https://github.com/datacamp/sqlwhat
https://github.com/datacamp/sqlwhat
fs le
Last synced: about 1 year ago
JSON representation
- Host: GitHub
- URL: https://github.com/datacamp/sqlwhat
- Owner: datacamp
- License: agpl-3.0
- Created: 2017-01-02T22:30:45.000Z (over 9 years ago)
- Default Branch: master
- Last Pushed: 2024-01-25T10:54:21.000Z (over 2 years ago)
- Last Synced: 2025-04-07T04:35:20.217Z (about 1 year ago)
- Topics: fs, le
- Language: Python
- Homepage: https://sqlwhat.readthedocs.io
- Size: 1.25 MB
- Stars: 4
- Watchers: 22
- Forks: 8
- Open Issues: 19
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- License: LICENSE
Awesome Lists containing this project
README
# sqlwhat
[](https://app.circleci.com/pipelines/github/datacamp/sqlwhat)
[](https://badge.fury.io/py/sqlwhat)
`sqlwhat` enables you to write Submission Correctness Tests (SCTs) for interactive SQL exercises on DataCamp.
- If you are new to teaching on DataCamp, check out https://authoring.datacamp.com.
- If you want to learn what SCTs are and how they work, visit [this article](https://authoring.datacamp.com/courses/exercises/technical-details/sct.html) specifically.
- For more information about writing SCTs for SQL exercises, consult https://sqlwhat.readthedocs.io.
## Installing
```
pip install sqlwhat # install from pypi
make install # install from source
```
## Reference
* API Docs: https://sqlwhat.readthedocs.io
* AST viewer: https://ast-viewer.datacamp.com
* Extensions: https://github.com/datacamp/sqlwhat-ext
## Raising issues with how SQL is parsed
Please raise an issue on the respsective parser repo:
* [antlr-tsql](https://github.com/datacamp/antlr-tsql)
* [antlr-psql](https://github.com/datacamp/antlr-plsql)
## Basic Use
```python
from sqlwhat.State import State # State holds info needed for tests
from sqlwhat.Reporter import Reporter
from sqlwhat.checks import * # imports all SCTs
from sqlalchemy import create_engine
code = "SELECT * FROM artists WHERE id < 100"
state = State(
student_code = code,
solution_code = code,
pre_exercise_code = "",
student_conn = create_engine('sqlite:///'),
solution_conn = create_engine('sqlite:///'),
student_result = {'id': [1,2,3], 'name': ['greg', 'jon', 'martha']},
solution_result = {'id': [1,2,3], 'name': ['toby', 'keith', 'deb']},
reporter = Reporter()
)
# test below passes, since code is equal for student and solution
has_equal_ast(state)
# test below raises a TestFail error, since 'name' col of results
# doesn't match between student and solution results
check_result(state)
# shows error data
state.reporter.build_payload()
# can also be done using a chain
from sqlwhat.sct_syntax import Ex
Ex(state).check_result()
```
## Running unit tests
```bash
pytest -m "not backend"
```
If you also want to run the backend tests, you need to set a `GITHUB_TOKEN` environment variable with access to the (private) `sqlbackend` repository.
After this, you can:
```bash
make install
pytest
```
### Rules of testing
1. Running queries is the backend's job
2. If a test doesn't run queries, it doesn't need the backend
3. Very few tests should run queries
## Building Docs
Install sqlwhat and run ..
```
cd docs
make html
```