GnuTests.yml

name: GnuTests

# spell-checker:ignore (abbrev/names) CodeCov gnulib GnuTests Swatinem
# spell-checker:ignore (jargon) submodules
# spell-checker:ignore (libs/utils) autopoint chksum gperf lcov libexpect pyinotify shopt texinfo valgrind libattr libcap taiki-e
# spell-checker:ignore (options) Ccodegen Coverflow Cpanic Zpanic
# spell-checker:ignore (people) Dawid Dziurla * dawidd dtolnay
# spell-checker:ignore (vars) FILESET SUBDIRS XPASS

# * note: to run a single test => `REPO/util/run-gnu-test.sh PATH/TO/TEST/SCRIPT`

on:
  pull_request:
  push:
    branches:
      - main

permissions:
  contents: read

# End the current execution if there is a new changeset in the PR.
concurrency:
  group: ${{ github.workflow }}-${{ github.ref }}
  cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}

jobs:
  gnu:
    permissions:
      actions: read  # for dawidd6/action-download-artifact to query and download artifacts
      contents: read  # for actions/checkout to fetch code
      pull-requests: read  # for dawidd6/action-download-artifact to query commit hash
    name: Run GNU tests
    runs-on: ubuntu-24.04
    steps:
    - name: Initialize workflow variables
      id: vars
      shell: bash
      run: |
        ## VARs setup
        outputs() { step_id="${{ github.action }}"; for var in "$@" ; do echo steps.${step_id}.outputs.${var}="${!var}"; echo "${var}=${!var}" >> $GITHUB_OUTPUT; done; }
        # * config
        path_GNU="gnu"
        path_GNU_tests="${path_GNU}/tests"
        path_UUTILS="uutils"
        path_reference="reference"
        outputs path_GNU path_GNU_tests path_reference path_UUTILS
        #
        repo_default_branch="${{ github.event.repository.default_branch }}"
        repo_GNU_ref="v9.5"
        repo_reference_branch="${{ github.event.repository.default_branch }}"
        outputs repo_default_branch repo_GNU_ref repo_reference_branch
        #
        SUITE_LOG_FILE="${path_GNU_tests}/test-suite.log"
        ROOT_SUITE_LOG_FILE="${path_GNU_tests}/test-suite-root.log"
        TEST_LOGS_GLOB="${path_GNU_tests}/**/*.log" ## note: not usable at bash CLI; [why] double globstar not enabled by default b/c MacOS includes only bash v3 which doesn't have double globstar support
        TEST_FILESET_PREFIX='test-fileset-IDs.sha1#'
        TEST_FILESET_SUFFIX='.txt'
        TEST_SUMMARY_FILE='gnu-result.json'
        TEST_FULL_SUMMARY_FILE='gnu-full-result.json'
        outputs SUITE_LOG_FILE ROOT_SUITE_LOG_FILE TEST_FILESET_PREFIX TEST_FILESET_SUFFIX TEST_LOGS_GLOB TEST_SUMMARY_FILE TEST_FULL_SUMMARY_FILE
    - name: Checkout code (uutil)
      uses: actions/checkout@v4
      with:
        path: '${{ steps.vars.outputs.path_UUTILS }}'
    - uses: dtolnay/rust-toolchain@master
      with:
        toolchain: stable
        components: rustfmt
    - uses: Swatinem/rust-cache@v2
      with:
        workspaces: "./${{ steps.vars.outputs.path_UUTILS }} -> target"
    - name: Checkout code (GNU coreutils)
      uses: actions/checkout@v4
      with:
        repository: 'coreutils/coreutils'
        path: '${{ steps.vars.outputs.path_GNU }}'
        ref: ${{ steps.vars.outputs.repo_GNU_ref }}
        submodules: false

    - name: Override submodule URL and initialize submodules
      # Use github instead of upstream git server
      run: |
        git submodule sync --recursive
        git config submodule.gnulib.url https://github.com/coreutils/gnulib.git
        git submodule update --init --recursive --depth 1
      working-directory: ${{ steps.vars.outputs.path_GNU }}

