Options¶
Every option rainbow-fmt reads, with its values, its default and what it
does. This page is generated from the option declarations
(python -m rainbow_fmt.options.reference; a test fails when it is out of
date). How the levels of configuration combine, how to set an option for
one language or for some files, and the inline directives are in
configuration.md.
Core¶
[core]: layout options every language reads. Any of them may also be set
in a [language.X] section, for that language only.
| Option | Values | Default | Description |
|---|---|---|---|
core.max_width |
integer ≥ 1 | 80 |
The column limit lines are fitted into. |
core.indent_style |
"space", "tab" |
"space" |
Indent with spaces or tabs. |
core.indent_size |
integer ≥ 0 | 4 |
Spaces per indentation level. |
core.tab_width |
integer ≥ 1 | 4 |
Columns a tab counts as when measuring width. |
core.line_ending |
"lf", "crlf", "preserve" |
"lf" |
Line ending of the output; preserve uses the input's first line ending. |
core.max_blank_lines |
integer ≥ 0 | 1 |
Blank lines kept between members, at most (none are added). |
core.max_width, given
result = combine(first_argument, second_argument, third_argument)
with max_width = 40:
result = combine(
first_argument,
second_argument,
third_argument,
)
with max_width = 80:
result = combine(first_argument, second_argument, third_argument)
core.indent_style, given
def total(items):
result = 0
for item in items:
result += item
return result
with indent_style = "space":
def total(items):
result = 0
for item in items:
result += item
return result
with indent_style = "tab":
def total(items):
result = 0
for item in items:
result += item
return result
core.indent_size, given
def total(items):
result = 0
for item in items:
result += item
return result
with indent_size = 2:
def total(items):
result = 0
for item in items:
result += item
return result
with indent_size = 4:
def total(items):
result = 0
for item in items:
result += item
return result
core.tab_width, given
if ready:
result = combine(alpha, beta, gamma)
with tab_width = 4:
if ready:
result = combine(alpha, beta, gamma)
with tab_width = 8:
if ready:
result = combine(
alpha, beta, gamma
)
core.line_ending, given
{"a": 1,
"b": 2}
with line_ending = "lf":
'{"a": 1, "b": 2}\n'
with line_ending = "crlf":
'{"a": 1, "b": 2}\r\n'
with line_ending = "preserve":
'{"a": 1, "b": 2}\r\n'
core.max_blank_lines, given
select 1;
select 2;
select 3;
with max_blank_lines = 0:
select 1;
select 2;
select 3;
with max_blank_lines = 1:
select 1;
select 2;
select 3;
with max_blank_lines = 2:
select 1;
select 2;
select 3;
Shared¶
[shared]: options several languages read. A language may give one another
default (listed under the language) or narrow it to fewer values, in
which case the option belongs to the language and [shared] values do not
apply to it.
| Option | Values | Default | Description |
|---|---|---|---|
shared.trailing_comma |
"never", "multiline" |
"never" |
Whether a comma follows the last item of a list broken over several lines. |
shared.trailing_comma, given
names = ["alpha_value_one", "beta_value_two", "gamma_value_three", "delta_value_four"]
with trailing_comma = "never":
names = [
"alpha_value_one",
"beta_value_two",
"gamma_value_three",
"delta_value_four"
]
with trailing_comma = "multiline":
names = [
"alpha_value_one",
"beta_value_two",
"gamma_value_three",
"delta_value_four",
]
Languages¶
[language.X]: the options of each built-in language pack. A plugin pack
documents its own.
JSON¶
trailing_comma narrows shared.trailing_comma: [shared] values do not apply to JSON.
