happytails.ai

CSV deterministic sampler

Open CSV deterministic sampler on Happytails.ai. This view has the same page content and links, with explicit tool capabilities. Read the processing notes before using a tool.

Tool capabilities

{
  "id": "csv-deterministic-sampler",
  "category": "data-tools",
  "built": true,
  "requirements": "Paste CSV, choose a whole-number sample size and enter a seed. Identical input, size and seed produce the same selected records. Changing the input order changes the sample. Records are sampled without replacement and exported in their original source order. The added source-record column identifies the original one-based data-record position, excluding the header. A quoted multiline record counts once. Choose an unused name for this column. Download the sample and the JSON report, which records the seed and the versioned sampling algorithm. This is deterministic convenience sampling for reviewing data or making repeatable test subsets. It is not cryptographic randomness or a certified experimental allocation method. CSV strings are preserved in the preview; default formula protection changes flagged exported strings with an apostrophe prefix.",
  "implementation": {
    "fields": [
      {
        "key": "delimiter",
        "label": "Input CSV separator",
        "value": ",",
        "type": "select",
        "values": [
          ",",
          ";",
          "\t"
        ]
      },
      {
        "key": "size",
        "label": "Sample records",
        "value": 5,
        "type": "number",
        "min": 1,
        "max": 10000,
        "step": 1
      },
      {
        "key": "seed",
        "label": "Reproducible seed",
        "value": "happytails",
        "type": "text"
      },
      {
        "key": "sourceColumn",
        "label": "Added source-record column",
        "value": "_source_row",
        "type": "text"
      },
      {
        "key": "exportMode",
        "label": "Spreadsheet formula handling",
        "value": "Protect formulas",
        "type": "select",
        "values": [
          "Protect formulas",
          "Preserve original cells"
        ]
      }
    ],
    "live": true,
    "presentation": "metrics",
    "note": "Sample records without replacement and retain source order. Identical input, seed and sample size reproduce the same output. The report identifies the versioned algorithm. This is not cryptographic randomness or certified study allocation. Use a header row with unique column names. Up to 1,000,000 input characters, 10,000 data records, 100 columns and 100,000 cells. Strings and spaces are preserved. Output CSV uses commas and quotes every cell. Protect formulas prefixes risky output cells with an apostrophe; choose Preserve original cells only for trusted machine processing. Runs on your device.",
    "experience": {
      "automatic": true,
      "single": false,
      "inputMaxLength": 1000000,
      "inputMode": "text",
      "label": "CSV",
      "placeholder": "name,score\nAda,98\nSam,92",
      "help": "Include a header row. Quote cells that contain commas or new lines. Results update as you type. Up to 1,000,000 characters.",
      "helpParts": [
        "Include a header row. Quote cells that contain commas or new lines.",
        "Results update as you type.",
        "Up to 1,000,000 characters."
      ],
      "action": null
    },
    "mode": "automatic",
    "interactive": false
  },
  "execution": "Browser interface",
  "api": null,
  "browser_agent": {
    "supported": true,
    "name": "run_current_tool",
    "discovery": "WebMCP on the human page in a supporting browser",
    "verification": "See audit/API-AUDIT.md; availability is not verification"
  }
}

Page guide

Create a repeatable sample of CSV records using a seed. Sample without replacement and retain each record's source number.

How to use CSV deterministic sampler
  1. Paste the CSV and choose a sample size no larger than the input record count.
  2. Enter a seed and an unused source-record column name.
  3. Review the selected records, then download the sample and processing report.

Useful next steps

CSV row filterCSV formula injection checkerDataset value mapperCSV to JSON

What to expect

Create a repeatable sample of CSV records using a seed. Sample without replacement and retain each record's source number. Paste CSV, choose a whole-number sample size and enter a seed. Identical input, size and seed produce the same selected records. Changing the input order changes the sample. Records are sampled without replacement and exported in their original source order. The added source-record column identifies the original one-based data-record position, excluding the header. A quoted multiline record counts once. Choose an unused name for this column. Download the sample and the JSON report, which records the seed and the versioned sampling algorithm. This is deterministic convenience sampling for reviewing data or making repeatable test subsets. It is not cryptographic randomness or a certified experimental allocation method. CSV strings are preserved in the preview; default formula protection changes flagged exported strings with an apostrophe prefix.

Try a small example

id,name
001,Ana
002,Bea
003,Cam
004,Dan
005,Eli