    - name: Retrieve reference artifacts
      uses: dawidd6/action-download-artifact@v6
      # ref: <https://github.com/dawidd6/action-download-artifact>
      continue-on-error: true ## don't break the build for missing reference artifacts (may be expired or just not generated yet)
      with:
        workflow: GnuTests.yml
        branch: "${{ steps.vars.outputs.repo_reference_branch }}"
        # workflow_conclusion: success ## (default); * but, if commit with failed GnuTests is merged into the default branch, future commits will all show regression errors in GnuTests CI until o/w fixed
        workflow_conclusion: completed ## continually recalibrates to last commit of default branch with a successful GnuTests (ie, "self-heals" from GnuTest regressions, but needs more supervision for/of regressions)
        path: "${{ steps.vars.outputs.path_reference }}"
    - name: Install dependencies
      shell: bash
      run: |
        ## Install dependencies
        sudo apt-get update
        sudo apt-get install -y autoconf autopoint bison texinfo gperf gcc g++ gdb python3-pyinotify jq valgrind libexpect-perl libacl1-dev libattr1-dev libcap-dev libselinux1-dev attr
    - name: Add various locales
      shell: bash
      run: |
        ## Add various locales
        echo "Before:"
        locale -a
        ## Some tests fail with 'cannot change locale (en_US.ISO-8859-1): No such file or directory'
        ## Some others need a French locale
        sudo locale-gen
        sudo locale-gen --keep-existing fr_FR
        sudo locale-gen --keep-existing fr_FR.UTF-8
        sudo locale-gen --keep-existing sv_SE
        sudo locale-gen --keep-existing sv_SE.UTF-8
        sudo locale-gen --keep-existing en_US
        sudo locale-gen --keep-existing ru_RU.KOI8-R

        sudo update-locale
        echo "After:"
        locale -a
    - name: Build binaries
      shell: bash
      run: |
        ## Build binaries
        cd '${{ steps.vars.outputs.path_UUTILS }}'
        bash util/build-gnu.sh --release-build
    - name: Run GNU tests
      shell: bash
      run: |
        ## Run GNU tests
        path_GNU='${{ steps.vars.outputs.path_GNU }}'
        path_UUTILS='${{ steps.vars.outputs.path_UUTILS }}'
        bash "${path_UUTILS}/util/run-gnu-test.sh"
    - name: Run GNU root tests
      shell: bash
      run: |
        ## Run GNU root tests
        path_GNU='${{ steps.vars.outputs.path_GNU }}'
        path_UUTILS='${{ steps.vars.outputs.path_UUTILS }}'
        bash "${path_UUTILS}/util/run-gnu-test.sh" run-root
    - name: Extract testing info into JSON
      shell: bash
      run : |
        ## Extract testing info into JSON
        path_UUTILS='${{ steps.vars.outputs.path_UUTILS }}'
        python ${path_UUTILS}/util/gnu-json-result.py ${{ steps.vars.outputs.path_GNU_tests }} > ${{ steps.vars.outputs.TEST_FULL_SUMMARY_FILE }}
    - name: Extract/summarize testing info
      id: summary
      shell: bash
      run: |
        ## Extract/summarize testing info
        outputs() { step_id="${{ github.action }}"; for var in "$@" ; do echo steps.${step_id}.outputs.${var}="${!var}"; echo "${var}=${!var}" >> $GITHUB_OUTPUT; done; }
        #
        path_UUTILS='${{ steps.vars.outputs.path_UUTILS }}'
        #
        SUITE_LOG_FILE='${{ steps.vars.outputs.SUITE_LOG_FILE }}'
        ROOT_SUITE_LOG_FILE='${{ steps.vars.outputs.ROOT_SUITE_LOG_FILE }}'
        ls -al ${SUITE_LOG_FILE} ${ROOT_SUITE_LOG_FILE}

