Sensibility

Finds and fixes syntax errors. This is experimental software.

NOTE: Please use the saner2018 tag if you're looking to replicate the results from our SANER 2018 paper. Find the replication data on archive.org See Citation for how to cite this paper. Use the following to clone Sensibility with the correct branch:
git clone --branch=saner2018 https://github.com/naturalness/sensibility.git

Requirements

Sensibility requires:

Python 3.6.

For fixing Java files:

Java 8 SE

For fixing JavaScript files:

Node.JS >= 6.0 and ZeroMQ (macOS: brew install zeromq; Ubuntu: apt install libzmq-dev)

Researchers that wish to mine more repositories or languages will require:

A running Redis server.

Researchers wishing to evaluate the current methods will require:

UnnaturalCode

Install

Activate a virtualenv, if that's your thing. Then,

pip install -e .

Usage

Once installed, there's one entry point to all the scripts and utilities included:

sensibility <SUBCOMMAND>

Most subcommands require the specification of a language, with the -l <LANGUAGE> option before the subcommand.

For example, to train Java models:

sensibility -l java train-list --help

Development

The following diagram visualizes the data flow. Subcommands to sensibility are in black rectangles; white ovals are products. Please contact the author to obtain the mistake database, which is only applicable when evaluating the Java models.

Tests

To run the tests, install tox using Pip, then run tox.

Mining repositories

You must create a GitHub OAuth token and save it as .token in the repository root.
Run redis-server on localhost on the default port.

Use sensibility mine find-repos to get a list of the top ~10k repos:

sensibility mine find-repos javascript | sort -u > javascript-repos.txt

Use bin/enqueue-repo to enqueue repos to download:

sensibility mine enqueue-repo < javascript-repos.txt

Start one or more downloaders. These will dequeue a repo from the running Redis server and download sources:
```
sensibility mine download
```

Evaluation

Type make experiments to train all of the models and evaluate each one. See libexec/experiments for more details.

Citation

If you use Sensibility in academic works, please use the following citation:

@inproceedings{santos2018,
    author={Santos, Eddie Antonio and
            Campbell, Joshua Charles and
            Patel, Dhvani and
            Hindle, Abram and
            Amaral, Jos{\'e} Nelson},
    booktitle={2018 {IEEE} 25th International Conference on Software Analysis, Evolution and Reengineering ({SANER})},
    title={Syntax and {Sensibility}: Using Language Models to Detect and Correct Syntax Errors},
    year={2018},
    month={Mar}}

License

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.