com.pervasive.datarush.schema

Class TextRecord

java.lang.Object

com.pervasive.datarush.schema.TextRecord

All Implemented Interfaces:

RecordTextSchema<TextDataType>, TextSchema

public class TextRecord
extends Object

A definition of a variable-width record type in a text file. Because fields are not necessarily fixed-size, additional information must be provided to a structured text reader or writer so that fields can be identified correctly.

Field Summary

Fields

Modifier and Type	Field and Description
static TextRecordDiscoverer	DEFAULT_AUTODISCOVER The default schema discoverer used for reading delimited text.
static TextRecordDiscoverer	TEXT_FIELD_DISCOVER A schema discoverer which treats all fields as raw text.
protected Map<String,TextField<T>>	textFieldsByName

Constructor Summary

Constructors

Constructor and Description
TextRecord() Defines an empty record.
TextRecord(TextConversionDefaults defaults) Defines an empty record with the specified text conversion behavior.
TextRecord(TextConversionDefaults defaults, Map<String,TextDataType> fieldTypes) Defines a record with the specified text conversion behavior and fields.

Method Summary

Methods

Modifier and Type	Method and Description
static TextRecord	convert(RecordTokenType type) Generates a text file schema from a token type.
static TextRecord	convert(RecordTokenType type, TextConversionDefaults defaults) Generates a text file schema from a token type, using the specified schema-wide default behavior.
static TextRecord	convert(RecordTokenType type, TextTypes.StringConversion behavior) Generates a text file schema from a token type, using the specified string conversion default behavior.
TokenFormatter	createFieldFormatter(String fieldName) Creates a new formatter for values of the specified field.
TokenParser	createFieldParser(String fieldName) Creates a new parser for values of the specified field.
void	defineField(String name, T type) Defines a new field in the record with the specified name and type.
void	defineField(String name, T type, FieldDomain domain) Defines a new field in the record with the specified name and type.
static TextRecordDiscoverer	extendDefault(List<TypePattern> additions) Creates a new discoverer which extends the default patterns.
TextConversionDefaults	getDefaults() Gets the default conversion behaviors for fields in the schema which do not explicitly define any.
int	getFieldCount() Gets the number of fields defined in the record.
FieldDomain	getFieldDomain(String name) Gets the domain of the specified field in the record.
List<String>	getFieldNames() Gets the names of the fields defined in the record.
T	getFieldType(String name) Gets the type of the specified fields in the record.
List<TextField<TextDataType>>	getTextFields()
RecordTokenType	getTokenType() Gets the type of the token representation of the text record.
boolean	isFieldDefined(String name) Indicates whether a field with the given name is already defined.
void	setFieldType(String name, T newType) Modifies the value conversion scheme for an existing field.

Methods inherited from class java.lang.Object

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

Field Detail

DEFAULT_AUTODISCOVER

public static final TextRecordDiscoverer DEFAULT_AUTODISCOVER

The default schema discoverer used for reading delimited text. This is the same discoverer as obtained by creating a new PatternBasedDiscovery.

TEXT_FIELD_DISCOVER

public static final TextRecordDiscoverer TEXT_FIELD_DISCOVER

A schema discoverer which treats all fields as raw text. When used with ReadDelimitedText, the resulting schema identifies the field values without interpreting them.

The output type produced is compatible for use as input to ParseTextFields.

textFieldsByName

protected final Map<String,TextField<T extends TextDataType>> textFieldsByName

Constructor Detail

TextRecord

public TextRecord()

Defines an empty record. Default text conversion behavior is used.

An empty record definition is not meaningful; at least one field must be defined before the schema is used.

See Also:

TextConversionDefaults

TextRecord

public TextRecord(TextConversionDefaults defaults)

Defines an empty record with the specified text conversion behavior.

An empty record definition is not meaningful; at least one field must be defined before the schema is used.

Parameters:

defaults - text conversion behavior to apply by default to the schema

TextRecord

public TextRecord(TextConversionDefaults defaults,
          Map<String,TextDataType> fieldTypes)

Defines a record with the specified text conversion behavior and fields. The field order is the iteration order of the provided field mapping.

Parameters:

defaults - default text conversion behavior for the schema

fieldTypes - the defined fields in the record schema default to the schema

Method Detail

convert

public static final TextRecord convert(RecordTokenType type)

Generates a text file schema from a token type. The names and order of fields in the resulting schema are the same as in the record token type. Default conversion for each token type will be used.

Parameters:

type - the template type for which to create a text schema

Returns:

a new schema which maps to the specified token type

convert

public static final TextRecord convert(RecordTokenType type,
                 TextTypes.StringConversion behavior)

Generates a text file schema from a token type, using the specified string conversion default behavior. The names and order of fields in the resulting schema are the same as in the record token type. Default conversion for each token type will be used, with strings following the requested conversion behavior.

