prettier-printer v1.1.4
≡ ▶ Prettier Printer ·

A pretty printing library for text documents that can be rendered to a desired maximum width. Basic features:
- Interactive documentation (the ▶ links)
- Functional API:
- Supports tree-shaking
- TypeScript typings
- Contract checking in non-production builds
- MIT license
As an example, the evaluation output in this live CodeSandbox example is formatted using this library.
≡ ▶ Contents
- Tutorial
- Reference
- Rendering documents
- Document constants
PP.line ~> docv1.0.0PP.lineBreak ~> docv1.0.0PP.softLine ~> docv1.0.0PP.softBreak ~> docv1.0.0
- Concatenating documents
- Lists of documents
- Lazy documents
PP.lazy(() => doc) ~> docv1.0.0
- Enclosing documents
PP.enclose([lhsDoc, rhsDoc], doc) ~> docv1.0.0- Document pair constants
PP.angles ~> ['<', '>']v1.0.0PP.braces ~> ['{', '}']v1.0.0- PP.brackets ~> ['[', ']'] v1.0.0
PP.dquotes ~> ['"', '"']v1.0.0PP.lineBreaks ~> [PP.lineBreak, PP.lineBreak]v1.1.0PP.lines ~> [PP.line, PP.line]v1.1.0PP.parens ~> ['(', ')']v1.0.0PP.spaces ~> [' ', ' ']v1.0.0PP.squotes ~> ["'", "'"]v1.0.0
- Alternative documents
PP.choice(wideDoc, narrowDoc) ~> docv1.0.0PP.group(doc) ~> docv1.0.0
- Nested documents
- Layout dependent documents
- Aligned documents
- Related Work
≡ ▶ Tutorial
To be done.
In the meanwhile, read Philip Wadler's paper A prettier printer.
≡ ▶ Reference
Typically one imports the library as:
import * as PP from 'prettier-printer'The examples also utilize Ramda, bound as R.
≡ ▶ Rendering documents
≡ ▶ PP.render(maxCols, doc) ~> string v1.0.0
PP.render renders the document to a string trying to keep the width of the
document within the specified maximum. A width of 0 means that there is no
maximum. See also PP.renderWith.
For example:
PP.render(
10,
PP.indent('-- ', PP.group(PP.intersperse(PP.line, ['Hello,', 'world!'])))
)
// -- Hello,
// -- world! ≡ ▶ PP.renderWith({text: (state, string) => state, line: state => state}, state, maxCols, doc) ~> state v1.0.0
PP.renderWith renders the document with the given actions text and line.
You can use this function to output the document without creating an
intermediate string of the whole document. See also PP.render.
≡ ▶ Document constants
Any string that doesn't contain '\n' or '\r' characters is considered as an
atomic document. For example, '' is an empty document and ' ' is a space.
≡ ▶ PP.line ~> doc v1.0.0
PP.line renders as a new line unless undone by PP.group in
which case PP.line renders as a space.
For example:
PP.render(20, ['Hello,', PP.line, 'world!'])
// Hello,
// world!PP.render(20, PP.group(['Hello,', PP.line, 'world!']))
// Hello, world! ≡ ▶ PP.lineBreak ~> doc v1.0.0
PP.lineBreak renders as a new line unless undone by PP.group in
which case PP.lineBreak renders as empty.
For example:
PP.render(20, ['Lol', PP.lineBreak, 'Bal'])
// Lol
// BalPP.render(20, PP.group(['Lol', PP.lineBreak, 'Bal']))
// LolBal ≡ ▶ PP.softLine ~> doc v1.0.0
PP.softLine renders as a space if the output fits and otherwise as a new line.
For example:
PP.render(
20,
PP.intersperse(
PP.softLine,
R.split(
/\s+/,
'Here is a paragraph of text that we will format to a desired width.'
)
)
)
// Here is a paragraph
// of text that we will
// format to a desired
// width. ≡ ▶ PP.softBreak ~> doc v1.0.0
PP.softBreak renders as empty if the output fits and otherwise as a new line.
For example:
PP.render(10, PP.intersperse(PP.softBreak, R.split(/\b/, 'this.method(rocks)')))
// this.
// method(
// rocks)≡ ▶ Concatenating documents
An array of documents is considered as a concatenation of documents. For
example, [] is an empty document and ['foo', 'bar'] is equivalent to
'foobar'.
≡ ▶ PP.append(rhsDoc, lhsDoc) ~> doc v1.0.0
PP.append reverse concatenates the documents.
For example:
PP.render(0, PP.append('bar', 'foo'))
// foobar ≡ ▶ PP.prepend(lhsDoc, rhsDoc) ~> doc v1.0.0
PP.prepend concatenates the documents.
For example:
PP.render(0, PP.prepend('foo', 'bar'))
// foobar≡ ▶ Lists of documents
≡ ▶ PP.intersperse(doc, [...docs]) ~> [...docs] v1.0.0
PP.intersperse puts the given separator document between each document in the
given list of documents.