| Option | Values | Default | Description |
|---|---|---|---|
language.json.trailing_comma |
"never" |
"never" |
JSON does not allow trailing commas (narrows shared.trailing_comma). |
language.json.object_wrap |
"preserve", "fit", "always" |
"preserve" |
When objects break: as written, only when too long, or always. |
language.json.align_values |
true, false |
false |
Align the values of a broken object whose values are all scalars. |
language.json.trailing_comma, given
{"items": [1, 2]}
with trailing_comma = "never":
{"items": [1, 2]}
language.json.object_wrap, given
{
"a": 1, "b": 2}
with object_wrap = "preserve":
{
"a": 1,
"b": 2
}
with object_wrap = "fit":
{"a": 1, "b": 2}
with object_wrap = "always":
{
"a": 1,
"b": 2
}
language.json.align_values, given
{
"name": "rainbow", "id": 1}
with align_values = false:
{
"name": "rainbow",
"id": 1
}
with align_values = true:
{
"name": "rainbow",
"id": 1
}
CSS¶
| Option | Values | Default | Description |
|---|---|---|---|
language.css.selector_list |
"one_per_line", "fit" |
"one_per_line" |
A rule's selectors: one per line, or on one line when they fit. |
language.css.rule_wrap |
"always", "fit", "preserve" |
"always" |
When a block of declarations breaks: always, when too long, or as written. |
language.css.last_semicolon |
"always", "never", "preserve" |
"always" |
Whether the last declaration of a block ends with a semicolon. |
language.css.selector_list, given
h1, h2, h3 {
margin: 0;
}
with selector_list = "one_per_line":
h1,
h2,
h3 {
margin: 0;
}
with selector_list = "fit":
h1, h2, h3 {
margin: 0;
}
language.css.rule_wrap, given
a { color: red; }
with rule_wrap = "always":
a {
color: red;
}
with rule_wrap = "fit":
a { color: red; }
with rule_wrap = "preserve":
a { color: red; }
language.css.last_semicolon, given
a {
color: red
}
with last_semicolon = "always":
a {
color: red;
}
with last_semicolon = "never":
a {
color: red
}
with last_semicolon = "preserve":
a {
color: red
}
Python¶
shared.trailing_comma defaults to "multiline" for Python.
| Option | Values | Default | Description |
|---|---|---|---|
language.python.bracket_wrap |
"magic_trailing_comma", "fit", "preserve" |
"magic_trailing_comma" |
When bracketed lists break: when too long or with a trailing comma, only when too long, or as written. |
language.python.binary_operator_break |
"before", "after" |
"before" |
Whether a broken expression breaks before or after its operators. |
language.python.definition_blank_lines |
"enforce", "cap", "preserve" |
"enforce" |
Blank lines around def and class: exactly the counts, at most the counts, or as written. |
language.python.top_level_blank_lines |
integer ≥ 0 | 2 |
Blank lines around top-level definitions (PEP 8: 2). |
language.python.nested_blank_lines |
integer ≥ 0 | 1 |
Blank lines around nested definitions (PEP 8: 1). |
language.python.inline_comment_spaces |
integer ≥ 1 | 2 |
Spaces before a comment that follows code (PEP 8: 2). |
language.python.statement_blank_lines |
"preserve", "separate" |
"preserve" |
Blank lines between the statements of a function: as written, or also around loops and after if statements. |
language.python.docstrings |
"preserve", "aligned" |
"preserve" |
Docstrings: as written, or the closing quotes on their own line and the lines after the first aligned under its first letter. |
language.python.bracket_hug |
true, false |
false |
A list whose items are all dicts hugs them: '[{', '}, {' and '}]'. |
language.python.bracket_wrap, given
items = [
1,
2,
]
with bracket_wrap = "magic_trailing_comma":
items = [
1,
2,
]
with bracket_wrap = "fit":
items = [1, 2]
with bracket_wrap = "preserve":
items = [
1,
2,
]
language.python.binary_operator_break, given
total = (first_value_in_the_sum + second_value_in_the_sum + third_value_in_the_sum + fourth_value)
with binary_operator_break = "before":
total = (
first_value_in_the_sum
+ second_value_in_the_sum
+ third_value_in_the_sum
+ fourth_value
)
with binary_operator_break = "after":
total = (
first_value_in_the_sum +
second_value_in_the_sum +
third_value_in_the_sum +
fourth_value
)
language.python.definition_blank_lines, given
x = 1
def f():
pass
def g():
pass
with definition_blank_lines = "enforce":
x = 1
def f():
pass
def g():
pass
with definition_blank_lines = "cap":
x = 1
def f():
pass
def g():
pass
with definition_blank_lines = "preserve":
x = 1
def f():
pass
def g():
pass
language.python.top_level_blank_lines, given
def f():
pass
def g():
pass
with top_level_blank_lines = 1:
def f():
pass
def g():
pass
with top_level_blank_lines = 2:
def f():
pass
def g():
pass
language.python.nested_blank_lines, given
class A:
def f(self):
pass
def g(self):
pass
with nested_blank_lines = 0:
class A:
def f(self):
pass
def g(self):
pass
with nested_blank_lines = 1:
class A:
def f(self):
pass
def g(self):
pass
language.python.inline_comment_spaces, given
x = 1 # one
with inline_comment_spaces = 1:
x = 1 # one
with inline_comment_spaces = 2:
x = 1 # one
language.python.statement_blank_lines, given
def f(items):
total = 0
for item in items:
total += item
if total:
print(total)
return total
with statement_blank_lines = "preserve":
def f(items):
total = 0
for item in items:
total += item
if total:
print(total)
return total
with statement_blank_lines = "separate":
def f(items):
total = 0
for item in items:
total += item
if total:
print(total)
return total
language.python.docstrings, given
def f():
"""First line.
