7 data headaches that eat developers' time. And the 1-line fixes for each.

These come from building a browser tool
that converts CSVs with 200k rows —
and learning every way data can fight back.

1. Excel semicolon exports.
   Your comma parser dies on the first row.
   Fix: detect the separator from the header line, not the extension.

2. UTF-8 BOM.
   One invisible character at the start of the file
   and your first column is named "\uFEFFname".
   Fix: strip BOM before parsing. Always.

3. Huge files freeze the tab.
   Loading 500k rows into memory at once
   kills the browser — and the user's patience.
   Fix: stream in chunks and process row by row.

4. Numbers that are actually strings.
   "1,234.56" with a comma breaks every calculation.
   Fix: parse numbers with locale in mind, never blindly.

5. Mixed encodings in one file.
   Half the file is UTF-8, half is Latin-1.
   Fix: detect encoding from the first bytes, then convert.

6. Empty rows and stray newlines.
   A trailing blank line crashes naive parsers.
   Fix: skip empty lines and trim before splitting.

7. "But it works on my machine."
   The file your parser handled perfectly
   fails on the user's real-world file.
   Fix: test against messy, real data — not clean examples.

One rule ties them all:
assume the data is broken. Then it never surprises you.

Save this for the next time a CSV ruins your day.

Which one of these hit you hardest?