tierwise 0.8.0__tar.gz

tierwise-0.8.0/.dockerignore

@@ -0,0 +1,13 @@
+.git
+.venv
+.superpowers
+.mypy_cache
+.ruff_cache
+.pytest_cache
+**/__pycache__
+*.db
+.env
+training/data
+models/router/model.onnx
+docs
+tests

tierwise-0.8.0/.env.example

@@ -0,0 +1,31 @@
+# Copy to .env and fill in the providers you want to use. Leave a key empty to disable that provider.
+ANTHROPIC_API_KEY=
+OPENAI_API_KEY=
+GEMINI_API_KEY=
+# Any OpenAI-compatible host works here (Groq, Together, DeepSeek, ...)
+OPENAI_BASE_URL=https://api.openai.com/v1
+# Local models: run `ollama serve`, then set this (leave unset to disable Ollama)
+# OLLAMA_BASE_URL=http://localhost:11434/v1
+
+# Model list with prices and capabilities, and where requests are logged
+TIERWISE_REGISTRY=config/registry.yaml
+TIERWISE_DB_URL=sqlite+aiosqlite:///tierwise.db
+
+# Reliability (defaults shown)
+# RETRIES=2
+# BREAKER_FAILURES=5
+# BREAKER_WINDOW_S=60
+# BREAKER_COOLDOWN_S=30
+# STREAM_HOLD_S=3.0
+# STREAM_HOLD_BYTES=4096
+
+# Semantic cache (needs models/router/model.onnx)
+# TIERWISE_CACHE=false   (opt-in: see the README's limits)
+# CACHE_SIMILARITY=0.95
+# CACHE_TTL_S=3600
+# CACHE_MAX_ENTRIES=10000
+# CACHE_MAX_TEMPERATURE=0.3
+
+# Gateway API keys (create with `tierwise keys create`)
+# TIERWISE_REQUIRE_KEY=false
+# KEYS_REFRESH_S=30

tierwise-0.8.0/.github/workflows/ci.yml

@@ -0,0 +1,54 @@
+name: CI
+
+on:
+  push:
+  pull_request:
+
+jobs:
+  check:
+    runs-on: ubuntu-latest
+    services:
+      postgres:  # the store tests also run against Postgres (TIERWISE_TEST_PG_URL)
+        image: postgres:16
+        env:
+          POSTGRES_USER: tierwise
+          POSTGRES_PASSWORD: tierwise
+          POSTGRES_DB: tierwise
+        ports:
+          - 5432:5432
+        options: >-
+          --health-cmd "pg_isready -h 127.0.0.1 -U tierwise -d tierwise"
+          --health-interval 5s
+          --health-timeout 5s
+          --health-retries 10
+    steps:
+      - uses: actions/checkout@v4
+      - uses: astral-sh/setup-uv@v6
+        with:
+          python-version: "3.12"
+      - run: uv sync
+      - run: uv run ruff check .
+      - run: uv run mypy src
+      - run: uv run pytest -q
+        env:
+          TIERWISE_TEST_PG_URL: postgresql+asyncpg://tierwise:tierwise@localhost:5432/tierwise
+      - name: wheel installs and serves without the repo  # the M8 exit check
+        run: |
+          uv build
+          uv venv "$RUNNER_TEMP/wheel"
+          uv pip install --python "$RUNNER_TEMP/wheel" dist/*.whl
+          mkdir "$RUNNER_TEMP/empty" && cd "$RUNNER_TEMP/empty"
+          "$RUNNER_TEMP/wheel/bin/tierwise" --help
+          "$RUNNER_TEMP/wheel/bin/python" "$GITHUB_WORKSPACE/scripts/wheel_smoke.py"
+
+  compose-smoke:  # the M7 exit check: the whole stack up, traffic through it, dashboard live
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: astral-sh/setup-uv@v6
+        with:
+          python-version: "3.12"
+      - run: docker compose -f deploy/docker-compose.yml up -d --build
+      - run: uv run python scripts/compose_smoke.py
+      - if: always()
+        run: docker compose -f deploy/docker-compose.yml down -v

