#pragma GCC target("sse,sse2,sse3,ssse3,sse4.1,sse4.2,avx,avx2,popcnt,tune=native")

#include <bits/stdc++.h>

typedef long long ll;
typedef unsigned int ui;
typedef unsigned long long ull;

namespace standard_utilities {
template <typename Ty>
inline Ty max(const Ty &a, const Ty &b) {
    return a > b ? a : b;
}

template <typename Ty>
inline Ty min(const Ty &a, const Ty &b) {
    return a < b ? a : b;
}

template <typename Ty>
inline void check_max(Ty &a, const Ty &b) {
    if (b > a)
        a = b;
}

template <typename Ty>
inline void check_min(Ty &a, const Ty &b) {
    if (b < a)
        a = b;
}

template <typename Ty>
inline void swap(Ty &a, Ty &b) {
    Ty c = a;
    a = b, b = c;
}
} // namespace standard_utilities

using namespace standard_utilities;

namespace Fast_IO {
const int MaxBuff = 1 << 15;
const int MaxOut = 1 << 24;

char b[MaxBuff], *S = b, *T = b;
char buff[MaxOut], *iter = buff;

inline void flush() {
    fwrite(buff, 1, iter - buff, stdout), iter = buff;
}

inline char get_char() {
    return (S == T && (T = (S = b) + fread(b, 1, MaxBuff, stdin), S == T) ? 0 : *S++);
}

inline void put_char(char c) {
    if (iter - buff >= 1 << 24 - 1)
        flush();

    *iter++ = c;
}

struct Fast_IO_template {
    ~Fast_IO_template() {
        flush();
    }

    inline Fast_IO_template &operator>>(char &c) {
        c = get_char();
        return *this;
    }

    inline Fast_IO_template &operator<<(char c) {
        put_char(c);
        return *this;
    }

    inline Fast_IO_template &operator>>(char *s) {
        char *iter = s;

        while (*iter = get_char(), *iter == ' ' || *iter == '\n' || *iter == '\r');

        while (*++iter = get_char(), *iter && *iter != ' ' && *iter != '\n' && *iter != '\r');

        *iter = 0;
        return *this;
    }

    inline Fast_IO_template &operator<<(int x) {
        static int stack[110];
        int O = 0;

        if (!x)
            put_char('0');
        else {
            if (x < 0)
                x = -x, put_char('-');

            for (; x; x /= 10)
                stack[++O] = x % 10;

            for (; O; put_char('0' + stack[O--]));
        }

        return *this;
    }
};
} // namespace Fast_IO

Fast_IO::Fast_IO_template io;

const int N = 1e6 + 10;

struct my_tuple {
    int x, y, z;

    inline bool operator < (const my_tuple &b) const {
        if (x != b.x)
            return x < b.x;

        if (y != b.y)
            return y < b.y;

        return z < b.z;
    }

    inline int operator[](const int b) {
        if (b == 0)
            return x;

        if (b == 1)
            return y;

        if (b == 2)
            return z;
    }

    inline bool operator == (const my_tuple &b) {
        return x == b.x && y == b.y && z == b.z;
    }

    inline bool operator != (const my_tuple &b) {
        return x != b.x || y != b.y || z != b.z;
    }
};

struct my_pair {
    int x, y;

    inline int &operator[](const int b) {
        return b == 0 ? x : y;
    }
};

namespace SAIS {
#define alloc(x, type, len) type *x = new type[len];

void induced_sort(int n, int m, int *s, bool *T, int *sa, int *cnt) {
    static int lptr[N], rptr[N];
    int tmp;
    lptr[0] = rptr[0] = 0;

    for (int i = 1; i <= m; ++i)
        lptr[i] = cnt[i - 1], rptr[i] = cnt[i] - 1;

    for (int i = 0; i <= n; ++i)
        sa[i] > 0 && !T[tmp = sa[i] - 1] ? sa[lptr[s[tmp]]++] = tmp : 0;

    for (int i = n; i >= 0; --i)
        sa[i] > 0 &&  T[tmp = sa[i] - 1] ? sa[rptr[s[tmp]]--] = tmp : 0;
}

bool compare(int *x, int *y, int n) {
    while (n--)
        if (*x++ != *y++)
            return 1;

