Refactoring

* Extract IsCharacterInSet to its own module.
* Extract PercentEncodedCharacterDecoder to its own module.

6 files changed, 250 insertions, 130 deletions

diff --git a/CMakeLists.txt b/CMakeLists.txt
index bfd19d7..ab5ca79 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -7,9 +7,13 @@ set(This Uri)
 
 set(Headers
     include/Uri/Uri.hpp
+    src/IsCharacterInSet.hpp
+    src/PercentEncodedCharacterDecoder.hpp
 )
 
 set(Sources
+    src/IsCharacterInSet.cpp
+    src/PercentEncodedCharacterDecoder.cpp
     src/Uri.cpp
 )
 
diff --git a/src/IsCharacterInSet.cpp b/src/IsCharacterInSet.cpp
new file mode 100644
index 0000000..82625e9
--- /dev/null
+++ b/src/IsCharacterInSet.cpp
@@ -0,0 +1,35 @@
+/**
+ * @file IsCharacterInSet.cpp
+ *
+ * This module contains the implementation of the
+ * Uri::IsCharacterInSet function.
+ *
+ * © 2018 by Richard Walters
+ */
+
+#include "IsCharacterInSet.hpp"
+
+namespace Uri {
+
+    bool IsCharacterInSet(
+        char c,
+        std::initializer_list< char > characterSet
+    ) {
+        for (
+            auto charInSet = characterSet.begin();
+            charInSet != characterSet.end();
+            ++charInSet
+        ) {
+            const auto first = *charInSet++;
+            const auto last = *charInSet;
+            if (
+                (c >= first)
+                && (c <= last)
+            ) {
+                return true;
+            }
+        }
+        return false;
+    }
+
+}
diff --git a/src/IsCharacterInSet.hpp b/src/IsCharacterInSet.hpp
new file mode 100644
index 0000000..f17460c
--- /dev/null
+++ b/src/IsCharacterInSet.hpp
@@ -0,0 +1,37 @@
+#ifndef URI_IS_CHARACTER_IN_SET_HPP
+#define URI_IS_CHARACTER_IN_SET_HPP
+
+/**
+ * @file IsCharacterInSet.hpp
+ *
+ * This module declares the Uri::IsCharacterInSet class.
+ *
+ * © 2018 by Richard Walters
+ */
+
+#include <initializer_list>
+
+namespace Uri {
+
+    /**
+     * This function determines whether or not the given character
+     * is in the given character set.
+     *
+     * @param[in] c
+     *     This is the character to check.
+     *
+     * @param[in] characterSet
+     *     This is the set of characters that are allowed.
+     *
+     * @return
+     *     An indication of whether or not the given character
+     *     is in the given character set is returned.
+     */
+    bool IsCharacterInSet(
+        char c,
+        std::initializer_list< char > characterSet
+    );
+
+}
+
+#endif /* URI_IS_CHARACTER_IN_SET_HPP */
diff --git a/src/PercentEncodedCharacterDecoder.cpp b/src/PercentEncodedCharacterDecoder.cpp
new file mode 100644
index 0000000..d3bc0d9
--- /dev/null
+++ b/src/PercentEncodedCharacterDecoder.cpp
@@ -0,0 +1,78 @@
+/**
+ * @file PercentEncodedCharacterDecoder.cpp
+ *
+ * This module contains the implementation of the
+ * Uri::PercentEncodedCharacterDecoder class.
+ *
+ * © 2018 by Richard Walters
+ */
+
+#include "IsCharacterInSet.hpp"
+#include "PercentEncodedCharacterDecoder.hpp"
+
+namespace Uri {
+
+    struct PercentEncodedCharacterDecoder::Impl {
+        /**
+         * This is the decoded character.
+         */
+        int decodedCharacter = 0;
+
+        /**
+         * This is the current state of the decoder's state machine.
+         * - 0: we haven't yet received the first hex digit.
+         * - 1: we received the first hex digit but not the second.
+         * - 2: we received both hex digits
+         */
+        size_t decoderState = 0;
+    };
+
+    PercentEncodedCharacterDecoder::~PercentEncodedCharacterDecoder() = default;
+    PercentEncodedCharacterDecoder::PercentEncodedCharacterDecoder(PercentEncodedCharacterDecoder&&) = default;
+    PercentEncodedCharacterDecoder& PercentEncodedCharacterDecoder::operator=(PercentEncodedCharacterDecoder&&) = default;
+
+    PercentEncodedCharacterDecoder::PercentEncodedCharacterDecoder()
+        : impl_(new Impl)
+    {
+    }
+
+    bool PercentEncodedCharacterDecoder::NextEncodedCharacter(char c) {
+        switch(impl_->decoderState) {
+            case 0: { // % ...
+                impl_->decoderState = 1;
+                impl_->decodedCharacter <<= 4;
