Data Table column ignored without closing delimiter

[SimonIpf]

Reqnroll Version

2025.2.359

Which test runner are you using?

xUnit 2

Test Runner Version Number

2.9.0

.NET Implementation

.NET 8.0

Test Execution Method

Visual Studio Test Explorer

Content of reqnroll.json configuration file

{
"$schema": "https://schemas.reqnroll.net/reqnroll-config-latest.json",
"trace": { "coloredOutput": false }
}

Issue Description

When binding columns a Data Table Reqnroll ignores the final column if it does not have a trailing delimited. In the example below InterestRate is set to zero.

It would be helpful if this could emit a warning or the auto-formatter correct this.

Steps to Reproduce

Given an interest rate 'ZeroInterest' exists with the following rates
        | EffectiveDate | InterestRate
        | 1999-01-01    |        10.00

Link to Repro Project

No response

[304NotModified]

Thanks for reporting!

But is confirmed, see #969

We have to find the source is the issue. It's possible that it's the Gherkin package (not a part of Reqnroll)

[304NotModified]

I think this is
cucumber/gherkin#22

[gasparnagy]

@304NotModified Thx for finding it. Let's close this issue here as "external" and track it in the Gherkin project.

[304NotModified]

I'm reopening this issue as I dont thinks this will be fixed the Gherkin package 😅. The issue alsmost 5 years old and without any activity.

I like to try if its possible for Reqnroll to give a warning for this, if isn't too complicated.

[gasparnagy]

@304NotModified There was no activity, because no one wanted to work on it. I think it is easier to fix it there and that is where the standard Gherkin behavior is maintained. So I would suggest sending a PR to that repository rather than workaround the behavior here.

Regarding my feeling, the problem might be somewhere in the token scanner, that would be hard to hack around in Reqnroll anyway.

[304NotModified]

I like to send a PR to the Gherkin repo, but I guess it should work for all languages? And that's hard. I could maybe handle 50%, but not all 😅

Regarding my feeling, the problem might be somewhere in the token scanner, that would be hard to hack around in Reqnroll anyway.

I think we get somewhere with only parsing the header :)
But I will try a Proof of Concept first. After the package changes. 🙃