Show HN: Datamorph – A clean JSON ⇄ CSV converter with auto-detect

(datamorphio.vercel.app)

22 points | by sumit_entr42 4 days ago

5 comments

cobertos 14 hours ago
Tried it on a 9.9MB json file I have. Seemed to handle the size fine, but the output was nothing like what I would want. I made a smaller version of what I uploaded in the below pastebin and the output.
The data is TLD data so it should have at least given me the "com", "net", "..." part
Input
https://pastebin.com/QAWeu8Xt
Output
```
    category,addedToListPrice,appliesToCustomers,amountRateType,amountRate,type
    fee_tax,false,eu,percent-varies-by-location,~,vat
    fee_tax,false,in,percent,15,gst
    fee_tax,false,au,percent,10,gst
    fee_tax,false,eu,percent-varies-by-location,~,vat
```
An explanation of _how_ it does the conversion would be nice when choosing a tool like this. As the "just works" didn't seem to work.

I tried converting this:

[

  {

    "name": "Avocado Dip",

    "carb": 2,

    "cholesterol": 5,

    "fiber": 0,

    "minerals": {

      "ca": 0,

      "fe": 0

    },

    "protein": 1,

    "sodium": 210,

    "vitamins": {

      "a": 0,

      "c": 0

    }

  }

]

And I got:

name,carb,cholesterol,fiber,minerals,protein,sodium,vitamins

Avocado Dip,2,5,0,"{""ca"":0,""fe"":0}",1,210,"{""a"":0,""c"":0}"

Which is not what I would expect.

zerratar 12 hours ago
{ "data": { "field1": "hello", "test": [ "what", "is", "this" ] }, "something": "yay" }
becomes
data,something [object Object],yay
I already knew complex types / nested objects would not work. but what makes json stronger than csv is that you can handle more complex data types and most people will use json that way too. Otherwise json would just be a less bloated xml, which is a less bloated csv. lol
kihyug6i 12 hours ago
Your api keys are public
[-]
- cluckindan 8 hours ago
  I love the smell of vibe coding in the morning.
louiechristie 9 hours ago
No unit tests.