Add Gemini Vertex AI integration with thinking budget support

.github/ISSUE_TEMPLATE/config.yml

@@ -24,4 +24,4 @@ contact_links:
     url: https://g.co/vulnz
     about: >
       To report a security issue, please use https://g.co/vulnz. The Google Security Team will
-      respond within 5 working days of your report on https://g.co/vulnz.
+      respond within 5 working days of your report on https://g.co/vulnz.

.github/workflows/check-infrastructure-changes.yml

@@ -0,0 +1,100 @@
+name: Protect Infrastructure Files
+
+on:
+  pull_request_target:
+    types: [opened, synchronize, reopened]
+  workflow_dispatch:
+
+permissions:
+  contents: read
+  pull-requests: write
+
+jobs:
+  protect-infrastructure:
+    if: github.event_name == 'workflow_dispatch' || github.event.pull_request.draft == false
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Check for infrastructure file changes
+        if: github.event_name == 'pull_request'
+        uses: actions/github-script@v7
+        with:
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+          script: |
+            // Get the PR author and check if they're a maintainer
+            const prAuthor = context.payload.pull_request.user.login;
+            const { data: authorPermission } = await github.rest.repos.getCollaboratorPermissionLevel({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              username: prAuthor
+            });
+
+            const isMaintainer = ['admin', 'maintain'].includes(authorPermission.permission);
+
+            // Get list of files changed in the PR
+            const { data: files } = await github.rest.pulls.listFiles({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              pull_number: context.payload.pull_request.number
+            });
+
+            // Check for infrastructure file changes
+            const infrastructureFiles = files.filter(file =>
+              file.filename.startsWith('.github/') ||
+              file.filename === 'pyproject.toml' ||
+              file.filename === 'tox.ini' ||
+              file.filename === '.pre-commit-config.yaml' ||
+              file.filename === '.pylintrc' ||
+              file.filename === 'Dockerfile' ||
+              file.filename === 'autoformat.sh' ||
+              file.filename === '.gitignore' ||
+              file.filename === 'CONTRIBUTING.md' ||
+              file.filename === 'LICENSE' ||
+              file.filename === 'CITATION.cff'
+            );
+
+            if (infrastructureFiles.length > 0 && !isMaintainer) {
+              // Check if changes are only formatting/whitespace
+              let hasStructuralChanges = false;
+              for (const file of infrastructureFiles) {
+                const additions = file.additions || 0;
+                const deletions = file.deletions || 0;
+                const changes = file.changes || 0;
+
+                // If file has significant changes (not just whitespace), consider it structural
+                if (additions > 5 || deletions > 5 || changes > 10) {
+                  hasStructuralChanges = true;
+                  break;
+                }
+              }
+
+              const fileList = infrastructureFiles.map(f => `  - ${f.filename} (${f.changes} changes)`).join('\n');
+
+              // Post a comment explaining the issue
+              await github.rest.issues.createComment({
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                issue_number: context.payload.pull_request.number,
+                body: `❌ **Infrastructure File Protection**\n\n` +
+                      `This PR modifies protected infrastructure files:\n\n${fileList}\n\n` +
+                      `Only repository maintainers are allowed to modify infrastructure files (including \`.github/\`, build configuration, and repository documentation).\n\n` +
+                      `**Note**: If these are only formatting changes, please:\n` +
+                      `1. Revert changes to \`.github/\` files\n` +
+                      `2. Use \`./autoformat.sh\` to format only source code directories\n` +
+                      `3. Avoid running formatters on infrastructure files\n\n` +
+                      `If structural changes are necessary:\n` +
+                      `1. Open an issue describing the needed infrastructure changes\n` +
+                      `2. A maintainer will review and implement the changes if approved\n\n` +
+                      `For more information, see our [Contributing Guidelines](https://github.com/google/langextract/blob/main/CONTRIBUTING.md).`
+              });
+
+              core.setFailed(
+                `This PR modifies ${infrastructureFiles.length} protected infrastructure file(s). ` +
+                `Only maintainers can modify these files. ` +
+                `Use ./autoformat.sh to format code without touching infrastructure.`
+              );
+            } else if (infrastructureFiles.length > 0 && isMaintainer) {
+              core.info(`PR modifies ${infrastructureFiles.length} infrastructure file(s) - allowed for maintainer ${prAuthor}`);
+            } else {
+              core.info('No infrastructure files modified');
+            }

