1: contract_info['contractor_name'] = lines[1].strip() # Extract date last updated date_elem = soup.find('div', class_='rpt77b') if date_elem and 'Date last updated:' in date_elem.get_text(): date_text = date_elem.get_text(strip=True) date_match = re.search(r'Date last updated:\s*(\d{2}/\d{2}/\d{2})', date_text) if date_match: contract_info['last_updated'] = date_match.group(1) # Extract contract details from the list details_list = soup.find('ul', class_='rpt77') if details_list: for li in details_list.find_all('li'): text = li.get_text(strip=True) if 'Job Description:' in text: # Extract job description from blockquote blockquote = li.find('blockquote') if blockquote: contract_info['job_description'] = blockquote.get_text(strip=True) elif 'Contract Accepted:' in text: contract_info['contract_accepted'] = text.split(':', 1)[1].strip() elif 'Stop Notice filing deadline:' in text: contract_info['stop_notice_deadline'] = text.split(':', 1)[1].strip() elif 'Expiration of Suit-filing period:' in text: contract_info['suit_filing_expiration'] = text.split(':', 1)[1].strip() elif 'Estimated Completion:' in text: completion_text = text.split(':', 1)[1].strip() contract_info['estimated_completion'] = completion_text elif 'Current Withhold Obligations:' in text: amount = re.search(r'\$([0-9,]+\.\d{2})', text) if amount: contract_info['current_withhold'] = float(amount.group(1).replace(',', '')) elif 'Curently Held:' in text: # Note: typo in original amount = re.search(r'\$([0-9,]+\.\d{2})', text) if amount: contract_info['currently_held'] = float(amount.group(1).replace(',', '')) elif 'Need to Hold:' in text: amount = re.search(r'\$([0-9,]+\.\d{2})', text) if amount: contract_info['need_to_hold'] = float(amount.group(1).replace(',', '')) return contract_info except Exception as e: self.logger.error(f"Error extracting contract details: {str(e)}") return contract_info def extract_payment_history(self, soup: BeautifulSoup) -> List[Dict[str, any]]: """Extract payment history from the payment table""" payments = [] try: # Find the payment history table payment_table = soup.find('table', class_='reporttablelft2') if not payment_table: self.logger.warning("Payment table not found") return payments # Find all data rows (skip header) rows = payment_table.find_all('tr')[1:] # Skip header row for row in rows: cells = row.find_all('td') if len(cells) >= 8: # Ensure we have enough columns # Skip total row if 'Total:' in cells[0].get_text(strip=True): continue payment = {} # Extract estimate number payment['estimate_number'] = cells[0].get_text(strip=True) # Extract detail and voucher links (IMPORTANT!) detail_link = cells[1].find('a') if detail_link: payment['detail_link'] = detail_link.get('href') voucher_link = cells[2].find('a') if voucher_link: payment['voucher_link'] = voucher_link.get('href') # Extract type payment['type'] = cells[3].get_text(strip=True) # Extract payment type payment['payment_type'] = cells[4].get_text(strip=True) # Extract release date payment['release_date'] = cells[5].get_text(strip=True) # Extract held amount held_text = cells[6].get_text(strip=True).replace('$', '').replace(',', '') try: payment['held_amount'] = float(held_text) if held_text else 0.0 except ValueError: payment['held_amount'] = 0.0 # Extract disbursed amount disbursed_text = cells[7].get_text(strip=True).replace('$', '').replace(',', '') try: payment['disbursed_amount'] = float(disbursed_text) if disbursed_text else 0.0 except ValueError: payment['disbursed_amount'] = 0.0 payments.append(payment) return payments except Exception as e: self.logger.error(f"Error extracting payment history: {str(e)}") return payments def extract_contact_info(self, soup: BeautifulSoup) -> Dict[str, str]: """Extract contact information for payment inquiries""" contact_info = {} try: # Look for contact information contact_section = soup.find(text="Please direct Payment Inquiries to:") if contact_section: parent = contact_section.parent if parent: # Extract email email_link = parent.find('a', href=re.compile(r'mailto:')) if email_link: contact_info['email'] = email_link.get('href').replace('mailto:', '').split('?')