Add code to check that scheme, if present, is legal

2 files changed, 127 insertions, 0 deletions

diff --git a/src/Uri.cpp b/src/Uri.cpp
index 8bc2f4a..da1b9f8 100644
--- a/src/Uri.cpp
+++ b/src/Uri.cpp
@@ -6,7 +6,9 @@
  * © 2018 by Richard Walters
  */
 
+#include <functional>
 #include <inttypes.h>
+#include <memory>
 #include <string>
 #include <Uri/Uri.hpp>
 #include <vector>
@@ -51,6 +53,83 @@ namespace {
         return true;
     }
 
+    /**
+     * This function takes a given "stillPassing" strategy
+     * and invokes it on the sequence of characters in the given
+     * string, to check if the string passes or not.
+     *
+     * @param[in] candidate
+     *     This is the string to test.
+     *
+     * @param[in] stillPassing
+     *     This is the strategy to invoke in order to test the string.
+     *
+     * @return
+     *     An indication of whether or not the given candidate string
+     *     passes the test is returned.
+     */
+    bool FailsMatch(
+        const std::string& candidate,
+        std::function< bool(char, bool) > stillPassing
+    ) {
+        for (const auto c: candidate) {
+            if (!stillPassing(c, false)) {
+                return true;
+            }
+        }
+        return !stillPassing(' ', true);
+    }
+
+    /**
+     * 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.
+     *
+     * @return
+     *      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 is returned.
+     */
+    std::function< bool(char, bool) > LegalSchemeCheckStrategy() {
+        auto isFirstCharacter = std::make_shared< bool >(true);
+        return [isFirstCharacter](char c, bool end){
+            if (end) {
+                return !*isFirstCharacter;
+            } else {
+                bool check;
+                if (*isFirstCharacter) {
+                    check = (
+                        (
+                            (c >= 'a')
+                            && (c <= 'z')
+                        ) || (
+                            (c >= 'A')
+                            && (c <= 'Z')
+                        )
+                    );
+                } else {
+                    check = (
+                        (
+                            (c >= 'a')
+                            && (c <= 'z')
+                        ) || (
+                            (c >= 'A')
+                            && (c <= 'Z')
+                        ) || (
+                            (c >= '0')
+                            && (c <= '9')
+                        )
+                        || (c == '+')
+                        || (c == '-')
+                        || (c == '.')
+                    );
+                }
+                *isFirstCharacter = false;
+                return check;
+            }
+        };
+    }
+
 }
 
 namespace Uri {
@@ -200,6 +279,15 @@ namespace Uri {
             rest = uriString;
         } else {
             impl_->scheme = uriString.substr(0, schemeEnd);
+            bool isFirstCharacter = true;
+            if (
+                FailsMatch(
+                    impl_->scheme,
+                    LegalSchemeCheckStrategy()
+                )
+            ) {
+                return false;
+            }
             rest = uriString.substr(schemeEnd + 1);
         }
 
diff --git a/test/src/UriTests.cpp b/test/src/UriTests.cpp
index 6ee57c4..b4be2d5 100644
--- a/test/src/UriTests.cpp
+++ b/test/src/UriTests.cpp
@@ -235,3 +235,42 @@ TEST(UriTests, ParseFromStringTwiceFirstUserInfoThenWithout) {
     ASSERT_TRUE(uri.ParseFromString("/foo/bar"));
     ASSERT_TRUE(uri.GetUserInfo().empty());
 }
+
+TEST(UriTests, ParseFromStringSchemeIllegalCharacters) {
+    const std::vector< std::string > testVectors{
+        {"://www.example.com/"},
+        {"0://www.example.com/"},
+        {"+://www.example.com/"},
+        {"@://www.example.com/"},
+        {".://www.example.com/"},
+        {"h@://www.example.com/"},
+    };
+    size_t index = 0;
+    for (const auto& testVector : testVectors) {
+        Uri::Uri uri;
+        ASSERT_FALSE(uri.ParseFromString(testVector)) << index;
+        ++index;
+    }
+}
+
+TEST(UriTests, ParseFromStringSchemeBarelyLegal) {
+    struct TestVector {
+        std::string uriString;
+        std::string scheme;
+    };
+    const std::vector< TestVector > testVectors{
+        {"h://www.example.com/", "h"},
+        {"x+://www.example.com/", "x+"},
+        {"y-://www.example.com/", "y-"},
+        {"z.://www.example.com/", "z."},
+        {"aa://www.example.com/", "aa"},
+        {"a0://www.example.com/", "a0"},
+    };
+    size_t index = 0;
+    for (const auto& testVector : testVectors) {
+        Uri::Uri uri;
+        ASSERT_TRUE(uri.ParseFromString(testVector.uriString)) << index;
+        ASSERT_EQ(testVector.scheme, uri.GetScheme());
+        ++index;
+    }
+}