rcol 0.0.4

Pandas DataFrame templates for REDCap instruments, with stacking and upload functionality.

rcol

rcol (RedCap Uni Oldenburg) is a Python package that provides Pandas DataFrame templates for REDCap instruments with stacking and upload functionality.

Tested on: Windows, Linux, macOS | Python: 3.8, 3.9, 3.10, 3.11, 3.12, 3.13

Installation

pip install rcol

Quick Start

from rcol.instruments import fal, ehi, bdi_ii
import pandas as pd

# Use individual instruments
print(f"FAL has {len(fal)} fields")
print(f"EHI has {len(ehi)} fields") 
print(f"BDI-II has {len(bdi_ii)} fields")

# Stack multiple instruments for REDCap upload
all_instruments = pd.concat([fal, ehi, bdi_ii], ignore_index=True)
print(f"Combined: {len(all_instruments)} total fields")

# Upload to REDCap (requires PyCap and API credentials)
from redcap import Project
project = Project(api_url, api_token)
project.import_metadata(all_instruments, import_format='df')

Available Instruments

• fal: Fragebogen zur Allgemeinen Leistungsfähigkeit (General Performance Questionnaire)
• ehi: Edinburgh Handedness Inventory
• bdi_ii: Beck Depression Inventory II
• moca: Montreal Cognitive Assessment
• tmt: Trail Making Test
• wnss: Weinstein's Noise Sensitivity Scale (WNSS-21)

Creating Custom Instruments

You can create custom instruments or extend existing ones without contributing to the package:

import pandas as pd
from rcol.instruments import fal, ehi
from redcap import Project

# Create a custom instrument from scratch
custom_instrument = pd.DataFrame({
    'field_name': ['record_id', 'custom_field_1', 'custom_field_2'],
    'field_label': ['Record ID', 'Custom Field 1', 'Custom Field 2'],
    'field_type': ['text', 'text', 'radio'],
    'form_name': ['custom_form', 'custom_form', 'custom_form'],
    'choices': ['', '', '1, Yes | 0, No']
})

# Add a new question to an existing instrument
fal_new_question = pd.DataFrame({
    'field_name': ['fal_like_redcap'],
    'field_label': ['Do you like REDCap?'],
    'field_type': ['radio'],
    'form_name': ['fal'],
    'choices': ['1, Yes | 0, No']
})

fal_extended = pd.concat([fal, fal_new_question], ignore_index=True)

# Combine everything and upload to REDCap
all_instruments = pd.concat([fal_extended, ehi, custom_instrument], ignore_index=True)

project = Project(api_url, api_token)
project.import_metadata(all_instruments, import_format='df')

See tutorial_custom_instruments.py for a complete guide with all REDCap metadata fields.

Contributing a New Instrument

1. Fork this repository

2. Add your instrument data in src/rcol/instruments.py:

   # Define your instrument fields
   my_instrument_data = [
       {
           "field_name": "record_id",
           "form_name": "my_instrument", 
           "field_type": "text",
           "field_label": "Record ID",
           # ... other REDCap metadata fields
       },
       # ... more fields
   ]
   
   # Create DataFrame
   my_instrument = pd.DataFrame(my_instrument_data)
   

3. Run the instrument test suite to validate your template. The shared tests automatically pick up every pandas.DataFrame exported from rcol.instruments and check for required REDCap metadata, non-empty field names, and duplicate protection:

   uv run --with pytest pytest -k instrument
   

   If you add custom validation that needs extra assertions, extend tests/test_templates.py accordingly.

4. Preview the documentation (optional). The MkDocs site renders the instrument tables directly from rcol.instruments. To check the output locally:

   uv run --group docs mkdocs serve
   

5. Submit a pull request. Every PR triggers the GitHub Actions CI workflow, which runs the instrument tests across Windows, Linux, and macOS with Python 3.8-3.13. Make sure the workflow badge stays green before requesting review.

Development

# Clone and install for development
git clone https://github.com/JuliusWelzel/rcol.git
cd rcol
uv sync

# Run tests
uv run pytest

# Build package
uv build

Documentation

The documentation site is powered by MkDocs and mkdocstrings, so instrumentation tables are rendered directly from rcol.instruments at build time.

# Serve docs locally
uv run --group docs mkdocs serve

# Build static site
uv run --group docs mkdocs build

License

MIT