# Real multi-hour soak — #581 query-only memory-leak reproducer.
#
# WHY THIS EXISTS (separate from _soak.yml / nightly-soak.yml):
#   The nightly path was structurally incapable of running a real long soak:
#     1. nightly-soak.yml passes duration_minutes: 240, but _soak.yml's
#        soak-quick / soak-asan jobs hard-cap `timeout-minutes: 30` (45 for
#        ASan). GitHub kills the job at 30 min → the "4h" soak NEVER ran past
#        30 min. (Fixed in _soak.yml too, but this workflow guarantees the
#        right budget for the long #581 run.)
#     2. scripts/soak-test.sh's default mode reindexes every 2 min;
#        index_repository triggers cbm_mem_collect (mimalloc page return),
#        which sweeps the query-only leak — masking #581 even on a long run.
#        This workflow drives CBM_SOAK_MODE=query-leak, which never reindexes
#        and never mutates files, so the leak can accumulate and be detected
#        by soak-test.sh's RSS slope / ratio / ceiling analysis.
#
# NON-GATING: workflow_dispatch + push to qa/soak-** only. Never a required
# check, never blocks a merge.
#
# CRITICAL: timeout-minutes = duration + 60. A 240-min soak gets ~300 min.
name: Soak (multi-hour #581)

on:
  workflow_dispatch:
    inputs:
      duration_minutes:
        description: 'Soak duration in minutes (default: 240 = 4h)'
        type: number
        default: 240
      mode:
        description: 'Soak mode (query-leak = #581 detector, no reindex/mutate)'
        type: choice
        options: ['default', 'query-leak']
        default: 'query-leak'
  # Iteration convenience: pushing a qa/soak-** branch starts a real run from
  # that branch's own copy of this file (no main merge needed). Non-gating.
  push:
    branches: ['qa/soak-**']

permissions:
  contents: read

jobs:
  # ── Unix: full matrix (linux amd64+arm64, darwin arm64+amd64) ──────────────
  soak-unix:
    strategy:
      fail-fast: false
      matrix:
        include:
          - os: ubuntu-latest
            cc: gcc
            cxx: g++
          - os: ubuntu-24.04-arm
            cc: gcc
            cxx: g++
          - os: macos-14
            cc: cc
            cxx: c++
          - os: macos-15-intel
            cc: cc
            cxx: c++
    runs-on: ${{ matrix.os }}
    # Fixed budget (NOT the 30 min that silently truncated nightly). 320 min covers
    # the 240-min default soak + build + analysis. `timeout-minutes` is evaluated at
    # workflow setup where the `inputs` context is null on push events, so an
    # inputs-based expression here is a startup failure — keep it a literal.
    # (A workflow_dispatch run with duration > ~250 min should bump this.)
    timeout-minutes: 320
    steps:
      - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0

      - name: Install deps (Linux)
        if: startsWith(matrix.os, 'ubuntu')
        run: sudo apt-get update && sudo apt-get install -y zlib1g-dev python3 git

      - name: Build (prod binary)
        run: scripts/build.sh CC=${{ matrix.cc }} CXX=${{ matrix.cxx }}

      - name: Soak
        env:
          # On push events there are no inputs → fall back to shell defaults
          # (240 min / query-leak) so a qa/soak-** push runs the real #581 soak.
          CBM_SOAK_MODE: ${{ inputs.mode || 'query-leak' }}
          DURATION_MINUTES: ${{ inputs.duration_minutes || '240' }}
        run: scripts/soak-test.sh build/c/codebase-memory-mcp "${DURATION_MINUTES}"

      - name: Upload metrics
        if: always()
        uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
        with:
          name: soak-${{ matrix.os }}-${{ inputs.mode || 'query-leak' }}
          path: soak-results/
          retention-days: 14

  # ── Windows: the platform #581 actually crashes on (50+ GB → crash) ───────
  soak-windows:
    runs-on: windows-latest
    timeout-minutes: 320
    steps:
      - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
      - uses: msys2/setup-msys2@66cd2cce69caa17b53920067426061ca1de3a884 # v2
        with:
          msystem: CLANG64
          path-type: inherit
          install: >-
            mingw-w64-clang-x86_64-clang
            mingw-w64-clang-x86_64-zlib
            mingw-w64-clang-x86_64-python3
            make
            git
            coreutils
      - name: Build (prod binary)
        shell: msys2 {0}
        run: scripts/build.sh CC=clang CXX=clang++
      - name: Soak
        shell: msys2 {0}
        env:
          CBM_SOAK_MODE: ${{ inputs.mode || 'query-leak' }}
          DURATION_MINUTES: ${{ inputs.duration_minutes || '240' }}
        run: |
          BIN=build/c/codebase-memory-mcp
          [ -f "${BIN}.exe" ] && BIN="${BIN}.exe"
          scripts/soak-test.sh "$BIN" "${DURATION_MINUTES}"
      - name: Upload metrics
        if: always()
        uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
        with:
          name: soak-windows-${{ inputs.mode || 'query-leak' }}
          path: soak-results/
          retention-days: 14