[0] # Extract phone phone_match = re.search(r'Phone:\s*\((\d{3})\)\s*(\d{3}-\d{4})', parent.get_text()) if phone_match: contact_info['phone'] = f"({phone_match.group(1)}) {phone_match.group(2)}" return contact_info except Exception as e: self.logger.error(f"Error extracting contact info: {str(e)}") return contact_info def parse_payment_date(self, date_str: str) -> datetime: """ Intelligently parse payment dates handling both 2-digit and 4-digit years Args: date_str: Date string in format "MM/DD/YY" or "MM/DD/YYYY" Returns: datetime object with correct year Note: For 2-digit years, assumes years 00-29 are 2000-2029 and years 30-99 are 1930-1999 This follows the "sliding window" approach used by many systems """ try: # First try 4-digit year format return datetime.strptime(date_str, '%m/%d/%Y') except ValueError: # If that fails, try 2-digit year format try: date_obj = datetime.strptime(date_str, '%m/%d/%y') # Apply sliding window logic for 2-digit years # Years 00-29 are interpreted as 2000-2029 # Years 30-99 are interpreted as 1930-1999 if date_obj.year < 2000: # If year is less than 2000, it's likely from the 2000s # Adjust based on current year context current_year = datetime.now().year if current_year >= 2020: # We're in 2020s, so adjust accordingly if date_obj.year < 30: # 00-29 date_obj = date_obj.replace(year=date_obj.year + 2000) else: # 30-99 date_obj = date_obj.replace(year=date_obj.year + 1900) else: # We're in earlier years, use standard sliding window if date_obj.year < 50: # 00-49 date_obj = date_obj.replace(year=date_obj.year + 2000) else: # 50-99 date_obj = date_obj.replace(year=date_obj.year + 1900) return date_obj except ValueError: raise ValueError(f"Unable to parse date: {date_str}. Expected format: MM/DD/YY or MM/DD/YYYY") def scrape_contract(self, contract_number: str) -> Dict[str, any]: """Scrape complete contract information from a contract number""" contract_url = f"{self.base_url}results.php?ctrnm={contract_number}" response = self._make_request(contract_url) if not response: return {} soup = BeautifulSoup(response.content, 'html.parser') # Extract all information contract_data = { 'url': contract_url, 'scraped_at': datetime.now().isoformat(), 'contract_details': self.extract_contract_details(soup), 'payment_history': self.extract_payment_history(soup), 'contact_info': self.extract_contact_info(soup) } # Calculate summary statistics if contract_data['payment_history']: total_disbursed = sum(p['disbursed_amount'] for p in contract_data['payment_history']) total_held = sum(p['held_amount'] for p in contract_data['payment_history']) contract_data['summary'] = { 'total_payments': len(contract_data['payment_history']), 'total_disbursed': total_disbursed, 'total_held': total_held, 'latest_payment_date': max((p['release_date'] for p in contract_data['payment_history'] if p['release_date']), default='N/A') } # Calculate monthly breakdown monthly_breakdown = {} for payment in contract_data['payment_history']: if payment['release_date'] and payment['release_date'] != 'N/A': try: # Parse date and extract month - handle both 2-digit and 4-digit years intelligently date_obj = self.parse_payment_date(payment['release_date']) month_key = date_obj.strftime('%Y-%m') if month_key not in monthly_breakdown: monthly_breakdown[month_key] = 0.0 monthly_breakdown[month_key] += payment['disbursed_amount'] except ValueError: continue contract_data['monthly_breakdown'] = monthly_breakdown return contract_data class CAS145Parser: """ Parser for California DOT CAS145 Progress Estimate Reports Converts CAS145 text format to GC Pay JSON format with proper calculations """ def __init__(self): self.data = { "prime_contract": {"items": [], "total": {}}, "change_orders": {"items": [], "total": {}}, "deductions": [], "extra_work_totals": { 'total_this_estimate': 0.0, 'total_previous_estimate': 0.0, 'total_to_date': 0.0 } } self.main_estimate_no = None def fetch_from_url(self, url: str) -> str: """Fetch content from URL""" try: response = requests.get(url, timeout=30) response.raise_for_status() return response.text except requests.RequestException as e: raise Exception(f"Failed to fetch URL: {e}") def parse_content(self, content: str) -> Dict: """Parse CAS145 content""" sections = self.split_sections(content) self.parse_extra_work(sections.get('extra_work', '')) self.parse_deductions(sections.get('deductions', '')) self.parse_project_items(sections.get('items', '')) self.calculate_totals() return self.data def split_sections(self, content: str) -> Dict[str, str]: """Split content into