Parameters:

type - the template type for which to create a text schema

behavior - the schema-wide default behavior for string conversion

Returns:

a new schema which maps to the specified token type

getTextFields

public List<TextField<TextDataType>> getTextFields()

convert

public static final TextRecord convert(RecordTokenType type,
                 TextConversionDefaults defaults)

Generates a text file schema from a token type, using the specified schema-wide default behavior. The names and order of fields in the resulting schema are the same as in the record token type. Default conversion for each token type will be used, following the requested behavior where applicable.

Parameters:

type - the template type for which to create a text schema

defaults - the schema-wide default settings for text conversion

Returns:

a new schema which maps to the specified token type

extendDefault

public static TextRecordDiscoverer extendDefault(List<TypePattern> additions)

Creates a new discoverer which extends the default patterns. The specified patterns take precedence over default patterns; this is equivalent to prepending the additions to the PatternBasedDiscovery.DEFAULT_PATTERNS.

Parameters:

additions - additional patterns to use in determining field types

Returns:

a new discoverer for automatically generating a schema for a delimited text file

getDefaults

public final TextConversionDefaults getDefaults()

Description copied from interface: RecordTextSchema

Gets the default conversion behaviors for fields in the schema which do not explicitly define any. These include settings such as the null indicator for values and whether string values are trimmed or preserved as is.

Specified by:

getDefaults in interface RecordTextSchema<T extends TextDataType>

Returns:

the default behavior for the schema

getTokenType

public final RecordTokenType getTokenType()

Description copied from interface: RecordTextSchema

Gets the type of the token representation of the text record. The fields of the resulting type are in the same order and have the same name as the fields in the schema.

Specified by:

getTokenType in interface RecordTextSchema<T extends TextDataType>

Returns:

the token type representing the schema

getFieldCount

public final int getFieldCount()

Description copied from interface: RecordTextSchema

Gets the number of fields defined in the record.

Specified by:

getFieldCount in interface RecordTextSchema<T extends TextDataType>

Returns:

the number of defined fields

getFieldNames

public final List<String> getFieldNames()

Description copied from interface: RecordTextSchema

Gets the names of the fields defined in the record. The names are returned in the order they were defined.

Specified by:

getFieldNames in interface RecordTextSchema<T extends TextDataType>

Returns:

a list of defined field names

isFieldDefined

public final boolean isFieldDefined(String name)

Description copied from interface: RecordTextSchema

Indicates whether a field with the given name is already defined.

Specified by:

isFieldDefined in interface RecordTextSchema<T extends TextDataType>

Parameters:

name - the field to check

Returns:

true if a field with the name is defined in the schema, false otherwise.

defineField

public final void defineField(String name,
               T type)

Defines a new field in the record with the specified name and type.

Parameters:

name - the name associated with the field. The name must be unique amongst all defined fields.

type - specifies how to convert between text and token values

defineField

public final void defineField(String name,
               T type,
               FieldDomain domain)

Defines a new field in the record with the specified name and type.

Parameters:

name - the name associated with the field. The name must be unique amongst all defined fields.

type - specifies how to convert between text and token values

getFieldType

public final T getFieldType(String name)

Description copied from interface: RecordTextSchema

Gets the type of the specified fields in the record. The names are returned in the order they were defined.

Specified by:

getFieldType in interface RecordTextSchema<T extends TextDataType>

Parameters:

name - the field for which to fetch the type

Returns:

the text type of the named field

getFieldDomain

public final FieldDomain getFieldDomain(String name)

Description copied from interface: RecordTextSchema

Gets the domain of the specified field in the record.

Specified by:

getFieldDomain in interface RecordTextSchema<T extends TextDataType>

Parameters:

name - the field for which to fetch the domain

Returns:

the domain of the named field

setFieldType

public final void setFieldType(String name,
                T newType)

Modifies the value conversion scheme for an existing field.

Parameters:

name - the field to modify. This field must exist in the schema.

newType - specifies how to convert between text and token values

createFieldParser

public final TokenParser createFieldParser(String fieldName)

Description copied from interface: RecordTextSchema

Creates a new parser for values of the specified field.

Specified by:

createFieldParser in interface RecordTextSchema<T extends TextDataType>

Parameters:

fieldName - the field for which to obtain a parser

Returns:

a parser for converting text values to token values for the field

createFieldFormatter

public final TokenFormatter createFieldFormatter(String fieldName)

Description copied from interface: RecordTextSchema

Creates a new formatter for values of the specified field.

Specified by:

createFieldFormatter in interface RecordTextSchema<T extends TextDataType>

Parameters:

fieldName - the field for which to obtain a formatter

Returns:

a formatter for converting token values to text values for the field