C++ jargons: Really CONFUSING

Hi folks,

We use these two C++ jargons very often while explaining the code architecture.

Code:

---


1. UNICODE and BYTECODE

and the other one is

2. EARLY BINDING  and LATE BINDING

---

As per my understanding, Early Binding and Late Binding are used during compilation may be similar to an Array (static memory allocation) and New (dynamic memory allocation).

BUT, I have no idead about Unicode and Bytecode.

I would really appreciate if anyone can provide a clear picture for both the above queries.

thanx in advance:)

-- Cavestine

Early binding means that it is clear at compile time which actual function a particular call refers to. This is the case for most functions. Late binding means it isn't clear until runtime, e.g. for virtual functions and function pointers.
DLL functions are bound at load time, so they're somewhere in-between.

Unicode is the name of a single, all-encompassing, universal character set, together with classification of these characters, a few algorithms for proper handling of their idiosyncrasies (such as the Unicode Bidi algorithm and the string collation algorithms) and a few mappings to byte streams (most notably UTF-8, UTF-16 and UTF-32). The actual character set is code-for-code identical to the Universal Character Set, standardized by ISO as ISO-10046 or something like that. (I can never remember the number.)
Unicode is further used in "The Unicode Consortium", which is the industry consortium that defines and evolves Unicode, and it is widely misused (especially in the Win32 world) to refer to the UTF-16 encoding. To say that a program "is Unicode" means that it uses wide strings.

You misheard (or someone misused) the term bytecode. The word refers to CPU-independent binary code such as Java Bytecode and CIL (.Net) Bytecode.

What was meant were single-byte encodings and multi-byte encodings. (The term "character set" is often used instead of encoding, but that's a misuse of the term.) Single-byte and multi-byte encodings stand in contrast to wide encodings. Single-byte encodings use a single byte for every character; they are thus very limited in the number of characters they can represent. ASCII, Windows-1252 and the ISO-8859-* family are single-byte encodings. Wide encodings use more than one byte for every character. Multi-byte encodings use a variable number of bytes for every character, from 1 up to as many as 4, depending on the encoding.
UTF-16 is a variable-width wide encoding (it uses one or two 16-bit units per character), UTF-32 is a fixed-width wide encoding (one 32-bit unit per character). UTF-8 is a multi-byte (or variable-width) "narrow" (the byte is the base) encoding (between 1 and 4 bytes per character) using a modified prefix byte scheme (you inspect the first byte to see if it is a character on its own or if follow-up bytes are part of it). Shift-JIS is a multi-byte encoding using a shift scheme (some bytes are not characters, but instead shift between one-byte and two-byte mode).

UNICODE is a definition of "wide characters" to allow international characters.

BYTECODE is, generally speaking, a way to define some sort of "machine code that isn't machine code", e.g. Java Bytecode.

--
Mats

FOLDOC (Free On Line Dictionary Of Computing) can be handy. Although, the definitions can sometimes be a bit too brief.

I didn't find "early binding or "late binding", but I did find "binding". (The definition was not as clear & complete as CornBee's.) There is half-a-page on Unicode with several links.