Add input provenance item helper

Signed-off-by: yoshifuminakamura <nakamura@riken.jp>

Author: yoshifuminakamura

docs/guides/add-app.md

@@ -130,20 +130,12 @@ portal の `/results/usage` では、この source provenance が各 app / syste
 Portal や review で dataset 名を見せたい場合だけ、任意の入力metadataで repo-relative path を補足できます。
 
 ```bash
-bk_record_input_info <<'EOF'
-{
-  "schema_version": 1,
-  "inputs": [
-    {
-      "dataset_id": "myapp-case0",
-      "kind": "repo-local-input",
-      "source": "source_info",
-      "repo_relative_path": "benchmarks/case0/input.dat",
-      "verification_status": "covered_by_source_commit"
-    }
-  ]
-}
-EOF
+bk_record_input_item \
+  --dataset-id myapp-case0 \
+  --kind repo-local-input \
+  --source source_info \
+  --repo-relative-path benchmarks/case0/input.dat \
+  --verification-status covered_by_source_commit
 ```
 
 ### pre-staged input と site-local 情報の扱い
@@ -163,29 +155,23 @@ site-local path や allocation / project ID は、それ自体を一律に secre
 
 pre-staged input を使う app では、「正しい場所にファイルがある」だけでは再現性の説明として不足します。
 可能であれば input directory と同じ場所に manifest を置き、run 前に manifest / digest を検証して、Result metadata へ dataset identity を残してください。
-app から実行時の入力metadataを渡す場合は、`scripts/bk_functions.sh` を source して `bk_record_input_info` を使ってください。
-app 側は共通層の受け渡し file path を意識せず、入力metadataの中身だけを定義します。
+app から実行時の入力metadataを渡す場合は、`scripts/bk_functions.sh` を source して `bk_record_input_item` などの共通helperを使ってください。
+app 側は `schema_version` や `inputs` 配列の形を組み立てず、入力metadataの中身だけを引数として渡します。
 
 最小例:
 
 ```bash
-bk_record_input_info <<'EOF'
-{
-  "schema_version": 1,
-  "inputs": [
-    {
-      "dataset_id": "myapp-case0",
-      "dataset_version": "2026-09",
-      "kind": "benchmark-input",
-      "manifest_digest": "sha256:<manifest-sha256>",
-      "verification_status": "verified"
-    }
-  ]
-}
-EOF
+bk_record_input_item \
+  --dataset-id myapp-case0 \
+  --dataset-version 2026-09 \
+  --kind pre-staged-file \
+  --source site-local \
+  --verification-status declared \
+  --recipe "how the benchmark input was prepared"
 ```
 
-ここで `manifest_digest` は manifest file だけの hash ではなく、manifest の中で dataset ID、version/revision、生成 recipe、期待 file list、size/hash などを説明できるようにしておくと後から追跡しやすくなります。
+`Verified` へ進める場合は、manifest file だけの hash ではなく、manifest の中で dataset ID、version/revision、生成 recipe、期待 file list、size/hash などを説明できるようにしておくと後から追跡しやすくなります。
+digest や公開URLなどの field を追加する場合も、app 側で Result JSON schema を直書きせず、共通helperまたは共通の受け渡し形式を拡張してください。
 公開 surface では detailed local path を出さず、dataset identity と検証状態を優先して見せる前提で設計してください。
 
 Portal の `/results/usage` では、通常の benchmark result に対する入力出自の状態を `Input Status` として表示します。

programs/genesis/run.sh

@@ -28,50 +28,16 @@ input=${header}.inp
 resultsdir=${SCRIPT_DIR}/results
 artifactsdir=${SCRIPT_DIR}/artifacts
 mkdir -p ${resultsdir}
+export BK_INPUT_INFO_FILE="${resultsdir}/input_info.json"
+export BK_INPUT_INFO_ITEMS_FILE="${resultsdir}/.input_info_items.jsonl"
+bk_reset_input_info
 #tmpdir="/vol0003/share/rccs-sdt/TMP_FugakuNEXT_CICD_LOG/${system}"
 #mkdir -p ${tmpdir}
 output="${resultsdir}/log_${header}.txt"
 stderr="${resultsdir}/log_${header}_err.txt"
 binary="spdyn"
 inputdir="../../../inputs/apoa1/"
 
