generate_slackdump_heatmap.py

import json
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from datetime import datetime, timedelta, timezone
import argparse

# This script can be useful when filling your freelancer timesheets,
# it generates a heatmap image with the messages/hour/day.
#
# Input file should be a valid json generated with:
# go run ./cmd/slackdump search messages "from:<@U014RHC5U4A> before:2024-04-01 after:2024-02-29"
#
# See: https://github.com/rusq/slackdump

# Parse command line arguments
parser = argparse.ArgumentParser(description="Generate a heatmap of Slack messages.")
parser.add_argument('filename', type=str, help='The JSON file containing Slack messages')
args = parser.parse_args()

# Set the timezone directly in the script
user_tz = timezone(timedelta(hours=1))  # CET (Central European Time), you can change this as needed

# Load JSON data from the file
with open(args.filename, 'r') as file:
    data_list = json.load(file)

# Extract messages from all items in the list
messages = []
for item in data_list:
    if 'sm' in item:
        messages.extend([msg for msg in item['sm'] if msg.get('type') in {'message', 'group', 'im'}])

# Convert timestamps to timezone-aware datetime
for message in messages:
    timestamp = float(message["ts"].split('.')[0])
    message["datetime"] = datetime.fromtimestamp(timestamp, tz=timezone.utc).astimezone(user_tz)

# Create a DataFrame
df = pd.DataFrame(messages)

# Extract date and hour from datetime
df['date'] = df['datetime'].dt.date
df['hour'] = df['datetime'].dt.hour

# Pivot table to create heat map data
heatmap_data = df.pivot_table(index='date', columns='hour', aggfunc='size', fill_value=0)

# Plot the heatmap
plt.figure(figsize=(16, 10))
sns.heatmap(heatmap_data, cmap="YlGnBu", linewidths=.5)
plt.title("Heatmap of Messages per Day and Hour")
plt.xlabel("Hour of Day")
plt.ylabel("Date")
plt.xticks(rotation=90)
plt.show()