RFC 8259 Quiz

Q1: An A2A endpoint accepts only a JSON object at the top level. What does RFC 8259 imply?

Multiple Choice

A. RFC 8259 permits only objects and arrays as JSON texts B. The endpoint may impose an object requirement as an application rule C. The endpoint must also accept every scalar because parsers cannot apply schemas

**Explanation:** A: RFC 8259 permits any serialized JSON value at the top level. B: An application can narrow the valid input beyond the general JSON grammar. C: RFC 8259 defines JSON syntax and does not prohibit schema validation after parsing.

Q2: A request contains {"role":"reader","role":"admin"}. What is the safest interoperable decision?

Multiple Choice

A. Reject the object before authorization logic uses it B. Always use the last value because RFC 8259 defines last-wins behavior C. Sort the two members and use the first one

**Explanation:** A: Object names should be unique and duplicate handling is unpredictable across implementations. B: Last-wins is common but is not the required behavior in RFC 8259. C: Member ordering does not resolve duplicate-name ambiguity and parsers expose ordering differently.

Q3: Can an A2A implementation assign meaning to the order of members in a JSON object?

Multiple Choice

A. Yes, because every parser preserves source order B. Yes, if the sender uses UTF-8 without a BOM C. No, use an array when order is part of the message meaning

**Explanation:** A: RFC 8259 notes that libraries differ on whether member ordering is visible. B: Encoding choice does not turn object-member order into an interoperable semantic contract. C: Array order is represented by element position while object-member order is not a safe dependency.

Q4: Which JSON array is valid under RFC 8259?

Multiple Choice

A. Only an array whose elements all have the same type B. An array containing a string, number, object, and null C. Only an array with at least one element

**Explanation:** A: RFC 8259 has no same-type requirement for array elements, so mixed values remain valid. B: Arrays may contain heterogeneous JSON values and each element follows the normal value grammar. C: The grammar permits an empty array as well as arrays with one or more elements.

Q5: A sender emits NaN as an unquoted JSON number. How should a conforming receiver classify it?

Multiple Choice

A. It is not valid JSON B. It is valid when the receiver uses IEEE 754 C. It is the JSON null literal

**Explanation:** A: NaN and Infinity are outside the RFC 8259 number grammar. B: IEEE 754 implementations can represent NaN but JSON syntax does not permit it. C: NaN is neither the lowercase null literal nor another JSON literal.

Q6: An agent identifier is the integer 9007199254740993 and must round-trip exactly across common runtimes. What is the better representation?

Multiple Choice

A. A JSON number with an exponent added B. A JSON number followed by a type comment C. A JSON string with application-level identifier rules

**Explanation:** A: Changing number notation does not make a value outside the exact binary64 integer range exact. B: JSON has no comment syntax and the numeric precision issue remains. C: A string avoids numeric rounding and lets the application define identifier syntax.

Q7: Which encoding rule applies to JSON exchanged between independent systems?

Multiple Choice

A. UTF-16 with a byte order mark is required B. Use UTF-8 and do not generate a leading BOM C. Encoding is selected by object-member order

**Explanation:** A: RFC 8259 requires UTF-8 for JSON exchanged outside a closed ecosystem. B: Networked JSON uses UTF-8 and generators must not add a BOM. C: Member ordering carries no encoding negotiation signal.

Q8: A parser receives the member names "a\\b" and "a\u005Cb". How should equality be determined?

Multiple Choice

A. They are always different because their source bytes differ B. They are equal only if object-member order also matches C. Decode escapes and compare the resulting code units

**Explanation:** A: Different escape spellings can encode the same string value. B: String equality does not depend on the position of the member in an object. C: RFC 8259 recommends code-unit comparison after converting the textual representation.

Q9: What should an implementation expect from a JSON string containing a lone UTF-16 surrogate?

Multiple Choice

A. Receiver behavior can be unpredictable across implementations B. Every receiver must replace it with U+FFFD C. It becomes a valid Unicode scalar value after UTF-8 encoding

**Explanation:** A: RFC 8259 describes differing lengths, values, and even runtime failures for such data. B: RFC 8259 does not define a universal replacement procedure. C: Encoding a lone surrogate does not turn it into a Unicode scalar value.

Q10: Where should an A2A service enforce limits on JSON size and nesting depth?

Multiple Choice

A. After authorization and business processing complete B. At the input boundary before expensive or security-sensitive processing C. Only in the sender because a conforming parser cannot reject deep input

**Explanation:** A: Late checks allow hostile input to consume parser and application resources first. B: RFC 8259 permits size and depth limits, and early enforcement bounds resource use. C: Parsers may impose limits on text size and maximum nesting depth.

Interoperable JSON parsing and generation

References

Q1: An A2A endpoint accepts only a JSON object at the top level. What does RFC 8259 imply?

Q2: A request contains {"role":"reader","role":"admin"}. What is the safest interoperable decision?

Q3: Can an A2A implementation assign meaning to the order of members in a JSON object?

Q4: Which JSON array is valid under RFC 8259?

Q5: A sender emits NaN as an unquoted JSON number. How should a conforming receiver classify it?

Q6: An agent identifier is the integer 9007199254740993 and must round-trip exactly across common runtimes. What is the better representation?

Q7: Which encoding rule applies to JSON exchanged between independent systems?

Q8: A parser receives the member names "a\\b" and "a\u005Cb". How should equality be determined?

Q9: What should an implementation expect from a JSON string containing a lone UTF-16 surrogate?

Q10: Where should an A2A service enforce limits on JSON size and nesting depth?