jev-align 0.1.0__tar.gz

jev_align-0.1.0/.github/workflows/publish.yml

@@ -0,0 +1,60 @@
+name: Publish to PyPI
+
+on:
+  release:
+    types: [published]
+
+concurrency:
+  group: publish-pypi
+  cancel-in-progress: false
+
+jobs:
+  publish:
+    name: Build and publish
+    runs-on: ubuntu-latest
+    environment:
+      name: pypi
+      url: https://pypi.org/p/jev-align
+    permissions:
+      contents: read
+      id-token: write
+
+    steps:
+      - name: Check out the release tag
+        uses: actions/checkout@v4
+
+      - name: Install uv
+        uses: astral-sh/setup-uv@v7
+        with:
+          enable-cache: true
+
+      - name: Install Python
+        run: uv python install 3.11
+
+      - name: Verify tag matches package version
+        shell: bash
+        run: |
+          package_version="$(uv version --short)"
+          tag_version="${GITHUB_REF_NAME#v}"
+          if [[ "${GITHUB_REF_NAME}" != v* || "${tag_version}" != "${package_version}" ]]; then
+            echo "Release tag ${GITHUB_REF_NAME} does not match package version ${package_version}."
+            exit 1
+          fi
+
+      - name: Install development dependencies
+        run: uv sync --locked --extra dev --python 3.11
+
+      - name: Run tests
+        run: uv run --no-active --quiet pytest
+
+      - name: Run lint
+        run: uv run --no-active --quiet ruff check src tests
+
+      - name: Build distributions
+        run: uv build
+
+      - name: Check distribution metadata
+        run: uvx twine check dist/*
+
+      - name: Publish distributions
+        uses: pypa/gh-action-pypi-publish@release/v1

jev_align-0.1.0/.gitignore

@@ -0,0 +1,12 @@
+.DS_Store
+.coverage
+.env
+.env.*
+!.env.example
+.envrc
+.jev-align/
+.pytest_cache/
+.venv/
+__pycache__/
+*.py[cod]
+dist/