More text."""
with docstrings = "preserve":
def f():
"""First line.
More text."""
with docstrings = "aligned":
def f():
"""First line.
More text.
"""
language.python.bracket_hug, given
rows = [{"name": "alpha", "value": 1}, {"name": "beta", "value": 2}, {"name": "gamma", "value": 3}]
with bracket_hug = false:
rows = [
{"name": "alpha", "value": 1},
{"name": "beta", "value": 2},
{"name": "gamma", "value": 3},
]
with bracket_hug = true:
rows = [{
"name": "alpha",
"value": 1,
}, {
"name": "beta",
"value": 2,
}, {
"name": "gamma",
"value": 3,
}]
JavaScript¶
shared.trailing_comma defaults to "multiline" for JavaScript.
| Option | Values | Default | Description |
|---|---|---|---|
language.javascript.semicolons |
"as_needed", "always", "preserve" |
"as_needed" |
Statement-final semicolons: only where a line would join the next, always, or as written. |
language.javascript.object_wrap |
"preserve", "fit" |
"preserve" |
When objects break: also when written with a line break after '{', or only when too long. |
language.javascript.bracket_spacing |
true, false |
true |
Spaces inside the braces of objects, patterns and import/export lists. |
language.javascript.binary_operator_break |
"after", "before" |
"after" |
Whether a broken expression breaks after or before its operators. |
language.javascript.return_semicolons |
"as_statements", "unless_last" |
"as_statements" |
With semicolons 'as_needed': a return statement followed by another statement in its block ends with ';'. |
language.javascript.bracket_hug |
true, false |
false |
An array whose items are all objects hugs them: '[{', '}, {' and '}]'. |
language.javascript.semicolons, given
const a = 1
const b = 2;
with semicolons = "as_needed":
const a = 1
const b = 2
with semicolons = "always":
const a = 1;
const b = 2;
with semicolons = "preserve":
const a = 1
const b = 2;
language.javascript.object_wrap, given
const point = {
x: 1, y: 2 };
with object_wrap = "preserve":
const point = {
x: 1,
y: 2,
}
with object_wrap = "fit":
const point = { x: 1, y: 2 }
language.javascript.bracket_spacing, given
const {x, y} = point;
with bracket_spacing = false:
const {x, y} = point
with bracket_spacing = true:
const { x, y } = point
language.javascript.binary_operator_break, given
const total = (firstValueInTheSum + secondValueInTheSum + thirdValueInTheSum + fourthValue);
with binary_operator_break = "after":
const total = (firstValueInTheSum +
secondValueInTheSum +
thirdValueInTheSum +
fourthValue)
with binary_operator_break = "before":
const total = (firstValueInTheSum
+ secondValueInTheSum
+ thirdValueInTheSum
+ fourthValue)
language.javascript.return_semicolons, given
function f() {
return compute()
cleanup()
}
with return_semicolons = "as_statements":
function f() {
return compute()
cleanup()
}
with return_semicolons = "unless_last":
function f() {
return compute();
cleanup()
}
language.javascript.bracket_hug, given
const rows = [{name: "alpha", value: 1}, {name: "beta", value: 2}, {name: "gamma", value: 3}];
with bracket_hug = false:
const rows = [
{ name: "alpha", value: 1 },
{ name: "beta", value: 2 },
{ name: "gamma", value: 3 },
]
with bracket_hug = true:
const rows = [{
name: "alpha",
value: 1,
}, {
name: "beta",
value: 2,
}, {
name: "gamma",
value: 3,
}]
TypeScript¶
shared.trailing_comma defaults to "multiline" for TypeScript.
