Field Definitions

Defining which fields should appear in the output and whether and how they need to be used in matching

fieldDefinition

This is a JSON array representing the fields from the source data to be used for matching, and the kind of matching they need.

Each field denotes a column from the input. Fields have the following JSON attributes:

fieldName

The name of the field from the input data schema

fields

To be defined later. For now, please keep this as the fieldName

dataType

Type of the column - string, integer, double, etc.

matchType

The way to match the given field. Multiple match types, separated by commas, can also be used. Here are the different types supported.

showConcise

Match Type	Description	Can be applied to
FUZZY	Broad matches with typos, abbreviations, and other variations.	string, integer, double, date
EXACT	No tolerance with variations, Preferable for country codes, pin codes, and other categorical variables where you expect no variations.	string
DONT_USE	Appears in the output but no computation is done on these. Helpful for fields like ids that are required in the output. DONT_USE fields are not shown to the user while labeling, if showConcise is set to true.	any
EMAIL	Matches only the id part of the email before the @ character	any
PINCODE	Matches pin codes like xxxxx-xxxx with xxxxx	string
NULL_OR___BLANK	By default Zingg marks matches as	string
TEXT	Compares words overlap between two strings.	string
NUMERIC	extracts numbers from strings and compares how many of them are same across both strings
NUMERIC_WITH_UNITS	extracts product codes or numbers with units, for example 16gb from strings and compares how many are same across both strings	string
ONLY_ALPHABETS_EXACT	only looks at the alphabetical characters and compares if they are exactly the same	string
ONLY_ALPHABETS_FUZZY	ignores any numbers in the strings and then does a fuzzy comparison	string