We want to process all of the exceptions we catch with the same process in our exception strategy. It is the one generated by apikit. We first send a lot with a custom component and then we generate json response. In the response with DataWeave we want to set the same statusCode. However, the statusCode do not seem to be a variable that is retrievable. Am I correct, or there is a good way to retrieve it?
<apikit:mapping-exception-strategy xmlns:apikit="http://www.mulesoft.org/schema/mule/apikit" name="svc0031_hotel-apiKitGlobalExceptionMapping">
<apikit:mapping statusCode="504">
<apikit:exception value="org.mule.api.transformer.TransformerMessagingException"/>
<flow-ref name="svc0031_manageErrors" doc:name="Manage Error"/>
</apikit:mapping>
<apikit:mapping statusCode="404">
<apikit:exception value="org.mule.api.transformer.TransformerMessagingException"/>
<flow-ref name="svc0031_manageErrors" doc:name="Manage Error"/>
</apikit:mapping>
</apikit:mapping-exception-strategy>
<sub-flow name="svc0031_manageErrors">
<set-payload value="#[groovy:message.exceptionPayload.rootException.message]" doc:name="Set BIP Payload"/>
<custom-transformer class="se.zystems.baseline.BaselineLogging" doc:name="Log BIP Outgoing">
<spring:property name="Level" value="ERROR"/>
<spring:property name="ObjectId" value="CatchErrors"/>
<spring:property name="TransactionStatus" value="failed"/>
</custom-transformer>
<dw:transform-message doc:name="Transform Message">
<dw:set-payload><![CDATA[%dw 1.0
%output application/json
---
{
status : 400,
message : payload,
code : 42,
more_info :"https://anypoint.mulesoft.com/apiplatform/nordic-choice-hotels"
}]]></dw:set-payload>
</dw:transform-message>
<logger level="INFO" doc:name="Logger"/>
</sub-flow>
Good,
If you create a variable with the http.status then you can use it in the mapper:
<flow name="test">
<set-variable variableName="httpStatus" value="#[message.inboundProperties['http.status']" doc:name="Variable"/>
<dw:transform-message doc:name="Transform Message">
<dw:set-payload><![CDATA[%dw 1.0
%output application/json
---
{
status : flowVars.httpStatus,
message : payload,
code : 42,
more_info :"https://anypoint.mulesoft.com/apiplatform/nordic-choice-hotels"
}]]></dw:set-payload>
</dw:transform-message>
</flow>
I found out where the statusCode value is stored thanks to the answer by Jesús Pablo Fernández
He was almost right in his answer, however, the statusCode is stored not in message.inboundProperties['http.status'], but in message.outboundProperties['http.status']. No need to even extract a variable, one can just access it in data transformer directly like this:
<dw:transform-message doc:name="Transform Message">
<dw:set-payload><![CDATA[%dw 1.0
%output application/json
---
{
status : outboundProperties['http.status'],
message : payload,
code : 42,
more_info :"https://anypoint.mulesoft.com/apiplatform/nordic-choice-hotels"
}]]></dw:set-payload>
</dw:transform-message>
Status codes defined in the APIKitExceptionStrategy will be those sent back in the response header as http.status given the corresponding exception to be raised.
For instance,
<apikit:mapping statusCode="504">
<apikit:exception value="org.mule.api.transformer.TransformerMessagingException"/>
<flow-ref name="svc0031_manageErrors" doc:name="Manage Error"/>
</apikit:mapping>
means that when a org.mule.api.transformer.TransformerMessagingException exception is thrown and caught by this exception block, then a 504 status code is sent back as a response header.
If you want to bypass the statusCode attribute in the APIKit exception block, you should have another exceptionStrategy block to catch your desired exception, and then you could set your status code like that
<set-property name="http.status" value="<the_desired_status, e.g. 401>" />
Hope it helps.
/Tony
Good Stanislav Ivanov,
I have a flow implemented with the APIKit, which is the code that I attach and whether there is property in the inboundProperties http.status, review it because if it should contain the value, if you do not have your flow may modify the value in a previous step.
I pointed him to directly access the outboundProperties in the mapper, I was not aware, thank you very much.