major sections""" sections = {} # Find section boundaries extra_work_start = content.find('S C H E D U L E O F E X T R A W O R K') deductions_start = content.find('S C H E D U L E O F D E D U C T I O N S') items_start = content.find('P R O J E C T R E C O R D - E S T I M A T E') # Extract sections if extra_work_start > 0 and deductions_start > 0: sections['extra_work'] = content[extra_work_start:deductions_start] if deductions_start > 0 and items_start > 0: sections['deductions'] = content[deductions_start:items_start] if items_start > 0: items_content = content[items_start:] sections['items'] = items_content return sections def set_estimate_number(self, estimate_no: str): """Set estimate number from URL filename""" self.main_estimate_no = estimate_no def parse_extra_work(self, section: str): """Parse Schedule of Extra Work section with improved flexibility for various formats""" if not section: return lines = section.split('\n') parsed_count = 0 # Initialize totals tracking self.data['extra_work_totals'] = { 'total_this_estimate': 0.0, 'total_previous_estimate': 0.0, 'total_to_date': 0.0 } logger.info(f"Parsing extra work section with {len(lines)} lines") for i, line in enumerate(lines): # Skip empty lines if not line.strip(): continue # Extract totals for reference (but don't parse as line items) line_lower = line.lower() if 'total this estimate' in line_lower: amount_match = re.search(r'([\d,]+\.\d{2})', line) if amount_match: self.data['extra_work_totals']['total_this_estimate'] = self.parse_currency(amount_match.group(1)) logger.debug(f"Extracted total this estimate: {self.data['extra_work_totals']['total_this_estimate']}") continue elif 'total previous estimate' in line_lower: amount_match = re.search(r'([\d,]+\.\d{2})', line) if amount_match: self.data['extra_work_totals']['total_previous_estimate'] = self.parse_currency(amount_match.group(1)) logger.debug(f"Extracted total previous estimate: {self.data['extra_work_totals']['total_previous_estimate']}") continue elif 'total to date' in line_lower: amount_match = re.search(r'([\d,]+\.\d{2})', line) if amount_match: self.data['extra_work_totals']['total_to_date'] = self.parse_currency(amount_match.group(1)) logger.debug(f"Extracted total to date: {self.data['extra_work_totals']['total_to_date']}") continue # Skip other total and summary lines if any(total_indicator in line_lower for total_indicator in [ 'total:', 'total ', 'subtotal', 'grand total' ]): logger.debug(f"Skipping total line {i+1}: {line.strip()}") continue logger.debug(f"Processing line {i+1}: {line.strip()}") # Try to parse the line using our helper method parsed_item = self.parse_extra_work_line(line) if parsed_item: self.data['change_orders']['items'].append(parsed_item) parsed_count += 1 logger.debug(f"Successfully parsed line {i+1} as {parsed_item['change_order_no']}: {parsed_item['description']}") continue # If the helper method didn't parse it, try some additional fallback patterns # Look for lines that might contain extra work information but don't fit standard patterns # Pattern: Lines with amounts and dates that might be extra work amount_date_match = re.search(r'([\d,]+\.\d{2})\s+(\d{2}/\d{2}/\d{2})', line) if amount_date_match and ('E.W.' in line or 'EXTRA' in line.upper()): amount = self.parse_currency(amount_date_match.group(1)) date = amount_date_match.group(2) # Extract any other meaningful information identifier = self.extract_identifier_from_line(line) cco_item = { "change_order_no": "EW_DATE", "revenue_code": "", "description": f"Extra Work by Date: {date} - {identifier}", "sub_item": "", "original_qty": 0, "this_period_qty": 0, "job_to_date_qty": 0, "units": "LS", "unit_price": amount, "original_amt": 0, "this_period_amt": amount, "job_to_date_amt": amount } self.data['change_orders']['items'].append(cco_item) parsed_count += 1 logger.debug(f"Parsed line {i+1} as EW_DATE: {cco_item['description']}") continue # Pattern: Lines that just have amounts and might be extra work # This is a very permissive fallback for lines that contain amounts if 'E.W.' in line or 'EXTRA' in line.upper(): amount_match = re.search(r'([\d,]+\.