Should your workflow start with CSV or JSON?
Understand practical tradeoffs for ingestion, transformation, export, and developer ergonomics.
Continue your journey with these related tools
Key Insights & Concepts
Comma Separated Values (CSV) is the cockroach of data formats—it survived mainframes, the Dot-com boom, and the Mobile era, and it will likely survive the AI revolution. It is the universal adapter for data.
A valid UTF-8 CSV often opens as gibberish in Microsoft Excel. Why?
Excel assumes strict ASCII unless told otherwise. To force Excel to recognize UTF-8 (e.g., emojis ✨, Chinese characters), you must include a hidden character at the very start of the file called the Byte Order Mark (BOM).
This is a critical vulnerability often overlooked. If you allow users to input data that is later exported to CSV, they can hack your admins.
An attacker sets their "Name" to:=cmd|' /C calc'!A0
When an admin exports the user list and opens it in Excel, Excel interprets the `=` as a formula and executes the command (opening Calculator, or downloading malware).
Sanitize fields starting with `=`, `+`, `-`, or `@`. Prepend a single quote `'` or tab `\t` to force Excel to treat it as text.
Junior developers often try to parse CSV using `string.split(',')`. This fails immediately on real-world data.
Name, "Role", "Quote"
"Doe, John", Admin, "He said ""Hello"" today"A parser needs to handle: 1. Quoted Fields: Commas inside quotes are data, not delimiters. 2. Escaped Quotes: Double quotes usually mean a single literal quote. 3. Newlines inside cells: A physical line break in the file doesn't always mean a new row.
Key Insights & Concepts