Related
I have an issue with a Mule Dataweave Transformer getting an error, this only occurs if :-
An enricher is present wrapping a json:json-to-object-transformer
<enricher target="#[flowVars.myVar]" doc:name="Message Enricher">
<json:json-to-object-transformer returnClass="java.util.HashMap" doc:name="JSON to Object"/>
</enricher>
A flowVar is involved in the Dataweave Transformer
The exception occurs on the 2nd dataweave :-
ERROR 2017-08-15 09:09:52,239 [amqpReceiver.02] org.mule.exception.RollbackMessagingExceptionStrategy:
********************************************************************************
Message : Exception while executing:
{bar=1, foo=0}
^
Unexpected character 'b' at index 1 (line 1, position 2), expected '"'
Payload : com.mulesoft.weave.reader.ByteArraySeekableStream#39e7a0d3
Element : /subscribe-to-changeFlow4/processors/4 # voa-009-mule-authorisation-search-api:test3.xml:58 (Transform Message)
Element XML : <dw:transform-message doc:name="Transform Message">
<dw:input-payload mimeType="application/json"></dw:input-payload>
<dw:set-payload>%dw 1.0%output application/json---{field1: payload.foo,field2: payload.bar,field3: flowVars.myFlowVar}</dw:set-payload>
</dw:transform-message>
--------------------------------------------------------------------------------
Root Exception stack trace:
com.mulesoft.weave.reader.json.JsonReaderException: Unexpected character 'b' at index 1 (line 1, position 2), expected '"'
at com.mulesoft.weave.reader.json.JsonTokenizer.fail(JsonTokenizer.scala:193)
at com.mulesoft.weave.reader.json.JsonTokenizer.require(JsonTokenizer.scala:190)
at com.mulesoft.weave.reader.json.JsonTokenizer.readString(JsonTokenizer.scala:74)
at com.mulesoft.weave.reader.json.JsonTokenizer.readObjectMembers(JsonTokenizer.scala:146)
at com.mulesoft.weave.reader.json.JsonTokenizer.com$mulesoft$weave$reader$json$JsonTokenizer$$readObject(JsonTokenizer.scala:140)
at com.mulesoft.weave.reader.json.JsonTokenizer$$anonfun$readValue$1.apply$mcV$sp(JsonTokenizer.scala:37)
What happens is as soon as a flowVar is added into the 2nd data weave, then the payload seems to show as a Map rather than JSON.
The enricher with nested json-to-object-transformer is required to take a payload and allow the JSON to be accessed. If the enricher is removed it works fine
The seek(0) is required to reset the stream after the json-to-object-transformer as Answered in my last question
if the flowVar is removed from the 2nd dataweave it works fine
Test Case
<flow name="subscribe-to-changeFlow4" processingStrategy="synchronous">
<amqp:inbound-endpoint numberOfChannels="1" ref="authorisationChangeQueueDef4" responseTimeout="10000" doc:name="authorisation-change-consumer" />
<dw:transform-message doc:name="Transform Message">
<dw:set-payload>
<![CDATA[%dw 1.0
%output application/json
---
{
foo:0,
bar:1
}]]>
</dw:set-payload>
</dw:transform-message>
<enricher target="#[flowVars.myVar]" doc:name="Message Enricher">
<json:json-to-object-transformer returnClass="java.util.HashMap" doc:name="JSON to Object"/>
</enricher>
<expression-component doc:name="Expression">
<![CDATA[payload.seek(0);]]>
</expression-component>
<set-variable variableName="myFlowVar" value="1234" doc:name="Variable" />
<dw:transform-message doc:name="Transform Message">
<dw:input-payload mimeType="application/json"/>
<dw:set-payload>
<![CDATA[%dw 1.0
%output application/json
---
{
field1: payload.foo,
field2: payload.bar,
field3: flowVars.myFlowVar
}
]]>
</dw:set-payload>
</dw:transform-message>
<logger level="INFO" message="#[message.payloadAs(java.lang.String)]" doc:name="Logger all "/>
</flow>
This is probably a bug in Mule and you can raise it in Mule jira
Which Mule runtime version are you using?
I tested locally and found this issue with Mule runtime 3.8.0 and Mule runtime 3.8.1:
but worked fine with Mule runtime 3.8.2 and Mule 3.8.5
Also tested with Mule 3.7.3 locally and it worked fine as below:-
I had similar problem, the DW processor messes up the payload. So I added an Object-to-String before DW transformer and that worked fine.
