getting "415:Media not supported" Error when passing pdf to IBM watson in Salesforce - pdf

Question

I am planning to integrate the IBM Watson Document Conversion service
with Salesforce.
From there I am unable to send my pdf file directly to Watson and I'm getting Media Type not supported.
I am also getting this error:
{
"code" : 500 ,
"error" : "Server Error" ,
"description" : "2017-07-18T06:02:19-04:00, Error WATSNGWERR-0x0113001c occurred when accessing https://gateway.watsonplatform.net/document-conversion/api/v1/convert_document?version=2015-12-15&config="{"conversion_target":"answer_units"}", Tran-Id: gateway-dp02-1967135880 - Watson Gateway Error"
}
Here is the code I'm using:
public class Resume {
String boundary = '----------------------------741e90d31eff';
public string id{get;set;}
public string content{get;set;}
Transient public Attachment att{set;get;}
public Resume(ApexPages.StandardController controller) {
id=ApexPages.currentPage().getParameters().get('id');
att=new Attachment();
att=[Select Id,ParentId, Name,body,ContentType From Attachment where ParentId=:id limit 1];
content=String.valueOf(att.body);
System.debug('---->' + content);
String header = '--' + boundary + '\nContent-Disposition: form-data; name="att"; filename="'+att.name+'";\nContent-Type: application/pdf';
String footer = '--' + boundary + '--';
String headerEncoded =
EncodingUtil.base64Encode(Blob.valueOf(header+'\r\n\r\n'));
String bodyEncoded = EncodingUtil.base64Encode(att.body);
Blob bodyBlob = null;
String last4Bytes =
bodyEncoded.substring(bodyEncoded.length() - 4, bodyEncoded.length());
while (headerEncoded.endsWith('=')){
header+=' ';
headerEncoded = EncodingUtil.base64Encode(Blob.valueOf(header+'\r\n\r\n'));
}
if (last4Bytes.endsWith('==')) {
last4Bytes = last4Bytes.substring(0,2) + '0K';
bodyEncoded = bodyEncoded.substring(0,bodyEncoded.length()-4) + last4Bytes;
String footerEncoded = EncodingUtil.base64Encode(Blob.valueOf(footer));
bodyBlob = EncodingUtil.base64Decode(headerEncoded + bodyEncoded + footerEncoded);
} else if (last4Bytes.endsWith('=')) {
last4Bytes = last4Bytes.substring(0,3) + 'N';
bodyEncoded = bodyEncoded.substring(0,bodyEncoded.length()-4) + last4Bytes;
footer = '\n' + footer;
String footerEncoded = EncodingUtil.base64Encode(Blob.valueOf(footer));
bodyBlob = EncodingUtil.base64Decode(headerEncoded + bodyEncoded + footerEncoded);
} else {
footer = '\r\n' + footer;
String footerEncoded = EncodingUtil.base64Encode(Blob.valueOf(footer));
bodyBlob = EncodingUtil.base64Decode(headerEncoded + bodyEncoded + footerEncoded);
}
String configAsString ='\"conversion_target:answer_units\"';
Http h = new Http();
HttpRequest request = new HttpRequest();
request.setMethod('POST');
request.setHeader('Content-Type','multipart/form-data; boundary=' + boundary);
String username= 'DOCUMENT-CONVERSION-USERNAME';
String password= 'DOCUMENT-CONVERSION-PASSWORD';
request.setHeader('Authorization', 'Basic ' + EncodingUtil.base64Encode(blob.valueOf(username + ':' + password)));
request.setEndpoint('https://gateway.watsonplatform.net/document-conversion/api/v1/convert_document?version=2015-12-15&config='+configAsString);
request.setBodyAsBlob(bodyBlob);
request.setCompressed(true);
HttpResponse response = h.send(request);
System.debug(response.getBody());
}
}

Answer 1

You are sending the config as a query parameter, but it should be in the body.
Here is the curl command that makes what you are trying to do:
curl -X POST \
-u "{username}":"{password}" \
-F config="{\"conversion_target\":\"answer_units\"}" \
-F "file=#sample.pdf;type=application/pdf" \
"https://gateway.watsonplatform.net/document-conversion/api/v1/convert_document?version=2015-12-15"
I also think there is an error in the way you are creating the body. My team built an SDK to use the Watson APIs in the Salesforce environment. I would suggest you take a look.
If you can't deploy the SDK to your Salesforce organization (It's a lot of code), copy the code from the IBMWatsonMultipartBody.cls class. It will help you encode an Attachment as base64 so that you can end it to Watson.
UPDATE: The Document Conversion service was deprecated but the features from that service were enhanced and migrated to the Discovery service