nimf

find-like, search for files in a directory hierarchy.

Features:

• Multithreaded and efficient (faster than find and fd)
• Mostly intuitive, and simple interface

Usage

Usage:
  f [optional-params] [input: string...]

Arguments:
  Entered `input` may be a pattern OR a path to search.

  The pattern will only match with the filename unless you include a "/".
  "/" and "$" also match the start and end of a path.

Options:
  -h, --help                               print this cligen-erated help
  --help-syntax                            advanced: prepend,plurals,..
  --version              bool       false  print version
  -a, --all              bool       false  Search all directories, including those ignored by default/your ".config/nimf/ignore.csv" file.
  -x=, --exclude=        strings    {}     Add patterns to ignore.
  -t=, --types=          filetypes  any    Select which file type(s) to match. File type may be any|{l}file|{l}directory|{f/d}link or a file extension.
  -e=, --execute=        strings    {}     Execute a command for each matching search result in parallel.
                                           Alternatively, end this argument with "+" to execute the command once with all results as arguments.
                                           Example: f .nim -e "$EDITOR"+
                                           The following placeholders are substituted before the command is executed:
                                           "{}":   path (of the current search result)
                                           "{/}":  basename
                                           "{//}": parent directory
                                           "{.}":  path without file extension
                                           "{/.}": basename without file extension
                                           Example: f .jpg$ -e 'convert {} {.}.png'
                                           If no placeholder is present, an implicit " {}" at the end is assumed.
  -d=, --max-depth=      int        0      Set a maximum of how deep in the directory tree to search.
  -l=, --limit=          int/ms/s   0      Limit the search to a number of results, alternatively end this argument with "ms" or "s" to limit the search based on time.
  -L, --follow-symlinks  bool       false  Enable traversing symlinks.
  -0, --null             bool       false  Separate search results (stdout) and split stdin with null characters "\0" instead of newlines "\n".
  -c, --color            flag       auto   Enable or disable colored printing. Default is based on the "NO_COLOR" environment variable.
  --hyperlink            flag       false  Enable clickable hyperlinks in supported terminals.

Benchmarks

375,000 file directory:

	No pattern	231,970 matches
f -ac=never	0.95s	0.55s
fd -uc=never	1.10s	0.73s
f -a	1.17s	0.84s
find	1.23s	0.96s
fd -u	1.29s	0.96s

1.75m file directory:

	No pattern	1,028,205 matches	14,075 matches
f -a	6.91s	4.16s	0.38s
find	7.30s	4.83s	1.25s
fd -u	9.62s	4.92s	0.96s

4m file directory (my root dir):

	No pattern	2,241,660 matches	3 matches
f -a	13.59s	9.54s	0.65s
find	15.32s	10.88s	2.68s
fd1	25.16s	11.32s	N/A

---

Footnotes

1. These measurements may be inaccurate, fd would not stop searching my drive (and pinning my CPU) so I had to watch for when it stopped printing results and ^C.
   As the 3 matches search finished printing its results much earlier than it finished searching, I could not measure it. But for brevity, fd (when lucky) finished printing the 3 results in around 1.5s. With this same measuring, f finishes in around 0.35s and find around 0.7s. ↩