Based on Boolean variable I want to get output. If variable is true null parameters should not visible in output and if false null parameters should be visible. How to make skipNullOn dynamic in mule dataweave script?
#swamy Thota not sure if you can do it in dataweave.
Alternate option i can think off is, use choice and 2 dataweaves, one will skip null and other will allow null.
Please use a flow variable , the use a choice and create two different mapping one for true condition and one for false.
It will be something as below
<?xml version="1.0" encoding="UTF-8"?>
<mule xmlns:encryption="http://www.mulesoft.org/schema/mule/encryption" xmlns:http="http://www.mulesoft.org/schema/mule/http"
<flow name="file2fileFlow">
<file:inbound-endpoint path="D:\Tushar\Training\DataWeave\in" moveToDirectory="D:\Tushar\Training\DataWeave\out" responseTimeout="10000" doc:name="File"/>
<set-variable variableName="test" value="#['a']" doc:name="Variable"/>
<choice tracking:enable-default-events="true" doc:name="Choice">
<when expression="#[flowVars.test] == 'a'">
<dw:transform-message doc:name="If Value True">
<dw:set-payload><![CDATA[%dw 1.0
%output application/java
---
{
}]]></dw:set-payload>
</dw:transform-message>
</when>
<otherwise>
<dw:transform-message doc:name="If Value false">
<dw:set-payload><![CDATA[%dw 1.0
%output application/java
---
{
}]]></dw:set-payload>
</dw:transform-message>
</otherwise>
</choice>
<logger level="INFO" doc:name="Logger"/>
<file:outbound-endpoint outputPattern="output.txt" responseTimeout="10000" doc:name="File"/>
</flow>
</mule>
You can load dw script from file and dynamic replace in text script skipNullOn value. Then, call the MEL function #[dw(<script>, [<output type>] ]
More examles you can see there: https://support.mulesoft.com/s/article/ka4340000004GbvAAE/Dynamic-DataWeave-Script
Hej,
I'm new in MULE, I try to learn "how to" following examples from Mule's sida: https://www.mulesoft.com/exchange#!/sending-csv-email-smtp?searchTerm=email
I think I did everything well, project is deploing good, but when file is taken exception is thrown and I'v got communicate:
Exception stack is:
1. Cannot coerce a :binary to a :array (com.mulesoft.weave.model.values.coercion.exception.UnsupportedTypeCoercionException)
com.mulesoft.weave.model.values.formatting.ArrayFormatTypeCoercionValue:31 (null)
2. Exception while executing:
payload map {
^
Cannot coerce a :binary to a :array (com.mulesoft.weave.mule.exception.WeaveExecutionException)
com.mulesoft.weave.mule.WeaveMessageProcessor:124 (null).
Abybody have idea what's happened. Anybody had same problem with that example?
Use the following mimeType in your DataWeave :- <dw:input-payload doc:sample="list_csv.csv" mimeType="application/csv"/>
So, the full code will be some thing like as follows:-
<smtp:gmail-connector doc:name="Gmail" name="Gmail" validateConnections="true"/>
<flow name="csv-to-smtpFlow">
<file:inbound-endpoint doc:name="Poll files" metadata:id="a75d5954-7fb4-4721-85f6-7ee66b793750" path="E:\test" responseTimeout="10000"/>
<dw:transform-message doc:name="Transform CSV to Map" metadata:id="84d175de-b3c6-4480-af9d-150e736f051e">
<dw:input-payload doc:sample="list_csv.csv" mimeType="application/csv"/>
<dw:set-payload><![CDATA[%dw 1.0
%output application/java
---
payload map {
name: $.name,
orderId: $.orderId,
pricePerUnit: $.pricePerUnit as :number,
units: $.units as :number,
totalPrice: ($.units * $.pricePerUnit ) as :number
}]]></dw:set-payload>
</dw:transform-message>
<logger doc:name="Log the processed objects" level="INFO" message="Processing objects: #[payload]"/>
<smtp:outbound-endpoint connector-ref="Gmail" doc:name="Send an email via SMTP" from="${mail.from}" host="${smtp.host}" password="${smtp.password}" port="587" responseTimeout="10000" subject="${mail.subject}" to="${mail.to}" user="${smtp.user}"/>
</flow>
This will solve your issue
I am receiving a JSON request with 3 attribute and one of the attribute is content of image which is base64 in string format. After receiving this request I am creating a Salesforce attachment using Salesforce connector.
