feat[back-compat]: Add a script to upload fixtures to a repo (#6993)

This is the final PR in initial creation of the back-compat testing rfc:

This PR updates the docs.
Add a script to upload, list and validate changes.

---------

Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk>

Author: joseph-isaacs

.github/workflows/compat-gen-upload.yml

@@ -3,18 +3,64 @@ name: Compat Fixture Upload
 on:
   workflow_dispatch:
     inputs:
-      version:
-        description: "Version to generate fixtures for (e.g. 0.62.0)"
+      git_ref:
+        description: "Git ref for version detection (e.g. v0.62.0). Defaults to HEAD."
+        required: false
+      confirm_upload:
+        description: "Type 'yes' to confirm upload after reviewing the dry-run output."
         required: true
+        default: "no"
 
 jobs:
-  upload-fixtures:
+  dry-run:
     runs-on: ubuntu-latest
+    permissions:
+      contents: read
+    outputs:
+      version: ${{ steps.detect.outputs.version }}
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - uses: dtolnay/rust-toolchain@stable
+
+      - uses: Swatinem/rust-cache@v2
+
+      - name: Detect version
+        id: detect
+        run: |
+          GIT_REF="${{ inputs.git_ref }}"
+          if [ -n "$GIT_REF" ]; then
+            TAG=$(git describe --tags --abbrev=0 "$GIT_REF")
+          else
+            TAG=$(git describe --tags --abbrev=0)
+          fi
+          VERSION="${TAG#v}"
+          echo "version=$VERSION" >> "$GITHUB_OUTPUT"
+          echo "Detected version: $VERSION (from tag: $TAG)"
+
+      - name: Dry run publish
+        run: |
+          GIT_REF="${{ inputs.git_ref }}"
+          CMD="python3 vortex-test/compat-gen/scripts/compat.py publish --dry-run"
+          if [ -n "$GIT_REF" ]; then
+            CMD="$CMD --git-ref $GIT_REF"
+          fi
+          $CMD
+
+  upload:
+    needs: dry-run
+    if: inputs.confirm_upload == 'yes'
+    runs-on: ubuntu-latest
+    environment: compat-upload
     permissions:
       id-token: write
       contents: read
     steps:
       - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
 
       - uses: dtolnay/rust-toolchain@stable
 
@@ -23,10 +69,14 @@ jobs:
       - name: Configure AWS credentials
         uses: aws-actions/configure-aws-credentials@v5
         with:
-          role-to-assume: arn:aws:iam::245040174862:role/GitHubBenchmarkRole
+          role-to-assume: "arn:aws:iam::245040174862:role/GitHubBenchmarkRole"
           aws-region: us-east-1
 
-      - name: Generate and upload fixtures
-        run: >
-          python3 vortex-test/compat-gen/scripts/upload.py
-          --version "${{ inputs.version }}"
+      - name: Upload fixtures for v${{ needs.dry-run.outputs.version }}
+        run: |
+          GIT_REF="${{ inputs.git_ref }}"
+          CMD="python3 vortex-test/compat-gen/scripts/compat.py publish --yes"
+          if [ -n "$GIT_REF" ]; then
+            CMD="$CMD --git-ref $GIT_REF"
+          fi
+          $CMD

.github/workflows/compat-test-weekly.yml

@@ -0,0 +1,34 @@
+name: Compat Validation
+
+on:
+  schedule:
+    - cron: "0 6 * * 1"  # Monday 6am UTC
+  workflow_dispatch:
+    inputs:
+      mode:
+        description: "Validation mode"
+        required: true
+        default: "last"
+        type: choice
+        options:
+          - last
+          - all
+
+env:
+  FIXTURES_URL: https://vortex-compat-fixtures.s3.amazonaws.com
+
+jobs:
+  compat-test:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: dtolnay/rust-toolchain@stable
+
+      - uses: Swatinem/rust-cache@v2
+
+      - name: Run compat tests
+        run: |
+          MODE="${{ inputs.mode || 'last' }}"
+          python3 vortex-test/compat-gen/scripts/compat.py check \
+            --mode "$MODE"

.github/workflows/compat-validation.yml

@@ -216,6 +216,7 @@ rustc-hash = "2.1"
 serde = "1.0.220"
 serde_json = "1.0.138"
 serde_test = "1.0.176"
+sha2 = "0.10"
 simdutf8 = "0.1.5"
 similar = "2.7.0"
 sketches-ddsketch = "0.3.0"

Cargo.lock

@@ -1,7 +1,7 @@
 [package]
 name = "vortex-compat"
 authors = { workspace = true }
-description = "Backward-compatibility fixture generation and testing for Vortex"
+description = "Backward-compatibility fixture generation and checking for Vortex"
 edition = { workspace = true }
 homepage = { workspace = true }
 include = { workspace = true }
@@ -16,21 +16,16 @@ version = { workspace = true }
 workspace = true
 
 [[bin]]
-name = "compat-gen"
+name = "vortex-compat"
 path = "src/main.rs"
 
-[[bin]]
-name = "compat-validate"
-path = "src/validate_main.rs"
-
 [dependencies]
 # Vortex crates
 vortex = { workspace = true, features = ["files", "tokio", "zstd"] }
 vortex-array = { workspace = true, features = ["_test-harness"] }
 vortex-buffer = { workspace = true }
 vortex-error = { workspace = true }
 vortex-session = { workspace = true }
-vortex-utils = { workspace = true }
 
 # TPC-H generation
 arrow-array = { workspace = true }
@@ -46,12 +41,12 @@ parquet = { workspace = true }
 futures = { workspace = true }
 tokio = { workspace = true, features = ["full"] }
 
-# HTTP fetching (for ClickBench fixture + compat-test S3 downloads)
+# HTTP fetching (for ClickBench fixture setup)
 reqwest = { workspace = true }
 
 # CLI + serialization
-chrono = { workspace = true, features = ["serde"] }
 clap = { workspace = true, features = ["derive"] }
 serde = { workspace = true, features = ["derive"] }
 serde_json = { workspace = true }
+sha2 = { workspace = true }
 tempfile = { workspace = true }