Trigger Tableau TDVT in connectors on snapshot publish and release (CON-264)

.github/scripts/mirror_tdvt.py

@@ -0,0 +1,148 @@
+#!/usr/bin/env python3
+"""Dispatch Tableau TDVT and mirror its result. Used by nightly.yml and release.yml."""
+
+from __future__ import annotations
+
+import argparse
+import json
+import os
+import subprocess
+import sys
+import time
+from datetime import datetime, timedelta, timezone
+
+TDVT_REPO = "ClickHouse/clickhouse-tableau-tdvt"
+WORKFLOW = "tableau-tdvt.yml"
+POLL_ATTEMPTS = 24
+POLL_INTERVAL_SEC = 15
+WATCH_INTERVAL_SEC = 30
+CREATED_FILTER_SKEW_SEC = 120
+
+
+def fail(message: str) -> None:
+    print(f"::error::{message}", file=sys.stderr)
+    sys.exit(1)
+
+
+def gh_json(*args: str) -> dict | list:
+    result = subprocess.run(["gh", *args], check=True, capture_output=True, text=True)
+    return json.loads(result.stdout)
+
+
+def gh_api_call(path: str, *fields: str) -> None:
+    args = ["gh", "api", path]
+    for field in fields:
+        flag = "-F" if field.startswith("client_payload[") else "-f"
+        args.extend([flag, field])
+    subprocess.run(args, check=True)
+
+
+def latest_run_id(repo: str) -> int:
+    data = gh_json(
+        "api",
+        f"repos/{repo}/actions/workflows/{WORKFLOW}/runs?event=repository_dispatch&per_page=1",
+    )
+    runs = data.get("workflow_runs", [])
+    return runs[0]["id"] if runs else 0
+
+
+def list_runs(repo: str, created_since: str) -> list[dict]:
+    data = gh_json(
+        "api",
+        f"repos/{repo}/actions/workflows/{WORKFLOW}/runs"
+        f"?event=repository_dispatch&created=>={created_since}&per_page=20",
+    )
+    return data.get("workflow_runs", [])
+
+
+def title_has_correlation(display_title: str, correlation_id: str) -> bool:
+    # TDVT run-name: "... [corr:{correlation_id}]" — match the bracketed token exactly.
+    return f"[corr:{correlation_id}]" in display_title
+
+
+def find_run_id(runs: list[dict], threshold: int, correlation_id: str) -> int | None:
+    matches = [
+        run
+        for run in runs
+        if run.get("id", 0) > threshold
+        and title_has_correlation(run.get("display_title", ""), correlation_id)
+    ]
+    return matches[0]["id"] if len(matches) == 1 else None
+
+
+def write_summary(version: str, run_url: str, info: dict) -> None:
+    path = os.environ.get("GITHUB_STEP_SUMMARY")
+    if not path:
+        return
+    lines = [
+        f"## Tableau TDVT - clickhouse-jdbc `{version}`",
+        "",
+        f"[Full TDVT run]({run_url})",
+        "",
+        f"**Conclusion: {info.get('conclusion', '')}**",
+        "",
+        "| Job | Result |",
+        "|---|---|",
+    ]
+    for job in info.get("jobs", []):
+        result = job.get("conclusion") or job.get("status") or ""
+        lines.append(f"| {job.get('name', '')} | {result} |")
+    with open(path, "a", encoding="utf-8") as summary:
+        summary.write("\n".join(lines) + "\n")
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("--event-type", required=True, choices=("jdbc-snapshot", "jdbc-release"))
+    parser.add_argument("--version", required=True)
+    parser.add_argument("--correlation-id", required=True)
+    parser.add_argument("--tdvt-repo", default=TDVT_REPO)
+    args = parser.parse_args()
+
+    if not os.environ.get("GH_TOKEN"):
+        fail("GH_TOKEN is required")
+
+    print(
+        f"Dispatching TDVT for clickhouse-jdbc {args.version} "
+        f"(correlation: {args.correlation_id})"
+    )
+
+    prev_id = latest_run_id(args.tdvt_repo)
+    created_since = (
+        datetime.now(timezone.utc) - timedelta(seconds=CREATED_FILTER_SKEW_SEC)
+    ).strftime("%Y-%m-%dT%H:%M:%SZ")
+    gh_api_call(
+        f"repos/{args.tdvt_repo}/dispatches",
+        f"event_type={args.event_type}",
+        f"client_payload[jdbc_version]={args.version}",
+        f"client_payload[correlation_id]={args.correlation_id}",
+    )
+
+    run_id = None
+    for _ in range(POLL_ATTEMPTS):
+        run_id = find_run_id(list_runs(args.tdvt_repo, created_since), prev_id, args.correlation_id)
+        if run_id is not None:
+            break
+        time.sleep(POLL_INTERVAL_SEC)
+
+    if run_id is None:
+        fail(f"TDVT run for correlation id '{args.correlation_id}' did not appear after dispatch")
+
+    run_url = f"https://github.com/{args.tdvt_repo}/actions/runs/{run_id}"
+    print(f"TDVT run: {run_url}")
+
+    subprocess.run(
+        ["gh", "run", "watch", str(run_id), "--repo", args.tdvt_repo, "--interval", str(WATCH_INTERVAL_SEC)],
+        check=False,
+    )
+    info = gh_json("run", "view", str(run_id), "--repo", args.tdvt_repo, "--json", "conclusion,jobs")
+    conclusion = info.get("conclusion") or ""
+    print(f"TDVT concluded: {conclusion}")
+    write_summary(args.version, run_url, info)
+
+    if conclusion != "success":
+        fail(f"Tableau TDVT did not pass (conclusion: {conclusion}) - {run_url}")
+
+
+if __name__ == "__main__":
+    main()