tierwise-0.8.0/.github/workflows/claude-code-review.yml

@@ -0,0 +1,45 @@
+name: Claude Code Review
+
+on:
+  pull_request:
+    types: [opened, synchronize, ready_for_review, reopened]
+    # Optional: Only run on specific file changes
+    # paths:
+    #   - "src/**/*.ts"
+    #   - "src/**/*.tsx"
+    #   - "src/**/*.js"
+    #   - "src/**/*.jsx"
+
+jobs:
+  claude-review:
+    # Optional: Filter by PR author
+    # if: |
+    #   github.event.pull_request.user.login == 'external-contributor' ||
+    #   github.event.pull_request.user.login == 'new-developer' ||
+    #   github.event.pull_request.author_association == 'FIRST_TIME_CONTRIBUTOR'
+
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      pull-requests: read
+      issues: read
+      id-token: write
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 1
+
+      - name: Run Claude Code Review
+        id: claude-review
+        uses: anthropics/claude-code-action@v1
+        with:
+          claude_code_oauth_token: ${{ secrets.CLAUDE_CODE_OAUTH_TOKEN }}
+          plugin_marketplaces: 'https://github.com/anthropics/claude-code.git'
+          plugins: 'code-review@claude-code-plugins'
+          prompt: '/code-review:code-review --comment ${{ github.repository }}/pull/${{ github.event.pull_request.number }}'
+          claude_args: '--allowedTools "mcp__github_inline_comment__create_inline_comment"'
+          # See https://github.com/anthropics/claude-code-action/blob/main/docs/usage.md
+          # or https://code.claude.com/docs/en/cli-reference for available options
+

tierwise-0.8.0/.github/workflows/claude.yml

@@ -0,0 +1,50 @@
+name: Claude Code
+
+on:
+  issue_comment:
+    types: [created]
+  pull_request_review_comment:
+    types: [created]
+  issues:
+    types: [opened, assigned]
+  pull_request_review:
+    types: [submitted]
+
+jobs:
+  claude:
+    if: |
+      (github.event_name == 'issue_comment' && contains(github.event.comment.body, '@claude')) ||
+      (github.event_name == 'pull_request_review_comment' && contains(github.event.comment.body, '@claude')) ||
+      (github.event_name == 'pull_request_review' && contains(github.event.review.body, '@claude')) ||
+      (github.event_name == 'issues' && (contains(github.event.issue.body, '@claude') || contains(github.event.issue.title, '@claude')))
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      pull-requests: read
+      issues: read
+      id-token: write
+      actions: read # Required for Claude to read CI results on PRs
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 1
+
+      - name: Run Claude Code
+        id: claude
+        uses: anthropics/claude-code-action@v1
+        with:
+          claude_code_oauth_token: ${{ secrets.CLAUDE_CODE_OAUTH_TOKEN }}
+
+          # This is an optional setting that allows Claude to read CI results on PRs
+          additional_permissions: |
+            actions: read
+
+          # Optional: Give a custom prompt to Claude. If this is not specified, Claude will perform the instructions specified in the comment that tagged it.
+          # prompt: 'Update the pull request description to include a summary of changes.'
+
+          # Optional: Add claude_args to customize behavior and configuration
+          # See https://github.com/anthropics/claude-code-action/blob/main/docs/usage.md
+          # or https://code.claude.com/docs/en/cli-reference for available options
+          # claude_args: '--allowed-tools Bash(gh pr *)'
+