\d{2})', line) if amount_match: amount = self.parse_currency(amount_match.group(1)) identifier = self.extract_identifier_from_line(line) cco_item = { "change_order_no": "EW_FALLBACK", "revenue_code": "", "description": f"Extra Work (Fallback): {identifier}", "sub_item": "", "original_qty": 0, "this_period_qty": 0, "job_to_date_qty": 0, "units": "LS", "unit_price": amount, "original_amt": 0, "this_period_amt": amount, "job_to_date_amt": amount } self.data['change_orders']['items'].append(cco_item) parsed_count += 1 logger.debug(f"Parsed line {i+1} as EW_FALLBACK: {cco_item['description']}") continue logger.info(f"Successfully parsed {parsed_count} extra work items out of {len(lines)} lines") logger.info(f"Extra work totals: This estimate: {self.data['extra_work_totals']['total_this_estimate']}, " f"Previous: {self.data['extra_work_totals']['total_previous_estimate']}, " f"To date: {self.data['extra_work_totals']['total_to_date']}") def parse_extra_work_line(self, line: str) -> Optional[Dict]: """ Helper method to parse individual extra work lines with various formats Args: line: The line to parse Returns: Parsed change order item or None if not a valid extra work line """ line = line.strip() if not line: return None logger.debug(f"Attempting to parse line: {line}") # Try different patterns to extract information patterns = [ # Pattern 1: Standard CCO format (r'(\d{3})\s+(\d{4})\s+([\d,]+\.\d{2})\s+E\.W\.\s+@\s+F\.A\.', 'CCO'), # Pattern 2: Continuation line (r'^\s+(\d{4})\s+([\d,]+\.\d{2})', 'CONT'), # Pattern 3: Extra work with amount (r'([\d,]+\.\d{2})\s+E\.W\.', 'EW'), # Pattern 4: Generic amount line (r'([\d,]+\.\d{2})', 'GENERIC') ] for pattern, cco_type in patterns: match = re.search(pattern, line) if match: logger.debug(f"Matched pattern {cco_type} for line: {line}") if cco_type == 'CCO': amount = self.parse_currency(match.group(3)) return { "change_order_no": f"{match.group(2)[0]}", "revenue_code": "", "description": f"CCO {match.group(1)}-{match.group(2)}: E.W. @", "sub_item": "", "original_qty": 0, "this_period_qty": 0, "job_to_date_qty": 0, "units": "LS", "unit_price": amount, "original_amt": 0, "this_period_amt": amount, "job_to_date_amt": amount } elif cco_type == 'CONT': amount = self.parse_currency(match.group(2)) return { "change_order_no": "CONT", "revenue_code": "", "description": f"Extra Work Continuation: {match.group(1)}", "sub_item": "", "original_qty": 0, "this_period_qty": 0, "job_to_date_qty": 0, "units": "LS", "unit_price": amount, "original_amt": 0, "this_period_amt": amount, "job_to_date_amt": amount } elif cco_type in ['EW', 'GENERIC']: amount = self.parse_currency(match.group(1)) # Extract identifier identifier = self.extract_identifier_from_line(line) return { "change_order_no": cco_type, "revenue_code": "", "description": f"Extra Work: {identifier}", "sub_item": "", "original_qty": 0, "this_period_qty": 0, "job_to_date_qty": 0, "units": "LS", "unit_price": amount, "original_amt": 0, "this_period_amt": amount, "job_to_date_amt": amount } logger.debug(f"No patterns matched for line: {line}") return None def extract_identifier_from_line(self, line: str) -> str: """ Extract meaningful identifier from a line Args: line: The line to extract identifier from Returns: Extracted identifier or "Unknown" """ line_parts = line.strip().split() for part in line_parts: # Look for 4-digit numbers (likely dates or report numbers) if re.match(r'^\d{4}$', part): return part # Look for 3-digit numbers (likely CCO numbers) elif re.match(r'^\d{3}$', part): return part # Look for alphanumeric codes elif re.match(r'^[A-Z0-9]+$', part): return part return "Unknown" def parse_deductions(self, section: str): """Parse Schedule of Deductions section""" if not section: return lines = section.split('\n') current_category = "" deductions_list = [] for line in lines: # Identify category if 'ADMINISTRATIVE' in line: current_category = "Administrative" elif 'LABOR COMPLIANCE' in line: current_category = "Labor Compliance" elif 'LIQUIDATED DAMAGES' in line: current_category = "Liquidated Damages" elif 'OVERBID ITEMS' in line: current_category = "Overbid Items" # Skip total lines if 'TOTAL DEDUCTIONS' in line: break # Parse deduction lines if current_category and re.search(r'[-\d,]+\.\d{2}', line): # Pattern to match deduction lines pattern = r'([A-Z][A-Z0-9\s/\-\.]+?)\s+([-\d,]+\.