cloc(1)

📦
version 2.10 · 14 versions
, in fedora 43, fedora 44

Okay so bestie, you found the cloc(1) page and I'm about to make it make sense — this is the User Commands manual for cloc, and it's lowkey iconic. Grab a strawberry 🍓 and let's go.

✨ NAME ✨

cloc — Count, or compute differences of, lines of source code and comments.

🦋 SYNOPSIS 🦋

cloc [options] <FILE|DIR> ...

💖 DESCRIPTION 💖

It counts — or computes the differences of — physical lines of source code in the files you hand it (and yeah, those files can be archives like compressed tarballs or zip files, or even git commit hashes or branch names 👑) and/or recursively below the directories you give it.

And the vibe? It's written entirely in Perl, using only modules from the standard distribution. No cap, that's the whole dependency story. 💅

🌸 OPTIONS 🌸

🎀 Input Options 🎀

To count standard input, use the special filename - and either --stdin-name=FILE to tell cloc the name of the file being piped in, or --force-lang=LANG to apply the LANG counter to all input. Fr fr, that's your stdin situation.

  • --extract-with=CMD — You only need this one if cloc can't figure out by itself how to extract the contents of the input file(s). Use CMD to extract binary archive files (e.g.: .tar.gz, .zip, .Z). Use the literal >FILE< as a stand-in for the actual file(s) to be extracted.

For example, to count lines of code in the input files gcc-4.2.tar.gz perl-5.8.8.tar.gz on Unix use:

--extract-with='gzip -dc >FILE< | tar xf -

or, if you have GNU tar:

--extract-with='tar zxf >FILE<'

and on Windows, use, for example:

--extract-with="\"c:\Program Files\WinZip\WinZip32.exe\" -e -o >FILE<
  • --list-file=FILE — Takes the list of file and/or directory names to process from FILE, which has one file/directory name per line. Only exact matches are counted; relative path names will be resolved starting from the directory where cloc is invoked. Set FILE to - to read file names from a STDIN pipe. See also --exclude-list-file, --config.
  • --diff-list-file=FILE — Takes the pairs of file names to be diff'ed from FILE, whose format matches the output of --diff-alignment. (Run with that option to see a sample.) The language identifier at the end of each line is ignored. This enables --diff mode and by-passes file pair alignment logic. See also --config.
  • --vcs=VCS — Invokes a system call to VCS to obtain a list of files to work on. If VCS is 'git', then it will invoke 'git ls-files' to get a file list and 'git submodule status' to get a list of submodules whose contents will be ignored. See also --git which accepts git commit hashes and branch names. If VCS is 'svn' then it will invoke 'svn list -R'. The primary benefit is that cloc will then skip files explicitly excluded by the versioning tool in question, ie, those in .gitignore or have the svn:ignore property. Alternatively VCS may be any system command that generates a list of files.
Note: cloc must be in a directory which can read the files as they are returned by VCS. cloc will not download files from remote repositories. 'svn list -R' may refer to a remote repository to obtain file names (and therefore may require authentication to the remote repository), but the files themselves must be local.

Setting VCS to 'auto' selects between 'git' and 'svn' (or neither) depending on the presence of a .git or .svn subdirectory below the directory where cloc is invoked.

  • --unicode — Checks binary files to see if they contain Unicode expanded ASCII text. Heads up bestie, this causes performance to drop noticeably. 😭

