Embercode: learn software engineering from zero

1. An AI wrote a C++ solution for a problem with 1,000,000 numbers. It reads with cin >> x in a loop, using default settings. The sample passes, but the judge reports time limit exceeded on the largest test. The algorithm is O(n). What is the first change to try?

Raise the stack size with ulimit -sAdd ios::sync_with_stdio(false); and cin.tie(nullptr); at the top of mainReplace the loop with a recursive functionPrint each result with endl

2. Which line is the bug, and what does it do? ``

python
import sys
data = sys.stdin.buffer.read().split()
if data[0] == "3":
    print("three")

``

data[0] is bytes, so comparing it to the string "3" is always False, and nothing is printedprint needs parentheses in Python 3split() must be called with "\n", or the first token is the whole inputread() must be given a size, or it returns an empty value

3. Which statements about ios::sync_with_stdio(false) are true?

Pick every answer that is true.

After it, a program that mixes cout and printf can print lines out of program orderIt makes cout and printf share one buffer, so their order never changesAfter it, pick one output family, cout or printf, and use it throughoutIt makes an interactive problem safe without any flush

4. An interactive program starts with cin.tie(nullptr);. It prints the query with cout << "? " << mid << '\n'; and then reads the judge's reply with cin. The judge reports TIMEOUT after 0 queries. Which change fixes it?

Raise the stack size with ulimit -sPrint the query with printf instead of coutAdd cout.flush(); after the line that prints the queryRemove #include <bits/stdc++.h> and include only <iostream>

5. On the lesson's test file of 1,000,000 integers, the Python input() loop took a median of 1.70 s, and sys.stdin.buffer.read().split() took 0.32 s. How many seconds does the second version save?

Type a number.

Fast I/O and the template