json deserialized

r0mm0n

unread,

Jan 17, 2022, 5:33:20 AM1/17/22

to KrakenD Community

Hello, it is possible to decompose json response with Krakend API gateway ?

For example:

I have response:
{"$type":"swagger_alfresco_dms_flow_search.k2RESTidentifier_ResultNode_DMS_FLOW[], swagger_alfresco_dms_flow_search, Version=0.0.0.0, Culture=neutral, PublicKeyToken=null","$values":[{"$type":"swagger_alfresco_dms_flow_search.k2RESTidentifier_ResultNode_DMS_FLOW, swagger_alfresco_dms_flow_search, Version=0.0.0.0, Culture=neutral, PublicKeyToken=null","version":"1","documentValid":"true","name":"MS TEAMS example1.docx","nodeId":"e64468d8-ea07-4a5b-ad1a-1115b372a3a0","title":"First doc","lastModifiedDate":"2021-10-07T22:11:27.87+02:00","createdDate":"2021-09-27T11:36:42.667+02:00","createdBy":"","comment":"This is an important document","mimeType":"application/vnd.openxmlformats-officedocument.wordprocessingml.document","mimeTypeName":"Microsoft Word 2007","path":"/Company Home/DMSFlow","lastVersion":true},{"$type":"swagger_alfresco_dms_flow_search.k2RESTidentifier_ResultNode_DMS_FLOW, swagger_alfresco_dms_flow_search, Version=0.0.0.0, Culture=neutral, PublicKeyToken=null","documentSubTypeId":"1","version":"1","documentValid":"true","name":"automat.jpg","nodeId":"f4e240cd-3c54-4578-b221-45d1d8ee036d","documentTypeId":"1","title":"Second doc","documentStatusId":"1","lastModifiedDate":"2021-09-27T13:35:15.012+02:00","createdDate":"2021-09-27T11:39:56.973+02:00","createdBy":"","comment":"This is an important document","mimeType":"image/jpeg","mimeTypeName":"JPEG Image","path":"/Company Home/DMSFlow","lastVersion":false}]}

I need to extract part A or part B from the given response.

part A)
{"$type":"swagger_alfresco_dms_flow_search.k2RESTidentifier_ResultNode_DMS_FLOW, swagger_alfresco_dms_flow_search, Version=0.0.0.0, Culture=neutral, PublicKeyToken=null","version":"1","documentValid":"true","name":"MS TEAMSexample1.docx","nodeId":"e64468d8-ea07-4a5b-ad1a-1115b372a3a0","title":"First doc","lastModifiedDate":"2021-10-07T22:11:27.87+02:00","createdDate":"2021-09-27T11:36:42.667+02:00","createdBy":"","comment":"This is an important document","mimeType":"application/vnd.openxmlformats-officedocument.wordprocessingml.document","mimeTypeName":"Microsoft Word 2007","path":"/Company Home/DMSFlow","lastVersion":true}

part B)
{"$type":"swagger_alfresco_dms_flow_search.k2RESTidentifier_ResultNode_DMS_FLOW, swagger_alfresco_dms_flow_search, Version=0.0.0.0, Culture=neutral, PublicKeyToken=null","documentSubTypeId":"1","version":"1","documentValid":"true","name":"automat.jpg","nodeId":"f4e240cd-3c54-4578-b221-45d1d8ee036d","documentTypeId":"1","title":"Second doc","documentStatusId":"1","lastModifiedDate":"2021-09-27T13:35:15.012+02:00","createdDate":"2021-09-27T11:39:56.973+02:00","createdBy":"","comment":"This is an important document","mimeType":"image/jpeg","mimeTypeName":"JPEG Image","path":"/Company Home/DMSFlow","lastVersion":false}

Thank you for your response

R.

Albert Lombarte

unread,

Jan 17, 2022, 6:22:15 AM1/17/22

to KrakenD Community, r0mm0n

Hi there,

Yes, what you want to achieve is actually possible. The different data manipulation attributes allow you to pick the data you want to get from a response. Given that your response contains an array inside "$values" then you need to work with the flatmap filter (https://www.krakend.io/docs/backends/flatmap/).

Here there is a working configuration that you can execute to see how it works, it picks part A) but you could pick only part B as well. Note that since I don't have the Alfresco server, I have mocked the data inside the /alfresco endpoint. The flatmap takes the result you want and deletes the rest of the content. The "status" is deleted too because it comes from the mocking endpoint.

{
    "version": 2,
    "endpoints": [
        {
            "endpoint": "/alfresco",
            "backend": [
                {
                    "host": [
                        "http://localhost:8080"
                    ],
                    "url_pattern": "/__health"
                }
            ],
            "extra_config": {
                "github.com/devopsfaith/krakend/proxy": {
                    "static": {
                        "strategy": "always",
                        "data": {

        },
        {
            "endpoint": "/extraction",
            "backend": {
                "host": [
                    "http://localhost:8080"
                ],
                "url_pattern": "/alfresco",
                "extra_config": {
                    "github.com/devopsfaith/krakend/proxy": {
                        "flatmap_filter": [
                            {
                                "type": "move",
                                "args": [
                                    "$values.0",
                                    "your_first_document"
                                ]
                            },
                            {
                                "type": "del",
                                "args": [
                                    "$values", "$type", "status"
                                ]
                            }
                        ]
                    }
                }
            }
        }
    ]
}

El dia dilluns, 17 de gener de 2022 a les 11:33:20 UTC+1, r0mm0n va escriure:

r0mm0n

unread,

Feb 11, 2022, 3:47:52 AM2/11/22

to KrakenD Community, alb...@krakend.io, r0mm0n

Hi, it works absolutely great.
I still have a question about array processing. When I have multiple objects in the array and I want to use the move function for all members of a collection ?
example:

"extra_config": {
"github.com/devopsfaith/krakend/proxy": {
"flatmap_filter": [
{
"type": "move",
"args": [

"list.entries.0-99",
"move-all99"
]
}
]
}
},

Thank You.

Reply all

Reply to author

Forward