Determinstic Vector Search

.git-blame-ignore-revs

@@ -4,3 +4,5 @@
 8ef6a0da56878177ff8d6880c92e8f7d0321d076
 7f8b7ffbcad2265b047a5e2195f76cc924028063
 2cbf261032dc8aca56c846971c090c991ac594a6
+# GITHUB#12655 Regen after google java formatter upgrade
+e59c607daf98bd0e0faf259ac9c9cf2e3cff5807

.gitattributes

@@ -1,5 +1,6 @@
 # Ignore all differences in line endings for the lock file.
 versions.lock text eol=lf
+versions.props text eol=lf
 
 # Gradle files are always in LF.
 *.gradle          text eol=lf

.github/actions/prepare-for-build/action.yml

@@ -0,0 +1,29 @@
+# This composite action is included in other workflows to have a shared setup
+# for java, gradle, caches, etc.
+
+name: Prepare Lucene build
+
+inputs:
+   java-version:
+     required: false
+     default: 17
+     description: "The default JDK version to set up."
+
+   java-distribution:
+     required: false
+     default: "temurin"
+     description: "The default JDK distribution type"
+
+runs:
+ using: "composite"
+ steps:
+    - name: Set up Java (${{ inputs.java-distribution }}, ${{ inputs.java-version }})"
+      uses: actions/setup-java@v4
+      with:
+        distribution: ${{ inputs.java-distribution }}
+        java-version: ${{ inputs.java-version }}
+        java-package: jdk
+
+    # This includes "smart" caching of the wrapper and dependencies.
+    - name: Set up Gradle
+      uses: gradle/actions/setup-gradle@v3

.github/workflows/mark-stale-PRs.yml

@@ -0,0 +1,50 @@
+# This workflow warns of PRs that have had no activity for a specified amount of time.
+#
+# For more information, see https://github.com/actions/stale
+name: Mark stale pull requests
+
+on:
+  # Run every day at 00:00 UTC
+  schedule:
+  - cron: '0 0 * * *'
+  # Or run on demand
+  workflow_dispatch:
+
+jobs:
+  stale:
+    runs-on: ubuntu-latest
+
+    permissions:
+      pull-requests: write
+
+    steps:
+      - uses: actions/stale@v5
+        with:
+          repo-token: ${{ secrets.GITHUB_TOKEN }}
+
+          days-before-pr-stale: 14
+          days-before-issue-stale: -1   # don't mark issues as stale
+          exempt-draft-pr: true         # don't mark draft PRs as stale
+
+          days-before-close: -1         # don't close stale PRs/issues
+
+          stale-pr-message: >
+            This PR has not had activity in the past 2 weeks, labeling it as stale.
+            If the PR is waiting for review, notify the dev@lucene.apache.org list.
+            Thank you for your contribution!
+
+          debug-only: false             # turn on to run the action without applying changes
+          operations-per-run: 500       # operations budget
+
+# The table shows the cost in operations of all combinations of stale / not-stale for a PR.
+# Processing a non-PR issue takes 0 operations, since we don't perform any action on it.
+#
+#                            +-----------------------+
+#            number of       |  state after workflow |
+#           operations       +-----------+-----------+
+#                            |   stale   | not stale |
+#     +----------+-----------+-----------+-----------+
+#     |  state   |   stale   |     3     |     4     |
+#     |  before  +-----------+-----------+-----------+
+#     | workflow | not stale |     5     |     1     |
+#     +----------+-----------+-----------+-----------+

.github/workflows/run-checks-all.yml

@@ -0,0 +1,67 @@
+name: "Run checks: all modules"
+
+on: 
+  workflow_dispatch:
+
+  pull_request:
+    branches:
+      - '*'
+
+  push:
+    branches:
+      - 'main'
+      - 'branch_9x'
+
+env:
+  GRADLE_ENTERPRISE_ACCESS_KEY: ${{ secrets.GE_ACCESS_TOKEN }}
+
+# We split the workflow into two parallel jobs for efficiency:
+# one is running all validation checks without tests,
+# the other runs all tests without other validation checks.
+
+jobs:
+
+  # This runs all validation checks without tests.
+  checks:
+    name: checks without tests (JDK ${{ matrix.java }} on ${{ matrix.os }})
+    timeout-minutes: 15
+
+    strategy:
+      matrix:
+        os: [ ubuntu-latest ]
+        java: [ '17' ]
+
+    runs-on: ${{ matrix.os }}
+
+    steps:
+      - uses: actions/checkout@v4
+      - uses: ./.github/actions/prepare-for-build
+
+      - name: Run gradle check (without tests)
+        run: ./gradlew check -x test -Ptask.times=true --max-workers 2
+
+
+  # This runs all tests without any other validation checks.
+  tests:
+    name: tests (JDK ${{ matrix.java }} on ${{ matrix.os }})
+    timeout-minutes: 30
+
+    strategy:
+      matrix:
+        # Operating systems to run on.
+        # windows-latest: fairly slow to build and results in odd errors (see LUCENE-10167)
+        # macos-latest: a tad slower than ubuntu and pretty much the same (?) so leaving out.
+        os: [ ubuntu-latest ]
+        java: [ '17' ]
+
+    runs-on: ${{ matrix.os }}
+
+    steps:
+      - uses: actions/checkout@v4
+      - uses: ./.github/actions/prepare-for-build
+
+      - name: Run gradle tests
+        run: ./gradlew test "-Ptask.times=true" --max-workers 2
+
+      - name: List automatically-initialized gradle.properties
+        run: cat gradle.properties