jev_align-0.1.0/AGENTS.md

@@ -0,0 +1,358 @@
+# AGENTS.md
+
+This repository contains `jev-align`, an interactive active-learning CLI for
+building AI Functions from a user's judgments. Coding agents may help configure
+and operate the workflow, but must preserve the human labeling loop.
+
+## Installation reference
+
+Requires Python 3.11 or newer. For a released build, prefer an isolated CLI
+installation:
+
+```shell
+uv tool install jev-align
+# Without uv:
+pip install jev-align
+```
+
+From a local checkout, use `uv tool install .`; after pulling or changing the
+source, refresh it with `uv tool install --force .`. To install straight from
+GitHub, use `uv tool install "git+https://github.com/sutro-sh/jev-align.git"`.
+For editable development, run `uv sync --extra dev` and launch with
+`uv run jeva`.
+
+Both `jeva` and `jev-align` invoke the same CLI. Interactive, non-editable
+installs check PyPI for newer releases at startup. In virtual environments the
+updater prefers `uv pip` when available and falls back to that environment's
+Python and pip. Set `JEVA_DISABLE_UPDATE_CHECK=1` to disable the check.
+
+## Operating the CLI for a user
+
+### Core rule
+
+The user supplies every label. Do not skip examples, silently infer labels, or
+accept an optimized definition on the user's behalf. A rationale is optional,
+but encourage one when it explains an important boundary or corrects the
+model's reasoning.
+
+### Before starting
+
+1. Confirm that `TYPESAFE_API_KEY` is available.
+2. Confirm a reflection provider is configured: `OPENAI_API_KEY`,
+   `ANTHROPIC_API_KEY`/`CLAUDE_API_KEY`, `GEMINI_API_KEY`, or the endpoint and
+   credentials required by a custom LiteLLM provider.
+3. Identify the intended CSV, Parquet, or JSONL file without modifying it.
+4. Establish the question, task type, input columns, and labels or score levels.
+5. If any choice would materially change the task semantics, ask the user.
+
+Never display secret values. It is enough to report whether a required key is
+configured.
+
+### Starting a run
+
+Use the guided home screen when the user wants to choose interactively:
+
+```shell
+jeva
+```
+
+Use flags when the setup is already known:
+
+```shell
+jeva optimize DATA \
+  --question "QUESTION" \
+  --column COLUMN
+```
+
+Relevant task shapes:
+
+- Binary: provide `--question`; optionally add concrete `--true-criteria` and
+  `--false-criteria`.
+- Multiclass: repeat `--class "NAME=DESCRIPTION"` for mutually exclusive
+  labels.
+- Multilabel: repeat `--class "NAME=DESCRIPTION"` and add `--multilabel`.
+- Score: repeat `--score-level "DESCRIPTION"` in lowest-to-highest order.
+
+The installed package includes preconfigured Hacker News, support-ticket, and
+agent-trace examples. The guided dataset picker lists those first, followed by
+CSV, Parquet, and JSONL files discovered below the current directory.
+
+Use `--batch-size` to choose the number of training annotations per round. The
+guided Advanced menu offers 5, 10, 15, or 20. Add `--holdout` only when the user
+wants a 20% reserved evaluation split; this adds 20% extra held-out annotations
+per round. Advanced also configures maximum GEPA metric calls, which defaults to
+300. For scripted runs, use `--max-metric-calls`; `--metric-budget` remains an
+alias.
+
+Use `--all-columns-concatenated` only when every field is useful. Prefer
+explicit `--column` values when IDs, timestamps, or metadata could distract the
+evaluator. The normal default is the first 1,000 rows or all rows for a smaller
+dataset; only set `--pool-size` when the user wants a different limit.
+
+Run the CLI in a real PTY when possible so arrow-key menus, progress displays,
+and prompts work correctly.
+
+### Reflection providers
+
+GEPA's reflection model is separate from the TypeSafe JEV evaluation model.
+Reflection uses LiteLLM model identifiers. OpenAI, Anthropic, and Gemini are
+listed automatically when their standard keys are present. For another
+provider, pass `--reflection-model provider/model`; in the wizard select
+**Choose a different model** and then **Enter a custom LiteLLM model**.
+
+Fireworks example:
+
+```shell
+export FIREWORKS_API_KEY="..."
+jeva optimize DATA \
+  --question "QUESTION" \
+  --column COLUMN \
+  --reflection-model \
+    "fireworks_ai/accounts/fireworks/models/llama-v3p1-8b-instruct"
+```
+
+For a local or hosted vLLM server exposing an OpenAI-compatible `/v1` API:
+
+```shell
+export HOSTED_VLLM_API_BASE="http://localhost:8000/v1"
+export HOSTED_VLLM_API_KEY="..." # Omit when the endpoint has no authentication.
+jeva optimize DATA \
+  --question "QUESTION" \
+  --column COLUMN \
+  --reflection-model "hosted_vllm/Qwen/Qwen3-8B"
+```
+
+For a generic OpenAI-compatible endpoint:
+
+```shell
+export OPENAI_API_BASE="http://localhost:8000/v1"
+export OPENAI_API_KEY="local" # Replace when the endpoint requires a real key.
+jeva optimize DATA \
+  --question "QUESTION" \
+  --column COLUMN \
+  --reflection-model "openai/Qwen/Qwen3-8B"
+```
+
+The provider/model identifier and environment variables must follow the
+[LiteLLM provider configuration](https://docs.litellm.ai/docs/providers).
+`TYPESAFE_API_KEY` remains required for JEV evaluation.
+
+### Labeling rounds
+
+For each displayed item:
+
+1. Relay the content and choices clearly if the user cannot see the terminal.
+2. Mention its uncertainty and whether it is a random audit sample.
+3. Ask the user for the label.
+4. Ask for an optional rationale, especially on ambiguous or surprising cases.
+5. Enter exactly what the user chose.
+
+Held-out cards are explicitly marked. Collect their labels normally, but never
+reuse their labels or rationales to guide task wording. The CLI keeps them out
+of GEPA and reports their score separately.
+
+There is no skip action. At the label picker, `b` removes the previous answer
+and moves backward. At the rationale prompt, `/back` returns to the current
+label picker; a literal `b` is valid rationale text. At the first item of a
+later round, `b` can rewind the prior optimization round after confirmation.
+
+For multilabel tasks, use Up/Down to navigate, Space to toggle, and Enter to
+confirm. An empty selection is a valid judgment.
+
+### Reviewing a GEPA proposal
+
+After optimization, summarize for the user:
+
+- The task metric before and after.
+- The fixed-pool certainty before and after.
+- Any regression, even when the task metric improved.
+- The material changes in the proposed definition.
+
+Then ask the user whether to accept, reject, or quit and resume later. Do not
+treat a higher training score as automatic approval. The score uses accumulated
+labels and is not a held-out generalization estimate.
+
+### Existing runs
+
+Browse saved AI Functions with:
+
+```shell
+jeva functions
+```
+
+Resume a known run directly with:
+
+```shell
+jeva optimize --resume .jev-align/runs/RUN_ID
+```
+
+Saved state, labels, prediction caches, GEPA artifacts, and rewind archives live
+inside the run directory. Treat them as application state: do not hand-edit,
+delete, or replace them. If a run is pending at the proposal screen, resume that
+decision rather than starting another optimization.
+
+Running an accepted function on another dataset writes JSONL results to
+`.jev-align/outputs/` without changing the saved function.
+
+### Continual learning from live calls
+
+Applications can call an accepted AI Function and record uncertain or randomly
+audited predictions for later human labeling. Install `jev-align` in the
+application environment and load the saved run with capture enabled:
+
+```python
+from jev_align import AIFunction
+
+is_aviation = AIFunction.load(
+    ".jev-align/runs/RUN_ID",
+    capture=True,
+)
+
+prediction = is_aviation(
+    title="Airport expansion",
+    text="A new runway opens next year.",
+)
+```
+
+Calls must use the input columns configured for the saved run and return a
+normalized `Prediction`, not a human label. Capture adds no model call and writes
+selected observations to `.jev-align/captures/` in the background. A recorder
+created by `capture=True` is flushed automatically at process shutdown. Call the
+AI Function's `close()` method or use it as a context manager when deterministic
+shutdown is required. This integration is currently synchronous.
+
+To continue learning, run `jeva functions`, open the matching AI Function, and
+select **Resume learning**, or run:
+
+```shell
+jeva optimize --resume .jev-align/runs/RUN_ID
+```
+
+When the CLI offers newly captured calls, let the user decide whether to import
+them. Every unique eligible call is imported; there is no capture-pool row cap.
+The model's recorded prediction is never treated as truth. The user must label
+every selected example, with an optional rationale, before GEPA can learn from
+it. Previously labeled captures remain in the evaluation pool, while only
+unlabeled inputs are eligible for another annotation batch.
+
+The accepted definition is loaded when `AIFunction.load(...)` runs. Restart the
+process or reload the AI Function to use a newly accepted version.
+Use `JEVA_CAPTURE_DIR` when the application and CLI need to share a capture
+directory outside the workspace.
+
+`AIFunction` applies the saved run's column selection, normalization, and
+concatenation. Its return value is a provider-neutral `Prediction`: binary tasks
+expose `probability`, multiclass tasks `choice` and `confidence`, multilabel
+tasks `label_probabilities`, and score tasks `score` and `confidence`. Runtime
+calls require `TYPESAFE_API_KEY`, but not a reflection-model key. Pending
+proposals are never loaded. A run with no accepted proposal uses its seed
+definition.
+
+Capture makes no additional JEV request and does not retry evaluations. By
+default it records predictions with ambiguity of at least 0.8 and a random 5%
+audit of the remainder. For binary tasks, 0.8 ambiguity corresponds to
+probabilities from 0.4 through 0.6. Each record includes the evaluated input,
+prediction, definition, model provenance, and selection reason, but none of
+those values constitute a human label.
+
+Capture uses JSONL files under `.jev-align/captures/`; it requires no database or
+server. Records enter a bounded queue and a background thread writes batches of
+up to 64, flushing about every 250 ms. Defaults are 256 queued records, 64 KiB
+per record, and 64 MiB per capture file. Full queues, oversized records, and
+exhausted file budgets increment `dropped`. Disk errors disable that writer, log