💅 Processing Options 💅

  • --autoconf — Counts .in files (as processed by GNU autoconf) of recognized languages. See also --no-autogen.
  • --by-file — Reports results for every source file encountered.
  • --by-file-by-lang — Reports results for every source file encountered in addition to reporting by language.
  • --config FILE — Reads command line switches from FILE instead of the default location of ~/.config/cloc/options.txt. The file should contain one switch, along with arguments (if any), per line. Blank lines and lines beginning with '#' are skipped. Options given on the command line take priority over entries read from the file. If a directory is also given with any of these switches: --list-file, --exclude-list-file, --read-lang-def, --force-lang-def, --diff-list-file and a config file exists in that directory, it will take priority over ~/.config/cloc/options.txt.
  • --count-and-diff SET1 SET2 — First performs direct code counts of source file(s) of SET1 and SET2 separately, then performs a diff of these. Inputs may be pairs of files, directories, or archives. If --out or --report-file is given, three output files will be created, one for each of the two counts and one for the diff. See also --diff, --diff-alignment, --diff-timeout, --ignore-case, --ignore-whitespace.
  • --diff SET1 SET2 — Computes differences in code and comments between source file(s) of SET1 and SET2. The inputs may be pairs of files, directories, or archives. Use --diff-alignment to generate a list showing which file pairs where compared. See also --count-and-diff, --diff-alignment, --diff-timeout, --ignore-case, --ignore-whitespace.
  • --diff-timeout N — Ignores files which take more than N seconds to process. Default is 10 seconds. Setting N to 0 allows unlimited time. (Large files with many repeated lines can cause Algorithm::Diff::sdiff() to take hours. 💀)
  • --docstring-as-code — cloc considers docstrings to be comments, but this is not always correct as docstrings represent regular strings when they appear on the right hand side of an assignment or as function arguments. This switch forces docstrings to be counted as code.
  • --follow-links — [Unix only] Follow symbolic links to directories (sym links to files are always followed).
  • --force-lang=LANG[,EXT] — Processes all files that have a EXT extension with the counter for language LANG. For example, to count all .f files with the Fortran 90 counter (which expects files to end with .f90) instead of the default Fortran 77 counter, use:
--force-lang="Fortran 90",f

If EXT is omitted, every file will be counted with the LANG counter. This option can be specified multiple times (but that is only useful when EXT is given each time). See also --script-lang, --lang-no-ext.

  • --force-lang-def=FILE — Loads language processing filters from FILE, then uses these filters instead of the built-in filters.
Note: languages which map to the same file extension (for example: MATLAB/Objective-C/MUMPS; Pascal/PHP; Lisp/OpenCL; Lisp/Julia; Perl/Prolog) will be ignored as these require additional processing that is not expressed in language definition files.

Use --read-lang-def to define new language filters without replacing built-in filters (see also --write-lang-def, --write-lang-def-incl-dup).

  • --git — Forces the inputs to be interpreted as git targets (commit hashes, branch names, et cetera) if these are not first identified as file or directory names. This option overrides the --vcs=git logic if this is given; in other words, --git gets its list of files to work on directly from git using the hash or branch name rather than from 'git ls-files'. This option can be used with --diff to perform line count diffs between git commits, or between a git commit and a file, directory, or archive. Use -v/--verbose to see the git system commands cloc issues.
  • --git-diff-rel — Same as --git --diff, or just --diff if the inputs are recognized as git targets. Only files which have changed in either commit are compared.
  • --git-diff-all — Git diff strategy #2: compare all files in the repository between the two commits.
  • --ignore-whitespace — Ignores horizontal white space when comparing files with --diff. See also --ignore-case.
  • --ignore-case — Ignores changes in case within file contents; consider upper- and lowercase letters equivalent when comparing files with --diff. See also --ignore-whitespace.
  • --ignore-case-ext — Ignores case of file name extensions.
Heads up: this will cause problems counting some languages (specifically, .c and .C are associated with C and C++; this switch would count .C files as C rather than C++ on *nix operating systems). File name case insensitivity is always true on Windows.
  • --ignore-regex — Ignores lines in source files that match the given Perl regular expression for the given language(s). This option can be specified multiple times. Language names are comma separated and are followed by the pipe character and the regular expression. Use * to match all languages. Examples:
--ignore-regex='C,Java,C++|^\s*[{};]\s*$'
--ignore-regex='*|DEBUG|TEST\s+ONLY'

These filters are applied after comments are removed. Use --strip-comments=EXT to create new files that show these filters applied. The primary use case is to ignore lines containing only braces, brackets, or puctuation.

