codecvt: Add class for encoding conversion

Add a `codecvt` class that can be used as facet for locale so that it's
possible to convert from internal UTF-8 encoding to other encodings such
as Windows ANSI codepage.

1 files changed, 57 insertions, 0 deletions

diff --git a/Source/cm_codecvt.hxx b/Source/cm_codecvt.hxx
new file mode 100644
index 0000000..5f16bd8
--- /dev/null
+++ b/Source/cm_codecvt.hxx
@@ -0,0 +1,57 @@
+/* Distributed under the OSI-approved BSD 3-Clause License.  See accompanying
+   file Copyright.txt or https://cmake.org/licensing for details.  */
+#ifndef cm_codecvt_hxx
+#define cm_codecvt_hxx
+
+#include <cmConfigure.h> // IWYU pragma: keep
+
+#include <locale>
+#include <vector>
+
+class codecvt : public std::codecvt<char, char, mbstate_t>
+{
+public:
+  enum Encoding
+  {
+    None,
+    UTF8,
+    ANSI
+  };
+
+#ifdef CMAKE_BUILD_WITH_CMAKE
+
+  codecvt(Encoding e);
+
+protected:
+  ~codecvt() CM_OVERRIDE;
+  bool do_always_noconv() const throw() CM_OVERRIDE;
+  result do_out(mbstate_t& state, const char* from, const char* from_end,
+                const char*& from_next, char* to, char* to_end,
+                char*& to_next) const CM_OVERRIDE;
+  result do_unshift(mbstate_t& state, char* to, char*,
+                    char*& to_next) const CM_OVERRIDE;
+  int do_max_length() const throw() CM_OVERRIDE;
+  int do_encoding() const throw() CM_OVERRIDE;
+
+private:
+  typedef struct
+  {
+    bool used;
+    unsigned char totalBytes;
+    unsigned char bytesLeft;
+    char bytes[4];
+  } State;
+
+  unsigned int findStateId() const;
+
+  bool m_noconv;
+  mutable std::vector<State> m_states;
+  mutable unsigned int m_lastState;
+#if defined(_WIN32)
+  unsigned int m_codepage;
+#endif
+
+#endif
+};
+
+#endif