tahoe-lafs/.codecov.yml

49 lines
1.7 KiB
YAML
Raw Normal View History

2020-08-22 12:14:08 +00:00
# Override defaults for codecov.io checks.
#
# Documentation is at https://docs.codecov.io/docs/codecov-yaml;
# reference is at https://docs.codecov.io/docs/codecovyml-reference.
#
# To validate this file, use:
#
# curl --data-binary @.codecov.yml https://codecov.io/validate
#
# Codecov's defaults seem to leave red marks in GitHub CI checks in a
# rather arbitrary manner, probably because of non-determinism in
# coverage (see https://tahoe-lafs.org/trac/tahoe-lafs/ticket/2891)
# and maybe because computers are bad with floating point numbers.
2020-08-22 12:14:08 +00:00
# Allow coverage percentage a precision of zero decimals, and round to
# the nearest number (for example, 89.957 to to 90; 89.497 to 89%).
# Coverage above 90% is good, below 80% is bad.
coverage:
round: nearest
range: 80..90
precision: 0
# Aim for a target test coverage of 90% in codecov/project check (do
# not allow project coverage to drop below that), and allow
# codecov/patch a threshold of 1% (allow coverage in changes to drop
# by that much, and no less). That should be good enough for us.
status:
project:
default:
target: 90%
threshold: 1%
patch:
default:
threshold: 1%
2020-12-14 15:36:45 +00:00
codecov:
# This is a public repository so supposedly we don't "need" to use an upload
# token. However, using one makes sure that CI jobs running against forked
# repositories have coverage uploaded to the right place in codecov so
2020-12-14 16:04:31 +00:00
# their reports aren't incomplete.
2020-12-14 15:36:45 +00:00
token: "abf679b6-e2e6-4b33-b7b5-6cfbd41ee691"
2020-12-17 17:40:20 +00:00
notify:
2020-12-17 18:08:33 +00:00
# The reference documentation suggests that this is the default setting:
# https://docs.codecov.io/docs/codecovyml-reference#codecovnotifywait_for_ci
# However observation suggests otherwise.
wait_for_ci: true