Checkstyle is configured using properties, which are string representations. This document describes how these string representations are mapped to typed properties.
This type represents a boolean.
The following string representations will map to true
:
yes
true
on
Anything else will map to false
.
This type represents a byte. The string representation is
parsed using the java.lang.Byte
class.
This type represents a set of bytes. The string representation
is parsed as a set of comma (',') separated bytes that are parsed
using the java.lang.Byte
class.
Alternatively, a property of this type can be supplied multiple times which is equivalent to a set of comma separated byte values.
This type represents a char. The string representation is
parsed using the java.lang.Character
class.
This type represents a set of chars. The string representation
is parsed as a set of comma (',') separated chars that are parsed
using the java.lang.Character
class.
Alternatively, a property of this type can be supplied multiple times which is equivalent to a set of comma separated char values.
This type represents a double. The string representation is
parsed using the java.lang.Double
class.
This type represents a set of doubles. The string representation
is parsed as a set of comma (',') separated doubles that are parsed
using the java.lang.Double
class.
Alternatively, a property of this type can be supplied multiple times which is equivalent to a set of comma separated integers. For example, the following:
<property name="tokens" value="0.42,0.666"/>
can instead be expressed as:
<property name="tokens" value="0.42"/> <property name="tokens" value="0.666"/>
This type represents a float. The string representation is
parsed using the java.lang.Float
class.
This type represents a set of floats. The string representation
is parsed as a set of comma (',') separated floats that are parsed
using the java.lang.Float
class.
Alternatively, a property of this type can be supplied multiple times which is equivalent to a set of comma separated float values.
This type represents an integer. The string representation is
parsed using the java.lang.Integer
class.
This type represents a set of integers. The string representation
is parsed as a set of comma (',') separated integers that are parsed
using the java.lang.Integer
class.
Alternatively, a property of this type can be supplied multiple times which is equivalent to a set of comma separated integers. For example, the following:
<property name="tokens" value="42,666"/>
can instead be expressed as:
<property name="tokens" value="42"/> <property name="tokens" value="666"/>
This type represents a long. The string representation is
parsed using the java.lang.Long
class.
This type represents a set of longs. The string representation
is parsed as a set of comma (',') separated longs that are parsed
using the java.lang.Long
class.
Alternatively, a property of this type can be supplied multiple times which is equivalent to a set of comma separated long values.
This type represents a short. The string representation is
parsed using the java.lang.Short
class.
This type represents a set of shorts. The string representation
is parsed as a set of comma (',') separated shorts that are parsed
using the java.lang.Short
class.
Alternatively, a property of this type can be supplied multiple times which is equivalent to a set of comma separated short values.
This type represents Java access modifiers.
public
protected
package
private
This type represents the policy for checking block statements. The following table describes the valid options:
Option | Definition |
text |
Require that there is some text in the block. For example:
catch (Exception ex) { // This is a bad coding practice } |
statement |
Require that there is a statement in the block. For example:
finally { lock.release(); } |
This type represents the policy for the styles for the ending parenthesis. The following table describes the valid options:
Option | Definition |
always |
Example:
@Deprecated() |
never |
Example:
@Deprecated |
ignore |
Anything goes. |
This type represents the policy for the styles for defining elements in an annotation. The following table describes the valid options:
Option | Definition |
expanded |
The expanded version is sometimes referred to as "named parameters"
in other languages.
Example:
@SuppressWarnings(value={"unchecked","unused",}) @SuppressWarnings({"unchecked","unused",}) |
compact |
This style can only be used when there is an element called 'value'
which is either the sole element or all other elements have default
values.
Example:
@SuppressWarnings({"unchecked","unused",}) @SuppressWarnings("unchecked") @SuppressWarnings(value = {"unchecked","unused",}) @SuppressWarnings(value = "unchecked") |
compact_no_array |
It is similar to the compact style but
single value arrays are flagged. With annotations a single value
array does not need to be placed in an array initializer.
Example:
@SuppressWarnings("unchecked") @MyAnnotation(someArray = "some value") @SuppressWarnings({"unchecked"}) @MyAnnotation(someArray = {"some value"}) |
ignore |
Anything goes. |
This type represents a local file. Unlike the uri type, which implies read only and not modifiable access to the resource, a property of this type indicates files whose contents can be modified by Checkstyle.
This type represents the policy for checking imports order. The following table describes the valid options:
Option | Definition |
top |
All static imports are at the top.
Groups for static import are defined by the property 'staticGroups'.
The blank line between groups is driven by the property 'separatedStaticGroups'.
For example:
import static a.b.C.*; import static x.y.Z.*; import a.b.D; import x.y.Z; |
above |
All static imports are above the local group. For example:
import static a.b.C.*; import a.b.D; import static x.y.Z.*; import x.y.Z; |
inflow |
All static imports are processed like non static
imports. For example:
import static a.b.C.*; import a.b.D; import x.y.Z; import static x.y.Z.*; |
under |
All static imports are under the local group. For example:
import a.b.D; import static a.b.C.*; import x.y.Z; import static x.y.Z.*; |
bottom |
All static imports are at the bottom.
Groups for static import are defined by the property 'staticGroups'.
The blank line between groups is driven by the property 'separatedStaticGroups'.