\d{2})\s+(\d{3})' match = re.search(pattern, line) if match: description = match.group(1).strip() amount = self.parse_currency(match.group(2)) est_no = match.group(3) deduction = { "estimate_no": est_no, "revenue_code": "", "description": f"{current_category}: {description}", "original": 0, "this_period": amount, "job_to_date": amount } deductions_list.append(deduction) self.data['deductions'] = [d for d in deductions_list if 'TOTAL DEDUCTIONS' not in d['description']] def parse_project_items(self, section: str): """Parse Project Record items table""" if not section: return lines = section.split('\n') current_item = None items = [] for i, line in enumerate(lines): # Skip header lines if 'ITEM' in line and 'DESCRIPTION' in line: continue if '---' in line: continue if not line.strip(): continue # Check if line starts with item number (3 digits) if re.match(r'^\d{3}\s', line): # Save previous item if exists if current_item: items.append(current_item) # Start new item current_item = self.parse_item_line(line) # Check for (F) marker on current or next line elif current_item: if '(F)' in line and len(line) < 10: current_item['sub_item'] = 'Y' # Add last item if current_item: items.append(current_item) self.data['prime_contract']['items'] = items def parse_item_line(self, line: str) -> Dict: """Parse a single item line from the project record""" item = { "item_number": "", "revenue_code": "", "description": "", "sub_item": "N", "quantities": { "original": 0.0, "this_period": 0.0, "percent_this_period": 0.0, "job_to_date": 0.0, "percent_jtd": 0.0 }, "units": "", "unit_price": 0.0, "amounts": { "original": 0.0, "this_period": 0.0, "job_to_date": 0.0, "moh": 0.0 } } # Parse by column positions (approximate) try: # Item number item['item_number'] = line[:3].strip().zfill(3) # Description desc_end = 45 for unit in ['LS', 'EA', 'LF', 'SF', 'SY', 'CY', 'TON', 'TN', 'LB', 'HR', 'SQFT', 'SQYD', 'CF', 'ACRE']: unit_pos = line[40:52].find(unit) if unit_pos > -1: desc_end = 40 + unit_pos break description = line[4:desc_end].strip() # Check for (F) marker - indicates subcontracted work if '(F)' in description: item['sub_item'] = 'Y' description = description.replace('(F)', '').strip() elif '(F)' in line: item['sub_item'] = 'Y' item['description'] = description # Find unit and parse numeric values remaining = line[desc_end:].strip() parts = remaining.split() if parts: # Unit item['units'] = parts[0] if parts[0] in ['LS', 'EA', 'LF', 'SF', 'SY', 'CY', 'TON', 'TN', 'LB', 'HR', 'SQFT', 'SQYD', 'CF', 'ACRE'] else '' # Parse numeric values nums = [] for part in parts[1:]: if re.match(r'^[\d,.-]+\.?\d*$', part): nums.append(self.parse_number(part)) # Assign values based on position if len(nums) > 0: item['unit_price'] = nums[0] if len(nums) > 1: item['amounts']['original'] = nums[1] # For LS items, original quantity is typically 1 if item['units'] == 'LS': item['quantities']['original'] = 1.0 else: # Back-calculate original quantity from amount and unit price if item['unit_price'] > 0: item['quantities']['original'] = item['amounts']['original'] / item['unit_price'] if len(nums) > 2: item['quantities']['this_period'] = nums[2] if len(nums) > 3: item['amounts']['this_period'] = nums[3] if len(nums) > 4: item['quantities']['job_to_date'] = nums[4] if len(nums) > 5: item['amounts']['job_to_date'] = nums[5] # Calculate percentages if item['quantities']['original'] > 0: if item['quantities']['this_period'] > 0: item['quantities']['percent_this_period'] = round( (item['quantities']['this_period'] / item['quantities']['original']) * 100, 2 ) if item['quantities']['job_to_date'] > 0: item['quantities']['percent_jtd'] = round( (item['quantities']['job_to_date'] / item['quantities']['original']) * 100, 2 ) except Exception as e: # Return item with defaults if parsing fails pass return item def parse_currency(self, value_str: str) -> float: """Parse currency string to float""" if not value_str: return 0.0 try: clean = value_str.replace(',', '').replace('$', '').strip() if clean.startswith('(') and clean.endswith(')'): clean = '-' + clean[1:-1] return float(clean) except: return 0.0 def parse_number(self, value_str: str) -> float: """Parse numeric string to float""" if not value_str: return 0.0 try: clean = value_str.replace(',', '').strip() return float(clean) except: return 0.0 def calculate_totals(self): """Calculate all totals from the parsed items""" # Prime contract totals prime_total = { "original": sum(item['amounts']['original'] for item in self.data['prime_contract']['items']), "this_period": sum(item['amounts']['this_period'] for