Class ParseConfiguration
- All Implemented Interfaces:
Serializable
,Cloneable
Models a series of parsing configurations that can be applied during document parsing
by MarkupParser
and its variants
SimpleMarkupParser
and DOMMarkupParser
.
Among others, the parameters that can be configured are:
- The parsing mode: XML or HTML.
- Whether to expect XML-well-formed code or not.
- Whether to perform automatic tag balancing or not.
- Whether we will allow parsing of markup fragments or just entire documents.
The htmlConfiguration()
and xmlConfiguration()
static methods act as starting points
for configuration. Once one of these pre-initialized configurations has been created, it can be
fine-tuned for the user's needs.
Note these configuration objects are mutable, so they should not be modified once they have been passed to a parser in order to initialize it.
Instances of this class can be cloned, so creating a variant of an already-tuned configuration is easy.
- Since:
- 2.0.0
- See Also:
-
Nested Class Summary
Nested ClassesModifier and TypeClassDescriptionstatic enum
Enumeration representing the possible actions to be taken with regard to element balancing:static enum
Enumeration used for determining the parsing mode, which will affect the parser's behaviour.static class
Class encapsulating the configuration parameters used for parsing and validating the "prolog" section of a markup document.static enum
Enumeration used for determining whether an element in the document prolog (DOCTYPE, XML Declaration) or the prolog itself should be allowed, required or even forbidden.static enum
Enumeration used for determining the behaviour the parser should have with respect to the presence and number of root elements in the parsed document. -
Field Summary
FieldsModifier and TypeFieldDescriptionprivate boolean
private static final ParseConfiguration
private static final ParseConfiguration
private ParseConfiguration.ParsingMode
private boolean
private static final long
private boolean
private boolean
private boolean
-
Constructor Summary
Constructors -
Method Summary
Modifier and TypeMethodDescriptionclone()
Returns the level of element balancing required at the document being parsed, enabling auto-closing of elements if needed.getMode()
Return the parsing mode to be used.Returns theParseConfiguration.PrologParseConfiguration
object determining the way in which prolog (XML Declaration, DOCTYPE) will be dealt with during parsing.This value determines whether it will be required that the document has a unique root element.static ParseConfiguration
Return an instance ofParseConfiguration
containing a valid configuration set for most HTML scenarios.boolean
Returns whether validations performed on the parsed document should be case sensitive or not (e.g.boolean
Returns whether unmatched close elements (those not matching any equivalent open elements) are allowed or not.boolean
Returns whether text fragments in markup can be split in more than one text node, if it occupies more than an entire buffer in size.boolean
Returns whether attributes should never appear duplicated in elements.boolean
Returns whether element attributes will be required to be well-formed from the XML standpoint.void
setCaseSensitive
(boolean caseSensitive) Specify whether validations performed on the parsed document should be case sensitive or not (e.g.void
setElementBalancing
(ParseConfiguration.ElementBalancing elementBalancing) Specify the level of element balancing required at the document being parsed, enabling auto-closing of elements if needed.void
Specify the parsing mode to be used.void
setNoUnmatchedCloseElementsRequired
(boolean noUnmatchedCloseElementsRequired) Specify whether unmatched close elements (those not matching any equivalent open elements) are allowed or not.void
setTextSplittable
(boolean textSplittable) Specify whether text fragments in markup can be split in more than one text node, if it occupies more than an entire buffer in size.void
setUniqueAttributesInElementRequired
(boolean uniqueAttributesInElementRequired) Returns whether attributes should never appear duplicated in elements.void
setUniqueRootElementPresence
(ParseConfiguration.UniqueRootElementPresence uniqueRootElementPresence) This value determines whether it will be required that the document has a unique root element.void
setXmlWellFormedAttributeValuesRequired
(boolean xmlWellFormedAttributeValuesRequired) Specify whether element attributes will be required to be well-formed from the XML standpoint.private static void
validateNotNull
(Object obj, String message) static ParseConfiguration
Return an instance ofParseConfiguration
containing a valid configuration set for most XML scenarios.
