Explain find code

What this find sample shows

start folder and tests

find . walks the current folder and everything below it, and each test after it narrows the list. -name "*.log" matched by file name and found ./debug.log and ./logs/app.log on a test project. -type d means folders and -type f means files.

-prune

-prune stops find from going into a folder it matched. The second command printed ./node_modules and ./web/node_modules and nothing from inside them. In the fourth, -path ./.git -prune -o ... skips .git and checks everything else.

-size and -mtime

-size +1M means bigger than 1 MB and found the 2 MB logs/big.bin. -mtime -7 means changed less than 7 days ago: it printed logs/big.bin and src/app.js, and skipped files last changed 20 and 30 days ago.

-exec and -delete

-exec wc -l {} + runs a command on the matches, with {} standing for the file names; the + hands them over in one go, so wc printed one line per file and a 3 total. The node_modules files were left out by -not -path. -delete removes what matched: it removed both empty folders under tmp, and then tmp too, once it was empty.

Good to know

• Run with GNU find 4.11 on a test project with code, logs, node_modules folders, a .git folder and empty folders.
• Quote patterns like "*.log". Unquoted, the shell filled in the .log files in the current folder first, and find stopped with find: paths must precede expression, naming one of those files, and asked about a possible unquoted pattern after -name.
• Run a find without -delete first and read the list. Nothing asks before -delete removes a file.

Watch out: -o means or, and an action such as -print only belongs to the test right before it. find . -name "*.log" -o -name "*.bin" -print printed only ./logs/big.bin. With parentheses, find . \( -name "*.log" -o -name "*.bin" \) -print printed all four matches.