epee: support unicode in parsed strings

3 files changed, 78 insertions, 22 deletions

diff --git a/contrib/epee/include/storages/parserse_base_utils.h b/contrib/epee/include/storages/parserse_base_utils.h
index b5c4138c5..fe53628a5 100644
--- a/contrib/epee/include/storages/parserse_base_utils.h
+++ b/contrib/epee/include/storages/parserse_base_utils.h
@@ -31,6 +31,9 @@
 #include <algorithm>
 #include <boost/utility/string_ref.hpp>
 
+#undef MONERO_DEFAULT_LOG_CATEGORY
+#define MONERO_DEFAULT_LOG_CATEGORY "serialization"
+
 namespace epee 
 {
 namespace misc_utils
@@ -62,6 +65,26 @@ namespace misc_utils
       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
     };
 
+    static const constexpr unsigned char isx[256] =
+    {
+     0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+     0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+     0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+        0,    1,    2,    3,    4,    5,    6,    7,    8,    9, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+     0xff,   10,   11,   12,   13,   14,   15, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+     0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+     0xff,   10,   11,   12,   13,   14,   15, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+     0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+     0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+     0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+     0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+     0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+     0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+     0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+     0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+     0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+    };
+
     inline bool isspace(char c)
     {
       return lut[(uint8_t)c] & 8;
@@ -162,6 +185,42 @@ namespace misc_utils
               val.push_back('\\');break;
             case '/':  //Slash character
               val.push_back('/');break;
+            case 'u':  //Unicode code point
+              if (buf_end - it < 4)
+              {
+                ASSERT_MES_AND_THROW("Invalid Unicode escape sequence");
+              }
+              else
+              {
+                uint32_t dst = 0;
+                for (int i = 0; i < 4; ++i)
+                {
+                  const unsigned char tmp = isx[(int)*++it];
+                  CHECK_AND_ASSERT_THROW_MES(tmp != 0xff, "Bad Unicode encoding");
+                  dst = dst << 4 | tmp;
+                }
+                // encode as UTF-8
+                if (dst <= 0x7f)
+                {
+                  val.push_back(dst);
+                }
+                else if (dst <= 0x7ff)
+                {
+                  val.push_back(0xc0 | (dst >> 6));
+                  val.push_back(0x80 | (dst & 0x3f));
+                }
+                else if (dst <= 0xffff)
+                {
+                  val.push_back(0xe0 | (dst >> 12));
+                  val.push_back(0x80 | ((dst >> 6) & 0x3f));
+                  val.push_back(0x80 | (dst & 0x3f));
+                }
+                else
+                {
+                  ASSERT_MES_AND_THROW("Unicode code point is out or range");
+                }
+              }
+              break;
             default:
               val.push_back(*it);
               LOG_PRINT_L0("Unknown escape sequence :\"\\" << *it << "\"");
diff --git a/contrib/epee/include/string_tools.h b/contrib/epee/include/string_tools.h
index da47b7d55..1be5eb5e1 100644
--- a/contrib/epee/include/string_tools.h
+++ b/contrib/epee/include/string_tools.h
@@ -59,26 +59,6 @@
 #pragma comment (lib, "Rpcrt4.lib")
 #endif
 
-static const constexpr unsigned char isx[256] =
-{
- 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
- 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
- 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
-    0,    1,    2,    3,    4,    5,    6,    7,    8,    9, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
- 0xff,   10,   11,   12,   13,   14,   15, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
- 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
- 0xff,   10,   11,   12,   13,   14,   15, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
- 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
- 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
- 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
- 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
- 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
- 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
- 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
- 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
- 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
-};
-
 namespace epee
 {
 namespace string_tools
@@ -99,10 +79,10 @@ namespace string_tools
       for(size_t i = 0; i < s.size(); i += 2)
       {
         int tmp = *src++;
-        tmp = isx[tmp];
+        tmp = epee::misc_utils::parse::isx[tmp];
         if (tmp == 0xff) return false;
         int t2 = *src++;
-        t2 = isx[t2];
+        t2 = epee::misc_utils::parse::isx[t2];
         if (t2 == 0xff) return false;
         *dst++ = (tmp << 4) | t2;
       }
diff --git a/tests/unit_tests/epee_utils.cpp b/tests/unit_tests/epee_utils.cpp
index 946731826..c5f9a42d0 100644
--- a/tests/unit_tests/epee_utils.cpp
+++ b/tests/unit_tests/epee_utils.cpp
@@ -946,3 +946,20 @@ TEST(parsing, number)
   epee::misc_utils::parse::match_number(i, s.end(), val);
   ASSERT_EQ(val, "+9.34e+03");
 }
+
+TEST(parsing, unicode)
+{
+  std::string bs;
+  std::string s;
+  std::string::const_iterator si;
+
+  s = "\"\""; si = s.begin(); ASSERT_TRUE(epee::misc_utils::parse::match_string(si, s.end(), bs)); ASSERT_EQ(bs, "");
+  s = "\"\\u0000\""; si = s.begin(); ASSERT_TRUE(epee::misc_utils::parse::match_string(si, s.end(), bs)); ASSERT_EQ(bs, std::string(1, '\0'));
+  s = "\"\\u0020\""; si = s.begin(); ASSERT_TRUE(epee::misc_utils::parse::match_string(si, s.end(), bs)); ASSERT_EQ(bs, " ");
+  s = "\"\\u1\""; si = s.begin(); ASSERT_FALSE(epee::misc_utils::parse::match_string(si, s.end(), bs));
+  s = "\"\\u12\""; si = s.begin(); ASSERT_FALSE(epee::misc_utils::parse::match_string(si, s.end(), bs));
+  s = "\"\\u123\""; si = s.begin(); ASSERT_FALSE(epee::misc_utils::parse::match_string(si, s.end(), bs));
+  s = "\"\\u1234\""; si = s.begin(); ASSERT_TRUE(epee::misc_utils::parse::match_string(si, s.end(), bs)); ASSERT_EQ(bs, "ሴ");
+  s = "\"foo\\u1234bar\""; si = s.begin(); ASSERT_TRUE(epee::misc_utils::parse::match_string(si, s.end(), bs)); ASSERT_EQ(bs, "fooሴbar");
+  s = "\"\\u3042\\u307e\\u3084\\u304b\\u3059\""; si = s.begin(); ASSERT_TRUE(epee::misc_utils::parse::match_string(si, s.end(), bs)); ASSERT_EQ(bs, "あまやかす");
+}