Knuth-Morris-Pratt string search: Difference between revisions
(julia example) |
(Added Wren) |
||
Line 129: | Line 129: | ||
Found <and> at: (one-based indices) [102, 129, 172] |
Found <and> at: (one-based indices) [102, 129, 172] |
||
Found <alfalfa> at: (one-based indices) [34, 88] |
Found <alfalfa> at: (one-based indices) [34, 88] |
||
</pre> |
|||
=={{header|Wren}}== |
|||
This is based on the code [https://www.geeksforgeeks.org/kmp-algorithm-for-pattern-searching/ here]. The examples used are the same as in the [[Boyer-Moore_string_search#Wren]] task. |
|||
<lang ecmascript>class KMP { |
|||
static search(haystack, needle) { |
|||
haystack = haystack.bytes.toList |
|||
needle = needle.bytes.toList |
|||
var hc = haystack.count |
|||
var nc = needle.count |
|||
var indices = [] |
|||
var i = 0 // index into haystack |
|||
var j = 0 // index into needle |
|||
var t = table_(needle) |
|||
while (i < hc) { |
|||
if (needle[j] == haystack[i]) { |
|||
i = i + 1 |
|||
j = j + 1 |
|||
} |
|||
if (j == nc) { |
|||
indices.add(i - j) |
|||
j = t[j-1] |
|||
} else if (i < hc && needle[j] != haystack[i]) { |
|||
if (j != 0) { |
|||
j = t[j-1] |
|||
} else { |
|||
i = i + 1 |
|||
} |
|||
} |
|||
} |
|||
return indices |
|||
} |
|||
static table_(needle) { |
|||
var nc = needle.count |
|||
var t = List.filled(nc, 0) |
|||
var i = 1 // index into table |
|||
var len = 0 // length of previous longest prefix |
|||
while (i < nc) { |
|||
if (needle[i] == needle[len]) { |
|||
len = len + 1 |
|||
t[i] = len |
|||
i = i + 1 |
|||
} else if (len != 0) { |
|||
len = t[len-1] |
|||
} else { |
|||
t[i] = 0 |
|||
i = i + 1 |
|||
} |
|||
} |
|||
return t |
|||
} |
|||
} |
|||
var texts = [ |
|||
"GCTAGCTCTACGAGTCTA", |
|||
"GGCTATAATGCGTA", |
|||
"there would have been a time for such a word", |
|||
"needle need noodle needle", |
|||
"InhisbookseriesTheArtofComputerProgrammingpublishedbyAddisonWesleyDKnuthusesanimaginarycomputertheMIXanditsassociatedmachinecodeandassemblylanguagestoillustratetheconceptsandalgorithmsastheyarepresented", |
|||
"Nearby farms grew a half acre of alfalfa on the dairy's behalf, with bales of all that alfalfa exchanged for milk." |
|||
] |
|||
var pats = ["TCTA", "TAATAAA", "word", "needle", "put", "and", "alfalfa"] |
|||
for (i in 0...texts.count) System.print("text%(i+1) = %(texts[i])") |
|||
System.print() |
|||
for (i in 0...pats.count) { |
|||
var j = (i < 5) ? i : i-1 |
|||
System.print("Found '%(pats[i])' in 'text%(j+1)' at indices %(KMP.search(texts[j], pats[i]))") |
|||
}</lang> |
|||
{{out}} |
|||
<pre> |
|||
text1 = GCTAGCTCTACGAGTCTA |
|||
text2 = GGCTATAATGCGTA |
|||
text3 = there would have been a time for such a word |
|||
text4 = needle need noodle needle |
|||
text5 = InhisbookseriesTheArtofComputerProgrammingpublishedbyAddisonWesleyDKnuthusesanimaginarycomputertheMIXanditsassociatedmachinecodeandassemblylanguagestoillustratetheconceptsandalgorithmsastheyarepresented |
|||
text6 = Nearby farms grew a half acre of alfalfa on the dairy's behalf, with bales of all that alfalfa exchanged for milk. |
|||
Found 'TCTA' in 'text1' at indices [6, 14] |
|||
Found 'TAATAAA' in 'text2' at indices [] |
|||
Found 'word' in 'text3' at indices [40] |
|||
Found 'needle' in 'text4' at indices [0, 19] |
|||
Found 'put' in 'text5' at indices [26, 90] |
|||
Found 'and' in 'text5' at indices [101, 128, 171] |
|||
Found 'alfalfa' in 'text6' at indices [33, 87] |
|||
</pre> |
</pre> |
Revision as of 16:36, 8 July 2022
< About Knuth-Morris-Pratt String Search Algorithm >
Emacs Lisp
<lang lisp> (defun kmp_compile_pattern (pattern)
"Compile pattern to DFA."