+                if (IsCharacterInSet(c, {'0','9'})) {
+                    impl_->decodedCharacter += (int)(c - '0');
+                } else if (IsCharacterInSet(c, {'A','F'})) {
+                    impl_->decodedCharacter += (int)(c - 'A') + 10;
+                } else {
+                    return false;
+                }
+            } break;
+
+            case 1: { // %[0-9A-F] ...
+                impl_->decoderState = 2;
+                impl_->decodedCharacter <<= 4;
+                if (IsCharacterInSet(c, {'0','9'})) {
+                    impl_->decodedCharacter += (int)(c - '0');
+                } else if (IsCharacterInSet(c, {'A','F'})) {
+                    impl_->decodedCharacter += (int)(c - 'A') + 10;
+                } else {
+                    return false;
+                }
+            } break;
+
+            default: break;
+        }
+        return true;
+    }
+
+    bool PercentEncodedCharacterDecoder::Done() const {
+        return (impl_->decoderState == 2);
+    }
+
+    char PercentEncodedCharacterDecoder::GetDecodedCharacter() const {
+        return (char)impl_->decodedCharacter;
+    }
+
+}
diff --git a/src/PercentEncodedCharacterDecoder.hpp b/src/PercentEncodedCharacterDecoder.hpp
new file mode 100644
index 0000000..b3e207d
--- /dev/null
+++ b/src/PercentEncodedCharacterDecoder.hpp
@@ -0,0 +1,85 @@
+#ifndef URI_PERCENT_ENCODED_CHARACTER_DECODER_HPP
+#define URI_PERCENT_ENCODED_CHARACTER_DECODER_HPP
+
+/**
+ * @file PercentEncodedCharacterDecoder.hpp
+ *
+ * This module declares the Uri::PercentEncodedCharacterDecoder class.
+ *
+ * © 2018 by Richard Walters
+ */
+
+#include <memory>
+#include <stddef.h>
+
+namespace Uri {
+
+    /**
+     * This class can take in a percent-encoded character,
+     * decode it, and also detect if there are any problems in the encoding.
+     */
+    class PercentEncodedCharacterDecoder {
+        // Lifecycle management
+    public:
+        ~PercentEncodedCharacterDecoder();
+        PercentEncodedCharacterDecoder(const PercentEncodedCharacterDecoder&) = delete;
+        PercentEncodedCharacterDecoder(PercentEncodedCharacterDecoder&&);
+        PercentEncodedCharacterDecoder& operator=(const PercentEncodedCharacterDecoder&) = delete;
+        PercentEncodedCharacterDecoder& operator=(PercentEncodedCharacterDecoder&&);
+
+        // Methods
+    public:
+        /**
+         * This is the default constructor.
+         */
+        PercentEncodedCharacterDecoder();
+
+        /**
+         * This method inputs the next encoded character.
+         *
+         * @param[in] c
+         *     This is the next encoded character to give to the decoder.
+         *
+         * @return
+         *     An indication of whether or not the encoded character
+         *     was accepted is returned.
+         */
+        bool NextEncodedCharacter(char c);
+
+        /**
+         * This method checks to see if the decoder is done
+         * and has decoded the encoded character.
+         *
+         * @return
+         *     An indication of whether or not the decoder is done
+         *     and has decoded the encoded character is returned.
+         */
+        bool Done() const;
+
+        /**
+         * This method returns the decoded character, once
+         * the decoder is done.
+         *
+         * @return
+         *     The decoded character is returned.
+         */
+        char GetDecodedCharacter() const;
+
+        // Properties
+    private:
+        /**
+         * This is the type of structure that contains the private
+         * properties of the instance.  It is defined in the implementation
+         * and declared here to ensure that it is scoped inside the class.
+         */
+        struct Impl;
+
+        /**
+         * This contains the private properties of the instance.
+         */
+        std::unique_ptr< struct Impl > impl_;
+    };
+
+}
+
+#endif /* URI_PERCENT_ENCODED_CHARACTER_DECODER_HPP */
diff --git a/src/Uri.cpp b/src/Uri.cpp
index f122c0a..208297d 100644
--- a/src/Uri.cpp
+++ b/src/Uri.cpp
@@ -6,6 +6,9 @@
  * © 2018 by Richard Walters
  */
 
