# pdf_parser.py
import re
from datetime import datetime
from PyPDF2 import PdfReader
from config import Config

def extract_text_from_pdf(pdf_path=None):
    if pdf_path is None:
        pdf_path = Config.PDF_PATH
    reader = PdfReader(pdf_path)
    full_text = ""
    for page in reader.pages:
        full_text += page.extract_text()
    return full_text

def parse_email_thread(text):
    """
    Splits the raw text into individual email records.
    Returns a list of dicts with keys: from, to, date, subject, body.
    """
    emails = []
    # Split by common email header patterns
    # We look for "From:" at the start of a line, but sometimes it's "From: " with spaces.
    parts = re.split(r'(?=^From:)', text, flags=re.MULTILINE)
    for part in parts:
        if not part.strip():
            continue
        from_match = re.search(r'^From:\s*(.*?)$', part, re.MULTILINE)
        to_match = re.search(r'^To:\s*(.*?)$', part, re.MULTILINE)
        # Try multiple date formats
        date_match = re.search(r'^(?:Sent|Date):\s*(.*?)$', part, re.MULTILINE)
        subject_match = re.search(r'^Subject:\s*(.*?)$', part, re.MULTILINE)
        # Body is everything after the headers – we can split on the first blank line after headers
        lines = part.splitlines()
        body_start = 0
        for i, line in enumerate(lines):
            if re.match(r'^(From|To|Sent|Date|Subject|Cc|Bcc|Reply-To):', line, re.I):
                continue
            if line.strip() == "":
                body_start = i + 1
                break
        body = "\n".join(lines[body_start:]) if body_start < len(lines) else ""
        
        emails.append({
            'from': from_match.group(1).strip() if from_match else "",
            'to': to_match.group(1).strip() if to_match else "",
            'date': date_match.group(1).strip() if date_match else "",
            'subject': subject_match.group(1).strip() if subject_match else "",
            'body': body.strip()
        })
    return emails

def extract_days_from_text(text):
    """Find a phrase like '95 days' and return the integer."""
    days_match = re.search(r'(\d+)\s*days', text)
    if days_match:
        return int(days_match.group(1))
    return None

def get_days_without_account(text):
    """Try to extract from text; otherwise fallback to config start date."""
    days = extract_days_from_text(text)
    if days is not None:
        return days
    from datetime import datetime
    start = Config.get_start_date()
    return (datetime.now() - start).days