Codeforces 601D. Acyclic Organic Compounds（四个愿望一次满足）

Posted 2020-10-17

　　trie合并的裸题...因为最多只有n个点，所以最多合并n次，复杂度$O(N*26)$。

#include<iostream>
#include<cstring>
#include<cstdlib>
#include<cstdio>
using namespace std;
const int maxn=500010, inf=1e9;
struct poi{int too, pre;}e[maxn<<1];
struct tjm{int nxt[26], size;}tree[maxn];
int n, x, y, ans, cnt, tot;
int last[maxn], c[maxn];
char s[maxn];
inline void read(int &k)
{
    int f=1; k=0; char c=getchar();
    while(c<‘0‘ || c>‘9‘) c==‘-‘ && (f=-1), c=getchar();
    while(c<=‘9‘ && c>=‘0‘) k=k*10+c-‘0‘, c=getchar();
    k*=f;
}
inline void add(int x, int y){e[++tot]=(poi){y, last[x]}; last[x]=tot;}
void merge(int &x, int y)
{
    if(!x || !y) {x+=y; return;}
    tree[x].size=1;
    for(int i=0;i<26;i++)
        merge(tree[x].nxt[i], tree[y].nxt[i]), tree[x].size+=tree[tree[x].nxt[i]].size;
}
void dfs(int x, int fa)
{
    tree[x].size=1;
    for(int i=last[x], too;i;i=e[i].pre)
    if((too=e[i].too)!=fa)
    {
        dfs(too, x); int tmp=tree[tree[x].nxt[s[e[i].too]-‘a‘]].size;
        merge(tree[x].nxt[s[e[i].too]-‘a‘], e[i].too);
        tree[x].size+=tree[tree[x].nxt[s[e[i].too]-‘a‘]].size-tmp; 
    }
    if(tree[x].size+c[x]>ans) ans=tree[x].size+c[x], cnt=1;
    else if(tree[x].size+c[x]==ans) cnt++;
}
int main()
{
    read(n);
    for(int i=1;i<=n;i++) read(c[i]); scanf("%s", s+1);
    for(int i=1;i<n;i++) read(x), read(y), add(x, y), add(y, x);
    dfs(1, 0); printf("%d\n%d", ans, cnt);
}

View Code

　　当然还可以直接上哈希+平衡树+启发式合并，用set自带去重就好了，复杂度$O(Nlog^2N)$。

#include<iostream>
#include<cstring>
#include<cstdlib>
#include<cstdio>
#include<set>
#define ll long long
using namespace std;
const int maxn=500010, inf=1e9;
const ll mod=233333333333333333;
typedef set<ll>::iterator ddq;
struct poi{int too, pre;}e[maxn<<1];
int n, x, y, tot, ans, ansnum;
int last[maxn], c[maxn], root[maxn];
ll hs[maxn];
char ch[maxn];
set<ll>s[maxn];
inline void read(int &k)
{
    int f=1; k=0; char c=getchar();
    while(c<‘0‘ || c>‘9‘) c==‘-‘ && (f=-1), c=getchar();
    while(c<=‘9‘ && c>=‘0‘) k=k*10+c-‘0‘, c=getchar();
    k*=f;
}
inline void add(int x, int y){e[++tot]=(poi){y, last[x]}; last[x]=tot;}
inline int merge(int x, int y)
{
    if(s[x].size()<s[y].size()) swap(x, y);
    for(ddq ity=s[y].begin();ity!=s[y].end();ity++) s[x].insert(*ity);
    s[y].clear(); return x;
}
void dfs(int x, int fa)
{
    hs[x]=(hs[fa]*27+ch[x]-‘a‘+1)%mod;
    s[x].insert(hs[x]); root[x]=x;
    for(int i=last[x], too;i;i=e[i].pre)
    if((too=e[i].too)!=fa)
    {
        dfs(too, x);
        root[x]=merge(root[x], root[too]);
    }
    int size=s[root[x]].size();
    if(size+c[x]>ans) ans=size+c[x], ansnum=1;
    else if(size+c[x]==ans) ansnum++;
}
int main()
{
    read(n);
    for(int i=1;i<=n;i++) read(c[i]);
    scanf("%s", ch+1);
    for(int i=1;i<n;i++) read(x), read(y), add(x, y), add(y, x);
    dfs(1, 0); printf("%d\n%d", ans, ansnum);
}

View Code

　　当然还可以hash之后用dsu on tree，复杂度$O(NlogN)$。

#include<iostream>
#include<cstring>
#include<cstdlib>
#include<cstdio>
#include<algorithm>
#define ll long long
using namespace std;
const int maxn=500010, inf=1e9;
const ll mod=233333333333333333;
struct poi{int too, pre;}e[maxn<<1];
int n, x, y, N, tot, sum, skip, ANS, ANSNUM;
int c[maxn], last[maxn], son[maxn], size[maxn], cnt[maxn], ans[maxn];
ll b[maxn], hs[maxn];
char s[maxn];
inline void read(int &k)
{
    int f=1; k=0; char c=getchar();
    while(c<‘0‘ || c>‘9‘) c==‘-‘ && (f=-1), c=getchar();
    while(c<=‘9‘ && c>=‘0‘) k=k*10+c-‘0‘, c=getchar();
    k*=f;
}
inline void add(int x, int y){e[++tot]=(poi){y, last[x]}; last[x]=tot;}
void dfs1(int x, int fa)
{
    size[x]=1; b[x]=hs[x]=(hs[fa]*27+s[x]-‘a‘+1)%mod;
    for(int i=last[x], too;i;i=e[i].pre)
    if((too=e[i].too)!=fa)
    {
        dfs1(too, x);
        size[x]+=size[too];
        if(size[too]>size[son[x]]) son[x]=too;
    }
}
void update(int x, int fa, int delta)
{
    if(delta==1) sum+=(!cnt[hs[x]]);
    cnt[hs[x]]+=delta;
    for(int i=last[x], too;i;i=e[i].pre)
    if((too=e[i].too)!=fa && too!=skip) update(too, x, delta);
}
void dfs2(int x, int fa, bool heavy)
{
    for(int i=last[x], too;i;i=e[i].pre)
    if((too=e[i].too)!=fa && too!=son[x]) dfs2(too, x, 0);
    if(son[x]) dfs2(son[x], x, 1), skip=son[x];
    update(x, fa, 1); ans[x]=sum; skip=0;
    if(!heavy) update(x, fa, -1), sum=0; 
    if(ans[x]+c[x]>ANS) ANS=ans[x]+c[x], ANSNUM=1;
    else if(ans[x]+c[x]==ANS) ANSNUM++;
}
int main()
{
    read(n);
    for(int i=1;i<=n;i++) read(c[i]);
    scanf("%s", s+1);
    for(int i=1;i<n;i++) read(x), read(y), add(x, y), add(y, x);
    dfs1(1, 0);
    N=n; sort(b+1, b+1+N); N=unique(b+1, b+1+N)-b-1;
    for(int i=1;i<=n;i++) hs[i]=lower_bound(b+1, b+1+N, hs[i])-b;
    dfs2(1, 0, 1); printf("%d\n%d", ANS, ANSNUM);
}

View Code

　　还可以hash之后写线段树合并，复杂度$O(NlogN)$。（真的懒得写了T T