Initial public release

Author: michal-lightly

.githooks/pre-commit

@@ -0,0 +1,14 @@
+#!/bin/bash
+
+# format easily fixable things
+make format
+make lint-fix
+
+# ensure no linting errors
+make lint
+
+# Check if any files were modified
+if ! git diff --quiet; then
+	echo "Staging formatted changes..."
+	git add -u
+fi

.github/copilot-instructions.md

@@ -0,0 +1,63 @@
+# Python coding guidelines
+
+## General guidelines
+
+When importing, import classes directly. For functions, import the containing module and call the function using dot notation.
+
+Inside each file, put public classes then public functions, then private classes, and finally private functions.
+
+For todos, use a format `# TODO({Name}, {month}/{year}): {full_sentence_comment}`, e.g. `# TODO(Michal, 08/2023): Address in the next PR.`.
+
+If using docstrings, use the google style guide and triple quotes. Use `Args:` and `Returns:` sections. Don't repeat the type in the description.
+Any example:
+```python
+def foo(bar: int) -> str:
+    """Converts an integer to a string.
+
+    Args:
+        bar: The bar to convert.
+
+    Returns:
+        The converted bar.
+    """
+    return str(bar)
+```
+
+For comments outside of docstrings, use full sentences and proper punctuation.
+E.g. `# This is a comment.` instead of `# this is a comment`.
+
+Avoid using `assert` outside of tests.
+
+Always use keyword arguments when calling functions, except for single-argument functions.
+
+Don't care about formatting, we use ruff for that.
+
+## Typing
+
+Always use type hints, such that mypy passes.
+
+Use newer syntax e.g. `list[int | None]` instead of `List[Optional[int]]`. When needing them, add `from __future__ import annotations` at the top of the file.
+
+Use abstract inputs and concrete outputs. See this example:
+```python
+def add_suffix_to_list(lst: Sequence[str], suffix: str) -> list[str]:
+    return [x + suffix for x in lst]
+```
+
+Use Sequence and Mapping instead of list and dict for immutable types. Import them from `collections.abc`.
+
+Be specific when ignoring type errors, e.g. `# type: ignore[no-untyped-call]` instead of `# type: ignore`.
+
+## Testing
+
+Always use pytest, never unittest.
+
+When testing a class named `MyClass`, put all tests under a class named `TestMyClass`.
+
+When testing a function or method, name it `test_{method_name_with_underscores}`.
+E.g. the test for `_internal_function` is named `test__internal_function`.
+E.g. the test for `MyClass.my_method` is named `TestMyClass.test_my_method`.
+
+When testing a special case of a function or method append a `__{special_case}` to the test name.
+E.g. the test for the function `compute_mean(arr: list[float])` for the empty array case
+should be named `test_compute_mean__empty_array`.

.github/pull_request_template.md

@@ -0,0 +1,12 @@
+## What has changed and why?
+
+(Delete this: Please include a summary of the change and which issue is fixed. Please also include relevant motivation and context. List any dependencies that are required for this change.)
+
+## How has it been tested?
+
+(Delete this: Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce. Please also list any relevant details for your test configuration.)
+
+## Did you update [CHANGELOG.md](../CHANGELOG.md)?
+
+- [ ] Yes
+- [ ] Not needed (internal change)

