@hystking
34 commits in the last 30 days
longest streak 23 dayscurrent 23
git reset --hard? They're probably still there.
git reflog git branch rescue <sha>
git reflog records everywhere HEAD has been. Find the SHA, park it on a branch, and you're back.git log -S "retryCount" --oneline
-S (the "pickaxe") lists commits where the number of occurrences of that string changed. Add -p to see the actual diff.curl -s -o /dev/null -w "%{http_code}\n" https://example.com-o /dev/null throws away the response, and -w prints only the format you ask for. It's handy for health checks in a loop.set -euo pipefail
-e exits on the first error, -u catches typos in variable names, and pipefail makes a pipeline fail if any stage fails, not just the last one.git switch -
cd -, it returns to the branch you checked out last. Run it again to toggle between the two.git merge - or git rebase -. The shorthand @{-1} works across almost every git command that takes a branch ref.find . -name "*.txt" -print0 | xargs -0 wc -l
-print0 ends each path with a null byte instead of a newline, and xargs -0 splits on that. Spaces and tabs in names are now harmless.git bisect start HEAD v1.4.0 git bisect run npm test
git bisect reset.lsof -ti :3000 | xargs kill
-t prints only the PIDs, so kill gets exactly what it needs. Swap in your port number. Handy for the "address already in use" moment.pbcopy < notes.txt
pbpaste goes the other way, dumping the clipboard to stdout. Pipe it into jq or grep and you're done.time:
time npm test
watch -n 2 ls -l dist/
watch reruns the command every 2 seconds and shows the latest output. On macOS, install it with brew install watch first.git commit --fixup <sha> GIT_SEQUENCE_EDITOR=true git rebase -i --autosquash <sha>~1
du -sh * | sort -h
sort -h understands human-readable sizes, so the biggest entries land at the bottom. Then go shame that node_modules.Ctrl+X Ctrl+E in bash. It opens the current command line in your $EDITOR, and saving and quitting runs the result. Multi-line pipelines are much easier to fix in vim than in a single prompt line.mount | column -t
column -t pads columns so they line up. Pipe any space-separated output through it and it stops looking like a crime scene.diff <(sort old.txt) <(sort new.txt)
<(...) is process substitution. The shell gives diff a fake file that holds each command's output. Works in bash and zsh.curl -I https://example.com
-I sends a HEAD request and prints only the headers. Good for spotting redirects, caching rules, and content types.sudo !!
!! expands to your last command, so the shell reruns it with sudo in front. Lazy, efficient, and a little smug.