V2 detects and repairs instances of configuration drift in Python snippets.
| Name | Version |
|---|---|
| Node.js | 12.1.0 |
| Docker | 18.09.2 |
| Docker Compose | 1.23.2 |
Install dependencies and link the V2 executable with
npm i
npm link
Then build Docker images and start services
v2 build
docker-compose up --detach
V2 requires the Docker daemon to be up and running. Additionally, the user must have permissions to create Docker containers.
To run V2 on a code snippet with logging information, run
v2 run --verbose <code-snippet>
V2 will attempt to create a Dockerfile describing a working environment configuration. If one cannot be found, it will present information about what instances of configuration drift were found and patched.
V2 may also be run from inside a Docker, but requires having the Docker socket bind mounted.
docker run --rm -it --mount type=bind,source=/var/run/docker.sock,target=/var/run/docker.sock -v "$(pwd):/data:ro" --network=<network> localhost:5000/v2/v2:latest v2 run /data
All examples are stored under examples/.
v2 run examples/bayes/bayes.py
Running on Bayes will produce the following Dockerfile.
FROM python:3.7
COPY . /app
RUN ["apt-get","update"]
RUN ["pip","install","the==0.1.5"]
RUN ["pip","install","PyYAML==5.1"]
RUN ["pip","install","numpy==1.16.3"]
RUN ["pip","install","scipy==1.2.1"]
RUN ["pip","install","Theano==0.8.2"]
RUN ["pip","install","Lasagne==0.1"]
CMD ["python","/app/bayes.py"]v2 run examples/guess_candidate_model/guess_candidate_model.py
No working environment can be found for Guess Candidate Model due to an error in the gist. However, two instances of configuration drift are still found.
{
"name": "NoWorkingEnvironmentFoundError",
"time": 271,
"validations": 12,
"metadata": [
{
"id": "0",
"fixedValidations": [
{
"status_code": "Failed",
"dependencies": {
"status_code": "Success",
"install_errors": []
},
"execution": {
"status_code": "Exception",
"exception_name": "ModuleNotFoundError",
"exception_message": "No module named 'sklearn.cross_validation'",
"exception_file_name": "/data/examples/guess_candidate_model/guess_candidate_model.py",
"exception_line_number": 4,
"exception_line": "from sklearn.cross_validation import train_test_split",
"exception_stack": [
[
"/data/examples/guess_candidate_model/guess_candidate_model.py",
4,
"<module>",
"from sklearn.cross_validation import train_test_split"
]
]
},
"validations": 1
},
{
"status_code": "Failed",
"dependencies": {
"status_code": "Success",
"install_errors": []
},
"execution": {
"status_code": "Exception",
"exception_name": "ModuleNotFoundError",
"exception_message": "No module named 'tensorflow'",
"exception_file_name": "/usr/local/lib/python3.7/site-packages/keras/backend/tensorflow_backend.py",
"exception_line_number": 5,
"exception_line": "import tensorflow as tf",
"exception_stack": [
[
"/data/examples/guess_candidate_model/guess_candidate_model.py",
5,
"<module>",
"from keras.preprocessing import sequence, text"
],
[
"/usr/local/lib/python3.7/site-packages/keras/__init__.py",
3,
"<module>",
"from . import utils"
],
[
"/usr/local/lib/python3.7/site-packages/keras/utils/__init__.py",
6,
"<module>",
"from . import conv_utils"
],
[
"/usr/local/lib/python3.7/site-packages/keras/utils/conv_utils.py",
9,
"<module>",
"from .. import backend as K"
],
[
"/usr/local/lib/python3.7/site-packages/keras/backend/__init__.py",
89,
"<module>",
"from .tensorflow_backend import *"
],
[
"/usr/local/lib/python3.7/site-packages/keras/backend/tensorflow_backend.py",
5,
"<module>",
"import tensorflow as tf"
]
]
},
"validations": 4
}
],
"mutations": [
{
"type": "decrement_semver_minor",
"changes": {
"package": "scikit-learn",
"from": "0.20.3",
"to": "0.19.2"
},
"iddfs": {
"mutatorIndex": 1
}
},
{
"type": "decrement_semver_major",
"changes": {
"package": "Keras",
"from": "2.2.4",
"to": "1.2.2"
},
"iddfs": {
"index": 1,
"mutatorIndex": 0
}
},
{
"type": "decrement_semver_major",
"changes": {
"package": "Keras",
"from": "1.2.2",
"to": "0.3.3"
},
"iddfs": {
"index": 1,
"mutatorIndex": 0
}
}
],
"checkpoint": {
"status_code": "Failed",
"dependencies": {
"status_code": "Success",
"install_errors": []
},
"execution": {
"status_code": "Exception",
"exception_name": "NameError",
"exception_message": "name 'labeled_sample' is not defined",
"exception_file_name": "/data/examples/guess_candidate_model/guess_candidate_model.py",
"exception_line_number": 27,
"exception_line": "X = [x[1] for x in labeled_sample]",
"exception_stack": [
[
"/data/examples/guess_candidate_model/guess_candidate_model.py",
27,
"<module>",
"X = [x[1] for x in labeled_sample]"
]
]
},
"validations": 0
},
"code": "NotRepairable",
"message": "Validation exception 'NameError' is not repairable."
},
...
],
"message": "Unable to find a working environment configuration",
"stack": [
"NoWorkingEnvironmentFoundError: Unable to find a working environment configuration",
" at V2.infer (/v2/index.js:359:15)",
" at process._tickCallback (internal/process/next_tick.js:68:7)"
]
}v2 run examples/mobi/mobi.py
Running on Mobi will produce the following Dockerfile.
FROM python:2.7
COPY . /app
RUN ["apt-get","update"]
RUN ["pip","install","six==1.12.0"]
RUN ["pip","install","docutils==0.14"]
RUN ["pip","install","pydocumentdb==2.3.3"]
RUN ["pip","install","Sphinx==1.4.5"]
CMD ["python","/app/mobi.py"]