csplit

split a big file (or stream) into smaller chunks based on pattern

Say we have a file like so:

This is the first section.
---
This is another section.
---
This is a final section.

To split it (on MacOS using default csplit), we do:

csplit file.txt /---/ {1}

which will result in 3 files xx00, xx01 and xx02 with contents:

# xx00
This is the first section.

# xx01
---
This is another section.

# xx02
---
This is a final section.

Here {1} is the increment count of splits to do. If we omit it, it’ll split the file in twain.

See man csplit for more options

On GNU coreutils, we may supply count as ”{*}” for however many splits required, but here we must pass a number (which can be arbitrarily large).