martino90/automatic_browser repository overview

This docker can be instrumented to automatically visit a list of URLs by means of Chrome Web Browser. It produces as output a set of HAR files (HTTP Archive, see http://www.softwareishard.com/blog/har-12-spec/⁠) for each visited URL. The HAR is generated by the chrome-har-capturer tool (see https://github.com/cyrus-and/chrome-har-capturer⁠). A HAR file contains the list of all the objects contained in page together with other statistics. Moreover it produces Tstat log files summarizing the TCP activity on the network (details in http://tstat.polito.it⁠). To run it, use this syntax (all paths must be absolute):

docker run \
-v <urls>:/opt/urls.txt \
-v <tstat_logs>:/opt/tstat_logs \
-v <hars>:/opt/hars  \
martino90/automatic_browser /opt/script/perform_experiment.sh \
[-p <pause>] [-a <max_attempts>]  [-t <timeout>]

Where:

• is the file containing the URLs to visit, one per line
• <tstat_logs> is the name of the directory where to store Tstat log files
• is the name of the directory where to store HAR files, one file for each visited page
• is the amount of seconds to wait after the onLoad event is triggered by Chrome. Default is 1.
• <max_attempts> is the maximum number of times the script tries to retrieve the page before giving up. Default is 3.
• is the maximum time in seconds the script waits for the page to load before giving up. Default is 60.