tierwise-0.8.0/.github/workflows/release.yml

@@ -0,0 +1,44 @@
+name: Release
+
+on:
+  push:
+    tags: ["v*"]
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: astral-sh/setup-uv@v6
+        with:
+          python-version: "3.12"
+      - name: tag matches the package version
+        run: |
+          version=$(uv version --short)
+          test "$GITHUB_REF_NAME" = "v$version" || { echo "tag $GITHUB_REF_NAME != v$version"; exit 1; }
+      - run: uv sync
+      - run: uv run pytest -q  # a PyPI version can't be re-uploaded: never publish untested
+      - run: uv build
+      - name: wheel installs and serves without the repo
+        run: |
+          uv venv "$RUNNER_TEMP/wheel"
+          uv pip install --python "$RUNNER_TEMP/wheel" dist/*.whl
+          mkdir "$RUNNER_TEMP/empty" && cd "$RUNNER_TEMP/empty"
+          "$RUNNER_TEMP/wheel/bin/python" "$GITHUB_WORKSPACE/scripts/wheel_smoke.py"
+      - uses: actions/upload-artifact@v4
+        with:
+          name: dist
+          path: dist/
+
+  publish:  # PyPI Trusted Publishing: no token is stored anywhere
+    needs: build
+    runs-on: ubuntu-latest
+    environment: pypi
+    permissions:
+      id-token: write
+    steps:
+      - uses: actions/download-artifact@v4
+        with:
+          name: dist
+          path: dist/
+      - uses: pypa/gh-action-pypi-publish@release/v1

tierwise-0.8.0/.gitignore

@@ -0,0 +1,14 @@
+.superpowers/
+__pycache__/
+*.py[cod]
+.venv/
+.env
+*.db
+.mypy_cache/
+.ruff_cache/
+.pytest_cache/
+dist/
+build/
+*.egg-info/
+models/router/model.onnx
+training/data/

