Repeated row names

[xinyaohuu]

When I converted the tables to markdown, I noticed repeated names, and it happened to many tables not just one. Do you have any suggestions?
Image:

Markdown output

[cau-git]

@xinyaohuu Thanks for the question. To clarify, in Markdown, this is actually expected behaviour, because the Employee Name / Address is detected as a multi-column cell with a span of 3. However, Markdown has no native way to represent cell spans, hence we resort to repeating the same values three times in this case. This makes sense when you need to address e.g. the column header from a data point in the table.

If you want to see an accurate representation considering cell spans, you can use the JSON output or export the tables to HTML instead. The output will look like this.

The minimal code to get the HTML tables is:

source = "your_table_doc.pdf"
converter = DocumentConverter()
result = converter.convert(source)

for table in result.document.tables:
    print("===============")
    print(table.export_to_html())

The screenshot above has this HTML source: sample_table.html.zip