• A column reference can refer to a column of String, Object, or Array type, which makes the FIND function useful for filtering data before it has been completely un-nested into tabular data.
• Returned value is from the beginning of the string, regardless of the string index value.
• If no match is found, the function returns a null value.
• If you need to determine if a value is in an array or not, you can use the MATCHES function, which returns a true/false response. See MATCHES Function.

You can also search a string from the right. For more information, see RIGHTFIND Function.

Column reference example:

Output: Searches the MyName column value for the string find this from the beginning of the value, ignoring case. If a match is found, the index value where the string is located is returned. 

String literal example:

Output: Searches the string Hello, World for the string lo, in a case-sensitive search, beginning at the third character in the string. Since the match is found at the fourth character, the value 3 is returned.

If example:

Output: Searches the SearchPool column value for the string FindIt from the beginning of the value (default). Default behavior is to not ignore case. If the string is found, the value found it is returned. Otherwise, the column is empty.

string_to_search

Name of the item to be searched. Valid values can be:

• String literals must be quoted ( 'Hello, World' ).
• Column reference to any type that can be inferred as a string, which encompasses all values.
• Function returning a string value.

Missing values generate the start-at parameter value.

• Multiple values and wildcards are not supported.

search_for

Column of strings, function returning a string, string literal or pattern to find. An input value can be a literal, 

• Missing string or column values generate the start-at parameter value.
  
  • String literals must be quoted ('Hello, World').
• Multiple values and wildcards are not supported.
• Column names are not supported.

ignore_case

If true, the FIND function ignores case when trying to match the string literal or pattern value.

Default value is false, which means that case-sensitive matching is performed by default.

start_at

Indicates the position in the column or string literal value at which to begin the search. For example, a value of 2 instructs the FIND function to begin searching from the third character in the column or string value.

• Value can be an integer, a function returning an integer, or a column containing integers. 

  • If a column name is specified, start_at can be different for each row.

  • If a constant integer value is specified, the start_at is same for all rows.
    

• Value must be a non-negative integer value.

• If this value is greater than the length of the string, then no match is possible.

Example - Locate product purchases in transaction stream

Source:

You have the simplified transaction stream listed below in which master information about a transaction (TransactionId and CustomerId) is paired with order detail information that is brought into the application as an array in the OrderDetail column. The array column contains information about product ID, quantity, and the type of transaction. 

The transaction stream includes test transactions, which are identified by the value TEST for TransType in the detail column. You want to remove these transactions early in the process, which should simplify your dataset and speed up its processing.

Transformation:

First, you must identify the records that contain the test transaction value. The following transform generates a new column containing true/false values for whether the value "TEST" appears in the OrderDetail transform.

When the step is added to the recipe, the find_OrderDetail column is generated, containing the index value returned by the FIND function. In this case, there is only one row that contains a value: 53 for the third transaction.

You can then add the following step to keep the rows where the FIND function returned a null value in the find_OrderDetail column:

Results:

You can delete the find_OrderDetail column at this time.