-
Field Details
-
serialVersionUID
private static final long serialVersionUID- See Also:
-
DEFAULT_HTML_PARSE_CONFIGURATION
-
DEFAULT_XML_PARSE_CONFIGURATION
-
mode
-
caseSensitive
private boolean caseSensitive -
textSplittable
private boolean textSplittable -
elementBalancing
-
noUnmatchedCloseElementsRequired
private boolean noUnmatchedCloseElementsRequired -
xmlWellFormedAttributeValuesRequired
private boolean xmlWellFormedAttributeValuesRequired -
uniqueAttributesInElementRequired
private boolean uniqueAttributesInElementRequired -
prologParseConfiguration
-
uniqueRootElementPresence
-
-
Constructor Details
-
ParseConfiguration
private ParseConfiguration()
-
-
Method Details
-
htmlConfiguration
Return an instance of
ParseConfiguration
containing a valid configuration set for most HTML scenarios.- Mode:
ParseConfiguration.ParsingMode.HTML
- Text splittable: false
- Element balancing:
ParseConfiguration.ElementBalancing.AUTO_CLOSE
- No unmatched close elements required: false
- Unique attributes in elements required: false
- Xml-well-formed attribute values required: false
- Unique root element presence:
ParseConfiguration.UniqueRootElementPresence.NOT_VALIDATED
- Validate Prolog: false
- Returns:
- a valid default configuration object for HTML parsing.
- Mode:
-
xmlConfiguration
Return an instance of
ParseConfiguration
containing a valid configuration set for most XML scenarios.- Mode:
ParseConfiguration.ParsingMode.XML
- Text splittable: false
- Element balancing:
ParseConfiguration.ElementBalancing.REQUIRE_BALANCED
- No unmatched close elements required: true
- Unique attributes in elements required: true
- Xml-well-formed attribute values required: true
- Unique root element presence:
ParseConfiguration.UniqueRootElementPresence.DEPENDS_ON_PROLOG_DOCTYPE
- Validate Prolog: true
- Prolog presence:
ParseConfiguration.PrologPresence.ALLOWED
- XML Declaration presence:
ParseConfiguration.PrologPresence.ALLOWED
- DOCTYPE presence:
ParseConfiguration.PrologPresence.ALLOWED
- Require DOCTYPE keyword to be uppercase: true
- Returns:
- a valid default configuration object for XML parsing.
- Mode:
-
getMode
Return the parsing mode to be used. Can be XML or HTML.
Depending on the selected mode parsers will behave differently, given HTML has some specific rules which are not XML-compatible (like void elements which might appear unclosed like <meta>.
- Returns:
- the parsing mode to be used.
-
setMode
Specify the parsing mode to be used. Can be XML or HTML.
Depending on the selected mode parsers will behave differently, given HTML has some specific rules which are not XML-compatible (like void elements which might appear unclosed like <meta>.
- Parameters:
mode
- the parsing mode to be used.
-
isCaseSensitive
public boolean isCaseSensitive()Returns whether validations performed on the parsed document should be case sensitive or not (e.g. attribute names, document root element name, element open vs close elements, etc.)
HTML requires this parameter to be false. Default for XML is true.
- Returns:
- whether validations should be case sensitive or not.
-
setCaseSensitive
public void setCaseSensitive(boolean caseSensitive) Specify whether validations performed on the parsed document should be case sensitive or not (e.g. attribute names, document root element name, element open vs close elements, etc.)
HTML requires this parameter to be false. Default for XML is true.
- Parameters:
caseSensitive
- whether validations should be case sensitive or not.
-
isTextSplittable
public boolean isTextSplittable()Returns whether text fragments in markup can be split in more than one text node, if it occupies more than an entire buffer in size.
Default is false.
- Returns:
- whether text fragments can be split or not.
-
setTextSplittable
public void setTextSplittable(boolean textSplittable) Specify whether text fragments in markup can be split in more than one text node, if it occupies more than an entire buffer in size.
Default is false.
- Parameters:
textSplittable
- whether text fragments can be split or not.
-
getElementBalancing
Returns the level of element balancing required at the document being parsed, enabling auto-closing of elements if needed.
