Merge pull request #28 from colorbox/929

929. Unique Email Addresses

Author: colorbox

929/step1.cpp

@@ -0,0 +1,28 @@
+/*
+Solve Time: 25:00
+
+Time Order: O(n)
+Space Order: O(n)
+
+方針は早くに定まっていたものの、C++の文字列操作が思うようにいかずに時間がかかってしまった。
+
+*/
+class Solution {
+public:
+  int numUniqueEmails(vector<string>& emails) {
+    set<string> unique_emails;
+    for (const string& email : emails) {
+      auto at_pos = email.find_first_of("@");
+      string local_part = email.substr(0, at_pos);
+      auto plus_pos = local_part.find_first_of("+");
+      if (plus_pos == std::string::npos) {
+        plus_pos = local_part.size();
+      }
+      string local_part_without_plus = local_part.substr(0, plus_pos);
+      std::erase(local_part_without_plus, '.');
+      email.replace(0, at_pos, local_part_without_plus);
+      unique_emails.insert(email);
+    }
+    return unique_emails.size();
+  }
+};

929/step2_1.cpp

@@ -0,0 +1,29 @@
+/*
+Time Order : O(n)
+Space Order : O(n)
+
+step1の改良版
+コメント使用ででローカルパートに`@`が含むものを対応するために`rfind`を利用。
+メールアドレス正規化処理をメソッド抽出。
+*/
+class Solution {
+ public:
+  int numUniqueEmails(vector<string>& emails) {
+    set<string> unique_emails;
+    for (auto email : emails) {
+      unique_emails.insert(canonicalize_email(email));
+    }
+    return unique_emails.size();
+  }
+
+ private:
+  string canonicalize_email(string email) {
+    const auto at_pos = email.rfind("@");
+    const string local_part = email.substr(0, at_pos);
+    const auto plus_pos = local_part.find("+");
+    string canonical_local_part = local_part.substr(0, plus_pos);
+    std::erase(canonical_local_part, '.');
+    email.replace(0, at_pos, canonical_local_part);
+    return email;
+  }
+};

929/step2_1_fix.cpp

@@ -0,0 +1,24 @@
+/*
+
+*/
+class Solution {
+ public:
+  int numUniqueEmails(vector<string>& emails) {
+    for (auto& email : emails) {
+      canonicalize_email(email);
+    }
+    set<string> unique_emails(emails.begin(), emails.end());
+    return unique_emails.size();
+  }
+
+ private:
+  void canonicalize_email(string& email) {
+    auto at_pos = email.rfind('@');
+    string local_part = email.substr(0, at_pos);
+    auto plus_pos = local_part.find('+');
+    string canonicalized_local_part = local_part.substr(0, plus_pos);
+    std::erase(canonicalized_local_part, '.');
+    string domain_part = email.substr(at_pos, email.size() - at_pos);
+    email = canonicalized_local_part + domain_part;
+  }
+};

929/step2_2.cpp

@@ -0,0 +1,33 @@
+/*
+Time Order : O(n)
+Space Order : O(n)
+
+ステートマシン的な解法
+*/
+class Solution {
+ public:
+  int numUniqueEmails(vector<string>& emails) {
+    set<string> unique_emails;
+    for (auto email : emails) {
+      unique_emails.insert(canonicalize_email(email));
+    }
+    return unique_emails.size();
+  }
+
+ private:
+  string canonicalize_email(string& email) {
+    string canonicalized_email = "";
+    auto at_pos = email.rfind('@');
+    string local_part = email.substr(0, at_pos);
+    for (auto c : local_part) {
+      if (c == '.') {
+        continue;
+      }
+      if (c == '+') {
+        break;
+      }
+      canonicalized_email += c;
+    }
+    return email.replace(0, at_pos, canonicalized_email);
+  }
+};

929/step2_3.cpp

@@ -0,0 +1,28 @@
+/*
+Time Order : O(n)
+Space Order : O(n)
+
+正規表現を使用したパターン
+local-partから`.`と`+`以降を削除する正規表現を使用している。
+*/
+class Solution {
+ public:
+  int numUniqueEmails(vector<string>& emails) {
+    set<string> unique_emails;
+    for (auto email : emails) {
+      unique_emails.insert(canonicalize_email(email));
+    }
+    return unique_emails.size();
+  }
+
+ private:
+  string canonicalize_email(string& email) {
+    auto at_pos = email.rfind('@');
+    string local_part = email.substr(0, at_pos);
+    std::regex dot_and_plus_suffix("\\.|\\+.*");
+    local_part = std::regex_replace(local_part, dot_and_plus_suffix, "");
+    
+    email.replace(0, at_pos, local_part);
+    return email;
+  }
+};

929/step3.cpp

@@ -0,0 +1,21 @@
+class Solution {
+ public:
+  int numUniqueEmails(vector<string>& emails) {
+    set<string> canonicalized_emails;
+    for (string email : emails) {
+      canonicalized_emails.insert(canonicalize_email(email));
+    }
+    return canonicalized_emails.size();
+  }
+
+ private:
+  string canonicalize_email(string email) {
+    size_t at_pos = email.rfind('@');
+    string local_part = email.substr(0, at_pos);
+    size_t plus_pos = local_part.find('+');
+    local_part = local_part.substr(0, plus_pos);
+    erase(local_part, '.');
+    email.replace(0, at_pos, local_part);
+    return email;
+  }
+};