Contact Us 1-800-596-4880

Excel Format

MIME Type: application/xlsx

ID: excel

An Excel workbook is a sequence of sheets. In DataWeave, this is mapped to an object where each sheet is a key. Only one table is allowed per Excel sheet. A table is expressed as an array of rows. A row is an object where its keys are the columns and the values the cell content.

Only .xlsx files are supported (Excel 2007). .xls files are not supported by Mule.

The DataWeave reader for Excel input supports the following parsing strategies:

  • In-Memory

  • Streaming

To understand the parsing strategies that DataWeave readers and writers can apply to this format, see DataWeave Parsing Strategies.

Excel Type Mapping

The following table shows how Excel types map to DataWeave types.

Excel Type DataWeave Type

String

String

Numeric

Number

Boolean

Boolean

Data

Date

Examples

The following examples show uses of the Excel format.

Example: Represent Excel in the DataWeave (dw) Format

This example shows how DataWeave represents an Excel workbook.

Input

The Excel workbook (Sheet1) serves as an input payload for the DataWeave source.

Sheet1:
.   | A    | B
1   | Id   | Name
2   | 123  | George
3   | 456  | Lucas

Source

The DataWeave script transforms the Excel input payload to the DataWeave (dw) format and MIME type.

%dw 2.0
output application/dw
---
payload

Output

The DataWeave output looks like this. You can select values the same way you select values in other objects.

{
    "Sheet1": [
        {
                "A": 123,
                "B": "George"
        },
        {
                "A": 456,
                "B": "Lucas"
        }
    ]
}

Example: Output an Excel Table

The following DataWeave script outputs an Excel table with the header and fields.

The body of this DataWeave script is a DataWeave object that defines the content of the Excel sheet. The name of the sheet, Sheet1, is the key of this object. The value is an array of objects. Each object in the array contains a collection of key-value pairs. The keys in each pair are treated as header values for the spreadsheet. The values in each pair are treated as data values for a row in the sheet.

The output directive indicates that the output is the Excel format and MIME type. The header=true setting indicates that the output includes the header values.

%dw 2.0
output application/xlsx header=true
---
{
  Sheet1: [
    {
      Id: 123,
      Name: George
    },
    {
      Id: 456,
      Name: Lucas
    }
  ]
}

For another example, see Look Up Data in an Excel (XLSX) File.

Example: Stream Excel Input

By default, the Excel reader stores input data from an entire file in-memory if the file is 1.5MB or less. If the file is larger than 1.5 MB, the process writes the data to disk. For very large files, you can improve the performance of the reader by setting a streaming property to true.

The following Configuration XML for a Mule application streams an Excel file and transforms it to JSON.

<http:listener-config
    name="HTTP_Listener_config"
    doc:name="HTTP Listener config" >
  <http:listener-connection host="0.0.0.0" port="8081" />
</http:listener-config>
<flow name="streaming_flow" >
  <http:listener
    doc:name="Listener"
    config-ref="HTTP_Listener_config"
    path="/"
    outputMimeType="application/xlsx; streaming=true"/>
  <ee:transform doc:name="Transform Message" >
    <ee:message >
      <ee:set-payload ><![CDATA[%dw 2.0
output application/json
---
payload."Sheet Name" map ((row) -> {
    foo: row.a,
    bar: row.b
})]]></ee:set-payload>
    </ee:message>
  </ee:transform>
</flow>

The example:

  • Configures the HTTP listener to stream the XLSX input by setting outputMimeType="application/xlsx; streaming=true". In the Studio UI, you can use the MIME Type on the listener to application/xlsx and the Parameters for the MIME Type to Key streaming and Value true.

  • Uses a DataWeave script in the Transform Message component to iterate over each row in the XLSX payload (an XLSX sheet called "Sheet Name") and select the values of each cell in the row (using row.a, row.b). It assumes columns named a and b and maps the values from each row in those columns into foo and bar, respectively.

Output

The following image shows the Excel table output.

A table in an Excel format displaying IDs and names

Configuration Properties

DataWeave supports the following configuration properties for Excel.

Reader Properties (for Excel)

The Excel format accepts properties that provide instructions for reading input data.

Parameter Type Default Description

header

Boolean

true

Indicates whether the Excel table contains headers. Valid values are true or false.

ignoreEmptyLine

Boolean

true

Indicates whether to ignore empty line. Valid values are true or false.

streaming

Boolean

false

Introduced in Mule 4.2.2: Streaming is intended for processing a large file. When streaming is enabled, the reader accesses each row sequentially, keeping one row in memory at a time, instead of making all data available at once. Streaming does not permit random access to rows in the file. Use only if the entries are accessed sequentially. Valid values are true or false.

tableOffset

String

null

The position of the first cell in the table (<Column><Row> example A1).

zipBombCheck

Boolean

true

If set to false, the zip bomb check is turned off. Valid values are true or false.

Writer Properties (for Excel)

The Excel format accepts properties that provide instructions for writing output data.

Parameter Type Default Description

bufferSize

Number

8192

Size of the writer buffer.

deferred

Boolean

false

When set to true, DataWeave generates the output as a data stream, and the script’s execution is deferred until it is consumed. Valid values are true or false.

header

Boolean

true

Indicates whether the Excel table contains headers. Valid values are true or false. When there are no headers, column names are used (for example, A, B, C, and so on).

ignoreEmptyLine

Boolean

true

Indicates whether to ignore empty lines. Valid values are true or false.

tableOffset

String

null

The position of the first cell in the table (<Column><Row> example A1).

zipBombCheck

Boolean

true

If set to false, the zip bomb check is turned off. Valid values are true or false.

Supported MIME Types (for Excel)

DataWeave supports the following MIME type.

MIME Type

application/vnd.openxmlformats-officedocument.spreadsheetml.sheet

application/xlsx