JREP is a general-purpose command line utility that takes the basic concept of GREP and transforms it into an infinitely more versatile tool fit for the modern world.
Basically I couldn't find a good GREP for windows and GREP itself kinda sucks so I did it myself. Excpect dumb bodges and messy code despite my efforts to keep both to a minimum.
JREP is released under the "Don't Be a Dick" public license. Please see dbad-license.org or LICENSE for more details.
All images used for testing have been provided by wikimedia under Crative Commons.
To install via PIP: pip install JREP
To install from GitHub: git clone https://github.com/Scripter17/JREP & pip install ./JREP
Check here for compatibility info
The current output of jrep --help:
For details, check below
usage: jrep [--help [topic]] [--enhanced-engine] [--file FILE [FILE ...]]
[--glob GLOB [GLOB ...]] [--include-dirs]
[--stdin-files | --stdin-globs | --stdin-anti-match-strings]
[--no-duplicates] [--no-name-duplicates]
[--name-regex Regex [Regex ...]]
[--name-anti-regex Regex [Regex ...]]
[--name-ignore-regex Regex [Regex ...]]
[--full-name-regex Regex [Regex ...]]
[--full-name-anti-regex Regex [Regex ...]]
[--full-name-ignore-regex Regex [Regex ...]]
[--name-glob Glob [Glob ...]] [--name-anti-glob Glob [Glob ...]]
[--name-ignore-glob Glob [Glob ...]]
[--full-name-glob Glob [Glob ...]]
[--full-name-anti-glob Glob [Glob ...]]
[--full-name-ignore-glob Glob [Glob ...]]
[--dir-name-regex Regex [Regex ...]]
[--dir-name-anti-regex Regex [Regex ...]]
[--dir-name-ignore-regex Regex [Regex ...]]
[--dir-full-name-regex Regex [Regex ...]]
[--dir-full-name-anti-regex Regex [Regex ...]]
[--dir-full-name-ignore-regex Regex [Regex ...]]
[--file-regex Regex [Regex ...]]
[--file-anti-regex Regex [Regex ...]]
[--file-ignore-regex Regex [Regex ...]]
[--match-regex Regex [Regex ...]]
[--match-anti-regex Regex [Regex ...]]
[--match-ignore-regex Regex [Regex ...]] [--sort SORT]
[--sort-dir SORT_DIR] [--sort-regex Regex [Regex ...]]
[--no-headers] [--print-dir-names] [--print-file-names]
[--print-full-paths] [--print-posix-paths] [--dont-print-matches]
[--print-match-offset] [--print-match-range]
[--replace Regex [Regex ...]] [--sub Regex [Regex ...]]
[--name-sub Regex [Regex ...]] [--dir-name-sub Regex [Regex ...]]
[--escape] [--count COUNT [COUNT ...]] [--limit LIMIT [LIMIT ...]]
[--depth-first] [--glob-root-dir GLOB_ROOT_DIR]
[--print-failed-files] [--json] [--no-warn] [--hard-warn]
[--weave-matches] [--strict-weave] [--no-exec]
[--pre-match-exec cmd] [--match-exec cmd]
[--if-match-exec-before cmd] [--if-match-exec-after cmd]
[--if-no-match-exec-after cmd] [--pre-file-exec cmd]
[--file-exec cmd] [--if-file-exec-before cmd]
[--if-file-exec-after cmd] [--if-no-file-exec-after cmd]
[--pre-dir-exec cmd] [--dir-exec cmd] [--if-dir-exec-before cmd]
[--if-dir-exec-after cmd] [--if-no-dir-exec-after cmd]
[--order ORDER [ORDER ...]] [--no-flush] [--force-flush]
[--verbose]
[Regex ...]
options:
--help [topic], -h [topic] show this help message and exit OR use
`--help [topic]` for help with [topic]
Global behaviour:
--enhanced-engine, -E Use alternate regex engine from
https://pypi.org/project/regex/
Files and regexes:
Regex Regex(es) to process matches for
(reffered to as "get regexes")
--file FILE [FILE ...], -f FILE [FILE ...]
A list of files to check
--glob GLOB [GLOB ...], -g GLOB [GLOB ...]
A list of globs to check
--include-dirs Process directories as files
--stdin-files, -F Treat STDIN as a list of files
--stdin-globs, -G Treat STDIN as a list of globs
--stdin-anti-match-strings Treat STDIN as a list of strings to
not match
Filters:
--no-duplicates, -D Don't print duplicate matches (See
also: --order)
--no-name-duplicates Don't process files whose names have
already been processed (takes --name-
sub, --print-full-paths and --print-
posix-paths)
--name-regex Regex [Regex ...], -t Regex [Regex ...]
