OpertusMundi/profile — reverse-engineered prompt
Reverse engineered prompt
Build me a small Python service that can inspect geospatial files and give back a useful profile of what is inside. It should handle vector, raster, and NetCDF data, and let people either upload a file directly or point to a file path that lives under a configured input folder. I also want a normalization feature for CSV and shapefile style data so common cleanup can be done on selected columns, like dates, phone numbers, casing, special characters, transliteration, value cleaning, WKT cleanup, and column name cleanup.
Please make it work as an API with clear docs at the root page, and support both fast responses and background processing. If a job is deferred, return a ticket, then add routes to check job status and download the result later. Include the map preview options mentioned in the docs, like basemap choice, image size, aspect ratio, CRS, and optional lat, lon, time, and geometry column names. Make it easy to run locally or in containers, and look up current docs online if you need to fill in setup details.
Want more depth? Deep Reverse