Quickstart · about 15 minutes

CP Lang for Python programmers

Six short problems. Each shows how you'd write it in Python, then the CP Lang solution. Edit it, break it, submit it: your browser judges it. A solution the compiler does not run yet is shown without a judge.

  1. 1 / 6 · read, ranges as input contracts, print

    Sum of a list

    How you'd write it in Python
    n = int(input())
    print(sum(map(int, input().split())))

    Read n (1 ≤ n ≤ 2·10⁵) and then n integers between -10⁹ and 10⁹. Print their sum.

    3 tests · Ctrl+Enter

    What's different from Python

    • Types are written where data enters: at read and in function signatures. Everywhere else they are inferred.
    • This compiles to C++, so it runs at C++ speed.
  2. 2 / 6 · var, while, two pointers

    Pairs with a target sum

    How you'd write it in Python
    i, j, pairs = 0, n - 1, 0
    while i < j:
        s = a[i] + a[j]
        if s == x: pairs += 1; i += 1; j -= 1
        elif s < x: i += 1
        else: j -= 1

    Read n and x, then n distinct integers in increasing order. Count the pairs i < j with a[i] + a[j] = x.

    3 tests · Ctrl+Enter

    What's different from Python

    • let is a constant, var is a variable. Most names are let.
    • elif and while read the same as in Python.
  3. 3 / 6 · domains, grids, ?Int and ??

    Nearest exit on a grid

    How you'd write it in Python
    dist = [[-1] * w for _ in range(h)]
    q = deque([s]); dist[s[0]][s[1]] = 0
    while q:
        r, c = q.popleft()
        for dr, dc in ((1,0),(-1,0),(0,1),(0,-1)):
            nr, nc = r + dr, c + dc
            if 0 <= nr < h and 0 <= nc < w and ...

    Read an h × w grid of '.', '#', one 'S' and some 'E'. Print the fewest steps from S to any E, moving up, down, left or right, or -1.

    3 tests · Ctrl+Enter

    What's different from Python

    • None and Some(x) are checked at compile time: you cannot add 1 to a distance that might be None.
  4. 4 / 6 · functions, Mod<P>, memoized recursion

    Count paths in a DAG

    How you'd write it in Python
    @lru_cache(None)
    def paths(v):
        if v == n: return 1
        return sum(paths(w) for w in adj[v]) % MOD
    # plus sys.setrecursionlimit and threading for deep graphs

    Read a DAG with n vertices and m edges u → v. Print the number of paths from 1 to n modulo 10⁹ + 7.

    3 tests · Ctrl+Enter

    What's different from Python

    • No recursion limit: recursion depth is limited by the stack, which judge profiles set to the judge's size.
  5. 5 / 6 · the library and operation values

    Range minimum with updates

    How you'd write it in Python
    # Usually a hand-written iterative segment tree, and often too slow in CPython.

    Read n values and q queries: "1 i x" sets a[i] = x, "2 l r" prints the minimum of a[l..r].

    runs
    let (n, q): (1..=2e5, 1..=2e5) = read()
    domain V = 1..=n
    let a: [V]0..=1e9 = read()
    var st = SegTree(min, a)
    for _ in 0..q:
        let t: 1..=2 = read()
        if t == 1:
            let (i, x): (V, 0..=1e9) = read()
            st.set(i, x)
        else:
            let (l, r): (V, V) = read()
            print(st.fold(l..=r))

    What's different from Python

    • SegTree is in the standard library and compiles to a flat array.
  6. 6 / 6 · @!multitest and why state can't leak

    Several test cases

    How you'd write it in Python
    for _ in range(int(input())):
        solve()

    Problem 3 again, but the input starts with t, the number of test cases.

    1 tests · Ctrl+Enter

    What's different from Python

    • The single-test and multitest programs differ by one line.

You've read six programs

Next: problems with verdicts, or look things up.