A productivity extension for Data Scientists and MLOps Engineers to streamline local data ingestion.
The Problem It Solves
- Data workflows often start with tedious boilerplate: calculating relative paths for CSVs across different environments, remembering whether you are in notebooks/ or src/, and typing out standard pandas imports repeatedly.
- CSV Path Importer turns data ingestion into a simple drag-and-drop motion, intelligently generating relative paths and dynamic variables so you can get straight to the analysis.
Core Features
- Intelligent Drag-and-Drop: Drag a .csv file into any Python script or Jupyter Notebook to instantly generate the necessary code.
- Smart Relative Pathing: Automatically calculates the correct relative path from the active file to the dataset, ensuring reproducibility across environments (bypassing hardcoded absolute path errors).
- Context-Aware Variable Numbering: Automatically increments variables (e.g., df_1, df_2 or file_path_1, file_path_2) preventing you from accidentally overwriting earlier datasets in the same session.
- Duplicate Import Prevention: Scans your Jupyter Notebook cells; if import pandas as pd already exists, it smartly omits redundant import blocks.
- Quick-Toggle Modes: Use the VS Code Command Palette to instantly switch between generating a full Pandas boilerplate or just a raw relative file path variable.
How to Use It
- Open a .py or .ipynb file.
- Find a .csv file in your VS Code Explorer.
- Hold the Shift key and drag the file directly into your code editor.
- The extension instantly generates the relative path and ingestion code.
To swap output modes: Open the Command Palette (Ctrl+Shift+P or Cmd+Shift+P), type CSV Path Importer: Switch Drop Template, and select either Pandas Boilerplate or Raw File Path Variable.
Built With
TypeScript * VS Code Extension API (DocumentDropEditProvider, WorkspaceConfiguration)
Node.js / ESBuild