Support for XLSX File Uploads in GeoNode

[Gpetrak]

📝 Introduction

User Story:
As a User, I want to upload Excel (.xlsx) files directly to GeoNode so that I can easily create geospatial layers from tabular data without manually converting them to CSV or Shapefiles first.

Problem Statement:
Currently, GeoNode primarily supports CSV for tabular uploads. However, many users work natively in Excel. Forcing a conversion to CSV often leads to encoding issues (UTF-8) or loss of formatting, making the data ingestion process friction-heavy and prone to errors.

---

✅ Expected Results (Acceptance Criteria)

• File Support: The upload interface should recognize and accept .xls or .xlsx file extensions.
• Sheet Selection:
  • If the workbook has multiple sheets, the system will read only the first sheet
  • The system should alert the user that the other sheets are ignored (even just a warning log)
  • If the first row of the xlsx file does not contains the headers, the system should raise an error
  • the system should ignore empty rows in the excel file if possible
• Geometry Mapping: The system should extract the geometry columns (e.g., Latitude/Longitude columns) similar to the CSV upload flow. Only point geometries are supported, and WKT fields will be ignored.
• Validation: Provide a clear error message to the user if the file is not in the correct format

---

🗒️ Notes & Technical Details

• Library Suggestion: Consider using pandas/geopandas or python-calamine for the backend parsing logic. Pandas is extremly powerfull, but it might be really heavy just for reading an excel file. python-calamine is faster and lighter but more complex to use.
• Re-use what we have: since we already have the full infrastructure to upload a CSV file, we might just consider to convert the XLSX file into a CSV and then use the existing upload pipeline. We already do this for other formats. We need to evaluate this before writing a new upload handler.
• Data Processing: To ensure the system handles large workbooks efficiently during the ingestion phase, the layer processing should be memory-efficient.
• Security: we should be carefull on the security: https://openpyxl.readthedocs.io/en/stable/#security
• The handler MUST be optional