I've been automating invoice processing for a side project, and one thing that always tripped me up was converting different formats to CSV. Whether it's a PDF from a vendor, an Excel sheet from a client, or an HTML table from a web app, you need a reliable way to extract data. Here's a bash script I use for quick conversions, plus a web tool for when I'm not in the terminal.
bash
!/bin/bash
Convert all invoices in a folder to CSV
Requires: python3, pandas, pdfplumber, openpyxl, lxml
for file in invoices/; do
ext="${file##.}"
case "$ext" in
pdf)
python3 -c "
import pdfplumber, pandas as pd
with pdfplumber.open('$file') as pdf:
rows = []
for page in pdf.pages:
table = page.extract_table()
if table:
rows.extend(table)
df = pd.DataFrame(rows[1:], columns=rows[0])
df.to_csv('${file%.}.csv', index=False)
print('Converted $file')
"
;;
xls|xlsx)
python3 -c "
import pandas as pd
df = pd.read_excel('$file', engine='openpyxl')
df.to_csv('${file%.}.csv', index=False)
print('Converted $file')
"
;;
html|htm)
python3 -c "
import pandas as pd
df = pd.read_html('$file')[0]
df.to_csv('${file%.*}.csv', index=False)
print('Converted $file')
"
;;
*)
echo "Skipping $file: unsupported format"
;;
esac
done
This script works great for batch processing, but if you need a quick, one-off conversion without setting up dependencies, SERPSpur's Invoice to CSV Converter handles PDF, XLS, XLSX, and HTML instantly—no code required.
Top comments (2)
Nice script! I've been using a similar approach but found that HTML tables with nested elements can trip up pandas read_html. Adding a fallback with BeautifulSoup for those edge cases saved me a lot of manual cleanup.
Great script! I've found that PDFs with scanned images can be tricky with pdfplumber—have you tried adding OCR support with pytesseract for those edge cases?