Skip to content

Text

Anything Argonaut doesn't recognise opens here - and it's not the poor relation that sounds like. It's built the same windowed way as every other view, so a multi-gigabyte log scrolls smoothly from the moment it opens, and the index behind it costs about 2.5 MB per gigabyte.

Reading

  • Immediate display at any size, with smooth scrolling from the first line to the last.
  • Wrapping at a width you choose. Pathologically long lines - several gigabytes of minified JSON on one line - are broken up so there's something useful to read, and a mark in the gutter shows where a line was broken rather than where it ended.
  • The caret's position in the status bar: byte offset, line, column and how much is selected. On a pathological line the column shows a dash rather than stall the window counting characters; the line number is always right.
  • The character under the caret is named, not just shown - U+2028 LINE SEPARATOR, U+00A0 NO-BREAK SPACE, U+200B ZERO WIDTH SPACE - which is how you find the invisible thing breaking your file.
  • Syntax colouring for JSON, JSONC and NDJSON, and for config files - YAML, INI, .properties, .env, .editorconfig and simple TOML. It's picked from the file name or the content, and you can change it or turn it off from the toolbar.
  • Value hints on values the colouring picks out, and on unambiguous things anywhere - links, ISO dates, JWTs, hex colours, Base64 pictures - in logs, comments and plain text.

Why a fixed wrap width?

Wrapping to the window would mean measuring every line against the font and screen and rebuilding the index on every resize. A chosen width keeps every file, however large, instant to scroll.

Editing

Edit in place, with undo and redo, then save or save as. Your edits are held on top of the file rather than in a copy of it, so editing a multi-gigabyte file is as quick as opening one.

Saving writes a copy beside the file and swaps it in, so a failed save never leaves you with half a file. Search pauses while there are unsaved edits.

Line endings

The status bar shows whether the file uses LF, CRLF or a mix. New lines you type follow the file's own convention - or your operating system's, in a new document - and pasted text is converted to match. To convert the whole file, choose Line endings from the Transform menu; the change is written when you save.

Transforms

The Transform menu encodes and decodes the whole document - JSON string, URL, Base64 - and runs recipes that turn a list of lines into a SQL IN (...) list, a JSON array or a comma-separated list, and back. See Text transforms.