DATASET_NOT_FOUND says my dataset is not declared, but the CSV is right there

Asked 12 Nov 20209,140 views2 answers

21votes

Trying to drive the login scenarios from a CSV instead of an Examples table. The file is at projects/demo-shop/data/common/users.csv and previewing it works:

bash

sdods data preview projects/demo-shop/data/common/users.csv

Six rows, columns id, username, password, role, displayName. Exactly what I want.

The scenario dies on the first step though:

text

✖ DATASET_NOT_FOUND: Dataset "users" is not declared for project demo-shop.
  hint: Known datasets: logins, posts. Add it under data.sources in sdods.project.yaml.

The step is:

gherkin

Given I load dataset "users" row 0

If the file is on disk and the CLI can read it, why does the runner care what the yaml says?

data config cli

asked by Dinuka Perera412 · 12 Nov 2020

2 answers

19votes✓
Accepted answer
The name in the step is a key in data.sources, not a file name. The provider looks users up in the resolved project config and never goes near the filesystem until it has a source spec, so a file with the right name in the right place is invisible to it.
The hint is telling you what somebody already declared. You have logins and posts, so either the CSV was declared under the name logins, or nothing points at it at all.
yaml
```
data:
  sources:
    users: { type: csv, path: 'data/{env}/users.csv', fallback: data/common/users.csv }
```
sdods data preview takes a path, not a dataset name, which is why it works without any of this — it is a file viewer, not the loader.
Test data and user pools
answered by Priya Venkateshmaintainer29,604 · 12 Nov 2020
6votes
Adding to that: after you edit the yaml, check what actually resolved rather than what you typed.
bash
```
sdods config show -p demo-shop -e staging
```
The merged project tree comes out with the data.sources block in it. I have twice edited a yaml in one project directory and run the suite in another.
answered by Marijn Kuiper18,840 · 15 Nov 2020

Have the same question?

Ask it with the command you ran and the output you got, and it will be answered here.

Ask a question

DATASET_NOT_FOUND says my dataset is not declared, but the CSV is right there

2 answers

Related