cljfmt

cljfmt is a tool for formatting Clojure code.

It can turn something like this:

( let [x 3
    y 4]
  (+ (* x x
  )(* y y)
  ))

Into nicely formatted Clojure code like this:

(let [x 3
      y 4]
  (+ (* x x) (* y y)))

Installation

Releases are published on the GitHub project. The native binaries are self-contained, so to install them simply place them on your path.

Usage

The cljfmt tool supports several different commands for checking source files.

Check and Fix

To check the formatting of your source files, use:

cljfmt check

If the formatting of any source file is incorrect, a diff will be supplied showing the problem, and what cljfmt thinks it should be.

If you want to check only a specific file, or several specific files, you can do that, too:

cljfmt check src/foo/core.clj

Once you've identified formatting issues, you can choose to ignore them, fix them manually, or let cljfmt fix them with:

cljfmt fix

As with the check task, you can choose to fix a specific file:

cljfmt fix src/foo/core.clj

Debugging

For inspecting what cljfmt is doing, one tool is to specify the --verbose flag, which will cause additional debugging output to be printed. There are also a few extra commands which can help understand what's happening.

The find command will print what files would be checked by cljfmt. It will print each file path to standard output on a new line:

cljfmt find [path...]

The config command will show what configuration settings cljfmt would use to process the specified files or files in the current directory:

cljfmt config [path]

Finally, version will show what version of the tool you're using:

cljfmt version

Configuration

The cljfmt tool comes with a sensible set of default configuration built-in, and reads additional configuration from .cljfmt files which may be placed in any directories to control cljfmt's behavior on files in those subtrees. These files are regular Clojure files which should contain a map of settings to use:

;; cljfmt configuration
{:max-blank-lines 3
 :file-ignore #{"checkouts" "target"}}

When cljfmt is run, it searches upwards in the filesystem to find parent configuration, and as it searches in directories it will merge in local config files. For example, in a tree like the following:

a
├── .cljfmt
└── b
    ├── c
    │   ├── .cljfmt
    │   └── foo.clj
    └── d
        ├── .cljfmt
        └── e
            └── bar.clj

Running cljfmt in directory c would use a/.cljfmt as the base configuration and would combine in the a/b/c/.cljfmt configuration to check foo.clj. Running it directly from directory e would look upwards and use the combination of a/.cljfmt and a/b/d/.cljfmt for bar.clj.

Configuration maps are merged together in depth-order, so that more local settings take precedence. As with Leiningen profiles, you can add metadata hints. If you want to override all existing indents, instead of just supplying new indents that are merged with the defaults, you can use the :replace hint:

{:indents ^:replace {#".*" [[:inner 0]]}}

File Settings

You can configure the way cljfmt looks for source files with the following settings:

• :file-pattern Pattern to match against filenames to determine which files to check. Includes all Clojure, ClojureScript, and cross-compiled files by default.

• :file-ignore Set of strings or patterns of files to ignore. Strings are matched against file and directory names exactly; patterns are matched against the entire (relative) file path. Ignored files will not be checked and ignored directories will not be recursed into.

Format Rules

cljfmt has many formatting rules, and these can be selectively enabled or disabled:

• :indentation? True if cljfmt should correct the indentation of your code.

• :remove-surrounding-whitespace? True if cljfmt should remove whitespace surrounding inner forms. This will convert ( foo ) to (foo).

• :remove-trailing-whitespace? True if cljfmt should remove trailing whitespace in lines. This will convert (foo) \n to (foo)\n.

• :insert-missing-whitespace? True if cljfmt should insert whitespace missing from between elements. This will convert (foo(bar)) to (foo (bar)).

• :remove-consecutive-blank-lines? True if cljfmt should collapse consecutive blank lines. Any runs of empty lines longer than :max-consecutive-blank-lines will be truncated to the configured limit. The default limit is 2. This will convert (foo)\n\n\n\n(bar) to (foo)\n\n\n(bar).

• :insert-padding-lines? Whether cljfmt should insert blank lines between certain top-level forms. Any multi-line form will be padded with at least :padding-lines empty lines between it and other non-comment forms. The defaults is 2 lines.

• :rewrite-namespaces? Whether cljfmt should rewrite namespace forms to standardize their layout.

• :single-import-break-width Control the threshold for breaking a single class import into a package import group. If the combined package and class name would be longer than this limit, it is represented as a group, otherwise it is inlined into a qualified class symbol.

Indentation rules

There are a few types of indentation rules that can be applied to forms. These are configured by a map of rule targets to indenters. Each rule is specified with either a symbol or a regular expression pattern. Rules are matched against forms in the following order:

1. Check the qualified form symbol against the rule, including namespace.
2. Check the name of the form symbol against the rule symbol.
3. If the rule is a pattern, match it against the form symbol string.

This ordering allows you to provide specific rules for overlapping symbols from different namespaces, e.g. differentiating d/catch from catch.

Inner rules

An :inner rule will apply a constant indentation to all elements at a fixed depth. So an indent rule:

{foo [[:inner 0]]}

Will indent all elements inside a foo form by two spaces:

(foo bar
  baz
  bang)

While an indent rule like:

{foo [[:inner 1]]}

Will indent all subforms one level in:

(foo bar
  baz
  (bang
    quz
    qoz))

Sometimes it's useful to limit indentation to one argument of the surrounding form. For example, letfn uses inner indentation only in its binding vector:

(letfn [(double [x]
          (* x 2))]   ;; special indentation here
  (let [y (double 2)
        z (double 3)]
    (println y
             z)))     ;; but not here

To achieve this, an additional index argument may be used:

{letfn [[:inner 2 0]]}

This will limit the inner indent to depth 2 in argument 0.

Block rules

A :block rule is a little smarter. This will act like an inner indent only if there's a line break before a certain number of arguments, otherwise it acts like a normal list form.

For example, an indent rule:

{foo [[:block 0]]}

Indents like this, if there are more than 0 arguments on the same line as the symbol:

(foo bar
     baz
     bang)

But indents at a constant two spaces otherwise:

(foo
  bar
  baz
  bang)

Stair rules

A :stair rule is similar to :block, except that it tries to indent test/expression clauses as pairs. This can be used as an alternative styling for case, cond, cond->, etc.The expression forms will be given an extra level of indentation if they are on their own line:

{cond [[:stair 0]]}

(cond
  a? :a
  b? :b)

(cond
  a?
    :a
  b?
    :b)

Ignoring Forms

By default, cljfmt will ignore forms which are wrapped in a (comment ...) form or preceeded by the discard macro #_. You can also optionally disable formatting rules from matching a form by tagging it with ^:cljfmt/ignore metadata - this is often useful for macros.