| Option | Values | Default | Description |
|---|---|---|---|
language.typescript.semicolons |
"as_needed", "always", "preserve" |
"as_needed" |
Statement-final semicolons: only where a line would join the next, always, or as written. |
language.typescript.object_wrap |
"preserve", "fit" |
"preserve" |
When objects break: also when written with a line break after '{', or only when too long. |
language.typescript.bracket_spacing |
true, false |
true |
Spaces inside the braces of objects, patterns and import/export lists. |
language.typescript.binary_operator_break |
"after", "before" |
"after" |
Whether a broken expression breaks after or before its operators. |
language.typescript.return_semicolons |
"as_statements", "unless_last" |
"as_statements" |
With semicolons 'as_needed': a return statement followed by another statement in its block ends with ';'. |
language.typescript.bracket_hug |
true, false |
false |
An array whose items are all objects hugs them: '[{', '}, {' and '}]'. |
language.typescript.semicolons, given
const a = 1
const b = 2;
with semicolons = "as_needed":
const a = 1
const b = 2
with semicolons = "always":
const a = 1;
const b = 2;
with semicolons = "preserve":
const a = 1
const b = 2;
language.typescript.object_wrap, given
const point = {
x: 1, y: 2 };
with object_wrap = "preserve":
const point = {
x: 1,
y: 2,
}
with object_wrap = "fit":
const point = { x: 1, y: 2 }
language.typescript.bracket_spacing, given
const {x, y} = point;
with bracket_spacing = false:
const {x, y} = point
with bracket_spacing = true:
const { x, y } = point
language.typescript.binary_operator_break, given
const total = (firstValueInTheSum + secondValueInTheSum + thirdValueInTheSum + fourthValue);
with binary_operator_break = "after":
const total = (firstValueInTheSum +
secondValueInTheSum +
thirdValueInTheSum +
fourthValue)
with binary_operator_break = "before":
const total = (firstValueInTheSum
+ secondValueInTheSum
+ thirdValueInTheSum
+ fourthValue)
language.typescript.return_semicolons, given
function f() {
return compute()
cleanup()
}
with return_semicolons = "as_statements":
function f() {
return compute()
cleanup()
}
with return_semicolons = "unless_last":
function f() {
return compute();
cleanup()
}
language.typescript.bracket_hug, given
const rows = [{name: "alpha", value: 1}, {name: "beta", value: 2}, {name: "gamma", value: 3}];
with bracket_hug = false:
const rows = [
{ name: "alpha", value: 1 },
{ name: "beta", value: 2 },
{ name: "gamma", value: 3 },
]
with bracket_hug = true:
const rows = [{
name: "alpha",
value: 1,
}, {
name: "beta",
value: 2,
}, {
name: "gamma",
value: 3,
}]
HTML¶
| Option | Values | Default | Description |
|---|---|---|---|
language.html.whitespace_sensitivity |
"css", "strict", "ignore" |
"css" |
Where whitespace around content matters: in inline elements (as CSS renders it), everywhere, or nowhere. |
language.html.bracket_same_line |
true, false |
false |
Keep the '>' of a tag broken over several lines on the last attribute's line. |
language.html.void_elements |
"no_slash", "slash", "preserve" |
"no_slash" |
Void elements (br, img, input …): ' ', ' ', or as written. |
language.html.whitespace_sensitivity, given
<p><a href="#one">one</a><a href="#two">two</a><a href="#three">three</a><a href="#four">four</a><a href="#five">five</a></p>
with whitespace_sensitivity = "css":
<p>
<a href="#one">one</a><a href="#two">two</a><a href="#three">three</a><a
href="#four"
>four</a><a href="#five">five</a>
</p>
with whitespace_sensitivity = "strict":