.github/workflows/end2end_test.yml

@@ -0,0 +1,206 @@
+name: End2End Test
+
+on:
+  push:
+    branches:
+      - main
+  pull_request:
+  workflow_dispatch:
+
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: true
+
+env:
+  NODE_OPTIONS: --max-old-space-size=4096
+  DATASET_PATH: $(pwd)/datasets
+  E2E_BASE_URL: http://localhost:8001
+  LIGHTLY_STUDIO_LICENSE_KEY: ${{ secrets.MUNDIG_LICENSE_KEY }}
+
+jobs:
+  end-to-end-test:
+    name: End2End Test
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout Code
+        uses: actions/checkout@v4
+        with:
+          submodules: recursive
+
+      # Authenticate to download lightly-edge-sdk.
+      - name: Authenticate with GCP
+        uses: 'google-github-actions/auth@v2'
+        with:
+          project_id: boris-250909
+          credentials_json: ${{ secrets.GCP_LIGHTLY_STUDIO_CI_READONLY }}
+
+      ### Caching rules for the workflow ###
+      - name: Cache UI app build
+        id: cache-ui-build
+        uses: actions/cache@v4
+        with:
+          path: |
+            lightly_studio_view/build
+            lightly_studio/src/lightly_studio/dist_lightly_studio_view_app
+          key: ui-build-${{ hashFiles('lightly_studio_view/src/**', 'lightly_studio_view/public/**', 'lightly_studio_view/package.json', 'lightly_studio_view/package-lock.json') }}
+
+      - name: Cache example dataset
+        id: cache-example-dataset
+        uses: actions/cache@v4
+        with:
+          path: lightly_studio/datasets/coco-128
+          key: coco-example-dataset
+
+      - name: Cache lightly_studio app build
+        id: cache-python-build
+        uses: actions/cache@v4
+        with:
+          path: lightly_studio/dist
+          key: python-build-${{ hashFiles('lightly_studio/src/**', 'lightly_studio/pyproject.toml', 'lightly_studio_view/src/**', 'lightly_studio_view/public/**', 'lightly_studio_view/package.json', 'lightly_studio_view/package-lock.json') }}
+
+      - name: Cache e2e-tests
+        id: cache-e2e-tests
+        uses: actions/cache@v4
+        with:
+          path: lightly_studio_view/playwright-report
+          key: python-build-${{ hashFiles('lightly_studio/e2e-tests/index_dataset_for_end2end_ui_tests.py', 'lightly_studio/src/**', 'lightly_studio/pyproject.toml', 'lightly_studio_view/e2e/**', 'lightly_studio_view/src/**', 'lightly_studio_view/public/**', 'lightly_studio_view/package.json', 'lightly_studio_view/package-lock.json') }}
+
+      ### Setup steps for the workflow ###
+      - name: Set Up Python
+        if: steps.cache-python-build.outputs.cache-hit != 'true' || steps.cache-e2e-tests.outputs.cache-hit != 'true'
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.8"
+
+      - name: Set Up uv
+        if: steps.cache-python-build.outputs.cache-hit != 'true' || steps.cache-e2e-tests.outputs.cache-hit != 'true'
+        uses: astral-sh/setup-uv@v6
+        with:
+          version: "0.8.17"
+          enable-cache: true
+
+      - name: Set Up Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: '22.11.0'
+          cache: 'npm'
+          cache-dependency-path: lightly_studio_view/package-lock.json
+
+      - name: Install dependencies
+        if: steps.cache-e2e-tests.outputs.cache-hit != 'true' || steps.cache-python-build.outputs.cache-hit != 'true'
+        working-directory: lightly_studio_view
+        run: npm ci
+
+      # Cache Playwright browsers to avoid re-downloading each run
+      - name: Cache Playwright browsers
+        id: cache-playwright-browsers
+        uses: actions/cache@v4
+        with:
+          path: ~/.cache/ms-playwright
+          key: playwright-${{ runner.os }}-${{ hashFiles('lightly_studio_view/package-lock.json') }}
+
+      # Always ensure system packages for browsers are present on Ubuntu
+      - name: Install Playwright system dependencies
+        if: steps.cache-e2e-tests.outputs.cache-hit != 'true'
+        working-directory: lightly_studio_view
+        run: npx playwright install-deps chromium
+
+      # Only download browser binaries when cache misses; otherwise reuse cache
+      - name: Install Playwright browsers