+#include "IsCharacterInSet.hpp"
+#include "PercentEncodedCharacterDecoder.hpp"
+
 #include <functional>
 #include <inttypes.h>
 #include <memory>
@@ -81,41 +84,6 @@ namespace {
     }
 
     /**
-     * This function determines whether or not the given character
-     * is in the given character set.
-     *
-     * @param[in] c
-     *     This is the character to check.
-     *
-     * @param[in] characterSet
-     *     This is the set of characters that are allowed.
-     *
-     * @return
-     *     An indication of whether or not the given character
-     *     is in the given character set is returned.
-     */
-    bool IsCharacterInSet(
-        char c,
-        std::initializer_list< char > characterSet
-    ) {
-        for (
-            auto charInSet = characterSet.begin();
-            charInSet != characterSet.end();
-            ++charInSet
-        ) {
-            const auto first = *charInSet++;
-            const auto last = *charInSet;
-            if (
-                (c >= first)
-                && (c <= last)
-            ) {
-                return true;
-            }
-        }
-        return false;
-    }
-
-    /**
      * This function returns a strategy function that
      * may be used with the FailsMatch function to test a scheme
      * to make sure it is legal according to the standard.
@@ -133,9 +101,9 @@ namespace {
             } else {
                 bool check;
                 if (*isFirstCharacter) {
-                    check = IsCharacterInSet(c, { 'a','z', 'A','Z' });
+                    check = Uri::IsCharacterInSet(c, { 'a','z', 'A','Z' });
                 } else {
-                    check = IsCharacterInSet(c, { 'a','z', 'A','Z', '0','9', '+','+', '-','-', '.','.' });
+                    check = Uri::IsCharacterInSet(c, { 'a','z', 'A','Z', '0','9', '+','+', '-','-', '.','.' });
                 }
                 *isFirstCharacter = false;
                 return check;
@@ -144,93 +112,6 @@ namespace {
     }
 
     /**
-     * This class can take in a percent-encoded character,
-     * decode it, and also detect if there are any problems in the encoding.
-     */
-    class PercentEncodedCharacterDecoder {
-        // Methods
-    public:
-        /**
-         * This method inputs the next encoded character.
-         *
-         * @param[in] c
-         *     This is the next encoded character to give to the decoder.
-         *
-         * @return
-         *     An indication of whether or not the encoded character
-         *     was accepted is returned.
-         */
-        bool NextEncodedCharacter(char c) {
-            switch(decoderState_) {
-                case 0: { // % ...
-                    decoderState_ = 1;
-                    decodedCharacter_ <<= 4;
-                    if (IsCharacterInSet(c, {'0','9'})) {
-                        decodedCharacter_ += (int)(c - '0');
-                    } else if (IsCharacterInSet(c, {'A','F'})) {
-                        decodedCharacter_ += (int)(c - 'A') + 10;
-                    } else {
-                        return false;
-                    }
-                } break;
-
-                case 1: { // %[0-9A-F] ...
-                    decoderState_ = 2;
-                    decodedCharacter_ <<= 4;
-                    if (IsCharacterInSet(c, {'0','9'})) {
-                        decodedCharacter_ += (int)(c - '0');
-                    } else if (IsCharacterInSet(c, {'A','F'})) {
-                        decodedCharacter_ += (int)(c - 'A') + 10;
-                    } else {
-                        return false;
-                    }
-                } break;
-
-                default: break;
-            }
-            return true;
-        }
-
-        /**
-         * This method checks to see if the decoder is done
-         * and has decoded the encoded character.
-         *
-         * @return
-         *     An indication of whether or not the decoder is done
-         *     and has decoded the encoded character is returned.
-         */
-        bool Done() const {
-            return (decoderState_ == 2);
-        }
-
-        /**
-         * This method returns the decoded character, once
-         * the decoder is done.
-         *
-         * @return
-         *     The decoded character is returned.
-         */
-        char GetDecodedCharacter() const {
-            return (char)decodedCharacter_;
-        }
-
-        // Properties
-    private:
-        /**
-         * This is the decoded character.
-         */
-        int decodedCharacter_ = 0;
-
-        /**
-         * This is the current state of the decoder's state machine.
-         * - 0: we haven't yet received the first hex digit.
-         * - 1: we received the first hex digit but not the second.
-         * - 2: we received both hex digits
-         */
-        size_t decoderState_ = 0;
-    };
-
-    /**
      * This method checks and decodes the given path segment.
      *
      * @param[in,out] segment
@@ -246,16 +127,16 @@ namespace {
         segment.clear();
         size_t decoderState = 0;
         int decodedCharacter = 0;
-        PercentEncodedCharacterDecoder pecDecoder;
+        Uri::PercentEncodedCharacterDecoder pecDecoder;
         for (const auto c: originalSegment) {
             switch(decoderState) {
                 case 0: { // default
                     if (c == '%') {
-                        pecDecoder = PercentEncodedCharacterDecoder();
+                        pecDecoder = Uri::PercentEncodedCharacterDecoder();
                         decoderState = 1;
                     } else {
                         if (
-                            IsCharacterInSet(
+                            Uri::IsCharacterInSet(
                                 c,
                                 {
                                     // unreserved
@@ -309,16 +190,16 @@ namespace {
         queryOrFragment.clear();
         size_t decoderState = 0;
         int decodedCharacter = 0;
-        PercentEncodedCharacterDecoder pecDecoder;
+        Uri::PercentEncodedCharacterDecoder pecDecoder;
         for (const auto c: originalQueryOrFragment) {
             switch(decoderState) {
                 case 0: { // default
                     if (c == '%') {
-                        pecDecoder = PercentEncodedCharacterDecoder();
+                        pecDecoder = Uri::PercentEncodedCharacterDecoder();
                         decoderState = 1;
                     } else {
                         if (
-                            IsCharacterInSet(
+                            Uri::IsCharacterInSet(
                                 c,
                                 {
                                     // unreserved