import sys
from collections import defaultdict
import numpy as np
from scipy.sparse.csgraph import connected_components, minimum_spanning_tree
from scipy.sparse import csr_matrix

input = sys.stdin.buffer.readline
MAX = 10 ** 10

N, M, K = map(int, input().split())
if not M: print(0);exit()

ABC = np.array([tuple(map(int, input().split())) for _ in range(M)], dtype=np.int64)
ABC[:, :2] -= 1

ans = ABC[:, 2].sum()
if K:
    E = np.array([int(input()) - 1 for _ in range(K)])
    frm = ABC[E, 0]
    to = ABC[E, 1]
    ans -= ABC[E, 2].sum()
    connection = csr_matrix(([1] * K, (frm, to)), shape=(N, N))
    compress_N, labels = connected_components(connection)
    ABC[:, :2] = labels[ABC[:, :2]]
else:
    compress_N = N

d = defaultdict(lambda: MAX)
for a, b, c in ABC:
    if a == b:
        continue
    if a > b: a, b = b, a
    d[(a, b)] = min(d[(a, b)], c)

length, frm, to = [], [], []
for (a, b), c in d.items():
    length.append(c)
    frm.append(a)
    to.append(b)
matr = csr_matrix((length, (frm, to)), shape=(compress_N, compress_N))
T = minimum_spanning_tree(matr)
ans -= int(T.sum())
print(ans)