  • --include-submodules — When using --vcs=git, include files in git submodules.
  • --lang-no-ext=LANG — Counts files without extensions using the LANG counter. This option overrides internal logic for files without extensions (where such files are checked against known scripting languages by examining the first line for "#!"). See also --force-lang, --script-lang.
  • --max-file-size=MB — Skips files larger than "MB" megabytes when traversing directories. By default, "MB"=100. cloc's memory requirement is roughly twenty times larger than the largest file so running with files larger than 100 MB on a computer with less than 2 GB of memory will cause problems.
Note: this check does not apply to files explicitly passed as command line arguments.
  • --no-autogen[=list] — Ignores files generated by code-production systems such as GNU autoconf. To see a list of these files (then exit), run with --no-autogen list See also --autoconf.
  • --no-recurse — Counts files in the given directories without recursively descending below them.
  • --only-count-files — Only counts files by language. Blank, comment, and code counts will be zero.
  • --original-dir — Only effective in combination with --strip-comments. Writes the stripped files to the same directory as the original files.
  • --read-binary-files — Processes binary files in addition to text files.
This is usually a bad idea and should only be attempted with text files that have embedded binary data. 🙈

okay that's the first chunk, keep scrolling bestie, there's more 👑

  • --read-lang-def=FILE — pulls brand new language processing filters out of FILE and merges them with the ones cloc already knows. If FILE defines a language cloc already knows, cloc's own definition wins, periodt. Wanna override cloc's takes? That's --force-lang-def's whole job. (peep --write-lang-def too 💅)
  • --script-lang=LANG,S — counts every file that invokes S as a #! scripting language using the counter for language LANG. So like, files starting with #!/usr/local/bin/perl5.8.8 get counted with the Perl counter when you do this:
        --script-lang=Perl,perl5.8.8

The language name is case insensitive but the script language executable name S must have the right case — no cap, it's picky like that 😭. This option can be specified multiple times. See also --force-lang.

  • --sdir=DIR — use DIR as the scratch directory instead of letting File::Temp pick where to go. Heads up bestie: files written here are not removed at the end of the run (unlike File::Temp's whole thing). 🙈
  • --skip-leading=N[,ext] — skips the first N lines of each file. And if you also hand it a comma separated list of extensions, it only skips lines from those file types. Example:
        --skip-leading=10,cpp,h

will skip the first ten lines of *.cpp and *.h files. Obsessed, honestly — perfect for ignoring boilerplate text 🎀