+        if: steps.cache-e2e-tests.outputs.cache-hit != 'true' && steps.cache-playwright-browsers.outputs.cache-hit != 'true'
+        working-directory: lightly_studio_view
+        run: npx playwright install chromium
+
+      - name: Download example dataset
+        if: steps.cache-example-dataset.outputs.cache-hit != 'true'
+        run: |
+          mkdir -p lightly_studio/datasets/coco-128
+          git clone --depth=1 https://github.com/lightly-ai/dataset_examples_studio.git tmp_dataset_repo
+          cp -r tmp_dataset_repo/coco_subset_128_images/* lightly_studio/datasets/coco-128/
+          rm -rf tmp_dataset_repo
+      
+      - name: Install lightly_studio python dependencies
+        if: steps.cache-python-build.outputs.cache-hit != 'true'
+        working-directory: lightly_studio
+        run: make install-optional-deps
+
+      - name: Export schema from backend
+        if: steps.cache-python-build.outputs.cache-hit != 'true'
+        working-directory: lightly_studio
+        run: make export-schema
+
+      ### Buildging and steps for the workflow ###
+      - name: Build the ui application
+        if: steps.cache-ui-build.outputs.cache-hit != 'true' || steps.cache-python-build.outputs.cache-hit != 'true'
+        working-directory: lightly_studio_view
+        run: |
+          rm -rf build
+          npm run build
+
+      - name: Copy the built ui app to lightly_studio source folder
+        if: steps.cache-ui-build.outputs.cache-hit != 'true' || steps.cache-python-build.outputs.cache-hit != 'true'
+        run: |
+          rm -rf lightly_studio/src/lightly_studio/dist_lightly_studio_view_app
+          cp -r lightly_studio_view/build/ lightly_studio/src/lightly_studio/dist_lightly_studio_view_app
+      
+      - name: Index dataset with using lightly_studio pypi package
+        if: steps.cache-e2e-tests.outputs.cache-hit != 'true'
+        working-directory: lightly_studio
+        # TODO(Michal, 10/2025): Starting up the server redownloads the embedding model. Consider caching it.
+        run: |
+          uv run e2e-tests/index_dataset_for_end2end_ui_tests.py > server.log 2>&1 &
+          echo "SERVER_PID=$!" >> $GITHUB_ENV
+        id: server
+
+      - name: Wait for the server to start
+        if: steps.cache-e2e-tests.outputs.cache-hit != 'true'
+        timeout-minutes: 5
+        run: |
+          echo "Waiting for the server to become available..."
+          timeout=360
+          elapsed=0
+          interval=5
+          
+          while [ $elapsed -lt $timeout ]; do
+            if curl -s -f ${E2E_BASE_URL}/healthz > /dev/null 2>&1; then
+              echo "✅ Server is up and running after ${elapsed} seconds!"
+              break
+            fi
+            echo "⏳ Server not ready yet, waiting... (${elapsed}/${timeout} seconds)"
+            sleep $interval
+            elapsed=$((elapsed + interval))
+          done
+          
+          if [ $elapsed -ge $timeout ]; then
+            echo "❌ Timed out waiting for server to become available"
+            exit 1
+          fi
+
+      ### Running end-to-end tests ###
+      - name: Run end-to-end tests for indexed dataset
+        if: steps.cache-e2e-tests.outputs.cache-hit != 'true'
+        working-directory: lightly_studio_view
+        run: npx playwright test --reporter=html --trace=on
+        timeout-minutes: 10
+      
+      ### Cleanup steps for the workflow ###
+      - name: Show server logs after failure
+        if: failure() && steps.cache-e2e-tests.outputs.cache-hit != 'true'
+        working-directory: lightly_studio
+        run: |
+          echo "Server logs:"
+          cat server.log
+
+      - name: Upload Playwright report after failure
+        uses: actions/upload-artifact@v4
+        if: failure()
+        with:
+          name: Upload Playwright Report
+          path: lightly_studio_view/playwright-report
+          retention-days: 5
+
+      - name: Stop lightly_studio application running in background
+        if: always() && steps.cache-e2e-tests.outputs.cache-hit != 'true'
+        run: |
+          kill ${{ env.SERVER_PID }} || true