.github/workflows/nightly.yml

@@ -21,6 +21,8 @@ jobs:
     name: "Build and Publish Nightly Snapshot"
     runs-on: "ubuntu-latest"
     timeout-minutes: 20
+    outputs:
+      jdbc_version: ${{ steps.publish-version.outputs.jdbc_version }}
     steps:
       - name: Check out Git repository
         uses: actions/checkout@v4
@@ -63,6 +65,12 @@ jobs:
           #   -e 's|^\(    <version>\).*\(</version>\)$|\1${{ env.CHC_VERSION }}-SNAPSHOT\2|' \
           #   -e 's|${parent.groupId}|com.clickhouse|g' -e 's|${project.parent.groupId}|com.clickhouse|g' '{}' \;
           find . -type f -name "simplelogger.*" -exec rm -fv '{}' \;
+      - name: Record published snapshot version
+        id: publish-version
+        run: |
+          VERSION=$(sed -n 's|.*<revision>\(.*\)</revision>.*|\1|p' pom.xml | head -1)
+          echo "Published snapshot version: $VERSION"
+          echo "jdbc_version=$VERSION" >> "$GITHUB_OUTPUT"
       - name: Release Snapshot
         uses: samuelmeuli/action-maven-publish@v1
         with:
@@ -84,3 +92,29 @@ jobs:
           gpg_passphrase: ${{ secrets.GPG_PASSPHRASE }}
           nexus_username: ${{ secrets.SONATYPE_TOKEN_USER }}
           nexus_password: ${{ secrets.SONATYPE_TOKEN }}
+  # Mirror Tableau TDVT after publish (non-blocking).
+  tdvt:
+    name: "Tableau TDVT"
+    needs: nightly
+    if: ${{ needs.nightly.result == 'success' && startsWith(github.repository, 'ClickHouse/') }}
+    runs-on: "ubuntu-latest"
+    timeout-minutes: 150 # the TDVT matrix (latest/head/cloud, single-threaded) runs ~1h; allow for a queue
+    steps:
+      - name: Check out Git repository
+        uses: actions/checkout@v4
+      - name: Mint a token for the TDVT repo (dispatch + read the run)
+        id: tdvt-token
+        uses: actions/create-github-app-token@v3
+        with:
+          app-id: ${{ secrets.WORKFLOW_AUTH_PUBLIC_APP_ID }}
+          private-key: ${{ secrets.WORKFLOW_AUTH_PUBLIC_PRIVATE_KEY }}
+          owner: ${{ github.repository_owner }}
+          repositories: "clickhouse-tableau-tdvt"
+      - name: Run Tableau TDVT and mirror its result
+        env:
+          GH_TOKEN: ${{ steps.tdvt-token.outputs.token }}
+        run: |
+          python3 .github/scripts/mirror_tdvt.py \
+            --event-type jdbc-snapshot \
+            --version "${{ needs.nightly.outputs.jdbc_version }}" \
+            --correlation-id "${{ github.repository }}#${{ github.run_id }}-${{ github.run_attempt }}"

.github/workflows/release.yml

@@ -85,3 +85,29 @@ jobs:
             clickhouse-r2dbc/target/clickhouse*.jar \
             client-v2/target/client-v2*.jar \
             jdbc-v2/target/jdbc-v2*.jar
+  # Mirror Tableau TDVT after release (non-blocking).
+  tdvt:
+    name: "Tableau TDVT"
+    needs: release
+    if: ${{ needs.release.result == 'success' && startsWith(github.repository, 'ClickHouse/') }}
+    runs-on: "ubuntu-latest"
+    timeout-minutes: 150 # the TDVT matrix (latest/head/cloud, single-threaded) runs ~1h; allow for a queue
+    steps:
+      - name: Check out Git repository
+        uses: actions/checkout@v4
+      - name: Mint a token for the TDVT repo (dispatch + read the run)
+        id: tdvt-token
+        uses: actions/create-github-app-token@v3
+        with:
+          app-id: ${{ secrets.WORKFLOW_AUTH_PUBLIC_APP_ID }}
+          private-key: ${{ secrets.WORKFLOW_AUTH_PUBLIC_PRIVATE_KEY }}
+          owner: ${{ github.repository_owner }}
+          repositories: "clickhouse-tableau-tdvt"
+      - name: Run Tableau TDVT and mirror its result
+        env:
+          GH_TOKEN: ${{ steps.tdvt-token.outputs.token }}
+        run: |
+          python3 .github/scripts/mirror_tdvt.py \
+            --event-type jdbc-release \
+            --version "$(cat VERSION)" \
+            --correlation-id "${{ github.repository }}#${{ github.run_id }}-${{ github.run_attempt }}"

.gitignore

@@ -65,3 +65,7 @@ performance/jmh-simple-results.json
 *.key
 *.srl
 *.csr
+
+# Python
+__pycache__/
+*.pyc