Board index » delphi » Re: Suggested Alternative Unicode Implementation (for Rudy+ misc others)
Jolyon Smith
Delphi Developer |
Re: Suggested Alternative Unicode Implementation (for Rudy+ misc others)2008-03-11 04:08:17 AM delphi199 In article <XXXX@XXXXX.COM>, XXXX@XXXXX.COM says... QuoteJolyon Smith writes: codepages is vastly less than that provided by 32-bits it costs nothing to make use of some of that space to encode some potentially useful information. The key point was that underlying all the syntactic sugar, there would have remained one single fully self-describing string type (excepting WideString for a moment). I guess I figured that consistency at that level would make the RTL implementation easier at least. Also, you might for example consider that the length information already stored with each String unnecessary duplication of redundant information, given that the length can be determined simply by detecting the offset of the null terminator. ;) QuoteAre you really need to deal with ANSI strings in different codepage :) Quote>In fact I don't think there would be many - if any - "checks" necessary If the underlying implementation of the string type is consistent across all encodings (which I considered a likely "Good Thing") then that space would potentially be present but unused for UTF string, so why not use it? Who knows when it might come in useful? :) |