Upload soil moisture data with missing depth(s)

[ekosmas]

Is this possible?

I have tried to skip a depth value, e.g. soil_moisture_60:

const postData = JSON.stringify([{
  dataset_id: "test_dataset",
  date: "2025-09-30T16:56:05.054Z",
  soil_moisture_10: 7.3,
  soil_moisture_20: 2.4,
  soil_moisture_30: 2.2,
  soil_moisture_40: 1.3,
  soil_moisture_50: 1.1,
  rain: 0,
  temperature: 32.1,
  humidity: 30.02
}]);

but an error is returned:

Response: {
  detail: [
    {
      type: 'missing',
      loc: [Array],
      msg: 'Field required',
      input: [Object]
    }
  ]
}

I have also tried to add empty value(s), e.g. for soil_moisture_60:

const postData = JSON.stringify([{
  dataset_id: "test_dataset",
  date: "2025-09-30T16:56:05.054Z",
  soil_moisture_10: 7.3,
  soil_moisture_20: 2.4,
  soil_moisture_30: 2.2,
  soil_moisture_40: 1.3,
  soil_moisture_50: 1.1,
  soil_moisture_60: null,
  rain: 0,
  temperature: 32.1,
  humidity: 30.02
}]);

but again an error is returned:

Response: {
  detail: [
    {
      type: 'float_type',
      loc: [Array],
      msg: 'Input should be a valid number',
      input: null
    }
  ]
}

[fedjo]

Hi @ekosmas, great catch. Which API call are you using to upload the data? Please give us more details on how you try to POST the data.

[ekosmas]

Hi @fedjo. I am sending a post request on this path: /api/v1/dataset/. Please find below the complete test code.

const { HOST_IP, TOKEN } = require('./constants');

const postData = JSON.stringify([{
  dataset_id: "test_dataset",
  date: "2025-09-30T16:56:05.054Z",
  soil_moisture_10: 7.3,
  soil_moisture_20: 2.4,
  soil_moisture_30: 2.2,
  soil_moisture_40: 1.3,
  soil_moisture_50: 1.1,
  soil_moisture_60: null,
  rain: 0,
  temperature: 32.1,
  humidity: 30.02
}]);

const options = {
  hostname: `${HOST_IP}`,
  port: 4999,
  path: '/api/v1/dataset/',
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    'Content-Length': Buffer.byteLength(postData),
    'Authorization': `Bearer ${TOKEN}`
  }
};

const req = http.request(options, (res) => {
  let data = '';
  res.on('data', (chunk) => {
    data += chunk;
  });
  res.on('end', () => {
    console.log('Response:', JSON.parse(data));
  });
});

req.on('error', (error) => {
  console.error('Error:', error);
});

req.write(postData);
req.end();

[fedjo]

Tested the script. You are right the implementation is buggy. It currently accepts a column with value 0. That way the service internally ignores this column. So instead of your payload use sth like

[{
  dataset_id: "test_dataset",
  date: "2025-09-30T16:56:05.054Z",
  soil_moisture_10: 7.3,
  soil_moisture_20: 2.4,
  soil_moisture_30: 2.2,
  soil_moisture_40: 1.3,
  soil_moisture_50: 1.1,
  soil_moisture_60: 0,
  rain: 0,
  temperature: 32.1,
  humidity: 30.02
}]

It is absolutely not the best way to skip columns. We are currently working on improving this.

From user side, which approach whould be best suited for you? To set the column value null or just omit the column?

[ekosmas]

Adding zero may mislead the analysis, since zero moisture is reported for the corresponding depth. It is important to explain whether the (revised) analysis can handle truly missing depth values, and not just zeros.

Reporting null values for missing depths may be preferable, although omitting the columns is also acceptable.

[prske]

Hi @ekosmas

Thanks for testing out and reporting.
It was designed to be "zero" if some depth(s) do not have measurement, but now I see it could lead to some confusion.
We will update this API request to be more robust to this kind of requests.

Regards