For example:
import a.b.D; import x.y.Z; import static a.b.C.*; import static x.y.Z.*; |
This type represents policy on placement of the Javadoc content. The following table describes the valid options:
Option | Definition |
FIRST_LINE |
Represents the policy for Javadoc content starts from the same line
as /** .
Example:
/** Summary text. * More details. */ public void method(); |
SECOND_LINE |
Represents the policy for Javadoc content starts from the next line
after /** .
Example:
/** * Summary text. * More details. */ public void method(); |
This type represents the policy for checking the placement of a
left curly brace ('{'
). The following table
describes the valid options:
Option | Definition |
eol |
The brace must always be on the end of the line. For example:#
if (condition) { ... |
nl |
The brace must always be on a new line. For example:
if (condition) { ... |
nlow |
If the statement/expression/declaration connected to the brace spans multiple lines,
then apply nl rule. Otherwise apply the eol rule.
nlow is a mnemonic for "new line on wrap".
For the example above Checkstyle will enforce:
if (condition) { ... if (condition1 && condition2 && condition3 && condition4) { ... |
This type represents the policy for line returns. The following table describes the valid options:
Option | Definition |
crlf |
Windows-style |
cr |
Mac-style |
lf |
Unix-style |
lf_cr_crlf |
lf, cr or crlf |
system |
system default |
This type represents the policy for padding with white space. The following table describes the valid options:
Option | Definition |
nospace |
Do not pad. For example, method(a, b);
|
space |
Ensure padding. For example,
method( a, b );
|
This type represents a regular expression. The string representation is parsed using java.util.regex package.
This type represents a set of patterns. The string representation is parsed as a set of forward slash ('/') separated patterns.
This type represents a regular expression. The string representation is parsed using java.util.regex package.
This type represents the policy for checking the placement of a
right curly brace ('}'
) in blocks but not blocks of expressions.
For right curly brace of expression blocks please follow issue
#5945. The following
table describes the valid options:
Option | Definition |
alone |
The brace must be alone on the line. For example:
try { ... } finally { ... } |
alone_or_singleline |
The brace must be alone on the line, yet
single-line format of block is allowed.
For example:
// Brace is alone on the line try { ... } finally { ... } // Single-line format of block public long getId() { return id; } |
same |
Works like alone_or_singleline but the brace should be on the same line
as the next part of a multi-block statement
(one that directly contains multiple blocks: if/else-if/else or try/catch/finally).
If no next part of a multi-block statement present, brace must be alone on line.
It also allows single-line format of multi-block statements.
Examples: public long getId() {return id;} // this is OK, it is single line // try-catch-finally blocks try { ... } catch (Exception ex) { // this is OK ... } finally { // this is OK ... } try { ... } // this is NOT OK, not on the same line as the next part of a multi-block statement catch (Exception ex) { ... } // this is NOT OK, not on the same line as the next part of a multi-block statement finally { ... } // if-else blocks if (a > 0) { ... } else { // this is OK ... } if (a > 0) { ... } // this is NOT OK, not on the same line as the next part of a multi-block statement else { ... } if (a > 0) { ... } int i = 5; // NOT OK, no next part of a multi-block statement, so should be alone Thread t = new Thread(new Runnable() { @Override public void run() { ... } // this is OK, should be alone as next part of a multi-block statement is absent }); // this case is out of scope of RightCurly Check (see issue #5945) if (a > 0) { ... } // OK, single-line multi-block statement if (a > 0) { ... } else { ... } // OK, single-line multi-block statement if (a > 0) { ... } else { ... } // OK, single-line multi-block statement |
This type represents a Java scope. Checks use this to determine which methods/fields/classes will be examined by it's logic. The valid options are:
nothing
public
protected
package
private
anoninner
Using a specific scope means not only will that modifier be examined, but also all the
modifiers listed above it in the previous list. Specifying public
means only
items with public
modifiers are checked. Specifying protected
means only public
and protected
modifiers are checked.
If you wish to only validate items with private
modifiers and ignore any
others, then you must set the exclude scope property, if available, to the scope above
it in the table. In this case you would exclude package
.
This type represents the severity level of a check violation. The valid options are:
ignore
info
warning
error
This type represents a string. The literal string representation is used.
This type represents a set of strings. The string representation is parsed as a set of comma (',') separated strings.
Alternatively, a property of this type can be supplied multiple times which is equivalent to a set of comma separated strings. For example, the following:
<property name="tokens" value="DIV_ASSIGN,PLUS_ASSIGN"/>
can instead be expressed as:
<property name="tokens" value="DIV_ASSIGN"/> <property name="tokens" value="PLUS_ASSIGN"/>
This type represents the policy for the styles for the trailing array comma. The following table describes the valid options:
Option | Definition |
always |
Example:
@SuppressWarnings(value={"unchecked","unused",}) |
never |
Example:
@SuppressWarnings(value={"unchecked","unused"}) |
ignore |
Anything goes. |
This type represents a URI. The string representation is parsed using a custom routine to analyze what type of URI the string is. It can be a URL, regular file, or a resource path. It will try loading the path as a URL first, then as a file that must exist, and then finally as a resource on the classpath.
This type represents the policy for wrapping lines. The following table describes the valid options:
Option | Definition |
nl |
The token must be on a new line. For example:
someVariable = aBigVariableNameToMakeThings + "this may work" + lookVeryInteresting; |
eol |
The token must be at the end of the line. For example:
someVariable = aBigVariableNameToMakeThings + "this may work" + lookVeryInteresting; |