Xteink-X4-crosspoint-reader/lib/Utf8/Utf8.cpp

#include "Utf8.h"

int utf8CodepointLen(const unsigned char c) {
  if (c < 0x80) return 1;          // 0xxxxxxx
  if ((c >> 5) == 0x6) return 2;   // 110xxxxx
  if ((c >> 4) == 0xE) return 3;   // 1110xxxx
  if ((c >> 3) == 0x1E) return 4;  // 11110xxx
  return 1;                        // fallback for invalid
}

uint32_t utf8NextCodepoint(const unsigned char** string) {
  if (**string == 0) {
    return 0;
  }

  const int bytes = utf8CodepointLen(**string);
  const uint8_t* chr = *string;
  *string += bytes;

  if (bytes == 1) {
    return chr[0];
  }

  uint32_t cp = chr[0] & ((1 << (7 - bytes)) - 1);  // mask header bits

  for (int i = 1; i < bytes; i++) {
    cp = (cp << 6) | (chr[i] & 0x3F);
  }

  return cp;
}

uint32_t utf8NextCodepointWithLigatures(const unsigned char** string) {
  const uint32_t cp = utf8NextCodepoint(string);
  if (cp != 'f') {
    return cp;
  }

  // Save position after 'f' for potential rollback
  const unsigned char* afterF = *string;

  const uint32_t next = utf8NextCodepoint(string);
  if (next == 'f') {
    // ff — check for ffi or ffl
    const unsigned char* afterFF = *string;
    const uint32_t third = utf8NextCodepoint(string);
    if (third == 'i') {
      return 0xFB03;  // ffi
    }
    if (third == 'l') {
      return 0xFB04;  // ffl
    }
    // Just ff, roll back the third character
    *string = afterFF;
    return 0xFB00;  // ff
  }
  if (next == 'i') {
    return 0xFB01;  // fi
  }
  if (next == 'l') {
    return 0xFB02;  // fl
  }

  // No ligature match, roll back to after 'f'
  *string = afterF;
  return 'f';
}

size_t utf8RemoveLastChar(std::string& str) {
  if (str.empty()) return 0;
  size_t pos = str.size() - 1;
  while (pos > 0 && (static_cast<unsigned char>(str[pos]) & 0xC0) == 0x80) {
    --pos;
  }
  str.resize(pos);
  return pos;
}

// Truncate string by removing N UTF-8 characters from the end
void utf8TruncateChars(std::string& str, const size_t numChars) {
  for (size_t i = 0; i < numChars && !str.empty(); ++i) {
    utf8RemoveLastChar(str);
  }
}