.github/workflows/check-linked-issue.yml

@@ -0,0 +1,90 @@
+name: Require linked issue with community support
+
+on:
+  pull_request_target:
+    types: [opened, edited, synchronize, reopened]
+  workflow_dispatch:
+
+permissions:
+  contents: read
+  issues: read
+  pull-requests: write
+
+jobs:
+  enforce:
+    if: github.event_name == 'workflow_dispatch' || github.event.pull_request.draft == false
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Verify linked issue
+        if: github.event_name == 'pull_request'
+        uses: nearform-actions/github-action-check-linked-issues@v1.2.7
+        with:
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+          comment: true
+          exclude-branches: main
+          custom-body: |
+            No linked issues found. Please add "Fixes #<issue-number>" to your pull request description.
+
+            Per our [Contributing Guidelines](https://github.com/google/langextract/blob/main/CONTRIBUTING.md#pull-request-guidelines), all PRs must:
+            - Reference an issue with "Fixes #123" or "Closes #123"
+            - The linked issue should have 5+ 👍 reactions
+            - Include discussion demonstrating the importance of the change
+
+            Use GitHub automation to close the issue when this PR is merged.
+
+      - name: Check community support
+        if: github.event_name == 'pull_request'
+        uses: actions/github-script@v7
+        with:
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+          script: |
+            // Check if PR author is a maintainer
+            const prAuthor = context.payload.pull_request.user.login;
+            const { data: authorPermission } = await github.rest.repos.getCollaboratorPermissionLevel({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              username: prAuthor
+            });
+
+            const isMaintainer = ['admin', 'maintain'].includes(authorPermission.permission);
+
+            const body = context.payload.pull_request.body || '';
+            const match = body.match(/(?:Fixes|Closes|Resolves)\s+#(\d+)/i);
+
+            if (!match) {
+              core.setFailed('No linked issue found');
+              return;
+            }
+
+            const issueNumber = Number(match[1]);
+            const { repository } = await github.graphql(`
+              query($owner: String!, $repo: String!, $number: Int!) {
+                repository(owner: $owner, name: $repo) {
+                  issue(number: $number) {
+                    reactionGroups {
+                      content
+                      users {
+                        totalCount
+                      }
+                    }
+                  }
+                }
+              }
+            `, {
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              number: issueNumber
+            });
+
+            const reactions = repository.issue.reactionGroups;
+            const thumbsUp = reactions.find(g => g.content === 'THUMBS_UP')?.users.totalCount || 0;
+
+            core.info(`Issue #${issueNumber} has ${thumbsUp} 👍 reactions`);
+
+            const REQUIRED_THUMBS_UP = 5;
+            if (thumbsUp < REQUIRED_THUMBS_UP && !isMaintainer) {
+              core.setFailed(`Issue #${issueNumber} needs at least ${REQUIRED_THUMBS_UP} 👍 reactions (currently has ${thumbsUp})`);
+            } else if (isMaintainer && thumbsUp < REQUIRED_THUMBS_UP) {
+              core.info(`Maintainer ${prAuthor} bypassing community support requirement (issue has ${thumbsUp} 👍 reactions)`);
+            }