.github/workflows/run-checks-gradle-upgrade.yml

@@ -0,0 +1,96 @@
+name: "Run checks: gradle upgrade"
+
+on:
+  workflow_dispatch:
+
+  pull_request:
+    branches:
+      - 'main'
+      - 'branch_9x'
+    paths:
+      - '.github/workflows/run-checks-gradle-upgrade.yml'
+      - 'gradle/wrapper/**'
+
+  push:
+    branches:
+      - 'main'
+      - 'branch_9x'
+    paths:
+      - '.github/workflows/run-checks-gradle-upgrade.yml'
+      - 'gradle/wrapper/**'
+
+env:
+  GRADLE_ENTERPRISE_ACCESS_KEY: ${{ secrets.GE_ACCESS_TOKEN }}
+
+jobs:
+  gradleSanityCheck:
+    name: "Run tasks (java: ${{ matrix.java-version }}, alt-java: ${{ matrix.uses-alt-java }})"
+    timeout-minutes: 30
+
+    strategy:
+      matrix:
+        os: [ ubuntu-latest ]
+        java-version: [ '23-ea' ]
+        uses-alt-java: [ true, false ]
+
+    runs-on: ${{ matrix.os }}
+
+    env:
+      ALT_JAVA_DIR: /tmp/alt-java
+
+    steps:
+      - uses: actions/checkout@v4
+      - uses: ./.github/actions/prepare-for-build
+        with:
+          java-version: ${{ matrix.java-version }}
+
+      - name: Set up RUNTIME_JAVA_HOME variable
+        if: ${{ matrix.uses-alt-java }}
+        run: |
+          echo "All installed JDKs:"
+          set | grep "JAVA"
+
+          echo "Gradle's 'RUNTIME_JAVA_HOME' JDK:"
+          RUNTIME_JAVA_HOME_VAR=JAVA_HOME_`echo ${{ matrix.java-version }} | egrep --only "[0-9]+"`_X64
+          echo ${RUNTIME_JAVA_HOME_VAR} points at ${!RUNTIME_JAVA_HOME_VAR}
+
+          # Copy the JDK from its default location to /tmp so that it appears different to gradle.
+          rsync -av ${!RUNTIME_JAVA_HOME_VAR}/ ${{ env.ALT_JAVA_DIR }}/
+
+          # This sets the environment variable and makes it available for subsequent job steps.
+          # https://docs.github.com/en/actions/using-workflows/workflow-commands-for-github-actions#setting-an-environment-variable
+          echo "RUNTIME_JAVA_HOME=${{ env.ALT_JAVA_DIR }}" >> "$GITHUB_ENV"
+
+      - name: ./gradlew tidy
+        run: |
+          ./gradlew tidy
+          if [ ! -z "$(git status --porcelain)" ]; then
+          echo ":warning: **tidy left local checkout in modified state**" >> $GITHUB_STEP_SUMMARY
+          echo '```' >> $GITHUB_STEP_SUMMARY
+          git status --porcelain >> $GITHUB_STEP_SUMMARY
+          echo '```' >> $GITHUB_STEP_SUMMARY
+          git reset --hard && git clean -xfd .
+          fi
+
+      - name: ./gradlew regenerate
+        run: |
+          # add this package for generateEmojiTokenizationTestChecksumLoad.
+          sudo apt-get install libwww-perl
+          ./gradlew regenerate -x generateUAX29URLEmailTokenizerInternal --rerun-tasks
+          if [ ! -z "$(git status --porcelain)" ]; then 
+            echo ":warning: **regenerate left local checkout in modified state**" >> $GITHUB_STEP_SUMMARY 
+            echo '```' >> $GITHUB_STEP_SUMMARY
+            git status --porcelain >> $GITHUB_STEP_SUMMARY
+            echo '```' >> $GITHUB_STEP_SUMMARY
+            git reset --hard && git clean -xfd .
+          fi
+
+      - run: ./gradlew testOpts
+      - run: ./gradlew helpWorkflow
+      - run: ./gradlew licenses updateLicenses
+      - run: ./gradlew check -x test -Pvalidation.git.failOnModified=false
+      - run: ./gradlew assembleRelease mavenToLocal
+
+      # Conserve resources: only run these in non-alt-java mode.
+      - run: ./gradlew getGeoNames
+        if: ${{ !matrix.uses-alt-java }}