item in self.data['prime_contract']['items']), "job_to_date": sum(item['amounts']['job_to_date'] for item in self.data['prime_contract']['items']), "moh": 0.0 } self.data['prime_contract']['total'] = prime_total # Change order totals co_total = { "original": 0, "this_period": sum(item['this_period_amt'] for item in self.data['change_orders']['items']), "job_to_date": sum(item['job_to_date_amt'] for item in self.data['change_orders']['items']) } self.data['change_orders']['total'] = co_total # Calculate overall project totals including extra work overall_total = { "original": prime_total["original"], "this_period": prime_total["this_period"] + co_total["this_period"], "job_to_date": prime_total["job_to_date"] + co_total["job_to_date"], "extra_work_this_estimate": self.data.get('extra_work_totals', {}).get('total_this_estimate', 0.0), "extra_work_previous": self.data.get('extra_work_totals', {}).get('total_previous_estimate', 0.0), "extra_work_to_date": self.data.get('extra_work_totals', {}).get('total_to_date', 0.0) } self.data['overall_totals'] = overall_total # Store estimate number in data self.data['main_estimate_no'] = self.main_estimate_no # Log summary information logger.info(f"Calculated totals for estimate {self.main_estimate_no}:") logger.info(f" Prime contract: ${overall_total['original']:,.2f} original, ${overall_total['job_to_date']:,.2f} to date") logger.info(f" Change orders: ${co_total['this_period']:,.2f} this period, ${co_total['job_to_date']:,.2f} to date") logger.info(f" Extra work: ${overall_total['extra_work_this_estimate']:,.2f} this estimate, ${overall_total['extra_work_to_date']:,.2f} to date") logger.info(f" Overall: ${overall_total['job_to_date']:,.2f} total to date") def parse_from_url(self, url: str) -> Dict: """Parse a CAS145 file from URL""" try: # Extract estimate number from URL url_parts = urlparse(url) filename = os.path.basename(url_parts.path) if filename.endswith('.txt'): base_name = filename.replace('.txt', '') match = re.search(r'(\d{3})$', base_name) if match: estimate_no = match.group(1) self.set_estimate_number(estimate_no) # Fetch and parse content content = self.fetch_from_url(url) return self.parse_content(content) except Exception as e: logger.error(f"Error parsing CAS145 from URL {url}: {str(e)}") return {} # Initialize scrapers pets_scraper = CaltransPETSScraper() cas145_parser = CAS145Parser() # API Endpoints @app.get("/") async def root(): """Root endpoint with API information""" return { "message": "Caltrans Payment Tracking API", "version": "1.0.0", "endpoints": { "get_contract_payments": "/contracts/{contract_number}/payments", "get_monthly_payments_with_details": "/contracts/{contract_number}/payments/monthly/{month}", "health_check": "/health" }, "workflow": { "basic_payments": "Scrapes PETS system for payment history", "monthly_details": "Scrapes PETS + follows detail_link to get CAS145 estimate details" } } @app.get("/health") async def health_check(): """Health check endpoint""" return {"status": "healthy", "timestamp": datetime.now().isoformat()} @app.get("/contracts/{contract_number}/payments", response_model=ContractResponse) async def get_contract_payments(contract_number: str): """ Retrieve complete payment history for a contract (basic PETS data only) Args: contract_number: The contract number to search for (e.g., '033F0704') Returns: Complete contract information including payment history from PETS system """ try: logger.info(f"Fetching payment data for contract: {contract_number}") contract_data = pets_scraper.scrape_contract(contract_number) if not contract_data or not contract_data.get('contract_details'): raise HTTPException( status_code=404, detail=f"Contract {contract_number} not found or no data available" ) contract_details = contract_data.get('contract_details', {}) if not contract_details.get('contract_number'): contract_details['contract_number'] = contract_number payment_history = contract_data.get('payment_history', []) contact_info = contract_data.get('contact_info', {}) summary = contract_data.get('summary', { 'total_payments': 0, 'total_disbursed': 0.0, 'total_held': 0.0, 'latest_payment_date': 'N/A' }) monthly_breakdown = contract_data.get('monthly_breakdown', {}) return ContractResponse( contract_details=ContractDetails(**contract_details), payment_history=[PaymentRecord(**payment) for