+one warning, and populate `error` without failing successful predictions. Limits
+and files are per `Capture` instance, so create one inside each worker after
+forking. For custom thresholds or storage settings, pass a configured
+`Capture(...)` instance instead of `capture=True`; the caller then owns and must
+close it. Existing files are not rotated or deleted automatically. Closing drains
+the queue for up to five seconds; abrupt shutdown can lose buffered records.
+
+On resume, the CLI searches `.jev-align/captures/` in the current workspace and
+the run workspace, plus `JEVA_CAPTURE_DIR` when configured. It imports complete,
+matching records already on disk, deduplicates repeated inputs, excludes the
+original and holdout datasets, and ignores malformed, partial, or over-1-MiB
+lines. Discovery itself makes no API call. Approved captured inputs are also
+stored in the run's `captured-inputs.json`, so removing the raw logs does not
+remove them from that run.
+
+Captured inputs use the normal full-pool evaluation, prediction cache, and batch
+selection. Human labels from captures join training, while the original pool and
+holdout remain fixed. Reports show captured-pool uncertainty separately so the
+original fixed-pool history remains comparable. Pending GEPA proposals must be
+resolved before importing new captures, and every new proposal still requires
+explicit human acceptance.
+
+## Helping develop the repository
+
+The main modules are:
+
+- `src/jev_align/cli.py`: interactive flows, rendering, and commands.
+- `src/jev_align/session.py`: round lifecycle, caching, rewind, and decisions.
+- `src/jev_align/optimizer.py`: GEPA integration and task metrics.
+- `src/jev_align/backends.py`: provider-neutral backend contract and factory.
+- `src/jev_align/jev.py`: TypeSafe JEV adapter.
+- `src/jev_align/models.py`: persisted schemas and normalized predictions.
+- `src/jev_align/persistence.py`: run and label storage.
+- `src/jev_align/runtime.py`: synchronous callable interface for saved functions.
+- `src/jev_align/capture.py`: bounded, best-effort JSONL capture of live predictions.
+- `src/jev_align/captured_inputs.py`: discovery and deduplication of captured inputs.
+
+Capture files are unlabeled observations, not run state. Never treat their model
+predictions as human labels. Resuming a saved function offers to import matching
+captured calls from the current or run workspace's `.jev-align/captures/` and
+optional `JEVA_CAPTURE_DIR`. Resolve pending proposals first. Approved inputs
+are persisted in `captured-inputs.json` inside the run; treat this as application
+state. Human labels from captures join training, while the original fixed pool
+and holdout remain unchanged. Keep capture writes off the evaluation path and
+preserve bounded queues, file-size limits, and nonblocking overflow behavior.
+Original and captured pools share evaluation, prediction caching, and batch
+selection. Every unique eligible captured input remains in the active evaluation
+pool, including every labeled input. Filter labeled inputs only when selecting
+the next annotation batch. Report captured-pool uncertainty separately so
+original fixed-pool history stays comparable.
+
+TypeSafe JEV is currently the only included evaluation backend. Runs persist a
+provider-neutral backend configuration so additional adapters can be added;
+`--jev-model` remains a compatibility alias for `--backend-model`.
+
+Preserve these product invariants when making changes:
+
+- The human explicitly confirms every label.
+- Proposed prompts are shown as a diff before acceptance.
+- The accepted candidate seeds the next optimization round.
+- Full-pool uncertainty remains comparable across rounds.
+- Old run state remains loadable through explicit migrations.
+- Backends must declare supported task types and usable uncertainty signals.
+- Provider-specific SDK objects do not cross the backend boundary.
+
+Before handing off a code change, run:
+
+```shell
+uv run --no-active --quiet pytest
+uv run --no-active --quiet ruff check src tests
+```
+
+Do not perform live API evaluations unless the user explicitly asks; the normal
+test suite is designed to run without them.
+
+## Publishing a release
+
+`pyproject.toml` is the authoritative version source. Prepare a release with:
+
+```shell
+uv version 0.1.0
+git add pyproject.toml uv.lock
+git commit -m "release 0.1.0"
+make release VERSION=0.1.0
+```
+
+The Make target requires a clean worktree, verifies the configured version,
+runs tests and lint, builds both distributions, and runs `twine check`. It then
+pushes an annotated `v<VERSION>` tag and creates a GitHub Release with generated
+notes. Publishing is not performed from the developer machine.
+
+Publishing the GitHub Release triggers `.github/workflows/publish.yml`. The
+workflow repeats validation from the tagged commit and publishes through PyPI
+Trusted Publishing. Its PyPI publisher must be configured for the
+`sutro-sh/jev-align` repository, `publish.yml` workflow, and `pypi` environment.
+No PyPI token should be added to the repository or GitHub secrets.

jev_align-0.1.0/LICENSE

@@ -0,0 +1,201 @@
+                                 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.