    return 0;
}

void sais(int n, int m, int *s, int *sa) {
    --n;
    alloc(T, bool, n + 10);
    alloc(cnt, int, m + 10);
    alloc(lms, int, n + 10);
    alloc(tmp, int, n + 10);

    for (int i = n; i >= 0; --i) {
        T[i] = (i == n || s[i] < s[i + 1] || s[i] == s[i + 1] && T[i + 1]);
        ++cnt[s[i]];
    }

    tmp[0] = 0;

    for (int i = 1; i <= m; ++i)
        cnt[i] += cnt[i - 1], tmp[i] = cnt[i] - 1;

    int LMS = 0;

    for (int i = 1; i <= n; ++i)
        T[i] && !T[i - 1] ? lms[LMS++] = i : 0;

    memset(sa, -1, (n + 1) << 2);

    for (int i = 0; i < LMS; ++i)
        sa[tmp[s[lms[i]]]--] = lms[i];

    induced_sort(n, m, s, T, sa, cnt);
    alloc(len, int, n + 10);
    lms[LMS] = n + 1;

    for (int i = 0; i < LMS; ++i)
        len[lms[i]] = lms[i + 1] - lms[i] + 1;

    alloc(lab, int, n + 10);
    int m0 = 0;
    memset(lab, -1, (n + 1) << 2);

    for (int i = 1, p = -1, q = -1; i <= n; ++i)
        if ((q = sa[i]) > 0 && T[q] && !T[q - 1]) {
            if (p == -1 || len[p] != len[q] || compare(s + p, s + q, len[p]))
                ++m0;

            lab[q] = m0;
            p = q;
        }

    lab[n] = 0;
    alloc(s0, int, LMS + 10);
    alloc(sa0, int, LMS + 10);

    for (int i = 0, p = 0; i <= n; ++i)
        lab[i] >= 0 ? s0[p++] = lab[i] : 0;

    if (m0 + 1 == LMS)
        for (int i = 0; i < LMS; ++i)
            sa0[s0[i]] = i;
    else
        sais(LMS, m0, s0, sa0);

    tmp[0] = 0;

    for (int i = 1; i <= m; ++i)
        tmp[i] = cnt[i] - 1;

    memset(sa, -1, (n + 1) << 2);

    for (int i = LMS - 1; i >= 0; --i)
        sa[tmp[s[lms[sa0[i]]]]--] = lms[sa0[i]];

    induced_sort(n, m, s, T, sa, cnt);
}
} // namespace SAIS

inline int log2n(int n) {
    return 32 - __builtin_clz(n - 1);
}

namespace RMQ_table {
const int N = 2e6 + 20;
const int D_max = 21;
const int L = 20;
const int inf = 1 << 30;

int stack[L + 1];

template <typename T, int MAXN>
class ST {
private:
    int f[N / L][D_max], M[N / L], a[N + L], n, q, D;

    struct node {
        int state[L + 1], *a;

        int Qmin(int x, int y) {
            return a[x + __builtin_ctz(state[y] >> x)];
        }

        void init(int *_a) {
            int top = 0;
            a = _a;

            for (int i = 1; i <= L; ++i) {
                state[i] = state[i - 1];

                while (top && a[i] <= a[stack[top]])
                    state[i] -= 1 << stack[top], --top;

                stack[++top] = i;
                state[i] += 1 << i;
            }
        }
    } c[N / L];

public:
    void build(int _n, int *_a) {
        n = _n;

        for (int i = 1; i <= _n; ++i)
            a[i] = _a[i];

        D = int(log2(n)) + 1;
        int nn = n / L;
        M[0] = -1;

        for (int i = 1; i <= nn; ++i) {
            f[i][0] = inf;

            for (int j = 1; j <= L; ++j)
                f[i][0] = min(f[i][0], a[(i - 1) * L + j]);
        }

        for (int i = 1; i <= nn; ++i)
            M[i] = !(i & (i - 1)) ? M[i - 1] + 1 : M[i - 1];

        for (int j = 1; j <= D; ++j)
            for (int i = 1; i <= nn - (1 << j) + 1; ++i)
                f[i][j] = min(f[i][j - 1], f[i + (1 << (j - 1))][j - 1]);

