JSON Select



JSON Select
Please note the JSON Select plugin was designed to handle some special .json files that contain multiple-nested-layer structure with recursive columns. This type of .json files are often seen in output files from AI-OCR solutions such as Rossum and Xtracta.



Need help?

Technical contact to tech@argos-labs.com


May you search all operations,




Usage Guide with Sample JSON code.

= Example JSON
            {
                "page": 1,
                "per_page": 3,
                "total": 12,
                "total_pages": 4,
                "pagedata": {
                    "data": [
                        {
                            "id": 1,
                            "name": "cerulean",
                            "year": 2000,
                            "color": "#98B2D1",
                            "pantone_value": "We"
                        },
                        {
                            "id": 2,
                            "name": "fuchsia rose",
                            "year": 2001,
                            "color": "#C74375",
                            "pantone_value": "Freedom"
                        },
                        {
                            "id": 3,
                            "name": "true red",
                            "year": 2002,
                            "color": "#BF1932",
                            "pantone_value": "Soul"
                        }
                    ]
                },
                'jpn': {
                    "data": {
                        "shippingDay": "2019-04-02",
                        "mansionName": "新宿ビル"
                    },
                    "status": 200
                }







1) Execute with xpath "total"

 > result is "12"


2)Execute with xpath "pagedata/data[2]/id"

 > result is "3"


3) Execute with xpath "pagedata/data"

 > result are

  • "id","name","year","color","pantone_value"
  • 1,"cerulean",2000,"#98B2D1","We"
  • 2,"fuchsia rose",2001,"#C74375","Freedom"
  • 3,"true red",2002,"#BF1932","Soul"





4) Execute with xpath "pagedata/invalid" with "Raise Except Flag"

 > return Error


5) Execute with xpath "pagedata/invalid" without "Raise Except Flag"

 > return Error

==> no need this option.

* "Opt Output Format" is only "csv"


6) Execute with xpath "pagedata/data" with "Get length"

> result are "3"


7) Execute with xpath "pagedata/data" with "AND Filter", "id=2" (this "AND Filter" can be mupliple)

 > result are

  • "id","name","year","color","pantone_value"
  • 2,"fuchsia rose",2001,"#C74375","Freedom"


8) Execute with xpath "pagedata/data" with "OR Filter", "id=2", "OR Filter", "id=3"

  (this "AND Filter" can be mupliple)

 > result are

  • "id","name","year","color","pantone_value"
  • 2,"fuchsia rose",2001,"#C74375","Freedom"
  • 3,"true red",2002,"#BF1932","Soul"


* "Find all Opt" option is gathers all data matching key value.

* "Cell key" option is defined then this column value is valid (exists and not empty).

* "Exclude index" option is 1 and "Exclude value" is "Item" then in csv result first column value is "Item" rows are excluded.


Return Code

Code
Meaning
0Execution successful
2Execution failed


Additional feature (as of Oct 26, 2020)


Now you can specify the encoding format of your text file.




All Plugins