  • --skip-uniqueness — skips the file uniqueness check. Glow-up in performance ✨ but at the expense of counting files with identical contents multiple times (if such duplicates exist). Choose your fighter.
  • --stat — okay so some file systems (AFS, CD-ROM, FAT, HPFS, SMB) don't have directory 'nlink' counts that match the number of their subdirectories, which means cloc may undercount or completely skip the contents of those file systems 💀. This switch forces File::Find to stat directories to get the correct count. Downside: file search speed will decrease. See also --follow-links.
  • --stdin-name=FILE — counts lines streamed via STDIN as if they came from a file named FILE.
  • --strip-code=EXT — for each file processed, writes to the current directory a version of the file with blank and code lines removed. The name of each stripped file is the original file name with .EXT appended to it. It's written to the current directory unless --original-dir is on.
  • --strip-comments=EXT — for each file processed, writes to the current directory a version with blank and commented lines removed (in-line comments persist, just so you know). Each stripped file gets .EXT appended to its original name. Same deal: written to the current directory unless --original-dir is on.
  • --strip-str-comments — replaces comment markers embedded in strings with 'xx'. This attempts to work around a limitation in Regexp::Common::Comment where comment markers embedded in strings get seen as actual comment markers and not strings, often resulting in a 'Complex regular subexpression recursion limit' warning and incorrect counts 😭. Two disadvantages to using this switch, fr fr: 1/ code count performance drops, and 2/ code generated with --strip-comments will contain different strings where ever embedded comments are found.
  • --sum-reports — the input arguments are report files previously created with the --report-file option. Makes a cumulative set of results containing the sum of data from the individual report files. Math but make it cute 🧮✨
  • --timeout=N — ignores files that take more than N seconds to process at any of the language's filter stages. The default maximum number of seconds spent on a filter stage is the number of lines in the file divided by one thousand. Setting N to 0 allows unlimited time. See also --diff-timeout.
  • --processes=NUM — [Available only on systems with a recent version of the Parallel::ForkManager module. Not available on Windows.] Sets the maximum number of cores cloc uses. The default value of 0 disables multiprocessing. Multi-core era, slay 🔥
  • --unix — over-rides the operating system detection logic and runs in UNIX mode. See also --windows, --show-os.
  • --use-sloccount — if SLOCCount is installed, use its compiled executables c_count, java_count, pascal_count, php_count, and xml_count instead of cloc's counters. SLOCCount's compiled counters are substantially faster than cloc's and may give a performance improvement when counting projects with large files. BUT ⚠️ these cloc-specific features will not be available: --diff, --count-and-diff, --strip-comments, --unicode.
  • --windows — over-rides the operating system detection logic and runs in Microsoft Windows mode. See also --unix, --show-os.

🦋 Filter Options 🦋

  • --include-content=REGEX — only counts files containing text that matches the given regular expression.
  • --exclude-content=REGEX — excludes files containing text that matches the given regular expression.
  • --exclude-dir=DIR1[,DIR2 ...] — excludes the given comma separated directories from being scanned. Like:
        --exclude-dir=.cache,test

will skip all files that match "/.cache/" or "/test/" as part of their path. Directories named .bzr, .cvs, .hg, .git, and .svn are always excluded. This option only works with individual directory names so including file path separators is not allowed 🙅‍♀️. Use --fullpath and --not-match-d=REGEX to supply a regex matching multiple subdirectories.

  • --exclude-ext=EXT1[,EXT2 ...] — do not count files having the given file name extensions.
  • --exclude-lang=L1[,L2[...]] — excludes the given comma separated languages from being counted.
  • --exclude-list-file=FILE — ignores files and/or directories whose names appear in FILE. FILE should have one file name per line. Only exact matches are ignored; relative path names will be resolved starting from the directory where cloc is invoked. See also --list-file, --config.
  • --fullpath — modifies the behavior of --match-f or --not-match-f to include the file's path in the regex, not just the file's basename. (This does not expand each file to include its absolute path — instead it uses as much of the path as is passed in to cloc.)
  • --include-ext=<ext1[,ext2[...]]> — counts only languages having the given comma separated file extensions. Use --show-ext to see the recognized extensions.
  • --include-lang=L1[,L2 ...] — counts only the given comma separated, case-insensitive languages L1, L2, L3, et cetera.
  • --match-d=REGEX — only counts files in directories matching the Perl regex. For example
     --match-d='/(src|include)/'

only counts files in directory paths containing "/src/" or "/include/".

  • --not-match-d=REGEX — counts all files except in directories matching the Perl regex. Only the trailing directory name is compared, for example, when counting in "/usr/local/lib", only "lib" is compared to the regex. Add --fullpath to compare parent directories to the regex. Do not include file path separators at the beginning or end of the regex. This switch may be repeated.
  • --match-f=REGEX — only counts files whose basenames match the Perl regex. For example this only counts files that start with Widget or widget:
     --match-f='^[Ww]idget'

Add --fullpath to include parent directories in the regex instead of just the basename.