Possible values are:
ParseConfiguration.ElementBalancing.NO_BALANCING
: Do not perform element balancing checks at all. Events will be reported as they appear. There is no guarantee that a DOM tree can be built from the fired events though.ParseConfiguration.ElementBalancing.REQUIRE_BALANCED
: Require that elements are already correctly balanced in markup, throwing an exception if not. Note that when in HTML mode, this does not require the specification of optional tags such as <tbody>. Also note that this will automatically consider thesetNoUnmatchedCloseElementsRequired(boolean)
flag to be set to true.ParseConfiguration.ElementBalancing.AUTO_OPEN_CLOSE
: Auto open and close elements, which includes both those elements that, according to the HTML spec (when in HTML mode) have optional start or end tags (see http://www.w3.org/html/wg/drafts/html/master/syntax.html#optional-tags) and those that simply are unclosed at the moment a parent element needs to be closed (so their closing is forced). As an example of optional tags, the HTML5 spec establishes that <html>, <body> and <tbody> are optional, and that an <li> will close any currently open <li> elements. This is not really ill-formed code, but something allowed by the spec. All of these will be reported as auto-* events by the parser.ParseConfiguration.ElementBalancing.AUTO_CLOSE
: Equivalent toParseConfiguration.ElementBalancing.AUTO_OPEN_CLOSE
but not performing any auto-open operations, so that processing of HTML fragments is possible (no <html> or <body> elements are automatically added).
- Returns:
- the level of element balancing.
-
setElementBalancing
Specify the level of element balancing required at the document being parsed, enabling auto-closing of elements if needed.
Possible values are:
ParseConfiguration.ElementBalancing.NO_BALANCING
: Do not perform element balancing checks at all. Events will be reported as they appear. There is no guarantee that a DOM tree can be built from the fired events though.ParseConfiguration.ElementBalancing.REQUIRE_BALANCED
: Require that elements are already correctly balanced in markup, throwing an exception if not. Note that when in HTML mode, this does not require the specification of optional tags such as <tbody>. Also note that this will automatically consider thesetNoUnmatchedCloseElementsRequired(boolean)
flag to be set to true.ParseConfiguration.ElementBalancing.AUTO_OPEN_CLOSE
: Auto open and close elements, which includes both those elements that, according to the HTML spec (when in HTML mode) have optional start or end tags (see http://www.w3.org/html/wg/drafts/html/master/syntax.html#optional-tags) and those that simply are unclosed at the moment a parent element needs to be closed (so their closing is forced). As an example of optional tags, the HTML5 spec establishes that <html>, <body> and <tbody> are optional, and that an <li> will close any currently open <li> elements. This is not really ill-formed code, but something allowed by the spec. All of these will be reported as auto-* events by the parser.ParseConfiguration.ElementBalancing.AUTO_CLOSE
: Equivalent toParseConfiguration.ElementBalancing.AUTO_OPEN_CLOSE
but not performing any auto-open operations, so that processing of HTML fragments is possible (no <html> or <body> elements are automatically added).
- Parameters:
elementBalancing
- the level of element balancing.
-
getPrologParseConfiguration
Returns the
ParseConfiguration.PrologParseConfiguration
object determining the way in which prolog (XML Declaration, DOCTYPE) will be dealt with during parsing.- Returns:
- the configuration object.
-
isNoUnmatchedCloseElementsRequired
public boolean isNoUnmatchedCloseElementsRequired()Returns whether unmatched close elements (those not matching any equivalent open elements) are allowed or not.
- Returns:
- whether unmatched close elements will be allowed (false) or not (true).
-
setNoUnmatchedCloseElementsRequired
public void setNoUnmatchedCloseElementsRequired(boolean noUnmatchedCloseElementsRequired) Specify whether unmatched close elements (those not matching any equivalent open elements) are allowed or not.
- Parameters:
noUnmatchedCloseElementsRequired
- whether unmatched close elements will be allowed (false) or not (true).
-
isXmlWellFormedAttributeValuesRequired
public boolean isXmlWellFormedAttributeValuesRequired()Returns whether element attributes will be required to be well-formed from the XML standpoint. This means:
- Attributes should always have a value.
