These options determine how the different types of modules within a project will be treated.
module.noParse
RegExp | [RegExp]
RegExp | [RegExp] | function
(since webpack 3.0.0)
Prevent webpack from parsing any files matching the given regular expression(s). Ignored files should not have calls to import
, require
, define
or any other importing mechanism. This can boost build performance when ignoring large libraries.
noParse: /jquery|lodash/
// since webpack 3.0.0
noParse: function(content) {
return /jquery|lodash/.test(content);
}
module.rules
array
An array of Rules which are matched to requests when modules are created. These rules can modify how the module is created. They can apply loaders to the module, or modify the parser.
A Rule can be separated into three parts — Conditions, Results and nested Rules.
There are two input values for the conditions:
The resource: An absolute path to the file requested. It's already resolved according to the resolve
rules.
The issuer: An absolute path to the file of the module which requested the resource. It's the location of the import.
Example: When we import './style.css'
within app.js
, the resource is /path/to/style.css
and the issuer is /path/to/app.js
.
In a Rule the properties test
, include
, exclude
and resource
are matched with the resource and the property issuer
is matched with the issuer.
When using multiple conditions, all conditions must match.
Be careful! The resource is the resolved path of the file, which means symlinked resources are the real path not the symlink location. This is good to remember when using tools that symlink packages (likenpm link
), common conditions like/node_modules/
may inadvertently miss symlinked files.
Rule results are used only when the Rule condition matches.
There are two output values of a Rule:
These properties affect the loaders: loader
, options
, use
.
For compatibility also these properties: query
, loaders
.
The enforce
property affects the loader category. Whether it's a normal, pre- or post- loader.
The parser
property affects the parser options.
Nested rules can be specified under the properties rules
and oneOf
.
These rules are evaluated when the Rule condition matches.
Rule.enforce
Possible values: "pre" | "post"
Specifies the category of the loader. No value means normal loader.
There is also an additional category "inlined loader" which are loaders applied inline of the import/require.
All loaders are sorted in the order pre, inline, normal, post
and used in this order.
All normal loaders can be omitted (overridden) by prefixing !
in the request.
All normal and pre loaders can be omitted (overridden) by prefixing -!
in the request.
All normal, post and pre loaders can be omitted (overridden) by prefixing !!
in the request.
Inline loaders and !
prefixes should not be used as they are non-standard. They may be use by loader generated code.
Rule.exclude
Rule.exclude
is a shortcut to Rule.resource.exclude
. If you supply a Rule.exclude
option, you cannot also supply a Rule.resource
. See Rule.resource
and Condition.exclude
for details.
Rule.include
Rule.include
is a shortcut to Rule.resource.include
. If you supply a Rule.include
option, you cannot also supply a Rule.resource
. See Rule.resource
and Condition.include
for details.
Rule.issuer
A Condition
to match against the module that issued the request. In the following example, the issuer
for the a.js
request would be the path to the index.js
file.
index.js
import A from './a.js'
This option can be used to apply loaders to the dependencies of a specific module or set of modules.
Rule.loader
Rule.loader
is a shortcut to Rule.use: [ { loader } ]
. See Rule.use
and UseEntry.loader
for details.
Rule.loaders
This option is deprecated in favor ofRule.use
.
Rule.loaders
is an alias to Rule.use
. See Rule.use
for details.
Rule.oneOf
An array of Rules
from which only the first matching Rule is used when the Rule matches.
{
test: /.css$/,
oneOf: [
{
resourceQuery: /inline/, // foo.css?inline
use: 'url-loader'
},
{
resourceQuery: /external/, // foo.css?external
use: 'file-loader'
}
]
}
Rule.options
/ Rule.query
Rule.options
and Rule.query
are shortcuts to Rule.use: [ { options } ]
. See Rule.use
and UseEntry.options
for details.
Rule.query
is deprecated in favor ofRule.options
andUseEntry.options
.
Rule.parser
An object with parser options. All applied parser options are merged.
Parsers may inspect these options and disable or reconfigure themselves accordingly. Most of the default plugins interpret the values as follows:
false
disables the parser.true
or leaving it undefined
enables the parser.However, parser plugins may accept more than just a boolean. For example, the internal NodeStuffPlugin
can accept an object instead of true
to add additional options for a particular Rule.
