Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
238 changes: 125 additions & 113 deletions acl_string.py
Original file line number Diff line number Diff line change
@@ -1,122 +1,121 @@
class string:
def sa_is(s, upper):
n = len(s)
if n == 0:
return []
if n == 1:
return [0]
if n == 2:
if s[0] < s[1]:
return [0, 1]
else:
return [1, 0]
sa = [0] * n
ls = [0] * n
for i in range(n - 2, -1, -1):
ls[i] = ls[i + 1] if (s[i] == s[i + 1]) else (s[i] < s[i + 1])
sum_l = [0] * (upper + 1)
sum_s = [0] * (upper + 1)
for i in range(n):
if not (ls[i]):
sum_s[s[i]] += 1
else:
sum_l[s[i] + 1] += 1
for i in range(upper + 1):
sum_s[i] += sum_l[i]
if i < upper:
sum_l[i + 1] += sum_s[i]
def _sa_is(s, upper):
n = len(s)
if n == 0:
return []
if n == 1:
return [0]
if n == 2:
if s[0] < s[1]:
return [0, 1]
else:
return [1, 0]
sa = [0] * n
ls = [0] * n
for i in range(n - 2, -1, -1):
ls[i] = ls[i + 1] if (s[i] == s[i + 1]) else (s[i] < s[i + 1])
sum_l = [0] * (upper + 1)
sum_s = [0] * (upper + 1)
for i in range(n):
if not (ls[i]):
sum_s[s[i]] += 1
else:
sum_l[s[i] + 1] += 1
for i in range(upper + 1):
sum_s[i] += sum_l[i]
if i < upper:
sum_l[i + 1] += sum_s[i]

def induce(lms):
for i in range(n):
sa[i] = -1
buf = sum_s[:]
for d in lms:
if d == n:
continue
sa[buf[s[d]]] = d
buf[s[d]] += 1
buf = sum_l[:]
sa[buf[s[n - 1]]] = n - 1
buf[s[n - 1]] += 1
for i in range(n):
v = sa[i]
if v >= 1 and not (ls[v - 1]):
sa[buf[s[v - 1]]] = v - 1
buf[s[v - 1]] += 1
buf = sum_l[:]
for i in range(n - 1, -1, -1):
v = sa[i]
if v >= 1 and ls[v - 1]:
buf[s[v - 1] + 1] -= 1
sa[buf[s[v - 1] + 1]] = v - 1
def induce(lms):
for i in range(n):
sa[i] = -1
buf = sum_s[:]
for d in lms:
if d == n:
continue
sa[buf[s[d]]] = d
buf[s[d]] += 1
buf = sum_l[:]
sa[buf[s[n - 1]]] = n - 1
buf[s[n - 1]] += 1
for i in range(n):
v = sa[i]
if v >= 1 and not (ls[v - 1]):
sa[buf[s[v - 1]]] = v - 1
buf[s[v - 1]] += 1
buf = sum_l[:]
for i in range(n - 1, -1, -1):
v = sa[i]
if v >= 1 and ls[v - 1]:
buf[s[v - 1] + 1] -= 1
sa[buf[s[v - 1] + 1]] = v - 1

lms_map = [-1] * (n + 1)
m = 0
for i in range(1, n):
if not (ls[i - 1]) and ls[i]:
lms_map[i] = m
m += 1
lms = []
for i in range(1, n):
if not (ls[i - 1]) and ls[i]:
lms.append(i)
induce(lms)
if m:
sorted_lms = []
for v in sa:
if lms_map[v] != -1:
sorted_lms.append(v)
rec_s = [0] * m
rec_upper = 0
rec_s[lms_map[sorted_lms[0]]] = 0
for i in range(1, m):
l = sorted_lms[i - 1]
r = sorted_lms[i]
end_l = lms[lms_map[l] + 1] if (lms_map[l] + 1 < m) else n
end_r = lms[lms_map[r] + 1] if (lms_map[r] + 1 < m) else n
same = True
if end_l - l != end_r - r:
lms_map = [-1] * (n + 1)
m = 0
for i in range(1, n):
if not (ls[i - 1]) and ls[i]:
lms_map[i] = m
m += 1
lms = []
for i in range(1, n):
if not (ls[i - 1]) and ls[i]:
lms.append(i)
induce(lms)
if m:
sorted_lms = []
for v in sa:
if lms_map[v] != -1:
sorted_lms.append(v)
rec_s = [0] * m
rec_upper = 0
rec_s[lms_map[sorted_lms[0]]] = 0
for i in range(1, m):
l = sorted_lms[i - 1]
r = sorted_lms[i]
end_l = lms[lms_map[l] + 1] if (lms_map[l] + 1 < m) else n
end_r = lms[lms_map[r] + 1] if (lms_map[r] + 1 < m) else n
same = True
if end_l - l != end_r - r:
same = False
else:
while l < end_l:
if s[l] != s[r]:
break
l += 1
r += 1
if (l == n) or (s[l] != s[r]):
same = False
else:
while l < end_l:
if s[l] != s[r]:
break
l += 1
r += 1
if (l == n) or (s[l] != s[r]):
same = False
if not (same):
rec_upper += 1
rec_s[lms_map[sorted_lms[i]]] = rec_upper
rec_sa = string.sa_is(rec_s, rec_upper)
for i in range(m):
sorted_lms[i] = lms[rec_sa[i]]
induce(sorted_lms)
return sa
if not (same):
rec_upper += 1
rec_s[lms_map[sorted_lms[i]]] = rec_upper
rec_sa = _sa_is(rec_s, rec_upper)
for i in range(m):
sorted_lms[i] = lms[rec_sa[i]]
induce(sorted_lms)
return sa


def suffix_array_upper(s, upper):
assert 0 <= upper
for d in s:
assert 0 <= d and d <= upper
return string.sa_is(s, upper)
class String:
def __init__(self, s):
self._s = s

def suffix_array(s):
def suffix_array(self):
s = self._s
n = len(s)
if type(s) == str:
s2 = [ord(i) for i in s]
return string.sa_is(s2, 255)
else:
idx = list(range(n))
idx.sort(key=lambda x: s[x])
s2 = [0] * n
now = 0
for i in range(n):
if i and s[idx[i - 1]] != s[idx[i]]:
now += 1
s2[idx[i]] = now
return string.sa_is(s2, now)
if isinstance(s, str):
return _sa_is([ord(c) for c in s], 255)
idx = sorted(range(n), key=lambda x: s[x])
s2 = [0] * n
now = 0
for i in range(n):
if i and s[idx[i - 1]] != s[idx[i]]:
now += 1
s2[idx[i]] = now
return _sa_is(s2, now)

def lcp_array(s, sa):
def lcp_array(self, sa=None):
if sa is None:
sa = self.suffix_array()
s = self._s
n = len(s)
assert n >= 1
rnk = [0] * n
Expand All @@ -137,7 +136,8 @@ def lcp_array(s, sa):
lcp[rnk[i] - 1] = h
return lcp

def z_algorithm(s):
def z_algorithm(self):
s = self._s
n = len(s)
if n == 0:
return []
Expand All @@ -153,3 +153,15 @@ def z_algorithm(s):
i += 1
z[0] = n
return z

def __len__(self):
return len(self._s)

def __getitem__(self, idx):
return self._s[idx]

def __str__(self):
return str(self._s)

def __repr__(self):
return f"String({self._s!r})"
Loading
Loading