[BUG]: Fractions Lesson accepts incorrect integer/fraction inputs [Developer Build]

[Rd4dev]

Describe the bug

The Prototype Exploration -> Fraction Lesson
Qn: "What fraction represents half of something?" accepts incorrect answers.

This is caused by the test asset: test_exp_id_2.json

"rule_specs": [
  {
    "rule_type": "IsExactlyEqualTo",
    "inputs": {
      "f": { "isNegative": false, "wholeNumber": 0, "numerator": 1, "denominator": 2 }
    }
  },
  {
    "rule_type": "HasIntegerPartEqualTo",
    "inputs": { "x": 1 }
  }
]

• The first rule expects exactly 1/2.
• The second rule accepts any fraction with integer part = 1, including 1, 1 1/2, 1 3/4, 1 5/6, 1 99/100.
• This leads to incorrect answers being marked as correct, e.g., 1 and 1 5/6.

The second rule incorrectly allows fractions or integers that are not actually “half”, causing misalignment between the intended answer and the current rules.

Steps To Reproduce

1. Run the oppia_dev build of the app.
2. Open Prototype Explorations.
3. Open Fractions Lesson.
4. Navigate to the Question: “What fraction represents half of something?”
5. Enter the following inputs:
   • 1 → accepted (incorrect)
   • 1 5/6 →accepted (incorrect)
   • 1/2 → accepted (correct)
   • 0 1/2 → accepted (correct)

Expected Behavior

• To Only accept exactly 1/2
• Seek clarification from the team (if optionally equivalent fractions like 2/4, 50/100 should be considered).

Proposed Solution

Note from team: The developer version should mirror the production app as closely as possible.
And to revert the PR changes.

• Remove HasIntegerPartEqualTo from this answer group:
  • In the JSON.
  • In the textprotos
• Add proper verifications to validate the expected behavior and prevent regressions in future.

Screenshots/Videos

Ans: 1	Ans: 1 5/6	Ans: 1/2	Ans: 0 1/2

What device/emulator are you using?

Nexus 5X

Which Android version is your device/emulator running?

API 27

Which version of the Oppia Android app are you using?

d7ef1d5

Additional Context

No response

[Rd4dev]

@adhiamboperes -- I interpret this rule is useful in cases of partly correct mixed fraction feedback (Ref: #4591). I’m not sure if any test explorations currently use it this way — if so, it would be helpful to know if any test exploration changes are needed to verify this as to add more context to the required fix. Thanks.

[alaotach]

hey i want to work on this issue can you assign it to me?