From: 0xCDCDCDCD@gmx.at   
      
   On 26.01.2012 20:50, Daniel James wrote:   
   > In article, Martin B. wrote:   
   >>> This is pretty-much where I came in ... we need better tools for   
   >>> C++ *because* the current tools don't do semantic analysis of the   
   >>> sourcecode and we need tools that do.   
   >>   
   >> Do we? Certainly we *want* those tool, but apparently it isn't in   
   >> high enogh economic demand to get properly implemented.   
   >   
   > I could have put that better ...   
   >   
   > -- we do feel the lack of those tools, and   
   > would work more efficiently if we had them.   
   >   
   > ...   
   >> The problem as I see it is that, no matter how portable your code   
   >> is, you always target specific compilers. That means that unless the   
   >> refactoring tool -- which needs to "compile" the code -- is based   
   >> on the same compiler, there will be imperfections, to say the least.   
   >   
   > There shouldn't be "imperfections" ... as long as the IDE and the   
   > compiler both parse the language correctly. C++ is an ISO standard   
   > language, so there's no ambiguity ...   
   >   
      
   As the quote I put in my last mail says:   
      
    (...) this view is naïve,   
    rooted in the widely believed myth   
    that programming languages exist.   
      
   :-)   
      
      
   > That does mean, though, that IDEs that are written by compiler vendors   
   > should have an easier job of it ... they can use the parser information   
   > from the compiler to assist the IDE.   
   > ...   
   >> And also, thinking about it, I'd say that correct refactoring is even   
   >> harder that static analysis, because for static analysis to be useful   
   >> you can get away with interpreting only one compilation unit at a   
   >> time, but with refactoring you always need to interpret the whole   
   >> source base.   
   >   
   > I'd agree ... but don't forget that there are already tools that do a   
   > fairly good job of refactoring for Java and C# (for example),   
   > refactoring C++ isn't hard because refactoring is hard ... it's hard   
   > because parsing C++ is hard.   
   >   
      
   I assume when you say parsing, you mean making sense of the parsed AST(or   
   whatever)?   
      
   > I have hopes that Microsoft's apparent renewed interest in native code,   
   > brought on by their work towards Windows 8's "Metro" interface, will   
   > nudge them towards giving as much support to C++ in the IDE as they   
   > currently do for C# ... if only on their own platform.   
   >   
      
   It's not that MS isn't working on it, VS2010 has certainly gotten better at   
   the C++ intellisense front and VS11 should add some more improvements. It's   
   just that at this rate, we still won't have something powerful by 2020.   
      
   cheers,   
   Martin   
      
      
   --   
         [ See http://www.gotw.ca/resources/clcm.htm for info about ]   
         [ comp.lang.c++.moderated.    First time posters: Do this! ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)