run2_pdf-parser

Improved PDF Schedule Parsing

This advanced skill automates the extraction of schedules from visual PDFs. Instead of manual inference, it mathematically calculates time blocks by using baseline anchor lines.

Dynamic Time Calculation

By identifying the top-most timeline grid line (e.g., the 12:00 AM line), you can calculate any Y-coordinate's corresponding time.

import fitz

def parse_calendar(pdf_path):
    doc = fitz.open(pdf_path)
    page = doc.load_page(0)
    
    # Extract lines to find the 15-minute grid spacing
    lines = [d["rect"] for d in page.get_drawings() if d.get("fill") is None and d["rect"].height < 2]
    lines.sort(key=lambda r: r.y0)
    
    base_y = lines[0].y0  # The 12:00 AM line
    interval_y = lines[1].y0 - lines[0].y0  # Height of 15 minutes
    hour_height = interval_y * 4
    
    # Extract rectangles (appointments)
    # Exclude the background/grid lines and thin borders by checking width/height and colors
    blocks = []
    for d in page.get_drawings():
        rect = d["rect"]
        color = d.get("fill")
        if color and rect.width > 100:  # Assuming meeting blocks span the column width
            start_hour = (rect.y0 - base_y) / hour_height
            end_hour = (rect.y1 - base_y) / hour_height
            blocks.append({
                "start": start_hour,
                "end": end_hour,
                "color": color
            })
    return blocks

Identifying Flexible Slots

By checking the RGB values of the fill color, we can identify "blue-colored" blocks that denote flexible time and drop them from our strict occupied schedule list.