#!/bin/sh # This script should be run when the current output has been # certified to be entirely more correct than the previous output, # with no regressions. It makes the new output become the # baseline for future testing. # Generate the new test output # ./mktestout.sh rm -rf verified_output rm -rf temp1 temp2 # Make a copy in temp1 and rename to temp2. # cp -r test_output temp1 mv test_output temp2 # Gather headerdoc comments and modify the tree for temp2. # Copy the masterTOC file into temp1. This simulates the # environment used by testLinkResolver. # ../gatherHeaderDoc.pl temp2 cp temp2/masterTOC.html temp1 # Generate the list of expected diffs that should be # caused by gatherHeaderDoc and resolvelinks # ./diffhtml.sh temp1 temp2 > /dev/null mv htmlchanges lr_expected_htmlchanges # Make temp1 (the one on which gatherHeaderDoc/resolveLinks has NOT # been run) the official certified source. # mv temp1 verified_output # Clean up.... # rm -rf temp2