<p><a href="#one">one</a><a href="#two">two</a><a href="#three">three</a><a
href="#four"
>four</a><a href="#five">five</a></p>
with whitespace_sensitivity = "ignore":
<p>
<a href="#one">one</a><a href="#two">two</a><a href="#three">three</a>
<a href="#four">four</a><a href="#five">five</a>
</p>
language.html.bracket_same_line, given
<input type="text" name="username" placeholder="Your name" autocomplete="off" required>
with bracket_same_line = false:
<input
type="text"
name="username"
placeholder="Your name"
autocomplete="off"
required
>
with bracket_same_line = true:
<input
type="text"
name="username"
placeholder="Your name"
autocomplete="off"
required>
language.html.void_elements, given
<p>line<br/>break<br>here</p>
with void_elements = "no_slash":
<p>line<br>break<br>here</p>
with void_elements = "slash":
<p>line<br />break<br />here</p>
with void_elements = "preserve":
<p>line<br/>break<br>here</p>
Svelte¶
| Option | Values | Default | Description |
|---|---|---|---|
language.svelte.whitespace_sensitivity |
"css", "strict", "ignore" |
"css" |
Where whitespace around content matters: in inline elements (as CSS renders it), everywhere, or nowhere. |
language.svelte.bracket_same_line |
true, false |
false |
Keep the '>' of a tag broken over several lines on the last attribute's line. |
language.svelte.void_elements |
"no_slash", "slash", "preserve" |
"no_slash" |
Void elements (br, img, input …): ' ', ' ', or as written. |
language.svelte.logic_block_blank_lines |
"preserve", "separate" |
"preserve" |
Blank lines between the items of the markup: as written, or also around {#each} blocks and after {#if} blocks. |
language.svelte.whitespace_sensitivity, given
<p><a href="#one">one</a><a href="#two">two</a><a href="#three">three</a><a href="#four">four</a><a href="#five">five</a></p>
with whitespace_sensitivity = "css":
<p>
<a href="#one">one</a><a href="#two">two</a><a href="#three">three</a><a
href="#four"
>four</a><a href="#five">five</a>
</p>
with whitespace_sensitivity = "strict":
<p><a href="#one">one</a><a href="#two">two</a><a href="#three">three</a><a
href="#four"
>four</a><a href="#five">five</a></p>
with whitespace_sensitivity = "ignore":
<p>
<a href="#one">one</a><a href="#two">two</a><a href="#three">three</a>
<a href="#four">four</a><a href="#five">five</a>
</p>
language.svelte.bracket_same_line, given
<input type="text" name="username" placeholder="Your name" autocomplete="off" required>
with bracket_same_line = false:
<input
type="text"
name="username"
placeholder="Your name"
autocomplete="off"
required
>
with bracket_same_line = true:
<input
type="text"
name="username"
placeholder="Your name"
autocomplete="off"
required>
language.svelte.void_elements, given
<p>line<br/>break<br>here</p>
with void_elements = "no_slash":
<p>line<br>break<br>here</p>
with void_elements = "slash":
<p>line<br />break<br />here</p>
with void_elements = "preserve":
<p>line<br/>break<br>here</p>
language.svelte.logic_block_blank_lines, given
<h1>{title}</h1>
{#each items as item}
<p>{item}</p>
{/each}
{#if done}
<p>Done</p>
{/if}
<footer>{note}</footer>
with logic_block_blank_lines = "preserve":
<h1>{title}</h1>
{#each items as item}
<p>{item}</p>
{/each}
{#if done}
<p>Done</p>
{/if}
<footer>{note}</footer>
with logic_block_blank_lines = "separate":
<h1>{title}</h1>
{#each items as item}
<p>{item}</p>
{/each}
{#if done}
<p>Done</p>
{/if}
<footer>{note}</footer>
TOML¶
TOML has no options of its own.
YAML¶
core.indent_size defaults to 2 for YAML.
