Split out project into own column if it had been provided (with colon)

This commit is contained in:
benjamin melançon 2021-04-27 21:45:37 -04:00
parent afbb5dbe1e
commit 796284931e

View file

@ -1,4 +1,5 @@
import pandas as pd
import numpy as np
# import matplotlib.pyplot as plt
timelog = pd.read_csv("timelog-titled.csv")
@ -12,6 +13,10 @@ timelog["time"] = 30
timelog["date"] = timelog["started"].dt.tz_convert("US/Pacific").dt.date
timelog["day_of_week"] = pd.to_datetime(timelog["date"]).dt.day_name()
timelog['project'] = (np.where(timelog['description'].str.contains(': '), timelog['description'].str.split(': ', 1).str[0], None))
timelog['description'] = (np.where(timelog['description'].str.contains(': '), timelog['description'].str.split(': ', 1).str[1], timelog['description']))
# Condense duplicate entries by date, summing the minutes spent, and listing
# the first started and last recorded times for each task.
tl = timelog.groupby(["date", "description"]).agg({"time": 'sum', "started": 'min', "recorded": 'max'}).reset_index()