feat: add Boyer-Moore algorithm

src/searching/README.md

@@ -1,5 +1,11 @@
 # Searching
 
+## [Boyer-Moore algorithm](./src/bm_search.cairo)
+
+The Boyer-Moore algorithm is a string-searching algorithm that finds the position of a pattern in a string. It preprocesses the pattern to create two lookup tables: one for the bad character rule and one for the good suffix rule. The bad character rule shifts the pattern to align with the last occurrence of the mismatched character in the pattern. The good suffix rule shifts the pattern to align with the last occurrence of the suffix of the pattern that matches the suffix of the text.
+
+The Boyer-Moore algorithm has a best-case time complexity of O(n/m) and a worst-case time complexity of O(nm), where n is the length of the text and m is the length of the pattern. It is the most efficient string-searching algorithm in practice.
+
 ## [Binary search](./src/binary_search.cairo)
 
 The binary search algorithm is a simple search in an ordered array-like compound. It starts by comparing the value we are looking for to the middle of the array. If it's not a match, the function calls itself recursively on the right or left half of the array until it does(n't) find the value in the array.

src/searching/src/bm_search.cairo

@@ -0,0 +1,93 @@
+// The Boyer-Moore string search algorithm
+use dict::Felt252DictTrait;
+
+/// Find `pattern` in `text` and return the index of every match.
+/// * `text` - The text to search in.
+/// * `pattern` - The pattern to search for.
+/// # Returns
+/// * `Array<usize>` - The index of every match.
+fn bm_search(text: @ByteArray, pattern: @ByteArray) -> Array<usize> {
+    let mut positions: Array<usize> = array![]; // Array to store the indices of every match
+    let text_len = text.len(); // Length of the text
+    let pattern_len = pattern.len(); // Length of the pattern
+
+    // Check for invalid inputs or if the pattern is longer than the text
+    if text_len == 0 || pattern_len == 0 || pattern_len > text_len {
+        return positions;
+    }
+
+    let mut char_dict = felt252_dict_new::<
+        usize
+    >(); // Dictionary to store the last occurrence of each character in the pattern
+    let mut pattern_index = 0; // Index of the current character in the pattern
+
+    // Build the character dictionary
+    loop {
+        if pattern_index == pattern_len {
+            break;
+        }
+        let current_char = pattern.at(pattern_index).unwrap();
+        char_dict
+            .insert(
+                current_char.into(), pattern_index + 1
+            ); // Avoid 0 since felt252_dict initializes every entry to 0 by default
+        pattern_index += 1;
+    };
+
+    let mut shift: usize = 0; // Shift value for pattern matching
+
+    // Perform pattern matching
+    loop {
+        if shift > text_len - pattern_len {
+            break;
+        }
+
+        let mut pattern_index = pattern_len;
+
+        // Compare characters from right to left
+        loop {
+            if pattern_index == 0
+                || @pattern
+                    .at(pattern_index - 1)
+                    .unwrap() != @text
+                    .at(shift + pattern_index - 1)
+                    .unwrap() {
+                break;
+            }
+            pattern_index -= 1;
+        };
+
+        // If the pattern is found at the current shift position
+        if pattern_index == 0 {
+            positions.append(shift); // Add the current shift position to the positions array
+
+            // Calculate the next shift value
+            let add_to_shift = {
+                if shift + pattern_len < text_len {
+                    let next_char = text.at(shift + pattern_len).unwrap();
+                    let index = char_dict.get(next_char.into());
+                    if index == 0 {
+                        pattern_len + 1
+                    } else {
+                        pattern_len - index + 1
+                    }
+                } else {
+                    1
+                }
+            };
+            shift += add_to_shift;
+        } else {
+            let current_char = text.at(shift + pattern_index - 1).unwrap();
+            let index = char_dict.get(current_char.into());
+
+            // Calculate the next shift value based on the last occurrence of the current character in the pattern
+            if pattern_index <= (index + 1) {
+                shift += 1;
+            } else {
+                shift += pattern_index - index;
+            }
+        }
+    };
+
+    positions // Return the array of positions
+}

src/searching/src/lib.cairo

@@ -1,4 +1,5 @@
 mod binary_search;
+mod bm_search;
 mod dijkstra;
 