tierwise-0.8.0/BENCHMARK.md

@@ -0,0 +1,87 @@
+# tierwise routing benchmark
+
+**Headline:** on 3,943 held-out prompts, tierwise's deployed router keeps **95% of GPT-4's quality at 44% of its cost (56.0% saved)**. Random routing saves 36.6% at the same quality. The learned classifier on its own saves 60.0%; the deployed router gives up 4 points of that so that rules can still push hard prompts upward.
+
+Every number here comes from precomputed answers, so reproducing it costs nothing in API calls.
+
+## Setup
+
+Each prompt has an answer from a **cheap** and a **strong** model, with a known quality and cost for each. Every router gives each prompt a score, and a prompt goes to the strong model when its score is at or above a threshold *t*. Sweeping *t* from "everything cheap" to "everything strong" traces the router's cost/quality curve.
+
+| | Held-out (headline) | SPROUT (out-of-distribution) |
+|---|---|---|
+| **Data** | Test split of [`routellm/gpt4_dataset`](https://huggingface.co/datasets/routellm/gpt4_dataset) (Apache-2.0). Split by prompt hash, never trained on. | Test split of [`CARROT-LLM-Routing/SPROUT`](https://huggingface.co/datasets/CARROT-LLM-Routing/SPROUT). **No license is stated**; it is downloaded at eval time, nothing from it is committed, and only aggregates are reported. |
+| **Prompts** | 3,943 | 6,637 (0 dropped) |
+| **Cheap / strong** | Mixtral-8x7B / GPT-4 (`gpt-4-1106-preview`) | gpt-4o-mini / gpt-4o (`2024-08-06`) |
+| **Quality** | Cheap: GPT-4 judge's grade of Mixtral's answer, (score − 1) / 4. Strong: 1.0. **Assumption:** the judge graded only Mixtral, so GPT-4's own answer is the reference. | The dataset's per-model `score` in [0, 1], from an LLM judge, for both models. |
+| **Cost** | Prompt and answer length ÷ 4 chars per token × list price | The dataset's token counts × list price |
+| **Prices** (USD / 1M tokens, in / out) | GPT-4 10 / 30 (OpenAI); Mixtral 0.60 / 0.60 (Together AI) | gpt-4o 2.50 / 10; gpt-4o-mini 0.15 / 0.60 (OpenAI) |
+
+**The routers:**
+- **random:** a uniform random score, seed 0.
+- **rules:** the M3 rules scorer.
+- **learned:** the M4 classifier's raw P(strong model needed).
+- **hybrid:** what the gateway actually runs: `route()` with rules first, the learned router when rules are unsure, and difficulty = max(rules, learned).
+
+**Metrics:**
+- **Cost saved at X%:** the share of the all-strong cost saved at the cheapest point that reaches X% of all-strong quality. It is interpolated between neighbouring thresholds, which in practice means randomly mixing two thresholds. *Single threshold* is the conservative version: one fixed threshold.
+- **Misroutes:** prompts the cheap model failed (quality < 0.5) while the strong one didn't, that were still routed cheap at the single-threshold 95% point.
+- **AUC:** the area under quality-fraction over cost-fraction.
+
+## Results
+
+### Held-out: Mixtral-8x7B vs GPT-4
+
+![held-out Pareto curve](docs/benchmark/heldout_pareto.png)
+
+| Router | Saved @ 95% | @ 98% | @ 99% | Single threshold @ 95% | Misroutes @ 95% | AUC |
+|---|---|---|---|---|---|---|
+| random | 36.6% | 14.9% | 7.0% | 36.6% | 110 | 0.905 |
+| rules | 38.9% | 10.1% | 5.1% | 38.3% | 98 | 0.905 |
+| learned | **60.0%** | **28.3%** | **18.0%** | 60.0% | 68 | 0.930 |
+| **hybrid (deployed)** | **56.0%** | 20.2% | 9.4% | 55.9% | 73 | 0.920 |
+
+All-Mixtral already reaches 85.6% of GPT-4's quality at 2.7% of its cost.
+
+### SPROUT: gpt-4o-mini vs gpt-4o
+
+![SPROUT Pareto curve](docs/benchmark/sprout_pareto.png)
+
+| Router | Saved @ 95% | @ 98% | @ 99% | AUC |
+|---|---|---|---|---|
+| random | 93.2% | 35.2% | 21.0% | 0.909 |
+| rules | 93.2% | **58.5%** | **53.4%** | 0.915 |
+| learned | 93.2% | 51.7% | 40.9% | 0.914 |
+| hybrid (deployed) | 93.2% | 52.0% | 48.8% | 0.915 |
+
+**gpt-4o-mini alone reaches 95.2% of gpt-4o's quality** (mean judged score 0.807 vs 0.848) at 6.8% of its cost, so every router "saves" 93.2% at the 95% target just by sending everything cheap. The stricter targets are where the routers differ:
+- **All routers beat random.**
+- **Rules beat the learned classifier out of distribution.** The classifier learned which prompts *Mixtral* struggles with, in 2024-era data; gpt-4o-mini fails on different prompts.
+- **The hybrid's recovery depends on the target.** Because rules can raise its difficulty, at 99% it recovers most of the gap between learned and rules (48.8% vs 40.9% and 53.4%). At 98% it recovers almost none of it (52.0% vs 51.7% and 58.5%).
+
+## The deployed hybrid
+
+- **Who decides:** the rules router decides alone only when it is confident (confidence ≥ 0.8). On every other prompt the learned classifier decides, and rules can only raise its difficulty.
+- **The max rule:** this was added in M5 after the critical-set gate caught the classifier sending 4 of 10 labeled hard prompts (a proof, a derivation, an architecture design and a legal review) to basic-tier models. The rule costs 4 points on the held-out headline (60.0% → 56.0%), and it brings the out-of-distribution result closer to the rules router.
+- **Low confidence:** 42.0% of held-out prompts (61.6% on SPROUT) get a decision with confidence below 0.3, so the gateway raises their quality bar one step. That comes from `to_decision` measuring distance from a low threshold (0.15) in probability space. It is a deliberate upward bias, and the next thing to tune when real traffic exists.
+
+## Limitations
+
+- **Older model pairs.** Neither dataset contains tierwise's registry models (Claude 5.5, GPT-5.x, Gemini 3). The benchmark measures how well the router *ranks* prompts by how much they need a strong model, and that ranking is what carries over. The registry's `quality` values stay uncalibrated placeholders until M9's shadow labels exist for current models.
+- **Two models, single-turn.** The gateway chooses among many models and caches multi-turn prefixes, and neither shows up here. Cache-aware selection is covered by the M3 tests instead.
+- **Judge-based quality.** Both datasets grade answers with an LLM judge, and the held-out set assumes GPT-4's answer is always right.
+- **SPROUT's license is unstated.** It is used for aggregate numbers only.
+
+## Reproduce
+
+```bash
+uv sync --group train
+uv run python -m training.export_onnx
+uv run python -m training.prepare_data --limit 40000
+uv run python -m evaluation.run_eval --dataset heldout   # docs/benchmark/heldout.{json,png}
+uv run python -m evaluation.run_eval --dataset sprout    # downloads ~580 MB into the HF cache
+```
+
+The critical-set gate (`tests/evaluation/test_critical_gate.py`) checks that the 10 labeled hard prompts never route to a basic-tier model. Coverage differs by environment:
+- **CI** runs the rules path only, because the 90 MB ONNX encoder isn't there. The max(rules, learned) logic is covered separately by unit tests with fake routers.
+- **The learned path**, with the trained classifier, runs wherever the model files exist. Run it after every retrain.