-record_public_input_info() {
-    local input_source_commit="$1"
-    {
-        printf '{\n'
-        printf '  "schema_version": 1,\n'
-        printf '  "inputs": [\n'
-        printf '    {\n'
-        printf '      "dataset_id": '
-        bk_json_string "apoa1-p8"
-        printf ',\n'
-        printf '      "dataset_version": '
-        bk_json_string "$BRANCH"
-        printf ',\n'
-        printf '      "kind": "public-git",\n'
-        printf '      "source": "public_url",\n'
-        printf '      "public_url": '
-        bk_json_string "$REPO_URL"
-        printf ',\n'
-        printf '      "source_ref": '
-        bk_json_string "$BRANCH"
-        printf ',\n'
-        printf '      "resolved_commit": '
-        bk_json_string "$input_source_commit"
-        printf ',\n'
-        printf '      "repo_relative_path": '
-        bk_json_string "$dir_path"
-        printf ',\n'
-        printf '      "recipe": '
-        bk_json_string "${input}.sub generated from ${input}"
-        printf ',\n'
-        printf '      "verification_status": "public_source_commit"\n'
-        printf '    }\n'
-        printf '  ]\n'
-        printf '}\n'
-    } | bk_record_input_info
-}
-
 echo "[${REPO_DIR}] Running on system: $system"
 
 if [[ -d "${REPO_DIR}" ]]; then
@@ -96,7 +62,17 @@ else
 fi
 
 input_source_commit=$(git -C "${REPO_DIR}" rev-parse HEAD)
-record_public_input_info "$input_source_commit"
+bk_record_input_item \
+    --dataset-id apoa1-p8 \
+    --dataset-version "$BRANCH" \
+    --kind public-git \
+    --source public_url \
+    --public-url "$REPO_URL" \
+    --source-ref "$BRANCH" \
+    --resolved-commit "$input_source_commit" \
+    --repo-relative-path "$dir_path" \
+    --recipe "${input}.sub generated from ${input}" \
+    --verification-status public_source_commit
 
 if [[ ! -f "${artifactsdir}/spdyn" ]]; then
     echo "Error: spdyn does not exist."

programs/petsc-gmres/run.sh

@@ -18,6 +18,7 @@ ARTIFACT="${PWD}/artifacts/GMRES-PETSc"
 RESULTS_DIR="${PWD}/results"
 mkdir -p "${RESULTS_DIR}"
 : > "${RESULTS_DIR}/result"
+bk_reset_input_info
 
 if [[ ! -x "${ARTIFACT}" ]]; then
   echo "Required artifact not found or not executable: ${ARTIFACT}" >&2
@@ -115,6 +116,17 @@ if [[ ! -f "${DATA}" ]]; then
   exit 1
 fi
 
+matrix_name="$(basename "${DATA}" .dat)"
+bk_record_input_item \
+  --dataset-id "petsc-gmres-${matrix_name}" \
+  --dataset-version "${matrix_name}" \
+  --kind pre-staged-file \
+  --source site-local \
+  --result-exp "${matrix_name}" \
+  --parameter ksp_gmres_restart "${KSP_RESTART}" \
+  --verification-status declared \
+  --recipe "Stokes flow saddle-point matrix generated with Gmsh/FreeFEM and stored in PETSc binary format."
+
 if ! grep -q "^FOM: ranks=" "${logfile}" 2>/dev/null; then
   echo "petsc-gmres success marker not found" >&2
   echo "---- ${logfile} tail ----" >&2
@@ -132,7 +144,7 @@ bk_emit_result \
   --fom "${ksp_iter_time}" \
   --fom-unit s \
   --fom-version ksp_iter_time \
-  --exp "$(basename "${DATA}" .dat)" \
+  --exp "${matrix_name}" \
   --nodes "${nodes}" \
   --numproc-node "${numproc_node}" \
   --nthreads "${nthreads}" >> "${RESULTS_DIR}/result"

programs/salmon/run.sh

@@ -42,6 +42,7 @@ FUGAKU_RESTART_NML="Si-3-3-3-tddft.nml"
 
