Improve error message for missing operands

[mpkorstanje]

🤔 What's changed?

Improved the error message for cases where an operand is missing.
Unfortunately the parsing algorithm can't detect which operator is
missing operands so the error message must remain fairly generic.

⚡️ What's your motivation?

Fixes: #196

🏷️ What kind of change is this?

• 🐛 Bug fix (non-breaking change which fixes a defect)

📋 Checklist:

• I agree to respect and uphold the Cucumber Community Code of Conduct
• I've changed the behaviour of the code
  • I have added/updated tests to cover my changes.
• My change requires a change to the documentation.
  • I have updated the documentation accordingly.
• Users should know about my change
  • I have added an entry to the "Unreleased" section of the CHANGELOG, linking to this pull request.

[ehuelsmann]

Looking good!

[luke-hill]

This does look good. But I cannot trace why this custom pop method was ever used.

But given what you've written and the error examples I can clearly see that for each of them it would be almost impossible to work out "what" the problem is. i.e. for a and is the problem the and or the missing second item.

Overall it's a 👍 from me, but I can't quite trace / understand the logic

[clrudolphi]

Question: in the addition to the errors test data, I notice that the tests being added are adding a new error message rather than reusing an existing one.

The input a and is resulting in an error message about an incomplete expression rather than reusing the existing error message about a missing expected operand.

Why the difference?

[mpkorstanje]

Question: in the addition to the errors test data, I notice that the tests being added are adding a new error message rather than reusing an existing one.

Good point. Addressed by d2d19c5

[mpkorstanje]

This does look good. But I cannot trace why this custom pop method was ever used.

I don't know either. I suspect it was an over abundance of caution.