#include<iostream>
#include<stack>
#include<map>
#include<vector>
#include<string>
#include<cassert>
#include<unordered_map>
#include<queue>
#include<cstdint>
#include<cstring>
#include<limits.h>
#include<cmath>
#include<set>
#include<algorithm>
#include<iomanip>
#include<numeric>
#include<bitset>
using namespace std;
#define ll long long
#define f first
#define s second
#define pii pair<int,int>
#define ppii pair<int,pii>
#define vi vector<int>
#define pb push_back
#define all(x) x.begin(),x.end()
#define rall(x) x.rbegin(),x.rend()
#define F(n) for(int i=0;i<n;i++)
#define lb lower_bound
#define ub upper_bound
#define fastio ios::sync_with_stdio(false);cin.tie(NULL);
#pragma GCC optimize ("03,unroll-loops")
#define int long long
#define double long double
using namespace std;
const int mod = 998244353, mxn = 1e6+5, inf = 1e16, minf = -1e16, lg = 30;
//#undef int
int n, k, m, q;
void setIO(string name) {
ios_base::sync_with_stdio(0);
cin.tie(0);
freopen((name + ".in").c_str(), "r", stdin);
freopen((name + ".out").c_str(), "w", stdout);
}
pair<int,pii> v1[mxn+10], v2[mxn+10];
//time,before,point
struct seg {
int v[4*mxn+10], lazy1[4*mxn+10], lazy2[4*mxn+10];
void build(int l, int r, int pos) {
int mid = l + (r - l) / 2;
lazy2[pos] = minf;
if (l == r) return;
build(l, mid, pos * 2);
build(mid + 1, r, pos * 2 + 1);
}
void push(int l, int r, int pos) {
v[pos] = max(v[pos] + lazy1[pos], lazy2[pos]);
if (l != r) {
if (lazy2[pos * 2] != minf) lazy2[pos * 2] += lazy1[pos];
if (lazy2[pos * 2 + 1] != minf) lazy2[pos * 2 + 1] += lazy1[pos];
lazy1[pos * 2 + 1] += lazy1[pos];
lazy1[pos * 2] += lazy1[pos];
lazy2[pos * 2] = max(lazy2[pos * 2], lazy2[pos]);
lazy2[pos * 2 + 1] = max(lazy2[pos * 2 + 1], lazy2[pos]);
}
lazy2[pos] = minf;
lazy1[pos] = 0;
}
void updateadd(int l, int r, int ql, int qr, int val, int pos) {
push(l, r, pos);
if (l > qr || r < ql) return;
if (l >= ql && r <= qr) {
lazy1[pos] += val;
push(l, r, pos);
return;
}
int mid = l + (r - l) / 2;
updateadd(l, mid, ql, qr, val, pos * 2);
updateadd(mid + 1, r, ql, qr, val, pos * 2 + 1);
v[pos] = max(v[pos * 2], v[pos * 2 + 1]);
}
void updateset(int l, int r, int ql, int qr, int val, int pos) {
push(l, r, pos);
if (l > qr || r < ql) return;
if (l >= ql && r <= qr) {
lazy2[pos] = max(lazy2[pos], val);
push(l, r, pos);
return;
}
int mid = l + (r - l) / 2;
updateset(l, mid, ql, qr, val, pos * 2);
updateset(mid + 1, r, ql, qr, val, pos * 2 + 1);
v[pos] = max(v[pos * 2], v[pos * 2 + 1]);
}
int qry(int l, int r, int qpos, int pos) {
push(l, r, pos);
if (l == r) return v[pos];
int mid = l + (r - l) / 2;
if (qpos <= mid) return qry(l, mid, qpos, pos * 2);
return qry(mid + 1, r, qpos, pos * 2 + 1);
}
} t;
int getbound(int x, int add) {
int l = 0, r = n, pos = -1;
while (l <= r) {
int mid = l + (r - l) / 2;
if (v1[mid].f + add <= x) l = mid + 1, pos = max(pos, mid);
else r = mid - 1;
}
return pos;
}
int32_t main() {
fastio
//n^2 can use lazy seg to op
cin >> n >> m;
for (int i = 1; i <= n; i++) {
cin >> v1[i].f >> v1[i].s.f >> v1[i].s.s;
v1[i].f += v1[i-1].f;
}
for (int i = 1; i <= m; i++) {
cin >> v2[i].f >> v2[i].s.f >> v2[i].s.s;
v2[i].f += v2[i-1].f;
}
int ans1 = 0;
vector<pii> gap;
for (int i = 1; i <= n; i++) {
if (v1[i].f <= v1[i].s.f) {
gap.pb({v1[i].s.f - v1[i].f, i});
ans1 += v1[i].s.s;
}
}
t.build(0, n, 1);
sort(all(gap));
int cur = 0, bruh, where, x;
vector<int32_t> cut;
for (int j = 1; j <= m; j++) {
while (cur < gap.size() && gap[cur].f < v2[j].f) {
t.updateadd(0, n, 0, gap[cur].s - 1, -v1[gap[cur].s].s.s, 1);
if (v1[gap[cur].s].s.s < 0) cut.pb(gap[cur].s - 1);
cur++;
}
x = getbound(v2[j].s.f, v2[j].f);
if (x != -1) {
cut.pb(x);
t.updateadd(0, n, 0, x, v2[j].s.s, 1);
if (v2[j].s.s > 0) cut.pb(x);
}
for (auto i : cut) {
bruh = t.qry(0, n, i, 1);
t.updateset(0, n, i + 1, n, bruh, 1);
}
cut.clear();
}
t.push(0, n, 1);
cout << ans1 + t.v[1] << '\n';
}
/*
*/
