feat(diff): implemented the myers diff algorithm with longest common subsequence and printing the diff

.gitignore

@@ -1 +1,2 @@
 build/
+test/

include/myers.h

@@ -1,9 +1,108 @@
 #ifndef MYERS_H
 #define MYERS_H
 
-#include <string.h>
+#include "file.h"
 #include "action_list.h"
 
-ActionList* myers_diff(char**, char**, uint64_t, uint64_t);
+// minimum helper
+static inline uint64_t min(uint64_t a, uint64_t b) {
+    return (a < b) ? a : b;
+}
+
+// maxmium helper
+static inline uint64_t max(uint64_t a, uint64_t b) {
+    return (a > b) ? a : b;
+}
+
+// diffing function based on Myers Diff Algorithm with LCS for the maximum matching
+ActionList* myers_diff(File* old_version, File* new_version, uint64_t old_offset, uint64_t new_offset) {
+    uint64_t old_len = old_version->lines;
+    uint64_t new_len = new_version->lines;
+    ActionList* result = new_list();
+
+    // old file is empty
+    if (old_len == 0) {
+        for (uint64_t i = 0; i < new_len; i++) {
+            Action action = (Action){.type=INSERT, .line_original=old_offset, .line_changed=new_offset + i};
+            add_action(result, action);
+        }
+        return result;
+    }
+
+    // new file is empty
+    if (new_len == 0) {
+        for (uint64_t i = 0; i < old_len; i++) {
+            Action action = (Action){.type=DELETE, .line_original=old_offset + i, .line_changed=0};
+            add_action(result, action);
+        }
+        return result;
+    }
+
+    // the LCS table
+    uint64_t** lcs_table = calloc(old_len + 1, sizeof(uint64_t*));
+    if (!lcs_table) return result;
+
+    for (uint64_t i = 0; i <= old_len; i++) {
+        lcs_table[i] = calloc(new_len + 1, sizeof(uint64_t));
+        if (!lcs_table[i]) {
+            for (uint64_t j = 0; j < i; j++) free(lcs_table[j]);
+            free(lcs_table);
+            return result;
+        }
+    }
+
+    // filling the LCS table
+    for (uint64_t i = 1; i <= old_len; i++) {
+        for (uint64_t j = 1; j <= new_len; j++) {
+            if (strcmp(old_version->content[i-1], new_version->content[j-1]) == 0) {
+                lcs_table[i][j] = lcs_table[i-1][j-1] + 1;
+            } else {
+                lcs_table[i][j] = max(lcs_table[i-1][j], lcs_table[i][j-1]);
+            }
+        }
+    }
+
+    uint64_t i = old_len, j = new_len;
+
+    // actions need to be collected temporarily because they are in reverse order
+    Action* temp_actions = calloc(old_len + new_len, sizeof(Action));
+    uint64_t temp_count = 0;
+ 
+    if (!temp_actions) {
+        for (uint64_t k = 0; k <= old_len; k++) free(lcs_table[k]);
+        free(lcs_table);
+        return result;
+    }
+
+    while (i > 0 || j > 0) {
+        if (i > 0 && j > 0 && strcmp(old_version->content[i-1], new_version->content[j-1]) == 0) {
+            i--;
+            j--;
+        } else if (j > 0 && (i == 0 || lcs_table[i][j-1] >= lcs_table[i-1][j])) {
+            temp_actions[temp_count].type = INSERT;
+            temp_actions[temp_count].line_original = old_offset + i;
+            temp_actions[temp_count].line_changed = new_offset + j - 1;
+            temp_count++;
+            j--;
+        } else if (i > 0) {
+            temp_actions[temp_count].type = DELETE;
+            temp_actions[temp_count].line_original = old_offset + i - 1;
+            temp_actions[temp_count].line_changed = 0;
+            temp_count++;
+            i--;
+        }
+    }
+
+    // reversing the actions into the correct order
+    for (int64_t k = temp_count - 1; k >= 0; k--) {
+        add_action(result, temp_actions[k]);
+    }
+
+    free(temp_actions);
+    for (uint64_t k = 0; k <= old_len; k++) free(lcs_table[k]);
+    free(lcs_table);
+
+    return result;
+}
 
 #endif // MYERS_H

include/utilities.h

@@ -0,0 +1,65 @@
+#ifndef UTILITIES_H
+#define UTILITIES_H
+
+#include "file.h"
+#include "action_list.h"
+
+#define RESET "\033[0m"
+#define RED_BG "\033[41m"
+#define GREEN_BG "\033[42m"
+#define BLACK_FG "\033[30m"
+
+void visualize_diff(File* old_version, File* new_version, ActionList* actions) {
+    int* deleted_lines = calloc(old_version->lines, sizeof(int));
+    int* inserted_lines = calloc(new_version->lines, sizeof(int));
+
+    if (!deleted_lines || !inserted_lines) {
+        free(deleted_lines);
+        free(inserted_lines);
+        return;
+    }
+
+    for (uint64_t i = 0; i < actions->len; i++) {
+        if (actions->actions[i].type == DELETE) {
+            deleted_lines[actions->actions[i].line_original] = 1;
+        } else if (actions->actions[i].type == INSERT) {
+            inserted_lines[actions->actions[i].line_changed] = 1;
+        }
+    }
+
+    uint64_t old_idx = 0, new_idx = 0;
+
+    while (old_idx < old_version->lines || new_idx < new_version->lines) {
+        // DELETE
+        if (old_idx < old_version->lines && deleted_lines[old_idx]) {
+            printf("%s%s%4ld      | %s%s\n", RED_BG, BLACK_FG, old_idx+1, old_version->content[old_idx], RESET);
+            old_idx++;
+        }
+        // INSERT
+        else if (new_idx < new_version->lines && inserted_lines[new_idx]) {
+            printf("%s%s     %4ld | %s%s\n", GREEN_BG, BLACK_FG, new_idx+1, new_version->content[new_idx], RESET);
+            new_idx++;
+        }
+        // STAYS
+        else if (old_idx < old_version->lines && new_idx < new_version->lines) {
+            printf("%4ld %4ld | %s\n", old_idx+1, new_idx+1, old_version->content[old_idx]);
+            old_idx++;
+            new_idx++;
+        }
+        // DELETE
+        else if (old_idx < old_version->lines) {
+            printf("%s%s%4ld      | %s%s\n", RED_BG, BLACK_FG, old_idx+1, old_version->content[old_idx], RESET);
+            old_idx++;
+        }
+        // INSERT
+        else if (new_idx < new_version->lines) {
+            printf("%s%s     %4ld | %s%s\n", GREEN_BG, BLACK_FG, new_idx+1, new_version->content[new_idx], RESET);
+            new_idx++;
+        }
+    }
+
+    free(deleted_lines);
+    free(inserted_lines);
+}
+
+#endif // UTILITIES_H

src/myers.c

@@ -1,5 +0,0 @@
-#include "myers.h"
-
-ActionList* myers_diff(char** file1, char** file2, uint64_t offset1, uint64_t offset2) {
-    return NULL; 
-}