From 7dd49c6293172b494c78918507242cdb55d35137 Mon Sep 17 00:00:00 2001
From: Joel Klinghed <the_jk@spawned.biz>
Date: Sun, 21 Jan 2024 12:31:30 +0100
Subject: WIP

---
 utf/inc/utf16.hh | 12 +++++++-----
 utf/inc/utf32.hh | 12 +++++++-----
 utf/inc/utf8.hh  | 19 +++++++++++++++----
 3 files changed, 29 insertions(+), 14 deletions(-)

(limited to 'utf/inc')

diff --git a/utf/inc/utf16.hh b/utf/inc/utf16.hh
index 344b1a2..b9229bc 100644
--- a/utf/inc/utf16.hh
+++ b/utf/inc/utf16.hh
@@ -4,27 +4,29 @@
 #include "macros.hh"
 
 #include <cstdint>
-#include <string_view>
+#include <span>
 
 namespace utf {
 
-/* Read one unicode codepoint from UTF-16 BigEndian encoded data if possible.
+/**
+ * Read one unicode codepoint from UTF-16 BigEndian encoded data if possible.
  * If successfull offset is incremented to point to next codepoint.
  * Will fail:
  * - not enough data is left in data given offset, returns NEED_MORE.
  * - data is not valid UTF-16, ie. invalid or incomplete surrogate pairs,
  *   returns INVALID.
  */
-uint32_t HIDDEN read16be(std::string_view data, std::size_t& offset);
+uint32_t HIDDEN read16be(std::span<uint8_t const> data, std::size_t& offset);
 
-/* Read one unicode codepoint from UTF-16 LittleEndian encoded data if possible.
+/**
+ * Read one unicode codepoint from UTF-16 LittleEndian encoded data if possible.
  * If successfull offset is incremented to point to next codepoint.
  * Will fail:
  * - not enough data is left in data given offset, returns NEED_MORE.
  * - data is not valid UTF-16, ie. invalid or incomplete surrogate pairs,
  *   returns INVALID.
  */
-uint32_t HIDDEN read16le(std::string_view data, std::size_t& offset);
+uint32_t HIDDEN read16le(std::span<uint8_t const> data, std::size_t& offset);
 
 }  // namespace utf
 
diff --git a/utf/inc/utf32.hh b/utf/inc/utf32.hh
index 2d3088e..4ee5eac 100644
--- a/utf/inc/utf32.hh
+++ b/utf/inc/utf32.hh
@@ -4,25 +4,27 @@
 #include "macros.hh"
 
 #include <cstdint>
-#include <string_view>
+#include <span>
 
 namespace utf {
 
-/* Read one unicode codepoint from UTF-32 BigEndian encoded data if possible.
+/**
+ * Read one unicode codepoint from UTF-32 BigEndian encoded data if possible.
  * If successfull offset is incremented to point to next codepoint.
  * Will fail:
  * - not enough data is left in data given offset, returns NEED_MORE.
  * - data is not valid UTF-32, ie. outside valid ranges, returns INVALID.
  */
-uint32_t HIDDEN read32be(std::string_view data, std::size_t& offset);
+uint32_t HIDDEN read32be(std::span<uint8_t const> data, std::size_t& offset);
 
-/* Read one unicode codepoint from UTF-32 LittleEndian encoded data if possible.
+/**
+ * Read one unicode codepoint from UTF-32 LittleEndian encoded data if possible.
  * If successfull offset is incremented to point to next codepoint.
  * Will fail:
  * - not enough data is left in data given offset, returns NEED_MORE.
  * - data is not valid UTF-32, ie. outside valid ranges, returns INVALID.
  */
-uint32_t HIDDEN read32le(std::string_view data, std::size_t& offset);
+uint32_t HIDDEN read32le(std::span<uint8_t const> data, std::size_t& offset);
 
 }  // namespace utf
 
diff --git a/utf/inc/utf8.hh b/utf/inc/utf8.hh
index a3ea84a..7735ecd 100644
--- a/utf/inc/utf8.hh
+++ b/utf/inc/utf8.hh
@@ -4,18 +4,29 @@
 #include "macros.hh"
 
 #include <cstdint>
-#include <string_view>
+#include <span>
 
 namespace utf {
 
-/* Read one unicode codepoint from UTF-8 encoded data if possible.
- * If successfull offset is incremented to point to next codepoint.
+/**
+ * Read one unicode codepoint from UTF-8 encoded data if possible.
+ * If successful, offset is incremented to point to next codepoint.
  * Will fail:
  * - not enough data is left in data given offset, returns NEED_MORE.
  * - data is not valid UTF-8, this includes overlong encodings and
  *   invalid unicode code points, returns INVALID.
  */
-uint32_t HIDDEN read8(std::string_view data, std::size_t& offset);
+uint32_t HIDDEN read8(std::span<uint8_t const> data, std::size_t& offset);
+
+/**
+ * Write one unicode codepoint to UTF-8 encoded data if possible.
+ * If successful, offset is incremented to the end of the written data
+ * and true is returned.
+ * If not successful, offset is not incremented and false is returned.
+ * data is not modified.
+ */
+bool HIDDEN write8(uint32_t codepoint, std::span<uint8_t> data,
+                   std::size_t& offset);
 
 }  // namespace utf
 
-- 
cgit v1.2.3-70-g09d2