Locate the table before exporting it
The repository's synthetic examples/table-sample.md contains two tables: country revenue and a cash-flow extract. The first contains Spanish accents, decimal commas, an escaped pipe, and a pipe inside inline code. None should create an extra CSV column.
git clone https://github.com/faigvain-ang/md-table-to-csv.git
cd md-table-to-csv
python md_table_csv.py examples/table-sample.md --list
python md_table_csv.py examples/table-sample.md --table 1 --output out/revenue.csvPython 3.11+ is enough; there are no packages to install. The listing shows table 1 with headers Country, Revenue and Note and two body rows. Table numbering starts at 1, not 0.
Inspect the actual text output
Country,Revenue,Note
España,"1.234,50",audited | draft
México,"900,25",`A|B`The quote marks around 1.234,50 belong to CSV transport syntax. A CSV reader receives one cell containing that exact string. The tool does not assume that a comma is a decimal mark, remove thousands separators, or infer currency. In particular, a numeric conversion to 1234.50 is not part of export.
The Markdown escape in audited \| draft becomes a literal pipe. Inline-code backticks stay in the default output; add --text if you explicitly want the simple code/link wrappers removed. Export every supported table with --all --output-dir out/tables; the cash-flow table then becomes table-002.csv.
Open with the right import settings
Select UTF-8 and comma separation when importing. Spreadsheet auto-detection may still reinterpret dates or numbers, so choose text columns where preservation matters. CSV quoting does not stop formulas from executing; read the method and formula-safety note before opening untrusted cells.
The related KolmoPDF cloud converter can provide an earlier PDF-to-Markdown step, but this CLI starts with a local Markdown file and makes no cloud request. This sample was authored directly, not extracted from a customer PDF.
← Field notes