payment in payment_history], contact_info=ContactInfo(**contact_info), summary=PaymentSummary(**summary), monthly_breakdown=monthly_breakdown, scraped_at=contract_data.get('scraped_at', datetime.now().isoformat()) ) except HTTPException: raise except Exception as e: logger.error(f"Error processing contract {contract_number}: {str(e)}") raise HTTPException(status_code=500, detail=f"Internal server error: {str(e)}") @app.get("/contracts/{contract_number}/payments/monthly/{month}", response_model=MonthlyPaymentSummary) async def get_monthly_payments(contract_number: str, month: str): """ Retrieve detailed payment information for a specific month This endpoint: 1. Scrapes PETS system to get payment history 2. Filters payments for the specified month 3. For each payment in that month, follows the detail_link 4. Parses the detailed CAS145 estimate information using main.py parser Args: contract_number: The contract number to search for month: Month in YYYY-MM format (e.g., '2024-12' for December 2024) Note: Handles both 2-digit and 4-digit year formats in payment dates (e.g., "12/06/24" or "12/06/2024") Returns: Monthly payment summary with detailed CAS145 estimate data for each payment """ try: # Validate month format try: datetime.strptime(month, '%Y-%m') except ValueError: raise HTTPException( status_code=400, detail="Month must be in YYYY-MM format (e.g., '2024-12')" ) logger.info(f"Fetching detailed monthly payment data for contract: {contract_number}, month: {month}") contract_data = pets_scraper.scrape_contract(contract_number) if not contract_data or not contract_data.get('payment_history'): raise HTTPException( status_code=404, detail=f"Contract {contract_number} not found or no payment history available" ) monthly_payments = [] total_disbursed = 0.0 total_held = 0.0 for payment in contract_data['payment_history']: if payment['release_date'] and payment['release_date'] != 'N/A': try: date_obj = pets_scraper.parse_payment_date(payment['release_date']) payment_month = date_obj.strftime('%Y-%m') if payment_month == month: monthly_payments.append(payment) total_disbursed += payment['disbursed_amount'] total_held += payment['held_amount'] except ValueError: continue if not monthly_payments: raise HTTPException( status_code=404, detail=f"No payments found for contract {contract_number} in month {month}" ) detailed_monthly_payments = [] for payment in monthly_payments: detailed_payment = payment.copy() if payment.get('detail_link'): try: logger.info(f"Fetching detailed estimate for payment {payment['estimate_number']} from {payment['detail_link']}") parser = CAS145Parser() detailed_estimate = parser.parse_from_url(payment['detail_link']) if detailed_estimate: detailed_payment['detailed_estimate'] = detailed_estimate logger.info(f"Successfully parsed detailed estimate for payment {payment['estimate_number']}") else: logger.warning(f"No detailed estimate data found for payment {payment['estimate_number']}") detailed_payment['detailed_estimate'] = None except Exception as e: logger.error(f"Error fetching detailed estimate for payment {payment['estimate_number']}: {str(e)}") detailed_payment['detailed_estimate'] = None else: logger.warning(f"No detail_link found for payment {payment['estimate_number']}") detailed_payment['detailed_estimate'] = None detailed_monthly_payments.append(detailed_payment) return MonthlyPaymentSummary( month=month, payment_count=len(detailed_monthly_payments), total_disbursed=total_disbursed, total_held=total_held, payments=[DetailedPaymentRecord(**payment) for payment in detailed_monthly_payments] ) except HTTPException: raise except Exception as e: logger.error(f"Error processing monthly data for contract {contract_number}, month {month}: {str(e)}") raise HTTPException(status_code=500, detail=f"Internal server error: {str(e)}") if __name__ == "__main__": uvicorn.run( "main:app", host="0.0.0.0", port=8000, reload=True, log_level="info" )
from fastapi import FastAPI, HTTPException
from pydantic import BaseModel, Field
from typing import Dict, List, Optional, Any
import requests
from bs4 import BeautifulSoup
import re
import time
import logging
from urllib.parse import urlparse
from datetime import datetime
import uvicorn
import os
# Configure logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
# FastAPI app initialization
app = FastAPI(
