How to extract content type haeder

[#1]

How to extract only content type from request header?. Some time the content type can have additional values. For example -

application/json

application/json; charset=UTF-8

text/xml

text/xml;charset=UTF-16

application/xml;charset=ASCII

& so on..

In all above cases I need only content type (application/json or text/xml).

<ExtractVariables async="false" continueOnError="false" enabled="true" name="EV.ExtractGenericAttributes">
    <DisplayName>EV.ExtractGenericAttributes</DisplayName>
    <Source clearPayload="false">request</Source>
    <VariablePrefix/>
    <IgnoreUnresolvedVariables>false</IgnoreUnresolvedVariables>    
    <Header name="Content-Type">
        <Pattern ignoreCase="true">{contentFormat};</Pattern>
    </Header>   
</ExtractVariables>

However, it is not setting any value if the Content-Type is just text/xml or application/json.

[aswinsegu]

Hi @Sujnana Rai

I wont set any value if Content-Type is just text/xml or application/json because the pattern you are comparing is {contentFormat}; which both of them lack.

What you can do as alternative is extract whole Content-Type header and fetch only the first part after splitting it by ; operator with javascript callout

[#3]

I tried by splitting ; operator

<PatternignoreCase="true">{contentFormat};</Pattern>

and


<Pattern ignoreCase="true">{contentFormat};*</Pattern>

But it fails if Content-Type doesn't have semicolon (;) in it. Ex: application/json or text/xml

[anilsr]

@Aswin Segu , Your statement "I wont set any value if Content-Type is just text/xml or application/json because the pattern you are comparing is {contentFormat}; which both of them lack." , Is actually not a valid one. Text inside braces is used to specify the variable into which the extracted pattern value will be saved. See my answer above which should solve OP query.

[nmallesh]

Hi @Sujnana Rai

string.split(delimiter) method in javascript does not throw error if the delimiter is not in the string.

Cross verify if the string has the content-type value or undefined.

[anilsr]

@Nisha Mallesh , Using javascript comes at a cost of performance, Specially when you can achieve same using the oob policies.

[nmallesh]

@Anil Sagar very true, and I agree with your approach. I have just cleared her concern that there might not be an issue while splitting.

[anilsr · Accepted solution]

@Sujnana Rai , Actually issue is your pattern,

Instead of ,

<Pattern ignoreCase="true">{contentFormat};</Pattern>

Use,

<Pattern ignoreCase="true">{contentFormat};{metadata}</Pattern>

Again, above will work only if ; is present in content type header, Ideally it should be,

        <Pattern ignoreCase="true">{contentFormat};{metadata}</Pattern>
        <Pattern ignoreCase="true">{contentFormat}</Pattern>

Where you leverage,

• If two matched patterns, then the pattern specified first in the policy is used for extraction.

Below should work,

<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<ExtractVariables name="ExtractVariable-ContentType">
    <Source>request</Source>
    <Header name="Content-Type">
        <Pattern ignoreCase="true">{contentType};{metadata}</Pattern>
	<Pattern ignoreCase="true">{contentType}</Pattern>
    </Header>
    <VariablePrefix>clientrequest</VariablePrefix>
    <IgnoreUnresolvedVariables>true</IgnoreUnresolvedVariables>
</ExtractVariables>

Sample proxy attached below,

extracttype-rev3-2016-10-25.zip

Hope it helps, Keep us posted if any.