Toggl Track API JSON to CSV conversion

histogram.py

"""create a histogram of "hours per week" from timesheets"""

import csv
import sys
import matplotlib.pyplot as plt
from datetime import datetime, timedelta
import numpy

FNAME = "combined.csv"

with open(FNAME, "r", encoding="utf-8") as file:
    reader = csv.DictReader(file)
    data = [row for row in reader]

# sort
data = sorted(data, key= lambda r: datetime.strptime(r["start"][0:10], "%Y-%m-%d"))

first_date = datetime.strptime(data[0]["start"][0:10], "%Y-%m-%d")
isoy, isow, isod = first_date.isocalendar()
first_date_week = datetime.fromisocalendar(isoy, isow, 1)

dates_bins = [first_date_week]
seconds = [0]
for row in data:
    last_dt = dates_bins[-1]
    lisoy, lisow, lisod = last_dt.isocalendar()

    this_dt = datetime.strptime(row["start"][0:10], "%Y-%m-%d")
    sisoy, sisow, sisod = this_dt.isocalendar()
    duration_s = row["duration"]

    while sisoy > lisoy or sisow > lisow:
        next_dt = last_dt + timedelta(days=7)
        dates_bins.append(next_dt)
        seconds.append(0)
        last_dt = dates_bins[-1]
        lisoy, lisow, lisod = last_dt.isocalendar()

    seconds[-1] += int(duration_s)

last_date = datetime.strptime(data[-1]["start"][0:10], "%Y-%m-%d")
lisoy, lisow, lisod = last_date.isocalendar()
last_date_week = datetime.fromisocalendar(lisoy, lisow + 1, 1)

dates_bins.append(last_date_week)

print("dates_bins", dates_bins)
print("seconds", seconds)

hours = [s / 3600 for s in seconds]

plt.stairs(hours, dates_bins, fill=True)
plt.xlabel("Week")
plt.ylabel("Hours")
plt.grid(axis="y")

plt.show()

parse_toggl.sh

#!/bin/bash

# turns Toggl JSON API data into CSV format
#  currently ignores any array fields
#  you can override the TOGGL_KEYS env variable to set which keys you want in the CSV
# usage
#   get toggl data
#   https://developers.track.toggl.com/docs/api/time_entries/index.html
#   curl  https://api.track.toggl.com/api/v9/me/time_entries   -H "Content-Type: application/json"   -u <email>:<password> > toggl.json
#   curl  https://api.track.toggl.com/api/v9/me/time_entries   -H "Content-Type: application/json"   -u <api token>:api_token > toggl.json
#   get January entries, e.g.,
#    curl "https://api.track.toggl.com/api/v9/me/time_entries?start_date=2024-01-01&end_date=2024-02-01" -H "Content-Type: application/json" -u <api token>:api_token > 2024-01_toggl.json
#   script:
#   as argument:
#     ./parse_toggl.sh toggl.json
#   as pipe
#     cat toggl.json | /parse_toggl.sh
#   or just directly pipe curl output to the script (use "-s" with curl to suppress loading output to stderr)
# from https://gist.github.com/alifeee/5ccfdff391f1bfc894f6e21beb78996d

KEYS="${TOGGL_KEYS:-.id, .workspace_id, .project_id, .task_id, .billable, .start, .stop, .duration, .description, .duronly, .at, .server_deleted_at, .user_id, .uid, .wid, .pid}"

# print CSV headers
echo $KEYS | awk -F', ' '{for (i=1; i<NF; i++) {printf "%s,", substr($i, 2)}} END {printf "\n"}'

# print CSV content (parse JSON with jq)
cat $1 | jq -r '.[] | ['"${KEYS}"'] | @csv'