title="Caltrans Payment Tracking System",
description="API for retrieving contract payment history and detailed monthly estimates",
version="1.0.0"
)
# Pydantic Models
class PaymentRecord(BaseModel):
estimate_number: str = Field(..., description="Estimate number")
detail_link: Optional[str] = Field(None, description="Link to payment details")
voucher_link: Optional[str] = Field(None, description="Link to voucher")
type: str = Field(..., description="Payment type")
payment_type: str = Field(..., description="Payment classification")
release_date: str = Field(..., description="Payment release date")
held_amount: float = Field(..., description="Amount held")
disbursed_amount: float = Field(..., description="Amount disbursed")
class ContractDetails(BaseModel):
contract_number: str = Field(..., description="Contract number")
contractor_name: Optional[str] = Field(None, description="Contractor name")
last_updated: Optional[str] = Field(None, description="Last update date")
job_description: Optional[str] = Field(None, description="Job description")
contract_accepted: Optional[str] = Field(None, description="Contract acceptance date")
stop_notice_deadline: Optional[str] = Field(None, description="Stop notice deadline")
suit_filing_expiration: Optional[str] = Field(None, description="Suit filing expiration")
estimated_completion: Optional[str] = Field(None, description="Estimated completion date")
current_withhold: Optional[float] = Field(None, description="Current withhold amount")
currently_held: Optional[float] = Field(None, description="Currently held amount")
need_to_hold: Optional[float] = Field(None, description="Amount needed to hold")
class ContactInfo(BaseModel):
email: Optional[str] = Field(None, description="Contact email")
phone: Optional[str] = Field(None, description="Contact phone")
class PaymentSummary(BaseModel):
total_payments: int = Field(..., description="Total number of payments")
total_disbursed: float = Field(..., description="Total amount disbursed")
total_held: float = Field(..., description="Total amount held")
latest_payment_date: str = Field(..., description="Date of latest payment")
class Quantities(BaseModel):
original: float = Field(..., description="Original quantity")
this_period: float = Field(..., description="This period quantity")
percent_this_period: float = Field(..., description="Percentage this period")
job_to_date: float = Field(..., description="Job to date quantity")
percent_jtd: float = Field(..., description="Percentage job to date")
class Amounts(BaseModel):
original: float = Field(..., description="Original amount")
this_period: float = Field(..., description="This period amount")
job_to_date: float = Field(..., description="Job to date amount")
moh: float = Field(..., description="Materials on hand")
class LineItem(BaseModel):
item_number: str = Field(..., description="Item number")
revenue_code: str = Field(..., description="Revenue code")
description: str = Field(..., description="Work description")
sub_item: str = Field(..., description="Y=Subcontracted, N=Self-performed")
quantities: Quantities
units: str = Field(..., description="Unit of measurement")
unit_price: float = Field(..., description="Unit price")
amounts: Amounts
class DetailedPaymentRecord(BaseModel):
estimate_number: str = Field(..., description="Estimate number")
detail_link: Optional[str] = Field(None, description="Link to payment details")
voucher_link: Optional[str] = Field(None, description="Link to voucher")
type: str = Field(..., description="Payment type")
payment_type: str = Field(..., description="Payment classification")
release_date: str = Field(..., description="Payment release date")
held_amount: float = Field(..., description="Amount held")
disbursed_amount: float = Field(..., description="Amount disbursed")
detailed_estimate: Optional[Dict[str, Any]] = Field(None, description="Detailed CAS145 estimate data")
class MonthlyPaymentSummary(BaseModel):
month: str = Field(..., description="Month in YYYY-MM format")
payment_count: int = Field(..., description="Number of payments in month")
total_disbursed: float = Field(..., description="Total disbursed in month")
total_held: float = Field(..., description="Total held in month")
payments: List[DetailedPaymentRecord] = Field(..., description="Individual payments with detailed estimates")