utf8_alignment_example.dart

import 'dart:convert';

import 'dart:math';

String charToHex(int c) => c.toRadixString(16).padLeft(2, '0');
String charListToHex(List<int> s) => s.map(charToHex).join(' ');

String shortToHex(int c) => c.toRadixString(16).padLeft(4, '0');
String shortListToHex(List<int> s) => s.map(shortToHex).join(' ');

List<List<int>> splitList(List<int> input, int size) {
  final result = <List<int>>[];
  for (var i = 0; i <= (input.length - 1) ~/ size; i++) {
    var offset = i * size;
    result.add(input.sublist(offset, min(input.length, offset + size)));
  }
  return result;
}

List<int> joinList(List<List<int>> input) => input.expand((e) => e).toList(); 

void main() {
  final smiley = 'Hello, World!!! 😊';
  print('Smiley: $smiley');
  
  // Strings in Dart (and C#) are actually lists of UTF-16 code points:
  print('UTF-16: ${shortListToHex(smiley.codeUnits)}');

  // We can convert them to a list of UTF-8 bytes with utf8.encode:
  final smileyChars = utf8.encode(smiley);
  print('UTF-8: ${charListToHex(smileyChars)}');
  
  // When sending streams of data over the wire, the data gets chunked:
  final splitSmileyChars = splitList(smileyChars, 4);
  print('Split up: ${splitSmileyChars.map(charListToHex).join(' | ')}');
  
  // The naive solution to get our original string is to utf-8 decode each chunk
  // into strings and concatenate them together. This blows up if a code point
  // crosses the boundary of a split:
  final decodedSplitSmiley = splitSmileyChars.map(utf8.decode).toList();
  print('Strings: ${jsonEncode(decodedSplitSmiley)}');
  print('Back together: ${decodedSplitSmiley.join()}');
  
  // A more correct solution decodes the entire string at once:
  final joinedSmileyChars = joinList(splitSmileyChars);
  print('Back together v2: ${utf8.decode(joinedSmileyChars)}');
}