texera/codecov.yml at main · apache/texera · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements.  See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership.  The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License.  You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

# Codecov configuration for apache/texera. CI uploads 11 flags: frontend,
# amber, amber-integration, pyamber, agent-service, and the six platform
# services (config-service, access-control-service, file-service,
# computing-unit-managing-service, workflow-compiling-service,
# notebook-migration-service). Without an explicit config, defaults bite us
# in three places that this file fixes:
#   1. Single-stack PRs leave non-uploaded flags as `?` in the comment and
#      drop them from the rollup; carrying forward the base-branch report
#      backfills them.
#   2. Coverage of generated code (protobuf), build output, and the
#      tests themselves dilutes the "real" coverage figure. Ignore
#      those paths.
#   3. Default status checks fail on any project drop and require 100%
#      patch coverage, so non-test refactors get a red Codecov check
#      even when behavior is preserved. Allow 1% slack on project and
#      require 60% patch coverage.

flag_management:
  default_rules:
    # If the current PR didn't run the job for a given flag, inherit
    # the latest report on the base branch instead of showing `?`.
    carryforward: true

ignore:
  # Generated protobuf — line counts swamp real code.
  - "amber/src/main/python/proto/**"
  - "frontend/src/app/common/type/proto/**"
  - "**/src_managed/**"          # sbt JOOQ / scalapb-generated sources
  # Build / install output — never source-of-truth.
  - "**/target/**"
  - "frontend/dist/**"
  - "frontend/.angular/**"
  # Test files don't count toward source coverage.
  - "**/test_*.py"
  - "**/*.spec.ts"
  - "**/src/test/**"             # sbt test source roots
  # Third-party vendored sources. These mirror the corresponding entries
  # in .licenserc.yaml — we don't maintain them, so their coverage figure
  # carries no signal.
  - "common/workflow-operator/src/main/scala/com/kjetland/**"          # MIT JSON schema generator
  - "frontend/src/app/common/formly/array.type.ts"                     # MIT ngx-formly derived
  - "frontend/src/app/common/formly/object.type.ts"
  - "frontend/src/app/common/formly/multischema.type.ts"
  - "frontend/src/app/common/formly/null.type.ts"
  - "pyright-language-service/src/main.ts"                             # MIT monaco-languageclient derived
  - "pyright-language-service/src/language-server-runner.ts"
  - "pyright-language-service/src/server-commons.ts"

coverage:
  status:
    project:
      default:
        # Don't drop below the base-branch baseline, but tolerate
        # ≤1% jitter from re-runs / flaky tests / ignore-rule churn.
        target: auto
        threshold: 1%
    patch:
      default:
        # Require new/changed lines in a PR to be at least 60% covered.
        # Lower than a 100%-patch policy so legitimate refactors and
        # config-only changes aren't blocked, but high enough that net-new
        # code ships with tests. The 10% threshold softens the quantization
        # on tiny diffs (e.g. a 2-coverable-line change passes at 1 line /
        # 50%) — patch passes when coverage is >= target - threshold = 50%.
        target: 60%
        threshold: 10%

comment:
  # CI is label-gated: a single-stack PR (e.g. a frontend-only change)
  # uploads only its own flag. carryforward backfills the rest from main;
  # this surfaces those carried rows (rendered `<ø>`) instead of hiding
  # them, so every PR's table lists all flags.
  show_carryforward_flags: true