src/parsers.py

"""
Meeting note parsers for extracting structured data from markdown files.
"""
from pathlib import Path
from typing import List, Optional
from datetime import datetime
from pydantic import BaseModel, Field
import re


class ActionItem(BaseModel):
    """Represents an action item from a meeting."""
    task: str
    assignee: Optional[str] = None
    deadline: Optional[str] = None
    completed: bool = False


class MeetingNote(BaseModel):
    """Represents a parsed meeting note."""
    project_name: str
    title: str
    date: Optional[datetime] = None
    participants: List[str] = Field(default_factory=list)
    discussion: Optional[str] = None
    decisions: List[str] = Field(default_factory=list)
    action_items: List[ActionItem] = Field(default_factory=list)
    blockers: List[str] = Field(default_factory=list)
    file_path: str


class MeetingParser:
    """Parser for markdown meeting notes."""

    @staticmethod
    def parse_date(date_str: str) -> Optional[datetime]:
        """Parse date from various formats."""
        date_formats = [
            "%Y-%m-%d",
            "%d/%m/%Y",
            "%m/%d/%Y",
            "%B %d, %Y",
            "%b %d, %Y",
            "%Y/%m/%d"
        ]

        for fmt in date_formats:
            try:
                return datetime.strptime(date_str.strip(), fmt)
            except ValueError:
                continue
        return None

    @staticmethod
    def parse_action_item(line: str) -> Optional[ActionItem]:
        """Parse an action item line."""
        # Match patterns like:
        # - [ ] Task
        # - [x] Task
        # - [ ] Alice: Task by Jan 20
        # - [x] Bob: Task (by 2025-01-20)

        completed = False
        if "[x]" in line.lower() or "[✓]" in line or "[✔]" in line:
            completed = True

        # Remove checkbox markers
        line = re.sub(r'\[[ xX✓✔]\]', '', line).strip()
        line = line.lstrip('- ').strip()

        if not line:
            return None

        # Try to extract assignee
        assignee = None
        assignee_match = re.match(r'^([A-Za-z\s]+):\s*(.+)$', line)
        if assignee_match:
            assignee = assignee_match.group(1).strip()
            line = assignee_match.group(2).strip()

        # Try to extract deadline
        deadline = None
        deadline_patterns = [
            r'by\s+([A-Za-z]+\s+\d{1,2}(?:,\s+\d{4})?)',
            r'by\s+(\d{4}-\d{2}-\d{2})',
            r'\(by\s+([^)]+)\)',
        ]

        for pattern in deadline_patterns:
            deadline_match = re.search(pattern, line, re.IGNORECASE)
            if deadline_match:
                deadline = deadline_match.group(1).strip()
                line = re.sub(pattern, '', line, flags=re.IGNORECASE).strip()
                break

        return ActionItem(
            task=line,
            assignee=assignee,
            deadline=deadline,
            completed=completed
        )

    @staticmethod
    def parse(file_path: Path, project_name: str) -> Optional[MeetingNote]:
        """Parse a markdown meeting note file."""
        if not file_path.exists():
            return None

        content = file_path.read_text(encoding='utf-8')
        lines = content.split('\n')

        # Initialize fields
        title = file_path.stem.replace('-', ' ').replace('_', ' ').title()
        date = None
        participants = []
        discussion = []
        decisions = []
        action_items = []
        blockers = []

        current_section = None

        for line in lines:
            line_stripped = line.strip()

            # Skip empty lines
            if not line_stripped:
                continue

            # Check for title
            if line_stripped.startswith('# '):
                title = line_stripped[2:].strip()
                # Try to extract from "Meeting: X" format
                if title.lower().startswith('meeting:'):
                    title = title[8:].strip()
                continue

            # Check for metadata
            if line_stripped.lower().startswith('date:'):
                date_str = line_stripped[5:].strip()
                date = MeetingParser.parse_date(date_str)
                continue

            if line_stripped.lower().startswith('participants:'):
                participants_str = line_stripped[13:].strip()
                participants = [p.strip() for p in participants_str.split(',')]
                continue

            # Check for sections
            if line_stripped.startswith('## '):
                section_name = line_stripped[3:].strip().lower()
                if 'discussion' in section_name or 'notes' in section_name:
                    current_section = 'discussion'
                elif 'decision' in section_name:
                    current_section = 'decisions'
                elif 'action' in section_name or 'todo' in section_name or 'task' in section_name:
                    current_section = 'action_items'
                elif 'blocker' in section_name or 'issue' in section_name:
                    current_section = 'blockers'
                else:
                    current_section = 'discussion'
                continue

            # Add content to current section
            if current_section == 'discussion':
                discussion.append(line_stripped)
            elif current_section == 'decisions':
                if line_stripped.startswith('-') or line_stripped.startswith('*'):
                    decisions.append(line_stripped.lstrip('-*').strip())
            elif current_section == 'action_items':
                if '[' in line_stripped:
                    action_item = MeetingParser.parse_action_item(line_stripped)
                    if action_item:
                        action_items.append(action_item)
            elif current_section == 'blockers':
                if line_stripped.startswith('-') or line_stripped.startswith('*'):
                    blockers.append(line_stripped.lstrip('-*').strip())

        return MeetingNote(
            project_name=project_name,
            title=title,
            date=date,
            participants=participants,
            discussion='\n'.join(discussion) if discussion else None,
            decisions=decisions,
            action_items=action_items,
            blockers=blockers,
            file_path=str(file_path)
        )


def load_meetings_from_directory(data_dir: Path) -> List[MeetingNote]:
    """Load all meeting notes from a directory structure."""
    meetings = []

    if not data_dir.exists():
        return meetings

    # Expected structure: data_dir/project_name/meetings/*.md
    for project_dir in data_dir.iterdir():
        if not project_dir.is_dir():
            continue

        project_name = project_dir.name
        meetings_dir = project_dir / "meetings"

        if not meetings_dir.exists():
            continue

        for meeting_file in meetings_dir.glob("*.md"):
            meeting = MeetingParser.parse(meeting_file, project_name)
            if meeting:
                meetings.append(meeting)

    return meetings