Debug and Extend Cursor Queries

You are working in an AI-assisted coding interview. You may consult AI-generated suggestions, but you must validate them, explain your reasoning, and share your screen while you work. You are given a small in-memory database abstraction. A table is a list of rows, where each row is a dictionary from column name to value. Queries return a cursor object that supports incremental iteration. A simplified implementation is shown below: ```python class Cursor: def __init__(self, rows): self.rows = rows self.index = 0 def has_next(self): self.index += 1 return self.index < len(self.rows) def next(self): if not self.has_next(): return None return self.rows[self.index] class Table: def __init__(self): self.rows = [] def insert(self, row): self.rows.append(row) def query(self, predicate=lambda row: True): return Cursor([row for row in self.rows if predicate(row)]) ``` Answer the following four parts: 1. **Bug finding:** Identify the cursor iteration bug or bugs, explain the expected cursor semantics, and fix the implementation. 2. **New feature design:** Add support for one of the following features and describe the design: `limit(n)`, `order_by(column)`, or pagination with a stable resume token. 3. **Pull request review:** Review an AI-generated change that eagerly copies all rows on every query and stores mutable row references in the cursor. What correctness, performance, and maintainability concerns would you raise? 4. **Production debugging:** Suppose users report that some paginated queries return duplicate rows or skip rows while writes are happening concurrently. Explain how you would investigate, reproduce, instrument, and fix the issue.