.github/workflows/check-pr-size.yml

@@ -0,0 +1,44 @@
+name: Check PR size
+
+on:
+  pull_request_target:
+    types: [opened, synchronize, reopened]
+  workflow_dispatch:
+
+permissions:
+  contents: read
+  pull-requests: write
+
+jobs:
+  size:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Evaluate PR size
+        if: github.event_name == 'pull_request'
+        uses: actions/github-script@v7
+        with:
+          script: |
+            const pr = context.payload.pull_request;
+            const totalChanges = pr.additions + pr.deletions;
+
+            core.info(`PR contains ${pr.additions} additions and ${pr.deletions} deletions (${totalChanges} total)`);
+
+            const sizeLabel =
+              totalChanges < 50   ? 'size/XS' :
+              totalChanges < 150  ? 'size/S'  :
+              totalChanges < 600  ? 'size/M'  :
+              totalChanges < 1000 ? 'size/L'  : 'size/XL';
+
+            await github.rest.issues.addLabels({
+              ...context.repo,
+              issue_number: pr.number,
+              labels: [sizeLabel]
+            });
+
+            const MAX_LINES = 1000;
+            if (totalChanges > MAX_LINES) {
+              core.setFailed(
+                `This PR contains ${totalChanges} lines of changes, which exceeds the maximum of ${MAX_LINES} lines. ` +
+                `Please split this into smaller, focused pull requests.`
+              );
+            }

.github/workflows/ci.yaml

@@ -15,6 +15,7 @@
 name: CI
 
 on:
+  workflow_dispatch:
   push:
     branches: ["main"]
   pull_request:
@@ -28,7 +29,7 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        python-version: ["3.10", "3.11"]
+        python-version: ["3.10", "3.11", "3.12"]
     steps:
       - uses: actions/checkout@v4
 
@@ -42,6 +43,91 @@ jobs:
           python -m pip install --upgrade pip
           pip install -e ".[dev,test]"
 
-      - name: Run tox (lint + tests)
+      - name: Run unit tests and linting
         run: |
-          tox
+          PY_VERSION=$(echo "${{ matrix.python-version }}" | tr -d '.')
+          tox -e py${PY_VERSION},format,lint-src,lint-tests
+
+  live-api-tests:
+    needs: test
+    runs-on: ubuntu-latest
+    if: |
+      github.event_name == 'push' ||
+      (github.event_name == 'pull_request' &&
+       github.event.pull_request.head.repo.full_name == github.repository)
+
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Set up Python 3.11
+        uses: actions/setup-python@v4
+        with:
+          python-version: "3.11"
+
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install -e ".[dev,test]"
+
+      - name: Run live API tests
+        env:
+          GEMINI_API_KEY: ${{ secrets.GEMINI_API_KEY }}
+          LANGEXTRACT_API_KEY: ${{ secrets.GEMINI_API_KEY }}  # For backward compatibility
+          OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
+        run: |
+          if [[ -z "$GEMINI_API_KEY" && -z "$OPENAI_API_KEY" ]]; then
+            echo "::notice::Live API tests skipped - no provider secrets configured"
+            exit 0
+          fi
+          tox -e live-api
+
+  ollama-integration-test:
+    needs: test
+    runs-on: ubuntu-latest
+    if: github.event_name == 'pull_request'
+
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Detect file changes
+        id: changes
+        uses: tj-actions/changed-files@v46
+        with:
+          files: |
+            langextract/inference.py
+            examples/ollama/**
+            tests/test_ollama_integration.py
+            .github/workflows/ci.yaml
+
+      - name: Skip if no Ollama changes
+        if: steps.changes.outputs.any_changed == 'false'
+        run: |
+          echo "No Ollama-related changes detected – skipping job."
+          exit 0
+
+      - name: Set up Python 3.11
+        uses: actions/setup-python@v4
+        with:
+          python-version: "3.11"
+
+      - name: Launch Ollama container
+        run: |
+          docker run -d --name ollama \
+            -p 127.0.0.1:11434:11434 \
+            -v ollama:/root/.ollama \
+            ollama/ollama:0.5.4
+          for i in {1..20}; do
+            curl -fs http://localhost:11434/api/version && break
+            sleep 3
+          done
+
+      - name: Pull gemma2 model
+        run: docker exec ollama ollama pull gemma2:2b || true
+
+      - name: Install tox
+        run: |
+          python -m pip install --upgrade pip
+          pip install tox
+
+      - name: Run Ollama integration tests
+        run: tox -e ollama-integration