[libc] Template StringConverter pop function to avoid duplicate code (#152204)

Addressed TODO to template the StringConverter pop functions to have a
single implementation (combine popUTF8 and popUTF32 into a single
templated pop function)

Author: uzairnawaz

libc/src/__support/wchar/character_converter.cpp

@@ -132,12 +132,6 @@ ErrorOr<char32_t> CharacterConverter::pop_utf32() {
   return utf32;
 }
 
-size_t CharacterConverter::sizeAsUTF32() {
-  return 1; // a single utf-32 value can fit an entire character
-}
-
-size_t CharacterConverter::sizeAsUTF8() { return state->total_bytes; }
-
 ErrorOr<char8_t> CharacterConverter::pop_utf8() {
   if (isEmpty())
     return Error(-1);

libc/src/__support/wchar/character_converter.h

@@ -12,6 +12,7 @@
 #include "hdr/types/char32_t.h"
 #include "hdr/types/char8_t.h"
 #include "hdr/types/size_t.h"
+#include "src/__support/CPP/type_traits.h"
 #include "src/__support/common.h"
 #include "src/__support/error_or.h"
 #include "src/__support/wchar/mbstate.h"
@@ -31,14 +32,18 @@ class CharacterConverter {
   bool isEmpty();
   bool isValidState();
 
-  size_t sizeAsUTF32();
-  size_t sizeAsUTF8();
+  template <typename CharType> size_t sizeAs();
+  template <> size_t sizeAs<char8_t>() { return state->total_bytes; }
+  template <> size_t sizeAs<char32_t>() { return 1; }
 
   int push(char8_t utf8_byte);
   int push(char32_t utf32);
 
   ErrorOr<char8_t> pop_utf8();
   ErrorOr<char32_t> pop_utf32();
+  template <typename CharType> ErrorOr<CharType> pop();
+  template <> ErrorOr<char8_t> pop() { return pop_utf8(); }
+  template <> ErrorOr<char32_t> pop() { return pop_utf32(); }
 };
 
 } // namespace internal

libc/src/__support/wchar/mbsnrtowcs.h

@@ -36,7 +36,7 @@ LIBC_INLINE static ErrorOr<size_t> mbsnrtowcs(wchar_t *__restrict dst,
   StringConverter<char8_t> str_conv(reinterpret_cast<const char8_t *>(*src), ps,
                                     len, nmc);
   size_t dst_idx = 0;
-  ErrorOr<char32_t> converted = str_conv.popUTF32();
+  ErrorOr<char32_t> converted = str_conv.pop<char32_t>();
   while (converted.has_value()) {
     if (dst != nullptr)
       dst[dst_idx] = converted.value();
@@ -47,7 +47,7 @@ LIBC_INLINE static ErrorOr<size_t> mbsnrtowcs(wchar_t *__restrict dst,
       return dst_idx;
     }
     dst_idx++;
-    converted = str_conv.popUTF32();
+    converted = str_conv.pop<char32_t>();
   }
 
   if (converted.error() == -1) { // if we hit conversion limit

libc/src/__support/wchar/string_converter.h

@@ -12,6 +12,7 @@
 #include "hdr/types/char32_t.h"
 #include "hdr/types/char8_t.h"
 #include "hdr/types/size_t.h"
+#include "src/__support/CPP/type_traits.h"
 #include "src/__support/common.h"
 #include "src/__support/error_or.h"
 #include "src/__support/wchar/character_converter.h"
@@ -53,9 +54,7 @@ template <typename T> class StringConverter {
                   size_t srclen = SIZE_MAX)
       : cr(ps), src(s), src_len(srclen), src_idx(0), num_to_write(dstlen) {}
 
-  // TODO: following functions are almost identical
-  // look into templating CharacterConverter pop functions
-  ErrorOr<char32_t> popUTF32() {
+  template <typename CharType> ErrorOr<CharType> pop() {
     if (num_to_write == 0)
       return Error(-1);
 
@@ -64,42 +63,17 @@ template <typename T> class StringConverter {
       if (!src_elements_read.has_value())
         return Error(src_elements_read.error());
 
-      if (cr.sizeAsUTF32() > num_to_write) {
+      if (cr.sizeAs<CharType>() > num_to_write) {
         cr.clear();
         return Error(-1);
       }
 
       src_idx += src_elements_read.value();
     }
 
-    auto out = cr.pop_utf32();
-    if (out.has_value() && out.value() == L'\0')
-      src_len = src_idx;
-
-    num_to_write--;
-
-    return out;
-  }
-
-  ErrorOr<char8_t> popUTF8() {
-    if (num_to_write == 0)
-      return Error(-1);
-
-    if (cr.isEmpty() || src_idx == 0) {
-      auto src_elements_read = pushFullCharacter();
-      if (!src_elements_read.has_value())
-        return Error(src_elements_read.error());
-
-      if (cr.sizeAsUTF8() > num_to_write) {
-        cr.clear();
-        return Error(-1);
-      }
-
-      src_idx += src_elements_read.value();
-    }
-
-    auto out = cr.pop_utf8();
-    if (out.has_value() && out.value() == '\0')
+    ErrorOr<CharType> out = cr.pop<CharType>();
+    // if out isn't null terminator or an error
+    if (out.has_value() && out.value() == 0)
       src_len = src_idx;
 
     num_to_write--;

libc/src/__support/wchar/wcsnrtombs.h

@@ -39,7 +39,7 @@ wcsnrtombs(char *__restrict dest, const wchar_t **__restrict ptr_to_src,
       reinterpret_cast<const char32_t *>(*ptr_to_src), ps, dest_len,
       num_src_widechars);
   size_t dst_idx = 0;
-  ErrorOr<char8_t> converted = str_conv.popUTF8();
+  ErrorOr<char8_t> converted = str_conv.pop<char8_t>();
   while (converted.has_value()) {
     if (dest != nullptr)
       dest[dst_idx] = converted.value();
@@ -51,7 +51,7 @@ wcsnrtombs(char *__restrict dest, const wchar_t **__restrict ptr_to_src,
     }
 
     dst_idx++;
-    converted = str_conv.popUTF8();
+    converted = str_conv.pop<char8_t>();
   }
 
   if (dest != nullptr)