functional-tests.yml

name: Functional tests, gcc13

# In secp256k1 some assertions are not exactly equal when coverage is enabled
# and when it is not. Hence, it is better to run the two cases separately.

on:
  push:
    branches:
      - frost
  pull_request:
  # Allows you to run this workflow manually from the Actions tab
  workflow_dispatch:

jobs:
  functional-tests:
    runs-on: ubuntu-22.04
    # Use fedora:38 to compile using gcc-13
    container:
      image: fedora:38
    steps:
      - name: Install build dependencies
        run: |
          dnf install -y \
              autoconf \
              automake \
              gcc \
              gcovr \
              libtool \
              pkg-config
      - uses: actions/checkout@v3
        with:
          fetch-depth: 1
      - name: Run autogen
        run: ./autogen.sh
      - name: Functional tests with code coverage, no -Werror. Generate a summary of code coverage
        # When code coverage is enabled, the upstream secp256k1 generates a
        # warning -> we cannot use -Werror.
        continue-on-error: true
        id: tests_with_coverage
        run: |
          ./configure \
              SECP_CFLAGS="" \
              --enable-tests \
              --disable-exhaustive-tests \
              --disable-benchmark \
              --enable-coverage \
              --enable-examples \
              --enable-experimental \
              --enable-module-frost
          make -j
          make check
          gcovr --exclude 'src/bench*' --print-summary
      - name: Clean the build
        run: |
          make clean
          # make clean does not remove the compiled binaries, and we are going
          # to regenerate "tests"
          rm -f tests
      - name: Functional tests with -Werror, no code coverage
        continue-on-error: true
        id: tests_no_coverage
        run: |
          ./configure \
              SECP_CFLAGS="-Werror" \
              --enable-tests \
              --disable-exhaustive-tests \
              --disable-benchmark \
              --disable-coverage \
              --enable-examples \
              --enable-experimental \
              --enable-module-frost
          make -j
          ./tests
      - name: Summarize outcomes. Fail if any step failed.
        run: |
          # summary
          RED='\033[0;31m'
          GREEN='\033[0;32m'
          NC='\033[0m' # No Color

          echo -n "Functional tests with code coverage, no -Werror: "
          if [[ ${{ steps.tests_with_coverage.outcome }} == "success" ]]; then
            printf "${GREEN}SUCCESS${NC}\n"
          else
            printf "${RED}FAIL${NC} (${{ steps.tests_with_coverage.outcome }}), please check\n"
          fi

          echo -n "Functional tests with -Werror, no code coverage: "
          if [[ ${{ steps.tests_no_coverage.outcome }} == "success" ]]; then
            printf "${GREEN}SUCCESS${NC}\n"
          else
            printf "${RED}FAIL${NC} (${{ steps.tests_no_coverage.outcome }}), please check\n"
          fi

          if [[ ${{ steps.tests_with_coverage.outcome }} != "success" ]] || [[ ${{ steps.tests_no_coverage.outcome }} != "success" ]]; then
            exit 1
          fi