Package eu.bandm.tools.dtm

Class AttributeChecks

java.lang.Object

eu.bandm.tools.dtm.AttributeChecks

Direct Known Subclasses:

AttributeChecks_valueTemplates

public class AttributeChecks
extends Object

Sorts all attributes' NamespaceName from the SAX event attributes part and from the DTM definition into the appropriate diagnosis sets ok, missing, violateType, violateFixed, or undeclared), according to the type check results.

"check()" should be called for a particular SAX event with a fresh instance of this class.

By explicitly calling sendErrorMessages_Undecl_Type_Fixed(MessageReceiver), all undeclared, type-violating and fixed-violating attributes are reported to the message channel with the given location Missing attributes are not, but must be treated explicitly by the caller.
The location information in theses error messages is (opaquely) the one supplied on creation, because in SAX contexts it comes with only coarse granular anyhow.
This is separated from the check(), because in the meantime in particular contexts attribute errors can be healed (cf xslt "attribute value templates", "use-attribute-set" attribute, etc.)
// OLD The "ignore" set is introduced for the "use-attribute-set" attribute (and some more) from xsl. FIXME nötig !? besser einfach nach UNDEF, dann wieder raus !?!
TDOM has its own checks, compiled into the constructor code, but these are not so complete (currently eg. undeclared atts are not recognized),
ATTENTION: Attributes of type ID IDREF IDREFS ENTITY ENTITIES and NOTATION must follow the "name" production, see XML section 3.3.1.
Attributes of type NMTOKEN, NMTOKENS and ENUMERATION need only follow the nmtoken production.
A name is an nmtoken which STARTS with an alphabetic character or underscore "_" or colon ":". An NMTOKEN may additionally start with "." or "-" or a digit.

Field Summary

Fields

Modifier and Type	Field	Description
protected final Attributes	atts
protected final DTM.Element	elDef
protected final Location<XMLDocumentIdentifier>	loc
protected Set<NamespaceName>	missing
protected Map<NamespaceName,Integer>	nsn2index
protected Set<NamespaceName>	ok
protected Set<NamespaceName>	undeclared
protected Set<NamespaceName>	violateFixed
protected Set<NamespaceName>	violateType

Constructor Summary

Constructors

Constructor	Description
AttributeChecks(Attributes atts, DTM.Element elDef, Location<XMLDocumentIdentifier> loc)

Method Summary

Modifier and Type	Method	Description
static String	attValueByNamespaceName(Attributes atts, NamespaceName n)
boolean	check()
protected boolean	check(String value, boolean mustBeName)
protected void	check_more_tokens(String value, boolean mustBeName, NamespaceName elname, NamespaceName attname)	FIXME must be split by exactly ONE space FIXME move to util/xml/AttributeValues
protected void	check_one_token(String value, boolean mustBeName, NamespaceName elname, NamespaceName attname)
void	dump(PrintStream p, boolean silentMode)
protected void	dump1(PrintStream p, Set<NamespaceName> set, String text)
Integer	get_index(NamespaceName nn)
Location<XMLDocumentIdentifier>	get_loc()
Set<NamespaceName>	get_missing()
Set<NamespaceName>	get_ok()
Set<NamespaceName>	get_undeclared()
Set<NamespaceName>	get_violateFixed()
Set<NamespaceName>	get_violateType()
boolean	noConstructionProhibitingErrors()
boolean	noErrors()
void	sendErrorMessages_Undecl_Type_Fixed(MessageReceiver<SimpleMessage<XMLDocumentIdentifier>> msg)	The attributes recognized as erronuous in XML sense, can later be healed when meant in a derived system, like xslt.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

nsn2index

protected Map<NamespaceName,Integer> nsn2index

missing

protected Set<NamespaceName> missing

violateType

protected Set<NamespaceName> violateType

violateFixed

protected Set<NamespaceName> violateFixed

undeclared

protected Set<NamespaceName> undeclared

ok

protected Set<NamespaceName> ok

loc

protected final Location<XMLDocumentIdentifier> loc

atts

protected final Attributes atts

elDef

protected final DTM.Element elDef

Constructor Details

AttributeChecks

public AttributeChecks(Attributes atts,
 DTM.Element elDef,
 Location<XMLDocumentIdentifier> loc)

Method Details

attValueByNamespaceName

public static String attValueByNamespaceName(Attributes atts,
 NamespaceName n)

get_missing

public Set<NamespaceName> get_missing()

get_violateType

public Set<NamespaceName> get_violateType()

get_violateFixed

public Set<NamespaceName> get_violateFixed()

get_undeclared

public Set<NamespaceName> get_undeclared()

get_ok

public Set<NamespaceName> get_ok()

get_index

public Integer get_index(NamespaceName nn)

get_loc

public Location<XMLDocumentIdentifier> get_loc()

check

public boolean check()

noErrors

public boolean noErrors()

noConstructionProhibitingErrors

public boolean noConstructionProhibitingErrors()

check

protected boolean check(String value,
 boolean mustBeName)

check_one_token

protected void check_one_token(String value,
 boolean mustBeName,
 NamespaceName elname,
 NamespaceName attname)

check_more_tokens

protected void check_more_tokens(String value,
 boolean mustBeName,
 NamespaceName elname,
 NamespaceName attname)

FIXME must be split by exactly ONE space FIXME move to util/xml/AttributeValues

dump

public void dump(PrintStream p,
 boolean silentMode)

dump1

protected void dump1(PrintStream p,
 Set<NamespaceName> set,
 String text)

sendErrorMessages_Undecl_Type_Fixed

public void sendErrorMessages_Undecl_Type_Fixed(MessageReceiver<SimpleMessage<XMLDocumentIdentifier>> msg)

The attributes recognized as erronuous in XML sense, can later be healed when meant in a derived system, like xslt. Then the categorization will be revisited. Therefore the generation of error messages must be postponed. The MISSING attributes are still postponed!