  • --not-match-f=REGEX — counts all files except those whose basenames match the Perl regex. Add --fullpath to include parent directories in the regex instead of just the basename. This switch may be repeated.
  • --skip-archive=REGEX — ignores files that end with the given Perl regular expression. So if you give it
        --skip-archive='(zip|tar(\.(gz|Z|bz2|xz|7z))?)'

the code will skip files that end with .zip, .tar, .tar.gz, .tar.Z, .tar.bz2, .tar.xz, and .tar.7z. Bye 👋

  • --skip-win-hidden — on Windows, ignore hidden files. 🫥

💜 Debug Options 💜

  • --categorized=FILE — saves file sizes in bytes, identified languages and names of categorized files to FILE.
  • --counted=FILE — saves names of processed source files to FILE. See also --found, --ignored, --unique.
  • --diff-alignment=FILE — writes to FILE a list of files and file pairs showing which files were added, removed, and/or compared during a run with --diff. This switch forces the --diff mode on.
  • --explain=LANG — prints the filters used to remove comments for language LANG and exits. In some cases the filters refer to Perl subroutines rather than regular expressions, so an examination of the source code may be needed for further explanation. iykyk 🔍
  • --help — prints cloc's internal usage information and exits.
  • --found=FILE — saves names of every file found to FILE. See also --counted, --ignored, --unique.
  • --ignored=FILE — saves names of ignored files and the reason they were ignored to FILE. See also --counted, --found, --unique.
  • --print-filter-stages — prints to STDOUT processed source code before and after each filter is applied.
  • --show-ext[=EXT] — prints information about all known (or just the given) file extensions and exits.
  • --show-errors[=EXT] — used with --quiet, prints error messages (if any).
  • --show-lang[=LANG] — prints information about all known (or just the given) languages and exits.
  • --show-os — prints the value of the operating system mode and exits. See also --unix, --windows.
  • --quiet — suppresses all information messages except the final output. See also --shwow-errors. 🤫
  • --unique=FILE — saves names of ignored files and the reason they were ignored to FILE. See also --counted, --found, --ignored.
  • -v[=N] — turns on verbose with optional numeric value.
  • --verbose[=N] — long form of -v.
  • --version — prints the version of this program and exits.
  • --write-lang-def=FILE — writes to FILE the language processing filters then exits. Useful as a first step to creating custom language definitions. Note: languages which map to the same file extension will be excluded. See also --force-lang-def, --read-lang-def.
  • --write-lang-def-incl-dup=FILE — same as --write-lang-def, but includes duplicated extensions. This generates a problematic language definition file because cloc will refuse to use it until duplicates are removed 😬

✨ Output Options ✨

  • --3 — print third-generation language output. heads up bestie 🫶 this one can make report summation fail if some of your reports were made with this option and others were made without it. So like... be consistent, ok?
  • --by-percent X — instead of showing comment and blank line counts as plain counts, show them as percentages based on whatever X is in the denominator. And X is one of:
    • c — meaning lines of code
    • cm — meaning lines of code + comments
    • cb — meaning lines of code + blanks
    • cmb — meaning lines of code + comments + blanks

So like, say you use method c and your code has twice as many lines of comments as lines of code? Then the comment column value will be 200%. Wild, right? The code column stays a plain line count tho. 💅

