Make continuous deployment safe by comparing before and after webpage screenshots for each release. Depicted shows when any visual, perceptual differences are found. This is the ultimate, automated end-to-end test.
Starts deprecating phantomjs flags in favor of more general capture flags.
Tested this with run_combined.sh using a --capture_binary=python and capture_script pointing to capture.py then called run_url_pair_diff.sh with command_executor and desired_capabilities args. Screenshots were loaded from Browserstack and worked great.
Still more testing to do, so for now nothing really changes for anyone with this diff, but if things go well, will at least write a README section about using BrowserStack instead of phantomjs.
Starts deprecating
phantomjs
flags in favor of more generalcapture
flags. Tested this withrun_combined.sh
using a--capture_binary=python
andcapture_script
pointing tocapture.py
then calledrun_url_pair_diff.sh
withcommand_executor
anddesired_capabilities
args. Screenshots were loaded from Browserstack and worked great.Still more testing to do, so for now nothing really changes for anyone with this diff, but if things go well, will at least write a README section about using BrowserStack instead of phantomjs.