        if test -f "${SUITE_LOG_FILE}"
        then
            source ${path_UUTILS}/util/analyze-gnu-results.sh ${SUITE_LOG_FILE} ${ROOT_SUITE_LOG_FILE}
            if [[ "$TOTAL" -eq 0 || "$TOTAL" -eq 1 ]]; then
              echo "::error ::Failed to parse test results from '${SUITE_LOG_FILE}'; failing early"
              exit 1
            fi
            output="GNU tests summary = TOTAL: $TOTAL / PASS: $PASS / FAIL: $FAIL / ERROR: $ERROR / SKIP: $SKIP"
            echo "${output}"
            if [[ "$FAIL" -gt 0 || "$ERROR" -gt 0 ]]; then echo "::warning ::${output}" ; fi
            jq -n \
                  --arg date "$(date --rfc-email)" \
                  --arg sha "$GITHUB_SHA" \
                  --arg total "$TOTAL" \
                  --arg pass "$PASS" \
                  --arg skip "$SKIP" \
                  --arg fail "$FAIL" \
                  --arg xpass "$XPASS" \
                  --arg error "$ERROR" \
                  '{($date): { sha: $sha, total: $total, pass: $pass, skip: $skip, fail: $fail, xpass: $xpass, error: $error, }}' > '${{ steps.vars.outputs.TEST_SUMMARY_FILE }}'
            HASH=$(sha1sum '${{ steps.vars.outputs.TEST_SUMMARY_FILE }}' | cut --delim=" " -f 1)
            outputs HASH
          else
            echo "::error ::Failed to find summary of test results (missing '${SUITE_LOG_FILE}'); failing early"
            exit 1
          fi
          # Compress logs before upload (fails otherwise)
          gzip ${{ steps.vars.outputs.TEST_LOGS_GLOB }}
    - name: Reserve SHA1/ID of 'test-summary'
      uses: actions/upload-artifact@v4
      with:
        name: "${{ steps.summary.outputs.HASH }}"
        path: "${{ steps.vars.outputs.TEST_SUMMARY_FILE }}"
    - name: Reserve test results summary
      uses: actions/upload-artifact@v4
      with:
        name: test-summary
        path: "${{ steps.vars.outputs.TEST_SUMMARY_FILE }}"
    - name: Reserve test logs
      uses: actions/upload-artifact@v4
      with:
        name: test-logs
        path: "${{ steps.vars.outputs.TEST_LOGS_GLOB }}"
    - name: Upload full json results
      uses: actions/upload-artifact@v4
      with:
        name: gnu-full-result.json
        path: ${{ steps.vars.outputs.TEST_FULL_SUMMARY_FILE }}
    - name: Compare test failures VS reference
      shell: bash
      run: |
        ## Compare test failures VS reference
        have_new_failures=""
        REF_LOG_FILE='${{ steps.vars.outputs.path_reference }}/test-logs/test-suite.log'
        ROOT_REF_LOG_FILE='${{ steps.vars.outputs.path_reference }}/test-logs/test-suite-root.log'
        REF_SUMMARY_FILE='${{ steps.vars.outputs.path_reference }}/test-summary/gnu-result.json'
        REPO_DEFAULT_BRANCH='${{ steps.vars.outputs.repo_default_branch }}'
        path_UUTILS='${{ steps.vars.outputs.path_UUTILS }}'
        # https://github.com/uutils/coreutils/issues/4294
        # https://github.com/uutils/coreutils/issues/4295
        IGNORE_INTERMITTENT="${path_UUTILS}/.github/workflows/ignore-intermittent.txt"

        mkdir -p ${{ steps.vars.outputs.path_reference }}

        COMMENT_DIR="${{ steps.vars.outputs.path_reference }}/comment"
        mkdir -p ${COMMENT_DIR}
        echo ${{ github.event.number }} > ${COMMENT_DIR}/NR
        COMMENT_LOG="${COMMENT_DIR}/result.txt"

        # The comment log might be downloaded from a previous run
        # We only want the new changes, so remove it if it exists.
        rm -f ${COMMENT_LOG}
        touch ${COMMENT_LOG}

        compare_tests() {
          local new_log_file=$1
          local ref_log_file=$2
          local test_type=$3  # "standard" or "root"

          if test -f "${ref_log_file}"; then
            echo "Reference ${test_type} test log SHA1/ID: $(sha1sum -- "${ref_log_file}") - ${test_type}"
            REF_ERROR=$(sed -n "s/^ERROR: \([[:print:]]\+\).*/\1/p" "${ref_log_file}"| sort)
            CURRENT_RUN_ERROR=$(sed -n "s/^ERROR: \([[:print:]]\+\).*/\1/p" "${new_log_file}" | sort)
            REF_FAILING=$(sed -n "s/^FAIL: \([[:print:]]\+\).*/\1/p" "${ref_log_file}"| sort)
            CURRENT_RUN_FAILING=$(sed -n "s/^FAIL: \([[:print:]]\+\).*/\1/p" "${new_log_file}" | sort)
            echo "Detailled information:"
            echo "REF_ERROR = ${REF_ERROR}"
            echo "CURRENT_RUN_ERROR = ${CURRENT_RUN_ERROR}"
            echo "REF_FAILING = ${REF_FAILING}"
            echo "CURRENT_RUN_FAILING = ${CURRENT_RUN_FAILING}"