Examples (parser options by the default plugins):
parser: {
amd: false, // disable AMD
commonjs: false, // disable CommonJS
system: false, // disable SystemJS
harmony: false, // disable ES2015 Harmony import/export
requireInclude: false, // disable require.include
requireEnsure: false, // disable require.ensure
requireContext: false, // disable require.context
browserify: false, // disable special handling of Browserify bundles
requireJs: false, // disable requirejs.*
node: false, // disable __dirname, __filename, module, require.extensions, require.main, etc.
node: {...} // reconfigure node layer on module level
}
Rule.resource
A Condition
matched with the resource. You can either supply a Rule.resource
option or use the shortcut options Rule.test
, Rule.exclude
, and Rule.include
. See details in Rule
conditions.
Rule.resourceQuery
A Condition
matched with the resource query. This option is used to test against the query section of a request string (i.e. from the question mark onwards). If you were to import Foo from './foo.css?inline'
, the following condition would match:
{
test: /.css$/,
resourceQuery: /inline/,
use: 'url-loader'
}
Rule.rules
An array of Rules
that is also used when the Rule matches.
Rule.test
Rule.test
is a shortcut to Rule.resource.test
. If you supply a Rule.test
option, you cannot also supply a Rule.resource
. See Rule.resource
and Condition.test
for details.
Rule.use
A list of UseEntries which are applied to modules. Each entry specifies a loader to be used.
Passing a string (i.e. use: [ "style-loader" ]
) is a shortcut to the loader property (i.e. use: [ { loader: "style-loader "} ]
).
Loaders can be chained by passing multiple loaders, which will be applied from right to left (last to first configured).
use: [
'style-loader',
{
loader: 'css-loader',
options: {
importLoaders: 1
}
},
{
loader: 'less-loader',
options: {
noIeCompat: true
}
}
]
See UseEntry for details.
Condition
Conditions can be one of these:
{ test: Condition }
: The Condition must match. The convention is to provide a RegExp or array of RegExps here, but it's not enforced.
{ include: Condition }
: The Condition must match. The convention is to provide a string or array of strings here, but it's not enforced.
{ exclude: Condition }
: The Condition must NOT match. The convention is to provide a string or array of strings here, but it's not enforced.
{ and: [Condition] }
: All Conditions must match.
{ or: [Condition] }
: Any Condition must match.
{ not: [Condition] }
: All Conditions must NOT match.
Example:
{
test: /\.css$/,
include: [
path.resolve(__dirname, "app/styles"),
path.resolve(__dirname, "vendor/styles")
]
}
UseEntry
object
It must have a loader
property being a string. It is resolved relative to the configuration context
with the loader resolving options (resolveLoader).
It can have a options
property being a string or object. This value is passed to the loader, which should interpret it as loader options.
For compatibility a query
property is also possible, which is an alias for the options
property. Use the options
property instead.
Example:
{
loader: "css-loader",
options: {
modules: true
}
}
Note that webpack needs to generate a unique module identifier from the resource and all loaders including options. It tries to do this with a JSON.stringify
of the options object. This is fine in 99.9% of cases, but may be not unique if you apply the same loaders with different options to the resource and the options have some stringified values.
It also breaks if the options object cannot be stringified (i.e. circular JSON). Because of this you can have a ident
property in the options object which is used as unique identifier.
Avoid using these options as they are deprecated and will soon be removed.
These options describe the default settings for the context created when a dynamic dependency is encountered.
Example for an unknown
dynamic dependency: require
.
Example for an expr
dynamic dependency: require(expr)
.
Example for an wrapped
dynamic dependency: require("./templates/" + expr)
.
Here are the available options with their defaults:
module: {
exprContextCritical: true,
exprContextRecursive: true,
exprContextRegExp: false,
exprContextRequest: ".",
unknownContextCritical: true,
unknownContextRecursive: true,
unknownContextRegExp: false,
unknownContextRequest: ".",
wrappedContextCritical: false
wrappedContextRecursive: true,
wrappedContextRegExp: /.*/,
strictExportPresence: false // since webpack 2.3.0
}
You can use theContextReplacementPlugin
to modify these values for individual dependencies. This also removes the warning.
A few use cases:
wrappedContextCritical: true
.require(expr)
should include the whole directory: exprContextRegExp: /^\.\//
require("./templates/" + expr)
should not include subdirectories by default: wrappedContextRecursive: false
strictExportPresence
makes missing exports an error instead of warning