db_query / documentations /fnb_parser_doc.py
DavMelchi's picture
feat: improve streamlit docs and clean navigation icons
4d0d437
import streamlit as st
st.markdown(
"""
# F4NB Extractor Documentation
## 1. Objective
Extract site and sector information from F4NB Word documents (`.docx`) and export structured Excel output.
## 2. When to use this tool
Use this page when transforming design documents into tabular data for analysis and mapping.
## 3. Input files and accepted formats
- Required: one or multiple `.docx` files
- Sample file available: `samples/FN4B.docx`
## 4. Required fields in document content
Parser looks for common labels in table cells such as:
- code/site name
- locality/address
- coordinates (`X`, `Y`, `Z`)
- sector parameters (`Azimuth`, `Height`, `Tilt mecanique`, `Tilt electrique`)
## 5. Step-by-step usage
1. Open `Apps > F4NB Extractor`.
2. Upload one or more `.docx` files.
3. Click `Process`.
4. Review extracted dataframe.
5. Download Excel output.
6. Review map if coordinates are parsed.
## 6. Outputs generated
- extracted dataframe (site + sector rows)
- downloadable file `extracted_fnb.xlsx`
- optional map using converted decimal coordinates
## 7. Frequent errors and fixes
- No data extracted.
- Fix: verify input document follows expected table structure.
- Coordinate conversion fails.
- Fix: clean coordinate format and direction markers.
- Missing map points.
- Fix: ensure X/Y values are valid and parseable.
## 8. Minimal reproducible example
- Input: `samples/FN4B.docx`
- Action: upload and click `Process`.
- Expected result: sector rows displayed and Excel download available.
## 9. Known limitations
- Extraction is heuristic and template-dependent.
- Highly custom DOCX layouts may require parser updates.
- Map relies on successful coordinate conversion.
## 10. Version and update date
- Documentation version: 1.0
- Last update: 2026-02-23
"""
)