 mkdir -p "${RESULTS_DIR}"
 : > "${RESULTS_DIR}/result"
+bk_reset_input_info
 
 if [[ ! -x artifacts/salmon ]]; then
   echo "Required artifact not found or not executable: artifacts/salmon" >&2
@@ -103,6 +104,28 @@ if uses_prestaged_restart "${system}"; then
   # I/O instead of the benchmark itself.
   ln -s "${restart_dir}/restart" "${WORK_DIR}/restart"
   grep -Ein '^[[:space:]]*theory[[:space:]]*=' "${WORK_DIR}/${tddft_nml}" >&2 || true
+  case "${tddft_nml}" in
+    Si-3-3-3-tddft.nml)
+      salmon_dataset_id="salmon-si-3x3x3-folded-restart"
+      salmon_dataset_version="Si-3-3-3-tddft"
+      ;;
+    Si-2-2-2-tddft.nml)
+      salmon_dataset_id="salmon-si-2x2x2-folded-restart"
+      salmon_dataset_version="Si-2-2-2-tddft"
+      ;;
+    *)
+      salmon_dataset_id="salmon-folded-restart"
+      salmon_dataset_version="${tddft_nml%.nml}"
+      ;;
+  esac
+  bk_record_input_item \
+    --dataset-id "${salmon_dataset_id}" \
+    --dataset-version "${salmon_dataset_version}" \
+    --kind pre-staged-restart \
+    --source site-local \
+    --parameter input_set "${tddft_nml}" \
+    --verification-status declared \
+    --recipe "Offline ground-state data folded to a complex Gamma-point TDDFT restart; the run consumes the restart, pseudopotentials, and TDDFT input."
 else
   case "${system}" in
     RC_GH200|RC_GENOA)
@@ -137,6 +160,14 @@ else
   chmod +x "${WORK_DIR}/salmon"
   cp "${input_dir}"/* "${WORK_DIR}/"
   grep -Ein '^[[:space:]]*theory[[:space:]]*=' "${WORK_DIR}/Si-1-1-1.nml" "${WORK_DIR}/Si-1-1-1-tddft.nml" >&2 || true
+  bk_record_input_item \
+    --dataset-id salmon-si-1x1x1-archive \
+    --dataset-version Si-1-1-1 \
+    --kind pre-staged-archive \
+    --source site-local \
+    --parameter input_set "Si-1-1-1 GS/RT inputs" \
+    --verification-status declared \
+    --recipe "Archive containing SALMON ground-state and TDDFT input files; the run generates the restart consumed by TDDFT."
 fi
 cd "${WORK_DIR}"
 
@@ -311,8 +342,6 @@ if uses_prestaged_restart "${system}"; then
   run_rt_once () {
     local label="$1" logfile="rt_$1.log"
     touch .rt_start_marker
-    local t0 t1
-    t0=$(date +%s.%N)
     if uses_stdin_input "${system}"; then
       run_salmon_or_diagnose RT .rt_start_marker "${logfile}" ./salmon < "${tddft_nml}"
     else
@@ -321,7 +350,6 @@ if uses_prestaged_restart "${system}"; then
       # under pjsub's mpiexec, not all ranks.
       run_salmon_or_diagnose RT .rt_start_marker "${logfile}" -stdin "${tddft_nml}" ./salmon
     fi
-    t1=$(date +%s.%N)
     cp "${logfile}" "${RESULTS_DIR}/"
     if ! salmon_output_has_marker_since "${logfile}" .rt_start_marker; then
       echo "SALMON RT run failed (${label})" >&2

programs/scale-letkf/run.sh

@@ -1,10 +1,15 @@
 #!/bin/bash
+# shellcheck disable=SC2155
 set -e
 system="$1"
 nodes="$2"
-mkdir -p results && > results/result
+TOPDIR="${PWD}"
+mkdir -p results && : > results/result
 
 source "${PWD}/scripts/bk_functions.sh"
+export BK_INPUT_INFO_FILE="${TOPDIR}/results/input_info.json"
+export BK_INPUT_INFO_ITEMS_FILE="${TOPDIR}/results/.input_info_items.jsonl"
+bk_reset_input_info
 
 case "$system" in
   Fugaku|FugakuCN)
@@ -60,6 +65,29 @@ case "$system" in
         elapse_letkf=$elapse
 