(defun create-2d-array (x y init) (let ((arr1 (make-vector x nil))) (dotimes (i x)
(aset arr1 i (make-vector y init)) )
arr1 ) ) (let* ((patLen (length pattern))
(R 256)
(restartPos 0)
(dfa (create-2d-array R patLen 0)))
(aset (aref dfa (elt pattern 0)) 0 1)
(let ((patPos 0)) (while (progn (setq patPos (1+ patPos)) (< patPos patLen))
(dotimes (c R) (aset (aref dfa c) patPos (aref (aref dfa c) restartPos)) )
(aset (aref dfa (elt pattern patPos)) patPos (1+ patPos)) (setq restartPos (aref (aref dfa (elt pattern patPos)) restartPos) ) )
) dfa ) )
(defun kmp_search (pattern text)
"Pattern search with KMP algorithm." (let ((dfa (kmp_compile_pattern pattern)))
(let ((textPos 0) (patPos 0) (N (length text)) (M (length pattern))) (while (and (< textPos N) (< patPos M))
(setq patPos (aref (aref dfa (elt text textPos)) patPos)) (setq textPos (1+ textPos)) )
(if (= patPos M) (- textPos M) N ) ) ) )
</lang>
Julia
<lang ruby>"""
function kmp_table(W)
input:
an array of characters, W (the word to be analyzed)
output:
an array of integers, T (the table to be filled)
define variables:
an integer, i ← 2 (the current one-based position we are computing in T) an integer, j ← 0 (the additive to index i in W of the next character of the current candidate substring)
""" function kmp_table(W)
len = length(W) T = zeros(Int, len) # start with the second letter of W, looking for patterns in W i = 2 while i < len j = 0 while i + j <= len # avoid overshooting end with index # compute the longest proper prefix if W[i + j] == W[j + 1] T[i + j] = T[i + j - 1] + 1 else T[i + j] = 0 # back to start j += 1 break end j += 1 end # entry in T found, so begin at next starting point along W i += j end return T
end
"""
function kmp_search(S, W)
input:
an array of characters, S (the text to be searched) an array of characters, W (the word sought)
output:
an array of integers, P (positions in S at which W is found)
define variables (one based indexing in Julia differs from the Wikipedia example):
an integer, i ← 1 (the position of the current character in S) an integer, j ← 1 (the position of the current character in W) an array of integers, T (the table, computed elsewhere)
""" function kmp_search(S, W)
lenW, lenS = length(W), length(S) i, P = 1, Int[] T = kmp_table(W) # get pattern table while i <= lenS - lenW + 1 for j in 1:lenW if S[i + j - 1] != W[j] # pattern not found, so skip unnecessary inner loops i += T[j] + 1 @goto next_outer_loop end end # found pattern W in S, so add to output P push!(P, i) i += 1 @label next_outer_loop end return P
end
const text1 = "InhisbookseriesTheArtofComputerProgrammingpublishedbyAddisonWesleyDKnuthusesanimaginarycomputertheMIXanditsassociatedmachinecodeandassemblylanguagestoillustratetheconceptsandalgorithmsastheyarepresented" const text2 = "Nearby farms grew a half acre of alfalfa on the dairy's behalf, with bales of all that alfalfa exchanged for milk." const pat1, pat2, pat3 = "put", "and", "alfalfa"
println("Found <$pat1> at: (one-based indices) ", kmp_search(text1, pat1)) println("Found <$pat2> at: (one-based indices) ", kmp_search(text1, pat2)) println("Found <$pat3> at: (one-based indices) ", kmp_search(text2, pat3))
</lang>
- Output:
Found <put> at: (one-based indices) [27, 91] Found <and> at: (one-based indices) [102, 129, 172] Found <alfalfa> at: (one-based indices) [34, 88]
Wren
This is based on the code here. The examples used are the same as in the Boyer-Moore_string_search#Wren task. <lang ecmascript>class KMP {
static search(haystack, needle) { haystack = haystack.bytes.toList needle = needle.bytes.toList var hc = haystack.count var nc = needle.count var indices = [] var i = 0 // index into haystack var j = 0 // index into needle var t = table_(needle) while (i < hc) { if (needle[j] == haystack[i]) { i = i + 1 j = j + 1 } if (j == nc) { indices.add(i - j) j = t[j-1] } else if (i < hc && needle[j] != haystack[i]) { if (j != 0) { j = t[j-1] } else { i = i + 1 } } } return indices }
static table_(needle) { var nc = needle.count var t = List.filled(nc, 0) var i = 1 // index into table var len = 0 // length of previous longest prefix while (i < nc) { if (needle[i] == needle[len]) { len = len + 1 t[i] = len i = i + 1 } else if (len != 0) { len = t[len-1] } else { t[i] = 0 i = i + 1 } } return t }
}
var texts = [
"GCTAGCTCTACGAGTCTA", "GGCTATAATGCGTA", "there would have been a time for such a word", "needle need noodle needle",
"InhisbookseriesTheArtofComputerProgrammingpublishedbyAddisonWesleyDKnuthusesanimaginarycomputertheMIXanditsassociatedmachinecodeandassemblylanguagestoillustratetheconceptsandalgorithmsastheyarepresented",
"Nearby farms grew a half acre of alfalfa on the dairy's behalf, with bales of all that alfalfa exchanged for milk."
] var pats = ["TCTA", "TAATAAA", "word", "needle", "put", "and", "alfalfa"] for (i in 0...texts.count) System.print("text%(i+1) = %(texts[i])") System.print() for (i in 0...pats.count) {
var j = (i < 5) ? i : i-1 System.print("Found '%(pats[i])' in 'text%(j+1)' at indices %(KMP.search(texts[j], pats[i]))")
}</lang>
- Output:
text1 = GCTAGCTCTACGAGTCTA text2 = GGCTATAATGCGTA text3 = there would have been a time for such a word text4 = needle need noodle needle text5 = InhisbookseriesTheArtofComputerProgrammingpublishedbyAddisonWesleyDKnuthusesanimaginarycomputertheMIXanditsassociatedmachinecodeandassemblylanguagestoillustratetheconceptsandalgorithmsastheyarepresented text6 = Nearby farms grew a half acre of alfalfa on the dairy's behalf, with bales of all that alfalfa exchanged for milk. Found 'TCTA' in 'text1' at indices [6, 14] Found 'TAATAAA' in 'text2' at indices [] Found 'word' in 'text3' at indices [40] Found 'needle' in 'text4' at indices [0, 19] Found 'put' in 'text5' at indices [26, 90] Found 'and' in 'text5' at indices [101, 128, 171] Found 'alfalfa' in 'text6' at indices [33, 87]