        for (int i = 1; i <= nn + 1; ++i)
            c[i].init(a + (i - 1) * L);
    }

    inline int Qmin_ST(int x, int y) {
        int z = M[y - x + 1];
        return min(f[x][z], f[y - (1 << z) + 1][z]);
    }

    inline int rmq(int x, int y) {
        ++x, ++y;
        int xx = (x - 1) / L + 1, yy = (y - 1) / L + 1, res;

        if (xx + 1 <= yy - 1)
            res = Qmin_ST(xx + 1, yy - 1);
        else
            res = inf;

        if (xx == yy)
            res = min(res, c[xx].Qmin(x - (xx - 1) * L, y - (yy - 1) * L));
        else
            res = min(res, c[xx].Qmin(x - (xx - 1) * L, L)), res = min(res, c[yy].Qmin(1, y - (yy - 1) * L));

        return res;
    }
};
} // namespace RMQ_table

using RMQ_table::ST;

template <int N, template <typename, int> class RMQ_T>
class SA {
public:
    RMQ_T < int, N - 1 > rmq;
    int n, sa[N], rk[N], height[N];

    void build(int _n, char *str) {
        n = _n;
        alloc(s, int, n + 10);

        for (int i = 0; i < n; ++i)
            s[i] = str[i];

        SAIS::sais(n + 1, 26, s, sa);

        for (int i = 1; i <= n; ++i)
            rk[sa[i]] = i;

        for (int k = 0, i = 0, j; i < n; height[rk[i++]] = k)
            for (k ? k-- : 0, j = sa[rk[i] - 1]; s[i + k] == s[j + k]; ++k);

        rmq.build(n, height);
    }

    int lcp(int i, int j) {
        //printf("lcp(%d,%d)\n", i, j);
        if (i == j)
            return n - i;

        if (j == n)
            return 0;

        i = rk[i], j = rk[j];

        if (i > j)
            std::swap(i, j);

        return rmq.rmq(i, j - 1);
    }
};

template <int N, template <typename, int> class RMQ_T>
class ISA {
public:
    RMQ_T < int, N - 1 > rmq;
    int n, sa[N], rk[N], height[N];

    void build(int _n, char *str) {
        n = _n;
        alloc(s, int, n + 10);

        for (int i = 0; i < n; ++i)
            s[i] = str[n - i - 1];

        SAIS::sais(n + 1, 26, s, sa);

        for (int i = 1; i <= n; ++i)
            rk[sa[i]] = i;

        for (int k = 0, i = 0, j; i < n; height[rk[i++]] = k)
            for (k ? k-- : 0, j = sa[rk[i] - 1]; s[i + k] == s[j + k]; ++k);

        rmq.build(n, height);
        //for (int i = 2; i <= n + 1; ++i) printf("%d ", height[i]); printf("\n");
        //for (int i = 0; i < n; ++i) printf("%d ", rk[i]); printf("\n");
    }

    int lcs(int i, int j) {
        //++i, ++j;
        if (i <= 0 || j <= 0)
            return 0;

        if (i == j)
            return i;

        i = rk[n - i], j = rk[n - j];

        if (i > j)
            std::swap(i, j);

        //printf("rmq(%d,%d)", i, j);
        return rmq.rmq(i, j - 1);
    }
};

int n, nsv[N], nbv[N];
char s[N];
SA<N, ST> sa;
ISA<N, ST> isa;
std::vector<my_tuple> runs;

int get_lcs(int i, int j) {
    int low = 0, high = min(i, j);

    while (low < high) {
        int middle = low + high + 1 >> 1;

        if (sa.lcp(i - middle, j - middle) >= middle)
            low = middle;
        else
            high = middle - 1;
    }