         FOM=$(echo "$elapse_scale $elapse_letkf" | awk '{printf "%.3f\n", $1 + $2}')
+        bk_record_input_item \
+          --dataset-id scale-letkf-sc23-128x128-fugaku-config \
+          --kind repo-local-input \
+          --source source_info \
+          --result-exp SC23_128x128 \
+          --repo-relative-path test/benchmark.Fugaku_128x128 \
+          --verification-status covered_by_source_commit \
+          --recipe "Benchmark configuration copied from the source repository artifacts."
+        bk_record_input_item \
+          --dataset-id scale-letkf-scale-database \
+          --kind pre-staged-archive \
+          --source site-local \
+          --result-exp SC23_128x128 \
+          --verification-status declared \
+          --recipe "SCALE database archive expanded before the benchmark run."
+        bk_record_input_item \
+          --dataset-id scale-letkf-sc23-part1-20240410 \
+          --dataset-version 20240410 \
+          --kind pre-staged-archive \
+          --source site-local \
+          --result-exp SC23_128x128 \
+          --verification-status declared \
+          --recipe "SC23 SCALE-LETKF benchmark dataset archive expanded before the benchmark run."
         bk_emit_result --fom "$FOM" --fom-unit s --fom-version SCALE-LETKF --exp SC23_128x128 --nodes "$nodes"  --numproc-node 4 --nthreads 12 >> ../results/result
       ;;
       75)
@@ -102,6 +130,29 @@ case "$system" in
         elapse_letkf=$elapse
 
         FOM=$(echo "$elapse_scale $elapse_letkf" | awk '{printf "%.3f\n", $1 + $2}')
+        bk_record_input_item \
+          --dataset-id scale-letkf-sc23-1280x1280-fugaku-config \
+          --kind repo-local-input \
+          --source source_info \
+          --result-exp SC23_1280x1280 \
+          --repo-relative-path test/benchmark.Fugaku_1280x1280 \
+          --verification-status covered_by_source_commit \
+          --recipe "Benchmark configuration copied from the source repository artifacts."
+        bk_record_input_item \
+          --dataset-id scale-letkf-scale-database \
+          --kind pre-staged-archive \
+          --source site-local \
+          --result-exp SC23_1280x1280 \
+          --verification-status declared \
+          --recipe "SCALE database archive expanded before the benchmark run."
+        bk_record_input_item \
+          --dataset-id scale-letkf-sc23-part1-20240410 \
+          --dataset-version 20240410 \
+          --kind pre-staged-archive \
+          --source site-local \
+          --result-exp SC23_1280x1280 \
+          --verification-status declared \
+          --recipe "SC23 SCALE-LETKF benchmark dataset archive expanded before the benchmark run."
         bk_emit_result --fom "$FOM" --fom-unit s --fom-version SCALE-LETKF --exp SC23_1280x1280 --nodes "$nodes" --numproc-node 4 --nthreads 12 >> ../results/result
       ;;
     esac
@@ -160,6 +211,29 @@ case "$system" in
         elapse_letkf=$elapse
 
         FOM=$(echo "$elapse_scale $elapse_letkf" | awk '{printf "%.3f\n", $1 + $2}')
+        bk_record_input_item \
+          --dataset-id scale-letkf-sc23-128x128-rc-gh200-config \
+          --kind repo-local-input \
+          --source source_info \
+          --result-exp SC23_128x128 \
+          --repo-relative-path test/benchmark.RC_GH200_128x128 \
+          --verification-status covered_by_source_commit \
+          --recipe "Benchmark configuration copied from the source repository artifacts."
+        bk_record_input_item \
+          --dataset-id scale-letkf-scale-database \
+          --kind pre-staged-archive \
+          --source site-local \
+          --result-exp SC23_128x128 \
+          --verification-status declared \
+          --recipe "SCALE database archive expanded before the benchmark run."
+        bk_record_input_item \
+          --dataset-id scale-letkf-sc23-part1-20240410 \
+          --dataset-version 20240410 \
+          --kind pre-staged-archive \
+          --source site-local \
+          --result-exp SC23_128x128 \
+          --verification-status declared \
+          --recipe "SC23 SCALE-LETKF benchmark dataset archive expanded before the benchmark run."
         bk_emit_result --fom "$FOM" --fom-unit s --fom-version SCALE-LETKF --exp SC23_128x128 --nodes "$nodes" --numproc-node 12 --nthreads 1 >> ../results/result
       ;;
     esac