            # Compare failing and error tests
            for LINE in ${CURRENT_RUN_FAILING}
            do
              if ! grep -Fxq ${LINE}<<<"${REF_FAILING}"
              then
                if ! grep ${LINE} ${IGNORE_INTERMITTENT}
                then
                  MSG="GNU test failed: ${LINE}. ${LINE} is passing on '${REPO_DEFAULT_BRANCH}'. Maybe you have to rebase?"
                  echo "::error ::$MSG"
                  echo $MSG >> ${COMMENT_LOG}
                  have_new_failures="true"
                else
                  MSG="Skip an intermittent issue ${LINE} (fails in this run but passes in the 'main' branch)"
                  echo "::warning ::$MSG"
                  echo $MSG >> ${COMMENT_LOG}
                  echo ""
                fi
              fi
            done

            for LINE in ${REF_FAILING}
            do
              if ! grep -Fxq ${LINE}<<<"${CURRENT_RUN_FAILING}"
              then
                if ! grep ${LINE} ${IGNORE_INTERMITTENT}
                then
                  MSG="Congrats! The gnu test ${LINE} is no longer failing!"
                  echo "::warning ::$MSG"
                  echo $MSG >> ${COMMENT_LOG}
                else
                  MSG="Skipping an intermittent issue ${LINE} (passes in this run but fails in the 'main' branch)"
                  echo "::warning ::$MSG"
                  echo $MSG >> ${COMMENT_LOG}
                  echo ""
                fi
              fi
            done

            for LINE in ${CURRENT_RUN_ERROR}
            do
              if ! grep -Fxq ${LINE}<<<"${REF_ERROR}"
              then
                MSG="GNU test error: ${LINE}. ${LINE} is passing on '${REPO_DEFAULT_BRANCH}'. Maybe you have to rebase?"
                echo "::error ::$MSG"
                echo $MSG >> ${COMMENT_LOG}
                have_new_failures="true"
              fi
            done

            for LINE in ${REF_ERROR}
            do
              if ! grep -Fxq ${LINE}<<<"${CURRENT_RUN_ERROR}"
              then
                MSG="Congrats! The gnu test ${LINE} is no longer ERROR!"
                echo "::warning ::$MSG"
                echo $MSG >> ${COMMENT_LOG}
              fi
            done
          else
            echo "::warning ::Skipping ${test_type} test failure comparison; no prior reference test logs are available."
          fi
        }

        # Compare standard tests
        compare_tests '${{ steps.vars.outputs.path_GNU_tests }}/test-suite.log' "${REF_LOG_FILE}" "standard"

        # Compare root tests
        compare_tests '${{ steps.vars.outputs.path_GNU_tests }}/test-suite-root.log' "${ROOT_REF_LOG_FILE}" "root"

        if test -n "${have_new_failures}" ; then exit -1 ; fi
    - name: Upload comparison log (for GnuComment workflow)
      if: success() || failure() # run regardless of prior step success/failure
      uses: actions/upload-artifact@v4
      with:
        name: comment
        path: ${{ steps.vars.outputs.path_reference }}/comment/
    - name: Compare test summary VS reference
      if: success() || failure() # run regardless of prior step success/failure
      shell: bash
      run: |
        ## Compare test summary VS reference
        REF_SUMMARY_FILE='${{ steps.vars.outputs.path_reference }}/test-summary/gnu-result.json'
        if test -f "${REF_SUMMARY_FILE}"; then
          echo "Reference SHA1/ID: $(sha1sum -- "${REF_SUMMARY_FILE}")"
          mv "${REF_SUMMARY_FILE}" main-gnu-result.json
          python uutils/util/compare_gnu_result.py
        else
          echo "::warning ::Skipping test summary comparison; no prior reference summary is available."
        fi