mirror of
https://github.com/NoRedInk/noredink-ui.git
synced 2024-12-26 23:22:07 +03:00
37 lines
1.4 KiB
Bash
Executable File
37 lines
1.4 KiB
Bash
Executable File
#!/usr/bin/env bash
|
|
set -euo pipefail
|
|
|
|
JSON_FILE="${1:-}"
|
|
if test -z "$JSON_FILE"; then
|
|
echo "Please specify a report JSON file as the first argument."
|
|
exit 1
|
|
fi
|
|
|
|
jq -r -f script/axe-report.jq "$JSON_FILE"
|
|
|
|
# Hey there! Did this script tell you to check out this file because the
|
|
# expected error count went down? Well done! Just change this number to the new
|
|
# value.
|
|
TARGET_ERRORS=5
|
|
|
|
# ideally we'd fail on any failures, but we have had a bunch build up over time!
|
|
# So right now, we need to fail if the error count is not exactly what we
|
|
# expect. This failure reminds us to come back and ratchet down the number of
|
|
# failures to the correct value.
|
|
NUM_ERRORS="$(jq '.violations | map(.nodes | length) | add' "$JSON_FILE")"
|
|
if test "$NUM_ERRORS" -ne "$TARGET_ERRORS"; then
|
|
echo "got $NUM_ERRORS errors, but expected $TARGET_ERRORS."
|
|
echo
|
|
echo 'If it went down, hooray!'
|
|
echo "Check out ${0:-} and change the count to the reported value above."
|
|
echo
|
|
echo "If it went up, let's fix it instead."
|
|
echo "Since there are so many errors right now, a decent debugging strategy is:"
|
|
echo
|
|
echo " 1. save this output somewhere ('make axe-report > errors.new')"
|
|
echo " 2. undo your changes ('git stash' or 'checkout master')"
|
|
echo " 3. regenerate the log with 'make axe-report > errors.old'"
|
|
echo " 4. see what's new with 'diff -u errors.old errors.new'"
|
|
exit 1
|
|
fi
|