    //printf("lcs(%d,%d) : %d %d\n", i, j, isa.lcs(i, j), low);
    return low;
}

void add_run(int i, int j) {
    int length = j - i;
    int lcs = isa.lcs(i, j);

    if (lcs < length) {
        int lcp = sa.lcp(i, j);

        if (lcs + lcp >= length)
            runs.push_back((my_tuple) {
            i - lcs + 1, j + lcp, length
        });
    }
}

std::vector<int> a[N], a2[N], a3[N];
int b[N], c[N];

void my_sort_and_unique() {
    int m = runs.size();

    for (int i = 0; i < m; ++i)
        a[runs[i].z].push_back(i);

    for (int i = 1, tmp = 0; i <= n; ++i)
        for (int j = 0; j < a[i].size(); ++j)
            b[tmp++] = a[i][j];

    for (int i = 0; i < m; ++i)
        a2[runs[b[i]].y].push_back(b[i]);

    for (int i = 1, tmp = 0; i <= n; ++i)
        for (int j = 0; j < a2[i].size(); ++j)
            b[tmp++] = a2[i][j];

    for (int i = 0; i < m; ++i)
        a3[runs[b[i]].x].push_back(b[i]);

    for (int i = 1, tmp = 0; i <= n; ++i)
        for (int j = 0; j < a3[i].size(); ++j)
            b[tmp++] = a3[i][j];

    int tmp = 0;
    c[0] = b[0];

    for (int i = 1; i < m; ++i)
        if (runs[b[i - 1]] != runs[b[i]])
            c[++tmp] = b[i];

    io << tmp + 1 << '\n';

    for (int i = 0; i <= tmp; ++i)
        io << runs[c[i]].x << ' ' << runs[c[i]].y << ' ' << runs[c[i]].z << '\n';
}

int main() {
    io >> s;
    n = strlen(s);

    for (int i = 0; i < n; ++i)
        s[i] = s[i] - 'a' + 1;

    sa.build(n, s);
    isa.build(n, s);

    for (int i = n; i--;) {
        for (int &j = nsv[i] = i + 1; j < n && sa.rk[i] < sa.rk[j];)
            j = nsv[j];

        for (int &j = nbv[i] = i + 1; j < n && sa.rk[i] > sa.rk[j];)
            j = nbv[j];

        add_run(i, nsv[i]);

        if (nsv[i] != nbv[i])
            add_run(i, nbv[i]);
    }

    std::sort(runs.begin(), runs.end());
    runs.erase(std::unique(runs.begin(), runs.end()), runs.end());
    io << (int)runs.size() << '\n';

    for (auto a : runs)
        io << a.x << ' ' << a.y << ' ' << a.z << '\n';

    return 0;
}

In file included from /usr/include/x86_64-linux-gnu/c++/11/bits/gthr.h:148,
                 from /usr/include/c++/11/ext/atomicity.h:35,
                 from /usr/include/c++/11/bits/ios_base.h:39,
                 from /usr/include/c++/11/ios:42,
                 from /usr/include/c++/11/istream:38,
                 from /usr/include/c++/11/sstream:38,
                 from /usr/include/c++/11/complex:45,
                 from /usr/include/c++/11/ccomplex:39,
                 from /usr/include/x86_64-linux-gnu/c++/11/bits/stdc++.h:54,
                 from answer.code:3:
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:102:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
  102 | __gthrw(pthread_once)
      | ^~~~~~~
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:102:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:103:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
  103 | __gthrw(pthread_getspecific)
      | ^~~~~~~
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:103:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:104:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
  104 | __gthrw(pthread_setspecific)
      | ^~~~~~~
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:104:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:106:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
  106 | __gthrw(pthread_create)
      | ^~~~~~~
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:106:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:107:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
  107 | __gthrw(pthread_join)
      | ^~~~~~~
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:107:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:108:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
  108 | __gthrw(pthread_equal)
      | ^~~~~~~
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:108:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:109:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
  109 | __gthrw(pthread_self)
      | ^~~~~~~
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:109:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:110:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
  110 | __gthrw(pthread_detach)
      | ^~~~~~~
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:110:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:112:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
  112 | __gthrw(pthread_cancel)
      | ^~~~~~~
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:112:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:114:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
  114 | __gthrw(sched_yield)
      | ^~~~~~~
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:114:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:116:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
  116 | __gthrw(pthread_mutex_lock)
      | ^~~~~~~
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:116:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:117:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
  117 | __gthrw(pthread_mutex_trylock)
      | ^~~~~~~
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:117:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:119:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
  119 | __gthrw(pthread_mutex_timedlock)
      | ^~~~~~~
/usr/include/x86_64-linux-gnu/c++/11/bits/gthr-default.h:119:1: error: attribute value ‘tune=native’ was already specified in ‘target’ attribute
/usr/...