- Attribute values should be surrounded by double-quotes.
- Returns:
- whether attributes should be XML-well-formed or not.
-
setXmlWellFormedAttributeValuesRequired
public void setXmlWellFormedAttributeValuesRequired(boolean xmlWellFormedAttributeValuesRequired) Specify whether element attributes will be required to be well-formed from the XML standpoint. This means:
- Attributes should always have a value.
- Attribute values should be surrounded by double-quotes.
- Parameters:
xmlWellFormedAttributeValuesRequired
- whether attributes should be XML-well-formed or not.
-
isUniqueAttributesInElementRequired
public boolean isUniqueAttributesInElementRequired()Returns whether attributes should never appear duplicated in elements.
- Returns:
- whether attributes should never appear duplicated in elements.
-
setUniqueAttributesInElementRequired
public void setUniqueAttributesInElementRequired(boolean uniqueAttributesInElementRequired) Returns whether attributes should never appear duplicated in elements.
- Parameters:
uniqueAttributesInElementRequired
- whether attributes should never appear duplicated in elements.
-
getUniqueRootElementPresence
This value determines whether it will be required that the document has a unique root element.
If set to
ParseConfiguration.UniqueRootElementPresence.REQUIRED_ALWAYS
, then a document with more than one elements at the root level will never be considered valid. And ifParseConfiguration.PrologParseConfiguration.isValidateProlog()
is true and there is a DOCTYPE clause, it will be checked that the root name established at the DOCTYPE clause is the same as the document's element root.If set to
ParseConfiguration.UniqueRootElementPresence.DEPENDS_ON_PROLOG_DOCTYPE
, then:- If
ParseConfiguration.PrologParseConfiguration.isValidateProlog()
is false, multiple document root elements will be allowed. - If
ParseConfiguration.PrologParseConfiguration.isValidateProlog()
is true:- If there is a DOCTYPE clause, a unique element root will be required, and its name will be checked against the name specified at the DOCTYPE clause.
- If there is no DOCTYPE clause (even if it is forbidden), multiple document root elements will be allowed.
If set to
ParseConfiguration.UniqueRootElementPresence.NOT_VALIDATED
, then nothing will be checked regarding the name of the root element/s.Default value is
ParseConfiguration.UniqueRootElementPresence.DEPENDS_ON_PROLOG_DOCTYPE
.- Returns:
- the configuration value for validating the presence of a unique root element.
- If
-
setUniqueRootElementPresence
public void setUniqueRootElementPresence(ParseConfiguration.UniqueRootElementPresence uniqueRootElementPresence) This value determines whether it will be required that the document has a unique root element.
If set to
ParseConfiguration.UniqueRootElementPresence.REQUIRED_ALWAYS
, then a document with more than one elements at the root level will never be considered valid. And ifParseConfiguration.PrologParseConfiguration.isValidateProlog()
is true and there is a DOCTYPE clause, it will be checked that the root name established at the DOCTYPE clause is the same as the document's element root.If set to
ParseConfiguration.UniqueRootElementPresence.DEPENDS_ON_PROLOG_DOCTYPE
, then:- If
ParseConfiguration.PrologParseConfiguration.isValidateProlog()
is false, multiple document root elements will be allowed. - If
ParseConfiguration.PrologParseConfiguration.isValidateProlog()
is true:- If there is a DOCTYPE clause, a unique element root will be required, and its name will be checked against the name specified at the DOCTYPE clause.
- If there is no DOCTYPE clause (even if it is forbidden), multiple document root elements will be allowed.
If set to
ParseConfiguration.UniqueRootElementPresence.NOT_VALIDATED
, then nothing will be checked regarding the name of the root element/s.Default value is
ParseConfiguration.UniqueRootElementPresence.DEPENDS_ON_PROLOG_DOCTYPE
.- Parameters:
uniqueRootElementPresence
- the configuration value for validating the presence of a unique root element.
- If
-
clone
- Overrides:
clone
in classObject
- Throws:
CloneNotSupportedException
-
validateNotNull
-