If a file name matches all supplied
regexes, keep going. Otherwise
continue
--name-anti-regex Regex [Regex ...], -T Regex [Regex ...]
Like --name-regex but excludes file
names that match any of the supplied
regexes
--name-ignore-regex Regex [Regex ...]
Like --name-anti-regex but doesn't
contribute to --count *-failed-files
--full-name-regex Regex [Regex ...] Like --name-regex but for absolute
file paths (C:/xyz instead of xyz)
--full-name-anti-regex Regex [Regex ...]
Like --name-anti-regex but applied to
full file paths
--full-name-ignore-regex Regex [Regex ...]
Like --full-name-anti-regex but
doesn't contribute to --count
*-failed-files
--name-glob Glob [Glob ...] If a file name matches all supplied
globs, keep going. Otherwise continue
--name-anti-glob Glob [Glob ...] Like --name-glob but excludes file
names that match any of the supplied
globs
--name-ignore-glob Glob [Glob ...] Like --name-anti-glob but doesn't
contribute to --count *-failed-files
--full-name-glob Glob [Glob ...] Like --name-glob but for absolute file
paths (C:/xyz instead of xyz)
--full-name-anti-glob Glob [Glob ...]
Like --name-anti-glob but applied to
full file paths
--full-name-ignore-glob Glob [Glob ...]
Like --full-name-anti-glob but doesn't
contribute to --count *-failed-files
--dir-name-regex Regex [Regex ...] If a directory name matches all
supplied regexes, enter it. Otherwise
continue
--dir-name-anti-regex Regex [Regex ...]
Like --dir-name-regex but excludes
directories that match any of the
supplied regexes
--dir-name-ignore-regex Regex [Regex ...]
Like --dir-name-anti-regex but doesn't
contribute to --count total-failed-
dirs
--dir-full-name-regex Regex [Regex ...]
Like --dir-name-regex but for absolute
directory paths (C:/xyz instead of
xyz)
--dir-full-name-anti-regex Regex [Regex ...]
Like --dir-name-anti-regex but applied
to full directory paths
--dir-full-name-ignore-regex Regex [Regex ...]
Like --dir-full-name-anti-regex but
doesn't contribute to --count total-
failed-dirs
--file-regex Regex [Regex ...] Regexes to test file contents for
--file-anti-regex Regex [Regex ...] Like --file-regex but excludes files
that match of the supplied regexes
--file-ignore-regex Regex [Regex ...]
Like --file-anti-regex but doesn't
contribute to --count *-failed-files
--match-regex Regex [Regex ...] Groups are split along lone *. Matches
from the Nth get regex are tested with
the Nth group
--match-anti-regex Regex [Regex ...] Like --match-regex but excludes
matches that match any of the supplied
regexes
--match-ignore-regex Regex [Regex ...]
Like --match-anti-regex but doesn't
contribute to --count *-failed-matches
Sorting:
--sort SORT, -S SORT Sort files by ctime, mtime, atime,
name, or size. Prefix key with "r" to
reverse. A windows-esque "blockwise"
sort is also available. Run jrep
--help blockwise for more info
--sort-dir SORT_DIR --sort on a per-directory basis
--sort-regex Regex [Regex ...] Regexes to apply to file names keys
(like --replace) for purposes of
sorting (EXPERIMENTAL)
Output:
--no-headers, -H Don't print match: or file: before
lines
--print-dir-names, -d Print names of explored directories
--print-file-names, -n Print file names as well as matches
--print-full-paths, -p Print full file paths
--print-posix-paths, -P Replace \ with / when printing file
paths
--dont-print-matches, -N Don't print matches (use with --print-
file-names to only print names)
--print-match-offset, -o Print where the match starts in the
file as a hexadecimal number (ignores
-H)
--print-match-range, -O Print where the match starts and ends
in the file as a hexadecimal number
(implies -o)
Replace/Sub:
--replace Regex [Regex ...], -r Regex [Regex ...]
Regex replacement
--sub Regex [Regex ...], -R Regex [Regex ...]
re.sub argument pairs after --replace
is applied. Run jrep.py --help sub for
more info
--name-sub Regex [Regex ...] Applies --sub to file names. A lone *
separates subsitutions for y/z and
C:/x/y/z
--dir-name-sub Regex [Regex ...] --name-sub but for directory names
--escape, -e Escape back slashes, newlines,
carriage returns, and non-printable
characters
Misc.:
--count COUNT [COUNT ...], -c COUNT [COUNT ...]
Count match/file/dir per file, dir,
and/or total (Ex: --count fm dir-
files)
--limit LIMIT [LIMIT ...], -l LIMIT [LIMIT ...]
