From b851dcf9b90fb4dc49d2ced76857bfce9e67d9a4 Mon Sep 17 00:00:00 2001
From: apio <blobs.trading@gmail.com>
Date: Fri, 6 Jan 2023 21:01:37 +0100
Subject: [PATCH] libc: Implement mbstowcs() using Utf8StringDecoder

---
 libc/include/stdlib.h    |  5 +++++
 libc/src/stdlib.cpp      | 18 ++++++++++++++++++
 luna/include/luna/Utf8.h |  4 ++++
 luna/src/Utf8.cpp        | 25 +++++++++++++++++++------
 4 files changed, 46 insertions(+), 6 deletions(-)

diff --git a/libc/include/stdlib.h b/libc/include/stdlib.h
index 38cb7095..4e4b3e56 100644
--- a/libc/include/stdlib.h
+++ b/libc/include/stdlib.h
@@ -27,6 +27,8 @@ typedef struct
     long long rem;
 } lldiv_t;
 
+#define MB_CUR_MAX 4
+
 #ifdef __cplusplus
 extern "C"
 {
@@ -98,6 +100,9 @@ extern "C"
     void qsort(void*, size_t, size_t, int (*)(const void*, const void*));
     void* bsearch(const void*, const void*, size_t, size_t, int (*)(const void*, const void*));
 
+    /* Convert a multibyte character string to a wide character string. */
+    size_t mbstowcs(wchar_t* buf, const char* src, size_t max);
+
 #ifdef __cplusplus
 }
 #endif
diff --git a/libc/src/stdlib.cpp b/libc/src/stdlib.cpp
index 12f338a4..b84dda9c 100644
--- a/libc/src/stdlib.cpp
+++ b/libc/src/stdlib.cpp
@@ -1,5 +1,6 @@
 #include <limits.h>
 #include <luna/NumberParsing.h>
+#include <luna/Utf8.h>
 #include <stdlib.h>
 #include <sys/syscall.h>
 #include <unistd.h>
@@ -99,4 +100,21 @@ extern "C"
         syscall(SYS_exit);
         __builtin_unreachable();
     }
+
+    size_t mbstowcs(wchar_t* buf, const char* src, size_t max)
+    {
+        if (max == 0) return 0;
+
+        Utf8StringDecoder decoder(src);
+
+        auto rc = decoder.decode(buf, max);
+
+        if (rc.has_error()) return (size_t)-1;
+
+        size_t code_points = decoder.code_points().value_or(0);
+
+        if (code_points >= max) return max - 1;
+
+        return code_points;
+    }
 }
diff --git a/luna/include/luna/Utf8.h b/luna/include/luna/Utf8.h
index 280d29ed..2f5b1b45 100644
--- a/luna/include/luna/Utf8.h
+++ b/luna/include/luna/Utf8.h
@@ -18,6 +18,8 @@ class Utf8StringDecoder
     // The caller must ensure that 'buf' is at least code_points() + a NULL wide.
     Result<void> decode(wchar_t* buf) const;
 
+    Result<void> decode(wchar_t* buf, size_t max) const;
+
   private:
     const char* m_str;
     usize m_byte_length;
@@ -38,6 +40,8 @@ class Utf8StringEncoder
     // The caller must ensure that 'buf' is at least byte_length() + a NULL wide.
     Result<void> encode(char* buf) const;
 
+    Result<void> encode(char* buf, size_t max) const;
+
   private:
     const wchar_t* m_str;
     usize m_code_points;
diff --git a/luna/src/Utf8.cpp b/luna/src/Utf8.cpp
index 3a84df36..5ebb880d 100644
--- a/luna/src/Utf8.cpp
+++ b/luna/src/Utf8.cpp
@@ -32,7 +32,9 @@ static inline usize wide_char_length_as_utf8_unchecked(wchar_t c)
 
 static Result<void> encode_wide_char_as_utf8(wchar_t c, char* result, usize& len)
 {
-    len = TRY(wide_char_length_as_utf8(c));
+    usize utf8_len = TRY(wide_char_length_as_utf8(c));
+
+    if (utf8_len > len) { return err(EILSEQ); }
 
     u8* buf = (u8*)result;
 
@@ -144,11 +146,11 @@ Result<usize> Utf8StringDecoder::code_points() const
     return len;
 }
 
-Result<void> Utf8StringDecoder::decode(wchar_t* buf) const
+Result<void> Utf8StringDecoder::decode(wchar_t* buf, size_t max) const
 {
     const char* it = m_str;
 
-    while ((usize)(it - m_str) < m_byte_length)
+    while ((usize)(it - m_str) < m_byte_length && max--)
     {
         usize len = m_byte_length - (usize)(it - m_str); // Remaining space
         *buf = TRY(encode_utf8_as_wide_char(it, len));
@@ -161,6 +163,11 @@ Result<void> Utf8StringDecoder::decode(wchar_t* buf) const
     return {};
 }
 
+Result<void> Utf8StringDecoder::decode(wchar_t* buf) const
+{
+    return decode(buf, (size_t)-1);
+}
+
 Utf8StringEncoder::Utf8StringEncoder(const wchar_t* str) : m_str(str), m_code_points(wcslen(str))
 {
 }
@@ -179,15 +186,16 @@ Result<usize> Utf8StringEncoder::byte_length() const
     return len;
 }
 
-Result<void> Utf8StringEncoder::encode(char* buf) const
+Result<void> Utf8StringEncoder::encode(char* buf, size_t max) const
 {
     const wchar_t* it = m_str;
 
-    while (*it)
+    while (*it && max > 1)
     {
-        usize len = 0;
+        usize len = max - 1;
         TRY(encode_wide_char_as_utf8(*it, buf, len));
         buf += len;
+        max -= len;
         it++;
     }
 
@@ -196,6 +204,11 @@ Result<void> Utf8StringEncoder::encode(char* buf) const
     return {};
 }
 
+Result<void> Utf8StringEncoder::encode(char* buf) const
+{
+    return encode(buf, (size_t)-1);
+}
+
 Utf8StateDecoder::Utf8StateDecoder() : m_state_len(0), m_state_index(0)
 {
 }