  • --csv — write the results as comma separated values. iconic spreadsheet energy 📊
  • --csv-delimiter=C — use the character C as the delimiter for comma separated files instead of ,. This switch forces --csv to be on. no take-backsies ✨
  • --file-encoding=E — write output files using the E encoding instead of the default ASCII (E = 'UTF-7'). Examples: 'UTF-16', 'euc-kr', 'iso-8859-16'. Wanna see the known encodings? Run:
perl -MEncode -e 'print join("\n", Encode->encodings(":all")), "\n"'
  • --fmt=N — alternate text output format where N is a number from 1 to 5, or -1 to -5. 'total lines' means the sum of code, comment, and blank lines. Negative values are the same as the positive values but they retain — instead of deleting — the intermediate JSON file that gets written. The JSON file name is randomly generated unless --out/--report-file is given. The formats are:
1:  by language (same as cloc default output)
2:  by language with an extra column for total lines
3:  by file with language
4:  by file with a total lines column
5:  by file with language and a total lines column
  • --hide-rate — don't show line and file processing rates in the output header. This makes output deterministic. same input, same output, every single time, periodt. ⭐
  • --json — write the results in JavaScript Object Notation (JSON). for the girlies who pipe things into other things 🦋
  • --md — write the results as Markdown-formatted text.
  • --out=FILE — synonym for --report-file=FILE. two names, one job 💜
  • --progress-rate=N — show a progress update after every N files are processed (default N=100). Set N to 0 to suppress progress output — super useful when you're redirecting output to STDOUT. 🔥
  • --quiet — suppress all information messages except for the final report. shy girl mode 🙈
  • --report-file=FILE — write the results to FILE instead of standard output.
  • --summary-cutoff=X:N — aggregate to 'Other' the results having X lines below N, where X is one of:
    • c — meaning lines of code
    • f — meaning files
    • m — meaning lines of comments
    • cm — meaning lines of code + comments

Appending a percent sign to N changes the calculation from a straight count to a percentage. Ignored with --diff or --by-file. (iykyk)

  • --sql=FILE — write results as SQL CREATE and INSERT statements which can be read by a database program such as SQLite. If FILE is -, output is sent to STDOUT.
  • --sql-append — append SQL insert statements to the file specified by --sql and do not generate the table creation option.
  • --sql-project=NAME — use name as the project identifier for the current run. Only valid with the --sql option. 👑
  • --sql-style=STYLE — write SQL statements in the given style instead of the default SQLite format. Styles include Oracle and Named_Columns.
  • --sum-one — for plain text reports, show the SUM: output line even if only one input file is processed. one file still deserves its moment 🫶
  • --xml — write the results in XML.
  • --xsl[=FILE] — reference FILE as an XSL stylesheet within the XML output. If FILE is not given, writes a default stylesheet, cloc.xsl. This switch forces --xml to be on. ✨
  • --yaml — write the results in YAML.

🌈 EXAMPLES 🌈

Count the lines of code in the Perl 5.10.0 compressed tar file on a UNIX-like operating system:

cloc perl-5.10.0.tar.gz

Count the changes in files, code, and comments between Python releases 2.6.6 and 2.7:

cloc --diff Python-2.6.6.tar.bz  Python-2.7.tar.bz2

Wanna see how cloc aligns files for comparison between two code bases? Use the --diff-alignment=FILE option. Here the alignment information is written to "align.txt":

cloc --diff-alignment=align.txt gcc-4.4.0.tar.bz2  gcc-4.5.0.tar.bz2

Count file, code, and comment changes between two git commits:

cloc --git --diff b409850824 HEAD

Print the recognized languages:

cloc --show-lang

Remove comments from "foo.c" and save the result in "foo.c.nc" ("nc" is an arbitrary extension; used here to denote "no comments"):

cloc --strip-comments=nc foo.c

Additional examples can be found at <https://github.com/AlDanial/cloc>. 🦋

💖 ENVIRONMENT 💖

None. literally nothing here bestie, move along 💅

🎀 FILES 🎀

None. same vibe, empty page, no notes ✨

⭐ SEE ALSO ⭐

sloccount(1) — go say hi to the cousin 👑

👑 AUTHORS 👑

The cloc program was written by Al Danial <al.danial@gmail.com> and is Copyright (C) 2006-2026 <al.danial@gmail.com>.

The manual page was originally written by Jari Aalto <jari.aalto@cante.net>.

Both the code and documentation is released under the GNU GPL version 2 or (at your option) any later version. For more information about license, visit <http://www.gnu.org/copyleft/gpl.html>. 🫶

okay that's the whole page bestie, you ate that 💅 this manual is from cloc version dated 2026-07-20 (that's the source/version and date the page footer carries), filed under section cloc(1) — go count some lines, you absolute icon 🌸💖✨

📦 where this page comes from