Chapter 74. CSV DataFormat

Available as of Camel version 1.3

The CSV Data Format uses Apache Commons CSV to handle CSV payloads (Comma Separated Values) such as those exported/imported by Excel.

74.1. Options

The CSV dataformat supports 28 options which are listed below.

NameDefaultJava TypeDescription

formatRef

 

String

The reference format to use, it will be updated with the other format options, the default value is CSVFormat.DEFAULT

formatName

 

String

The name of the format to use, the default value is CSVFormat.DEFAULT

commentMarkerDisabled

false

Boolean

Disables the comment marker of the reference format.

commentMarker

 

String

Sets the comment marker of the reference format.

delimiter

 

String

Sets the delimiter to use. The default value is , (comma)

escapeDisabled

false

Boolean

Use for disabling using escape character

escape

 

String

Sets the escape character to use

headerDisabled

false

Boolean

Use for disabling headers

header

 

List

To configure the CSV headers

allowMissingColumnNames

false

Boolean

Whether to allow missing column names.

ignoreEmptyLines

false

Boolean

Whether to ignore empty lines.

ignoreSurroundingSpaces

false

Boolean

Whether to ignore surrounding spaces

nullStringDisabled

false

Boolean

Used to disable null strings

nullString

 

String

Sets the null string

quoteDisabled

false

Boolean

Used to disable quotes

quote

 

String

Sets the quote which by default is

recordSeparatorDisabled

 

String

Used for disabling record separator

recordSeparator

 

String

Sets the record separator (aka new line) which by default is new line characters (CRLF)

skipHeaderRecord

false

Boolean

Whether to skip the header record in the output

quoteMode

 

String

Sets the quote mode

ignoreHeaderCase

false

Boolean

Sets whether or not to ignore case when accessing header names.

trim

false

Boolean

Sets whether or not to trim leading and trailing blanks.

trailingDelimiter

false

Boolean

Sets whether or not to add a trailing delimiter.

lazyLoad

false

Boolean

Whether the unmarshalling should produce an iterator that reads the lines on the fly or if all the lines must be read at one.

useMaps

false

Boolean

Whether the unmarshalling should produce maps (HashMap)for the lines values instead of lists. It requires to have header (either defined or collected).

useOrderedMaps

false

Boolean

Whether the unmarshalling should produce ordered maps (LinkedHashMap) for the lines values instead of lists. It requires to have header (either defined or collected).

recordConverterRef

 

String

Refers to a custom CsvRecordConverter to lookup from the registry to use.

contentTypeHeader

false

Boolean

Whether the data format should set the Content-Type header with the type from the data format if the data format is capable of doing so. For example application/xml for data formats marshalling to XML, or application/json for data formats marshalling to JSon etc.

74.2. Marshalling a Map to CSV

The component allows you to marshal a Java Map (or any other message type that can be converted in a Map) into a CSV payload.

Considering the following body

Map<String, Object> body = new LinkedHashMap<>();
body.put("foo", "abc");
body.put("bar", 123);

and this Java route definition

from("direct:start")
    .marshal().csv()
    .to("mock:result");

or this XML route definition

<route>
    <from uri="direct:start" />
    <marshal>
        <csv />
    </marshal>
    <to uri="mock:result" />
</route>

then it will produce

abc,123

74.3. Unmarshalling a CSV message into a Java List

Unmarshalling will transform a CSV messsage into a Java List with CSV file lines (containing another List with all the field values).

An example: we have a CSV file with names of persons, their IQ and their current activity.

Jack Dalton, 115, mad at Averell
Joe Dalton, 105, calming Joe
William Dalton, 105, keeping Joe from killing Averell
Averell Dalton, 80, playing with Rantanplan
Lucky Luke, 120, capturing the Daltons

We can now use the CSV component to unmarshal this file:

from("file:src/test/resources/?fileName=daltons.csv&noop=true")
    .unmarshal().csv()
    .to("mock:daltons");

The resulting message will contain a List<List<String>> like…​

List<List<String>> data = (List<List<String>>) exchange.getIn().getBody();
for (List<String> line : data) {
    LOG.debug(String.format("%s has an IQ of %s and is currently %s", line.get(0), line.get(1), line.get(2)));
}

74.4. Marshalling a List<Map> to CSV

Available as of Camel 2.1

If you have multiple rows of data you want to be marshalled into CSV format you can now store the message payload as a List<Map<String, Object>> object where the list contains a Map for each row.

