scottsnapperlab / db_tools

Database Tools

A set of command line executable and script importable tools to aid the Snapper Lab in managing and combining RedCap, FreezerPro, and other databases.

• Free software: MIT license
• Documentation: https://db-tools.readthedocs.io.

Features

• TODO

Install for Development

1. Install and become familiar with conda/Anaconda.

2. Make sure that you have (at least) the following anaconda channels activated in your .condarc file (the order is kind of important too):

   • anaconda
   • r
   • conda-forge
   • bioconda
   • pandas
   • defaults

3. Fork the repository to your github by clicking the "Fork" button at the top right of this project's github page.

4. Clone your forked repo to your dev computer: git clone [email protected]:YOUR_GITHUB_NAME/db_tools.git.

5. Enter your freshly cloned Database Tools directory: cd db_tools.

6. Run make help to see most of the make targets available.

7. Running make install. This creates and registers a conda environment named db_tools. Into that conda environment, it installs all of the needed libraries to run and develop Database Tools.

8. To uninstall your dev environment just run make uninstall-conda-env. All traces of the environment should be erased.

9. Remember to activate the conda env before you try to use or interact with Database Tools or you will not have access to it. That is, unless you are running a make target. In most cases, the first step of the make target will be to activate the correct env for you.

Credits

This package was created with Cookiecutter and the xguse/cookiecutter-pypackage project template which is based on audreyr/cookiecutter-pypackage.

db_tools's Issues

in data_dict: othermedraae is Text/Number; suggest changing

othermedraae is medical side effects. I believe that they are some kind of medical ontology codes. But they are NOT really numbers.

However, I can see the need to validate them as if they are numbers in the redcap form to prevent people from including garbage text characters (13 H anybody?).

I suggest setting them to something like integer rather than number.

This helps by:

• keeping the form validation able to screen out letters and symbols
• allowing us to detect that they should not be represented by floating point numbers

Discuss.