 #[cfg(test)]

src/searching/src/tests.cairo

@@ -1,2 +1,3 @@
 mod binary_search_test;
+mod bm_search_test;
 mod dijkstra_test;

src/searching/src/tests/bm_search_test.cairo

@@ -0,0 +1,150 @@
+use alexandria_searching::bm_search::bm_search;
+
+
+// Check if two arrays are equal.
+/// * `a` - The first array.
+/// * `b` - The second array.
+/// # Returns
+/// * `bool` - True if the arrays are equal, false otherwise.
+fn is_equal(mut a: Span<u32>, mut b: Span<u32>) -> bool {
+    if a.len() != b.len() {
+        return false;
+    }
+    loop {
+        match a.pop_front() {
+            Option::Some(val1) => {
+                let val2 = b.pop_front().unwrap();
+                if *val1 != *val2 {
+                    break false;
+                }
+            },
+            Option::None => { break true; },
+        };
+    }
+}
+
+
+#[test]
+#[available_gas(5000000)]
+fn bm_search_test_1() {
+    // AABCAB12AFAABCABFFEGABCAB -> 41,41,42,43,41,42,31,32,41,46,41,41,42,43,41,42,46,46,45,47,41,42,43,41,42
+    let mut text: ByteArray = Default::default();
+    text.append_byte(0x41_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    text.append_byte(0x43_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    text.append_byte(0x31_u8);
+    text.append_byte(0x32_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x46_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    text.append_byte(0x43_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    text.append_byte(0x46_u8);
+    text.append_byte(0x46_u8);
+    text.append_byte(0x45_u8);
+    text.append_byte(0x47_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    text.append_byte(0x43_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    // ABCAB -> 41,42,43,41,42
+    let mut pattern: ByteArray = Default::default();
+    pattern.append_byte(0x41_u8);
+    pattern.append_byte(0x42_u8);
+    pattern.append_byte(0x43_u8);
+    pattern.append_byte(0x41_u8);
+    pattern.append_byte(0x42_u8);
+
+    let positions = bm_search(@text, @pattern);
+    let ground_truth: Array<usize> = array![1, 11, 20];
+    assert(is_equal(positions.span(), ground_truth.span()), 'invalid result');
+}
+
+#[test]
+#[available_gas(5000000)]
+fn bm_search_test_2() {
+    // AABCAB12AFAABCABFFEGABCAB -> 41,41,42,43,41,42,31,32,41,46,41,41,42,43,41,42,46,46,45,47,41,42,43,41,42
+    let mut text: ByteArray = Default::default();
+    text.append_byte(0x41_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    text.append_byte(0x43_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    text.append_byte(0x31_u8);
+    text.append_byte(0x32_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x46_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    text.append_byte(0x43_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    text.append_byte(0x46_u8);
+    text.append_byte(0x46_u8);
+    text.append_byte(0x45_u8);
+    text.append_byte(0x47_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    text.append_byte(0x43_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    // FFF -> 46,46,46
+    let mut pattern: ByteArray = Default::default();
+    pattern.append_byte(0x46_u8);
+    pattern.append_byte(0x46_u8);
+    pattern.append_byte(0x46_u8);
+
+    let positions = bm_search(@text, @pattern);
+    let ground_truth: Array<usize> = array![];
+    assert(is_equal(positions.span(), ground_truth.span()), 'invalid result');
+}
+
+#[test]
+#[available_gas(5000000)]
+fn bm_search_test_3() {
+    // AABCAB12AFAABCABFFEGABCAB -> 41,41,42,43,41,42,31,32,41,46,41,41,42,43,41,42,46,46,45,47,41,42,43,41,42
+    let mut text: ByteArray = Default::default();
+    text.append_byte(0x41_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    text.append_byte(0x43_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    text.append_byte(0x31_u8);
+    text.append_byte(0x32_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x46_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    text.append_byte(0x43_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    text.append_byte(0x46_u8);
+    text.append_byte(0x46_u8);
+    text.append_byte(0x45_u8);
+    text.append_byte(0x47_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    text.append_byte(0x43_u8);
+    text.append_byte(0x41_u8);
+    text.append_byte(0x42_u8);
+    // CAB -> 43,41,42
+    let mut pattern: ByteArray = Default::default();
+    pattern.append_byte(0x43_u8);
+    pattern.append_byte(0x41_u8);
+    pattern.append_byte(0x42_u8);
+
+    let positions = bm_search(@text, @pattern);
+    let ground_truth: Array<usize> = array![3, 13, 22];
+    assert(is_equal(positions.span(), ground_truth.span()), 'invalid result');
+}