BrowserActivity_DayofWeek

Extracts history activity for Safari and Chrome on OS X/macOS and plots proportional amount of activity for different days of the week and for different hours of a day (localized by time zone).

I don't really know what this says about my behaviour and would like to process the data more in future versions to gain more insight.

To use:

1. Download this script or git clone https://github.com/jasonrwang/BrowserActivity_DayofWeek.git so you have a local copy
2. Use Terminal to navigate to find main.py
3. Type in python main.py

Alternatively, if you download this as a ZIP to your Downloads folder, use python ~/Downloads/BrowserActivity_DayofWee-master/main.py from Terminal

Timezones:

Input into timezones.txt a comma separated list of timezones you were in. If you were only in one timezone, you will still need to use this.

Format: start_time,end_time,time_zone Times should be in ISO 8601 () format or 'now' (without quotations). Timezones should be in Olson time zones. Use 0,now,[Your timezone code] as a default.

Changelog

v0.4.2 Limited timezone support

• Added time zone support as defined in timezones.txt
• Can filter all data to a single time period as well if it is in the same time zone
• Timezones can be improved if it was easier to input!

v0.4.1 Hour of Day Analysis added

• Also analyzes and displays information about usage at different hours during a day
• Uses naive datetimes i.e. does not account for history in different timezones

v0.3 Chrome added, Rename

• Also gathers information from Google Chrome
• Processes sum of Chrome and Safari data
• Rename to BrowserActivity_DayofWeek from SafariActivity_DayofWeek

v0.2 Percent Use

• Displays output in percentage of total instead of raw value since it is more valuable and protects privacy

v0.1 Basic functionality added:

• Finds History.db file from ~/Library/Safari/ automatically
• Will catch if the database cannot be opened
• Converts NSDate timestamps for history data into day of the week
  • Note that time zone differences are probably not correct since Safari likely looks to the system to determine the time and travelling users (like me) don't always update this immediately
• Matlibplot histogram works but is ugly

P.S. This is my first python script! I have a feeling it could be much faster since I suspect I did not use numpy to its full capabilities. Suggestions are welcome.