For example:
PP.intersperse(',', ['a', 'b', 'c'])
// ['a', ',', 'b', ',', 'c'] ≡ ▶ PP.punctuate(sepDoc, [...docs]) ~> [...docs] v1.0.0
PP.punctuate concatenates the given separator after each document in the given
list of documents except the last.
For example:
PP.punctuate(',', ['a', 'b', 'c'])
// [ [ 'a', ',' ], [ 'b', ',' ], 'c' ]≡ ▶ Lazy documents
≡ ▶ PP.lazy(() => doc) ~> doc v1.0.0
PP.lazy creates a lazy document. The given thunk is only invoked as needed to
compute the document.
≡ ▶ Enclosing documents
≡ ▶ PP.enclose([lhsDoc, rhsDoc], doc) ~> doc v1.0.0
PP.enclose encloses the given document between the given pair of documents.
For example:
PP.render(0, PP.enclose(PP.parens, 'foo'))
// (foo)≡ ▶ Document pair constants
≡ ▶ PP.angles ~> ['<', '>'] v1.0.0
≡ ▶ PP.braces ~> ['{', '}'] v1.0.0
≡ ▶ PP.brackets ~> ['[', ']'] v1.0.0
≡ ▶ PP.dquotes ~> ['"', '"'] v1.0.0
≡ ▶ PP.lineBreaks ~> [PP.lineBreak, PP.lineBreak] v1.1.0
≡ ▶ PP.lines ~> [PP.line, PP.line] v1.1.0
≡ ▶ PP.parens ~> ['(', ')'] v1.0.0
≡ ▶ PP.spaces ~> [' ', ' '] v1.0.0
≡ ▶ PP.squotes ~> ["'", "'"] v1.0.0
≡ ▶ Alternative documents
≡ ▶ PP.choice(wideDoc, narrowDoc) ~> doc v1.0.0
PP.choice(wideDoc, narrowDoc) renders as the given wideDoc on a line if it
fits within the maximum width and otherwise as the narrowDoc.
PP.lines and PP.lineBreaks within the wideDoc
are undone like with PP.group.
For example:
PP.render(5, PP.choice('wide', 'narrow'))
// 'wide'PP.render(3, PP.choice('wide', 'narrow'))
// 'narrow'Note that usually the idea is that the narrow version can indeed be rendered more narrowly.
For example:
const hyphen = PP.choice('', ['-', PP.lineBreak])
PP.render(5, PP.intersperse(hyphen, ['hy', 'phen', 'at', 'ed']))
// hy-
// phen-
// ated ≡ ▶ PP.group(doc) ~> doc v1.0.0
PP.group allows PP.lines and PP.lineBreaks
within the given document to be undone if the result fits within the maximum
width. PP.group(doc) is equivalent to PP.choice(doc, doc).
≡ ▶ Nested documents
≡ ▶ PP.nest(string | number, doc) ~> doc v1.0.0
PP.nest increases the nesting after next new line by the given string or by
the given number of spaces.
For example:
PP.render(6, PP.nest(2, PP.group(PP.intersperse(PP.line, ['foo', 'bar']))))
// foo
// bar≡ ▶ Layout dependent documents
≡ ▶ PP.column(column => doc) ~> doc v1.0.0
PP.column allows a document to depend on the column at which the document
starts.
≡ ▶ PP.nesting(nesting => doc) ~> doc v1.0.0
PP.nesting allows a document to depend on the nesting after the next new line.
≡ ▶ Aligned documents
≡ ▶ PP.align(doc) ~> doc v1.0.0
PP.align creates a document such that the nesting of the document is aligned
to the current column.
For example:
PP.render(10, PP.group(['foo(', PP.align(['bar,', PP.line, 'baz']), ')']))
// foo(bar,
// baz) ≡ ▶ PP.hang(string | number, doc) ~> doc v1.0.0
PP.hang creates a document such that the document is nested by the given
string or number of spaces starting from the current column.
For example:
PP.render(10, PP.group(['foo(', PP.hang(2, ['bar,', PP.line, 'baz']), ')']))
// foo(bar,
// baz) ≡ ▶ PP.indent(string | number, doc) ~> doc v1.0.0
PP.indent creates a document such that the document is indented by the given
prefix or number of spaces starting from the current column.
PP.render(
20,
PP.nest(
2,
PP.group([
'A comment:',
PP.line,
PP.line,
PP.indent(
'-- ',
PP.intersperse(
PP.softLine,
R.split(/\s+/, 'This is the comment that you are looking for.')
)
)
])
)
)
// A comment:
//
// -- This is the
// -- comment that
// -- you are looking
// -- for.≡ ▶ Related Work
- Philip Wadler's paper A prettier printer describes the basic ideas and implementation.
- Text.PrettyPrint.Leijen is Daan Leijen's implementation with some extensions.
- Other prettier printer implementations by the author of this library:
- text.pretty-printing another JS implementation based on Wadler's paper. Marked as "Unmaintained".
- Prettier uses a similar pretty printing library underneath.