gem-hunter

sd is an intuitive find & replace CLI.

Why use it over any existing tools?

Painless regular expressions.   sd uses regex syntax that you already know from JavaScript and Python. Forget about dealing with quirks of sed or awk - get productive immediately.

String-literal mode.   Non-regex find & replace. No more backslashes or remembering which characters are special and need to be escaped.

Easy to read, easy to write.   Find & replace expressions are split up, which makes them easy to read and write. No more messing with unclosed and escaped slashes.

Smart, common-sense defaults.   Defaults follow common sense and are tailored for typical daily use.

While sed does a whole lot more, sd focuses on doing just one thing and doing it well. Here are some cherry-picked examples where sd shines.

Simpler syntax for replacing all occurrences:

• sd: sd before after
• sed: sed s/before/after/g

Replace newlines with commas:

• sd: sd -A '\n' ','
• sed: sed ':a;N;$!ba;s/\n/,/g'

Note: this requires -A (across mode) since \n is a cross-line pattern.

Extracting stuff out of strings containing slashes:

• sd: echo "sample with /path/" | sd '.*(/.*/)' '$1'

• sed: echo "sample with /path/" | sed -E 's/.*(\\/.*\\/)/\1/g'

  With sed, you can make it better with a different delimiter, but it is still messy:

  echo "sample with /path/" | sed -E 's|.*(/.*/)|\1|g'

In place modification of files:

• sd: sd before after file.txt

• sed: sed -i -e 's/before/after/g' file.txt

  With sed, you need to remember to use -e or else some platforms will consider the next argument to be a backup suffix.

Simple replacement on ~1.5 gigabytes of JSON

hyperfine --warmup 3 --export-markdown out.md \
  'sed -E "s/\"/'"'"'/g" *.json > /dev/null' \
  'sed    "s/\"/'"'"'/g" *.json > /dev/null' \
  'sd     "\"" "'"'"'"   *.json > /dev/null'

Result: ~2.35 times faster

Regex replacement on a ~55M json file:

hyperfine --warmup 3 --export-markdown out.md \
  'sed -E "s:(\w+):\1\1:g"    dump.json > /dev/null' \
  'sed    "s:\(\w\+\):\1\1:g" dump.json > /dev/null' \
  'sd     "(\w+)" "$1$1"      dump.json > /dev/null'

Result: ~11.93 times faster

Line-by-line vs across mode (1M lines, ~36MB file):

Line-by-line mode is ~2-3x slower than across mode but still faster than sed for regex replacements. The tradeoff is dramatically lower memory usage:

Install through cargo with cargo install sd, or through various package managers

1. String-literal mode. By default, expressions are treated as regex. Use -F or --fixed-strings to disable regex.

   > echo 'lots((([]))) of special chars' | sd -F '((([])))' ''
   lots of special chars

2. Basic regex use - let's trim some trailing whitespace

   > echo 'lorem ipsum 23   ' | sd '\s+$' ''
   lorem ipsum 23

3. Capture groups

   Indexed capture groups:

   > echo 'cargo +nightly watch' | sd '(\w+)\s+\+(\w+)\s+(\w+)' 'cmd: $1, channel: $2, subcmd: $3'
   cmd: cargo, channel: nightly, subcmd: watch

   Named capture groups:

   > echo "123.45" | sd '(?P<dollars>\d+)\.(?P<cents>\d+)' '$dollars dollars and $cents cents'
   123 dollars and 45 cents

   In the unlikely case you stumble upon ambiguities, resolve them by using ${var} instead of $var. Here's an example:

   > echo '123.45' | sd '(?P<dollars>\d+)\.(?P<cents>\d+)' '$dollars_dollars and $cents_cents'
    and
   
   > echo '123.45' | sd '(?P<dollars>\d+)\.(?P<cents>\d+)' '${dollars}_dollars and ${cents}_cents'
   123_dollars and 45_cents

4. Find & replace in a file

   > sd 'window.fetch' 'fetch' http.js

   That's it. The file is modified in-place.

   To preview changes:

   > sd -p 'window.fetch' 'fetch' http.js

5. Find & replace across project

   This example uses fd.

   Good ol' unix philosophy to the rescue.

   fd --type file --exec sd 'from "react"' 'from "preact"'

   Same, but with backups (consider version control).

   fd --type file --exec cp {} {}.bk \; --exec sd 'from "react"' 'from "preact"'

sd will interpret every argument starting with - as a (potentially unknown) flag. The common convention of using -- to signal the end of flags is respected:

$ echo "./hello foo" | sd "foo" "-w"
error: Found argument '-w' which wasn't expected, or isn't valid in this context

USAGE:
    sd [OPTIONS] <find> <replace-with> [files]...

For more information try --help
$ echo "./hello foo" | sd "foo" -- "-w"
./hello -w
$ echo "./hello --foo" | sd -- "--foo" "-w"
./hello -w

By default, sd processes input line by line. This means:

• Low memory usage (only one line in memory at a time)
• Streaming output for stdin (results appear before EOF)
• ^ and $ match the start/end of each line without phantom matches
• \s+$ trims trailing whitespace without eating newlines

If you need patterns to match across line boundaries (e.g. replacing \n or matching multi-line patterns), use the -A / --across flag:

> echo -e "hello\nworld" | sd -A '\n' ','
hello,world

To escape the $ character, use $$:

❯ echo "foo" | sd 'foo' '$$bar'
$bar