洛谷 P5487 【模板】线性递推+BM算法（BM+多项式取模）

传送门
这是一道板题。
要用到$BM$算法和常系数齐次线性递推。
代码：

#include<bits/stdc++.h>
#define ri register int
using namespace std;
const int rlen=1<<18|1;
inline char gc(){
	static char buf[rlen],*ib,*ob;
	(ib==ob)&&(ob=(ib=buf)+fread(buf,1,rlen,stdin));
	return ib==ob?-1:*ib++;
}
inline int read(){
	int ans=0;
	char ch=gc();
	while(!isdigit(ch))ch=gc();
	while(isdigit(ch))ans=((ans<<2)+ans<<1)+(ch^48),ch=gc();
	return ans;
}
typedef long long ll;
typedef vector<int> poly;
const int mod=998244353;
inline int add(int a,int b){return (a+=b)>=mod?a-mod:a;}
inline int dec(int a,int b){return (a-=b)<0?a+mod:a;}
inline int mul(int a,int b){return (ll)a*b%mod;}
inline void Add(int&a,int b){(a+=b)>=mod?a-=mod:a;}
inline void Dec(int&a,int b){(a-=b)<0?a+=mod:a;}
inline void Mul(int&a,int b){a=(ll)a*b%mod;}
inline int ksm(int a,int p){int ret=1;for(;p;p>>=1,Mul(a,a))if(p&1)Mul(ret,a);return ret;}
int invv[23],w[23],lim,tim;
vector<int>rev[23];
inline void init_w(){
	w[22]=ksm(3,(mod-1)>>23),invv[0]=1;
	for(ri i=21;~i;--i)w[i]=mul(w[i+1],w[i+1]);
	for(ri i=1,iv=mod+1>>1;i<23;++i)invv[i]=mul(invv[i-1],iv);
}
inline void init(const int&up){
	lim=1,tim=0;
	while(lim<up)lim<<=1,++tim;
	if(rev[tim].size())return;
	rev[tim].resize(lim);
	for(ri i=0;i<lim;++i)rev[tim][i]=(rev[tim][i>>1]>>1)|((i&1)<<(tim-1));
}
inline void ntt(poly&a,const int&type){
	for(ri i=0;i<lim;++i)if(i<rev[tim][i])swap(a[i],a[rev[tim][i]]);
	for(ri i=1,a0,a1,t=0;i<lim;i<<=1,++t)for(ri j=0,len=i<<1;j<lim;j+=len)
	for(ri k=0,mt=1;k<i;++k,Mul(mt,w[t]))a0=a[j+k],a1=mul(a[j+k+i],mt),a[j+k]=add(a0,a1),a[j+k+i]=dec(a0,a1);
	if(~type)return;
	reverse(++a.begin(),a.end());
	for(ri i=0;i<lim;++i)Mul(a[i],invv[tim]);
}
inline poly operator+(poly a,poly b){
	int n=a.size(),m=b.size();
	poly c(max(n,m));
	for(ri i=0;i<n;++i)Add(c[i],a[i]);
	for(ri i=0;i<m;++i)Add(c[i],b[i]);
	return c;
}
inline poly operator-(poly a,poly b){
	int n=a.size(),m=b.size();
	poly c(max(n,m));
	for(ri i=0;i<n;++i)Add(c[i],a[i]);
	for(ri i=0;i<m;++i)Dec(c[i],b[i]);
	return c;
}
inline poly operator*(poly a,poly b){
	int n=a.size(),m=b.size(),t=n+m-1;
	if(t<=128){
		poly c(t);
		for(ri i=0;i<n;++i)for(ri j=0;j<m;++j)Add(c[i+j],mul(a[i],b[j]));
		return c;
	}
	init(t);
	a.resize(lim),ntt(a,1);
	b.resize(lim),ntt(b,1);
	for(ri i=0;i<lim;++i)Mul(a[i],b[i]);
	return ntt(a,-1),a.resize(t),a;
}
inline poly poly_inv(poly a,int k){
	poly b(1,ksm(a[0],mod-2)),c;
	for(ri i=4,up=k<<2;i<up;i<<=1){
		init(i);
		c=a,c.resize(i>>1);
		b.resize(lim),ntt(b,1);
		c.resize(lim),ntt(c,1);
		for(ri j=0;j<lim;++j)Mul(b[j],dec(2,mul(b[j],c[j])));
		ntt(b,-1),b.resize(i>>1);
	}
	return b.resize(k),b;
}
inline poly operator/(poly a,poly b){
	int n=a.size(),m=b.size(),t=n-m+1;
	reverse(a.begin(),a.end());
	reverse(b.begin(),b.end()),b=poly_inv(b,t);
	return a=a*b,a.resize(t),reverse(a.begin(),a.end()),a;
}
inline poly operator%(poly a,poly b){if(a.size()<b.size())return a;poly c=a-a/b*b;return c.resize(b.size()),c;}
inline int Cayley_Hamilton(poly coe,int*a,int k,int n){
	poly md(k+1),f(2),g(1);
	md[k]=f[1]=g[0]=1;
	for(ri i=1;i<=k;++i)md[k-i]=coe[i]?mod-coe[i]:0;
	for(;n;n>>=1,f=f*f%md)if(n&1)g=g*f%md;
	int ret=0;
	for(ri i=0;i<k;++i)Add(ret,mul(a[i],g[i]));
	return ret;
}
const int N=10005;
int n,m,k=0,a[N];
namespace Berlekamp_Massey{
	poly R[N];
	int cnt=0,fail[N],det[N];
	inline void update(int n){
		int t=a[n];
		for(ri i=1,up=R[cnt].size();i<up;++i)Dec(t,mul(a[n-i],R[cnt][i]));
		det[n]=t;
		if(!t)return;
		fail[cnt]=n;
		if(!cnt){R[++cnt].resize(n+1);return;}
		Mul(t,ksm(det[fail[cnt-1]],mod-2));
		R[cnt+1].resize(n-fail[cnt-1]);
		R[cnt+1].push_back(t);
		for(ri i=1,up=R[cnt-1].size();i<up;++i)R[cnt+1].push_back(mul(mod-R[cnt-1][i],t));
		R[cnt+1]=R[cnt]+R[cnt+1];
		++cnt;
	}
	inline int solve(){
		for(ri i=1,up=R[cnt].size()-1;i<=up;++i)cout<<R[cnt][i]<<' ';
		puts("");
		return Cayley_Hamilton(R[cnt],a+1,R[cnt].size()-1,n);
	}
}
int main(){
	init_w();
	m=read(),n=read();
	for(ri i=1;i<=m;++i)a[i]=read(),Berlekamp_Massey::update(i);
	cout<<Berlekamp_Massey::solve();
	return 0;
}