Limit match/file/dir per file, dir,
and/or total (Ex: --limit filematch=1
total_dirs=5)
--depth-first Enter subdirectories before processing
files
--glob-root-dir GLOB_ROOT_DIR Root dir to run globs in (JANK)
--print-failed-files Print file names even if they fail
(Partially broken)
--json, -j Print output as JSON
--no-warn Don't print warning messages
--hard-warn Throw errors instead of warnings
--weave-matches, -w Weave regex matchdes (print first
results for each get regex, then
second results, etc.)
--strict-weave, -W Only print full weave sets
Exec:
--no-exec Don't run any exec functions. Useful
if using user input (STILL NOT SAFE)
--pre-match-exec cmd Command to run before printing each
match
--match-exec cmd Command to run after printing each
match
--if-match-exec-before cmd Command to run as soon as least one
match passes
--if-match-exec-after cmd Command to run at the end if at least
one match passed
--if-no-match-exec-after cmd Command to run at the end if at no
matches passed
--pre-file-exec cmd Command to run before printing each
file name
--file-exec cmd Command to run after printing each
file name
--if-file-exec-before cmd Command to run as soon as least one
file passes
--if-file-exec-after cmd Command to run at the end if at least
one file passed
--if-no-file-exec-after cmd Command to run at the end if at no
files passed
--pre-dir-exec cmd Command to run before printing each
dir name
--dir-exec cmd Command to run after printing each dir
name
--if-dir-exec-before cmd Command to run as soon as least one
dir passes
--if-dir-exec-after cmd Command to run at the end if at least
one dir passed
--if-no-dir-exec-after cmd Command to run at the end if at no
dirs passed
Debugging/Advanced:
--order ORDER [ORDER ...] The order in which modifications to
matches are applied. Run jrep --help
order for more info
--no-flush Improves speed by disabling manually
flushing the stdout buffer (ideal for
chaining commands)
--force-flush Always flush STDOUT (slow)
--verbose, -v Verbose info
The following have extended help that can be seen with --help [topic]: sub, blockwise, order, exec
These can be accessed by doing jrep --help [topic] where [topic] is the part in parenthesis
The easiest way to explain advanced uses of --sub is to give an example. So take --sub a ? b ? c d e f + x ? y z * ? t ? e d * abc xyz as an example.
What it means is the following
a ? b ? c d e f: If a match from get regex 0 matchesaand notb, replacecwithdandewithf+: New conditions but stay on the same get regexx ? y z: If a match from get regex 0 matchesx, replaceywithz*: Move on to the next get regex? t ? e d: If a match from get regex 1 does't matcht, replaceewithd*: Move on to the next get regexabc xyz: Replaceabcwithxyzwithout any conditions
Obviously 99% of use cases don't need conditionals at all so just doing --sub abc def * uvw xyz is sufficient
A generic sort function will think "file10.jpg" comes before "file2.jpg"
Windows, on the other hand, has code that treats the number part as a number
Blockwise sort mimics this behaviour by
- Splitting filenames into groups of number and non-number characters. Ex.
abc123def456.jpg->["abc", "123", "def", "456", ".jpg"] - When comparing 2 filenames, compare the first element ("block") of both name's lists according to the following two rules
- If either block is made of non-number characters, compare the two blocks as strings
- If both blocks are numbers, compare them as numbers
The end result is that file2.jpg is correctly placed before file10.jpg
--order determines the order of functions that process matches
- The default value for
--orderis replace, match-whole-lines, sub, stdin-anti-match-strings, match-regex, no-name-duplicates, no-duplicates, print-dir-name, print-name, print-match - Changing the order of
sub,replace, andmatch-whole-lineswill mostly "work" but the output will make next to no sense - The main purpose of this is to move
match-regexandno-duplicatesto earlier in the chain
Usage looks like --exec "echo {}" or just --exec "echo"
To use the filename/dir multiple times use {0} (this is for future proofing)
--match-exec/--exec: after printing matches
--pre-match-exec: before printing matches
--match-exec: after printing file names
--pre-match-exec: before printing file names
--if-file-exec-before: Run once before the first file is processed
--if-file-exec-after: Run once after the last file is processed
--if-no-file-exec-after: Run at the end if no file is ever processed
--dir-exec: after printing directory names
--pre-dir-exec: before printing directory names
--if-dir-exec-before: Run once before the first dir is processed
--if-dir-exec-after: Run once after the last dir is processed
--if-no-dir-exec-after: Run at the end if no dir is ever processed
Currently due to technical limitations I can only do automatic testing for Ubuntu 20.04, Windows Server 2022, and macOS Big Sur 11
The end goal is for JREP to 100% work on
- Windows 7 through 11
- The past decade of every major Linux distro
- The past decade of Mac machines
I don't think there's much (if any) platform-specific jank in JREP but it'll take a while to confirm that