74.5. File Poller of CSV, then unmarshaling

Given a bean which can handle the incoming data…​

MyCsvHandler.java

// Some comments here
public void doHandleCsvData(List<List<String>> csvData)
{
    // do magic here
}
  1. your route then looks as follows
<route>
        <!-- poll every 10 seconds -->
        <from uri="file:///some/path/to/pickup/csvfiles?delete=true&amp;consumer.delay=10000" />
        <unmarshal><csv /></unmarshal>
        <to uri="bean:myCsvHandler?method=doHandleCsvData" />
</route>

74.6. Marshaling with a pipe as delimiter

Considering the following body

Map<String, Object> body = new LinkedHashMap<>();
body.put("foo", "abc");
body.put("bar", 123);

and this Java route definition

// Camel version < 2.15
CsvDataFormat oldCSV = new CsvDataFormat();
oldCSV.setDelimiter("|");
from("direct:start")
    .marshal(oldCSV)
    .to("mock:result")

// Camel version >= 2.15
from("direct:start")
    .marshal(new CsvDataFormat().setDelimiter(&#39;|&#39;))
    .to("mock:result")

or this XML route definition

<route>
  <from uri="direct:start" />
  <marshal>
    <csv delimiter="|" />
  </marshal>
  <to uri="mock:result" />
</route>

then it will produce

abc|123

Using autogenColumns, configRef and strategyRef attributes inside XML # DSL

Available as of Camel 2.9.2 / 2.10 and deleted for Camel 2.15

You can customize the CSV Data Format to make use of your own CSVConfig and/or CSVStrategy. Also note that the default value of the autogenColumns option is true. The following example should illustrate this customization.

<route>
  <from uri="direct:start" />
  <marshal>
    <!-- make use of a strategy other than the default one which is 'org.apache.commons.csv.CSVStrategy.DEFAULT_STRATEGY' -->
    <csv autogenColumns="false" delimiter="|" configRef="csvConfig" strategyRef="excelStrategy" />
  </marshal>
  <convertBodyTo type="java.lang.String" />
  <to uri="mock:result" />
</route>

<bean id="csvConfig" class="org.apache.commons.csv.writer.CSVConfig">
  <property name="fields">
    <list>
      <bean class="org.apache.commons.csv.writer.CSVField">
        <property name="name" value="orderId" />
      </bean>
      <bean class="org.apache.commons.csv.writer.CSVField">
        <property name="name" value="amount" />
      </bean>
    </list>
  </property>
</bean>

<bean id="excelStrategy" class="org.springframework.beans.factory.config.FieldRetrievingFactoryBean">
  <property name="staticField" value="org.apache.commons.csv.CSVStrategy.EXCEL_STRATEGY" />
</bean>

74.7. Using skipFirstLine option while unmarshaling

Available as of Camel 2.10 and deleted for Camel 2.15

You can instruct the CSV Data Format to skip the first line which contains the CSV headers. Using the Spring/XML DSL:

<route>
  <from uri="direct:start" />
  <unmarshal>
    <csv skipFirstLine="true" />
  </unmarshal>
  <to uri="bean:myCsvHandler?method=doHandleCsv" />
</route>

Or the Java DSL:

CsvDataFormat csv = new CsvDataFormat();
csv.setSkipFirstLine(true);

from("direct:start")
  .unmarshal(csv)
.to("bean:myCsvHandler?method=doHandleCsv");

74.8. Unmarshaling with a pipe as delimiter

Using the Spring/XML DSL:

<route>
  <from uri="direct:start" />
  <unmarshal>
    <csv delimiter="|" />
  </unmarshal>
  <to uri="bean:myCsvHandler?method=doHandleCsv" />
</route>

Or the Java DSL:

CsvDataFormat csv = new CsvDataFormat();
CSVStrategy strategy = CSVStrategy.DEFAULT_STRATEGY;
strategy.setDelimiter('|');
csv.setStrategy(strategy);

from("direct:start")
  .unmarshal(csv)
  .to("bean:myCsvHandler?method=doHandleCsv");
CsvDataFormat csv = new CsvDataFormat();
csv.setDelimiter("|");

from("direct:start")
  .unmarshal(csv)
  .to("bean:myCsvHandler?method=doHandleCsv");
CsvDataFormat csv = new CsvDataFormat();
CSVConfig csvConfig = new CSVConfig();
csvConfig.setDelimiter(";");
csv.setConfig(csvConfig);

from("direct:start")
  .unmarshal(csv)
  .to("bean:myCsvHandler?method=doHandleCsv");

Issue in CSVConfig

It looks like that

CSVConfig csvConfig = new CSVConfig();
csvConfig.setDelimiter(';');

doesn’t work. You have to set the delimiter as a String!

74.9. Dependencies

To use CSV in your Camel routes you need to add a dependency on camel-csv, which implements this data format.

If you use Maven you can just add the following to your pom.xml, substituting the version number for the latest and greatest release (see the download page for the latest versions).

<dependency>
  <groupId>org.apache.camel</groupId>
  <artifactId>camel-csv</artifactId>
  <version>x.x.x</version>
</dependency>