| Option | Values | Default | Description |
|---|---|---|---|
language.yaml.sequence_indent |
"indent", "none" |
"indent" |
A sequence that is the value of a key: indented under the key, or at its level. |
language.yaml.sequence_indent, given
items:
- a
- b
with sequence_indent = "indent":
items:
- a
- b
with sequence_indent = "none":
items:
- a
- b
SQL¶
| Option | Values | Default | Description |
|---|---|---|---|
language.sql.keyword_case |
"preserve", "upper", "lower" |
"preserve" |
Keywords as written, in upper case, or in lower case. |
language.sql.clause_alignment |
"left", "right" |
"left" |
Clause keywords at the margin, or right-aligned so SELECT, FROM, JOIN, ON and WHERE end in one column (GROUP BY and HAVING in another) and items hang under the first. |
language.sql.keyword_case, given
select a from t where b = 1
with keyword_case = "preserve":
select a
from t
where b = 1
with keyword_case = "upper":
SELECT a
FROM t
WHERE b = 1
with keyword_case = "lower":
select a
from t
where b = 1
language.sql.clause_alignment, given
select a.id, count(*) from users a join orders b on a.id = b.user_id where a.active = 1 group by a.id having count(*) > 1 order by 2 desc
with clause_alignment = "left":
select a.id, count(*)
from users a
join orders b on a.id = b.user_id
where a.active = 1
group by a.id
having count(*) > 1
order by 2 desc
with clause_alignment = "right":
select a.id, count(*)
from users a
join orders b
on a.id = b.user_id
where a.active = 1
group by a.id
having count(*) > 1
order by 2 desc
Markdown¶
| Option | Values | Default | Description |
|---|---|---|---|
language.markdown.fenced_code |
"format", "preserve" |
"format" |
Fenced code in a language rainbow-fmt formats: formatted by that pack, or as written. |
language.markdown.table_width |
integer ≥ 1 | 180 |
The widest row an aligned pipe table may have; a wider table is written condensed, one space around each cell. |
language.markdown.fenced_code, given
```json
{"a":1,
"b":2}
```
with fenced_code = "format":
```json
{"a": 1, "b": 2}
```
with fenced_code = "preserve":
```json
{"a":1,
"b":2}
```
language.markdown.table_width, given
| name | value |
|---|---|
| alpha | 1 |
with table_width = 12:
| name | value |
| --- | --- |
| alpha | 1 |
with table_width = 180:
| name | value |
| ----- | ----- |
| alpha | 1 |
Django¶
| Option | Values | Default | Description |
|---|---|---|---|
language.django.whitespace_sensitivity |
"css", "strict", "ignore" |
"css" |
Where whitespace around content matters: in inline elements (as CSS renders it), everywhere, or nowhere. |
language.django.bracket_same_line |
true, false |
false |
Keep the '>' of a tag broken over several lines on the last attribute's line. |
language.django.void_elements |
"no_slash", "slash", "preserve" |
"no_slash" |
Void elements (br, img, input …): ' ', ' ', or as written. |
language.django.whitespace_sensitivity, given
<p><a href="#one">one</a><a href="#two">two</a><a href="#three">three</a><a href="#four">four</a><a href="#five">five</a></p>
with whitespace_sensitivity = "css":
<p>
<a href="#one">one</a><a href="#two">two</a><a href="#three">three</a><a
href="#four"
>four</a><a href="#five">five</a>
</p>
with whitespace_sensitivity = "strict":
<p><a href="#one">one</a><a href="#two">two</a><a href="#three">three</a><a
href="#four"
>four</a><a href="#five">five</a></p>
with whitespace_sensitivity = "ignore":
<p>
<a href="#one">one</a><a href="#two">two</a><a href="#three">three</a>
<a href="#four">four</a><a href="#five">five</a>
</p>
language.django.bracket_same_line, given
<input type="text" name="username" placeholder="Your name" autocomplete="off" required>
with bracket_same_line = false:
<input
type="text"
name="username"
placeholder="Your name"
autocomplete="off"
required
>
with bracket_same_line = true:
<input
type="text"
name="username"
placeholder="Your name"
autocomplete="off"
required>
language.django.void_elements, given
<p>line<br/>break<br>here</p>
with void_elements = "no_slash":
<p>line<br>break<br>here</p>
with void_elements = "slash":
<p>line<br />break<br />here</p>
with void_elements = "preserve":
<p>line<br/>break<br>here</p>
Files¶
[files]: which files a run formats and how (read for the working
directory, not per file).
| Option | Values | Default | Description |
|---|---|---|---|
files.exclude |
list of strings | [] |
Patterns (as in [[override]] files, relative to the configuration file) of files and directories a directory walk skips; a file named on the command line is always formatted. |
files.ignore_unknown |
true, false |
false |
Do not warn about explicitly named files of unknown type. |
files.editorconfig |
true, false |
true |
Read .editorconfig files (between presets and the project configuration). |
files.verify |
true, false |
true |
Verify that formatting keeps the meaning and is stable before writing a file. |
files.cache |
true, false |
true |
Skip files a previous run found formatted (with the same options and version). |
files.jobs |
integer ≥ 0 | 0 |
Worker processes for many files (0: one per CPU; 1: none); read for the working directory. |