tierwise-0.8.0/CHANGELOG.md

@@ -0,0 +1,20 @@
+# Changelog
+
+## 0.8.0 — 2026-11-02
+
+First public release. Versions track milestones: 0.8.0 is M8.
+
+- **M0–M1:** packaging, CI, OpenAI-compatible schema, Anthropic and OpenAI adapters, passthrough.
+- **M2:** Gemini and Ollama, streaming, model registry, cost accounting, SQLite request log.
+- **M3:** rules router, cache-aware selector, prefix affinity, `auto*` / `tier:*` models.
+- **M3.5:** reasoning effort chosen together with the model.
+- **M4:** learned router (MiniLM on ONNX + numpy classifier).
+- **M5:** eval harness, Pareto curves, `BENCHMARK.md`, critical-set CI gate.
+- **M6:** retry, failover, two-level circuit breakers, stream hold, opt-in semantic cache.
+- **M7:** API keys, rate limits, monthly budgets; Prometheus, OpenTelemetry, Grafana, docker-compose with Postgres.
+- **M8:**
+  - `pip install tierwise` works without a checkout: the registry and router artifacts ship in the wheel.
+  - `tierwise fetch-model` downloads the encoder at a pinned commit and checks its sha256.
+  - A cancelled request now releases its budget reservation.
+  - Releases publish to PyPI from a tag, and the Docker base image is pinned by digest.
+  - `scripts/demo.py`.

tierwise-0.8.0/Dockerfile

@@ -0,0 +1,24 @@
+FROM python:3.12-slim@sha256:aa943ce411423c7582288f18707ff66eddcbaeb510502d747a72aeb34f30d159
+
+COPY --from=ghcr.io/astral-sh/uv:0.5.11 /uv /usr/local/bin/uv
+ENV UV_COMPILE_BYTECODE=1 UV_LINK_MODE=copy UV_PROJECT_ENVIRONMENT=/opt/venv \
+    PATH="/opt/venv/bin:$PATH" PYTHONUNBUFFERED=1
+
+WORKDIR /app
+# dependencies first, so source edits don't reinstall them
+COPY pyproject.toml uv.lock README.md ./
+RUN uv sync --frozen --no-dev --no-install-project
+
+COPY src ./src
+COPY config ./config
+COPY deploy/registry.compose.yaml ./deploy/registry.compose.yaml
+COPY models/router ./models/router
+COPY scripts/fake_upstream.py ./scripts/fake_upstream.py
+RUN uv sync --frozen --no-dev
+
+RUN useradd --system --no-create-home tierwise
+USER tierwise
+
+ENV TIERWISE_HOST=0.0.0.0 TIERWISE_PORT=8000
+EXPOSE 8000
+CMD ["tierwise"]

tierwise-0.8.0/LICENSE

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.