Now when I test this via my local Anypoint Studio everything is fine, however after deploying to Cloudhub the fileContent (Image content) is getting truncated to 8192 bytes and creating the attachment as corrupted.
I can see API is receiving the content since message content length is 38820 (both local and Cloudhub), and it looks like message is getting lost after the Message Transformation, and code is as follows:
<processor-chain doc:name="Processor Chain">
<dw:transform-message doc:name="Transform Message">
<dw:set-payload><![CDATA[%dw 1.0
%output application/java
---
{
accountId : payload.accountId,
name : 'MyProfilePicture',
ContentType: payload.contentType,
fileContent : payload.fileContent
}]]></dw:set-payload>
</dw:transform-message>
<set-session-variable variableName="accountId" value="#[payload.accountId]" doc:name="Session Variable - accountId"/>
<set-session-variable variableName="contentType" value="#[payload.ContentType]" doc:name="Session Variable - contentType"/>
<logger message="payload.fileContent Before Conversion size #[payload.fileContent.length()], data:#[payload.fileContent] " level="INFO" doc:name="Logger"/>
<set-session-variable variableName="fileContent" value="#[(org.apache.commons.codec.binary.Base64.decodeBase64(payload.fileContent))]" mimeType="binary/octet-stream" doc:name="Session Variable - fileContent"/>
<dw:transform-message metadata:id="1676949a-75ba-4e51-82ac-b5788a8e87dc" doc:name="Transform Message">
<dw:input-payload/>
<dw:set-payload><![CDATA[%dw 1.0
%output application/java
---
[{
Name: 'MyProfilePicture',
ParentId: sessionVars.accountId,
ContentType: sessionVars.contentType,
Body: sessionVars.fileContent
}]]]></dw:set-payload>
</dw:transform-message>
<logger message="Payload after Transform Message: #[payload]" level="INFO" doc:name="Logger"/>
<sfdc:create config-ref="Salesforce__Cfg" type="Attachment" doc:name="Salesforce">
<sfdc:objects ref="#[payload]"/>
</sfdc:create>
</processor-chain>
Could anyone please help on this? why when I deploy to Cloudhub it is only processing 8192 bytes?
Thanks,
Thomas
This can be fixed by switching to the deprecated DataMapper for the huge base64 transforms.
I have a case, where I have to log the exception error in an object in Salesforce, where I have to also take the payload (the record) for which error is there. The error here will be the record level error and I am using a batch process for this. In the screenshot there is how we are processing the exception.
Following is the xml
<batch:step name="HandleFailedRecords_AccountSF-360" accept-policy="ONLY_FAILURES">
<set-payload value="#[getStepExceptions()]" doc:name="Set Payload"/>
<foreach doc:name="For Each" collection="#[payload.values()]">
<set-variable variableName="Record_level_error" value="#[payload]" doc:name="Record_level_error"/>
<dw:transform-message doc:name="Transform Message" metadata:id="9c2e408a-f530-4ffd-a205-a787c8bc94b2">
<dw:set-payload><![CDATA[%dw 1.0
%output application/java
---
payload map
{
SF_Object_Type__c: "Account"
}]]></dw:set-payload>
</dw:transform-message>
<sfdc:create config-ref="Salesforce__Basic_Authentication" type="Exception__c" doc:name="Salesforce">
<sfdc:objects ref="#[payload]"/>
</sfdc:create>
<logger message="#[payload]" level="INFO" doc:name="Logger"/>
<!-- <flow-ref name="HandleRecordFailure" doc:name="HandleRecordFailure"/> -->
</foreach>
</batch:step>
</batch:process-records>
Here I am able to get what exception I am getting, but what else I need is for which value it is getting errored out.
Bast way to handle properly exception handling in batch processing is to do a flow-ref to a normal flow with his proper exception handling, there you will be able to handle your exception in a standard mule way and do what you need in the exception catch flow.
Hope this helps
Regards