35 ms·

Memory model and byte order is enough to add breakage to stuff written in all of those languages.

• problems 9y ago
  I believe Java and C# enforce specific byte orders to get around this issue, so I doubt you'll see a problem there unless interacting with C++ libraries.
  • MichaelGG 9y ago
    Not if you're converting bytes into integers, for instance. E.g. https://msdn.microsoft.com/en-us/library/system.bitconverter.toint32(v=vs.110).aspx https://msdn.microsoft.com/en-us/library/system.bitconverter... - it's arch specific.
    • problems 9y ago
      Interesting, similar functions in the Java world are always big endian: http://docs.oracle.com/javase/6/docs/api/java/math/BigInteger.html#toByteArray%28%29 http://docs.oracle.com/javase/6/docs/api/java/math/BigIntege... Seems like a rather poor decision on Microsoft's part. Fortunately rather few things interact with data this way anyways, most serialization formats are text based or include endian information.