People mentioned in We Didnt start the fire by Billy Joel. Data from https://en.wikipedia.org/wiki/List_of_references_in_We_Didn%27t_Start_the_Fire and https://everyday-learning.org/we-didnt-start-the-fire-historical-references/

dtesf2.csv

fire.py

import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import datetime
from matplotlib.patches import Patch

# Load the dataset
combined = pd.read_csv("datesF2_with_category.csv")

# --- CLEANUP ---

# Normalize and fix category labels
combined['category'] = combined['category'].replace({
    "Sports": "Sport",
    "Books and Movies": "Entertainment"
}).str.strip()

combined['coarse_category'] = combined['category']

def normalize_date_string(date_str):
    try:
        if pd.isna(date_str):
            return pd.NaT
        date_str = str(date_str).strip()
        if date_str.isdigit() and len(date_str) == 4:
            return datetime.datetime.strptime(date_str + "-01-01", "%Y-%m-%d")
        return pd.to_datetime(date_str, errors='coerce')
    except:
        return pd.NaT

combined['event_date'] = combined['date'].apply(normalize_date_string)

# Add manual extras
extra_events = {
    "Richard Nixon": [datetime.datetime(1969, 1, 20)],
    "John F. Kennedy": [datetime.datetime(1963, 11,22)]
}
combined['extra_event_dates'] = combined['subject2'].map(extra_events).where(
    combined['subject2'].isin(extra_events), None
)

# Manual fix for Bob Dylan's album release
combined.loc[combined['subject2'] == 'Bob Dylan', 'event_date'] = pd.to_datetime("1962-03-19")

# Filter valid people
people_df = combined.dropna(subset=['birth_year'])
current_year = datetime.datetime.now().year
people_df['death_year'] = people_df['death_year'].fillna(current_year)

# --- COLORS ---
coarse_color_map = {
    "Politics": "#1f77b4",
    "Entertainment": "#ff7f0e",
    "Sport": "#2ca02c",
    "Other": "#7f7f7f"
}

# --- PLOT ---

# Adjust height dynamically
row_height = 0.35
num_people = len(people_df)
fig_height = max(10, row_height * num_people)

plt.figure(figsize=(14, fig_height))

for _, row in people_df.iterrows():
    birth = int(row['birth_year'])
    death = int(row['death_year'])
    subject = row['subject2']
    category = str(row['coarse_category']).strip()
    y = subject

    color = coarse_color_map.get(category, 'gray')
    plt.hlines(y=y, xmin=birth, xmax=death, color=color, linewidth=4)

    if pd.notna(row['event_date']):
        year = row['event_date'].year
        if birth <= year <= death:
            plt.plot(year, y, 'o', color='black')

    if isinstance(row['extra_event_dates'], list):
        for extra_date in row['extra_event_dates']:
            year = extra_date.year
            if birth <= year <= death:
                plt.plot(year, y, 'o', color='black')

# --- Labels and Legend ---

plt.axvline(x=1989, color='black', linestyle='--')
plt.text(1990, -1, "1989 (song release)", fontsize=10)

legend_elements = [
    Patch(facecolor=color, edgecolor='none', label=label)
    for label, color in coarse_color_map.items()
]
plt.legend(
    handles=legend_elements,
    title="Famous For",
    bbox_to_anchor=(1.02, 1),
    loc="upper left",
    borderaxespad=0,
    frameon=False
)
plt.tight_layout(rect=[0, 0, 0.85, 1])  # THEN:
plt.title("People in Billy Joel’s “We Didn’t Start the Fire”", fontsize=20, weight='bold', pad=20)
plt.text(0.5, 1.02, "Dot represents the time of event referenced in the lyrics",
         ha='center', va='bottom', transform=plt.gca().transAxes, fontsize=12)
plt.figtext(0.23, 0.063, "Data Wikipedia by @iamreddave", ha="center", fontsize=9, style='italic', color="gray")

# Axes
plt.xlabel("Year")
plt.ylabel("Person")
plt.grid(axis='x', linestyle='--', alpha=0.5)
plt.yticks(fontsize=8)
plt.gca().invert_yaxis()
plt.xlim(left=1860, right=2025)
plt.axvline(x=2025, color='gray', linestyle='--', linewidth=1)
plt.subplots_adjust(top=0.92)
plt.savefig("lifespans_compact.png", dpi=300)
plt.show()