From: daniel.kruegler@googlemail.com   
      
   Am 14.10.2012 04:28, schrieb ralph:[..]   
     > There is absolutely no reason for the OP to make "plain" or   
     > "standardize" that entry point. In fact he was doing the correct   
     > thing by including it.   
      
   If a person would like to discuss the effects of a program that   
   depends on this kind of entry point, I recommend to use a newsgroup   
   that is more appropriate for this kind of discussion.   
      
     > It wasn't a factor in this case, as the OP clearly indicated the   
     > compiler he was using ("VS10"), and the 'main()' declaration did   
     > not have any impact on his issue.   
      
   How can you know that in advance? Several problems that had been   
   discussed in this group had be problems that were based on assumptions   
   about program behaviour that is not guaranteed.   
      
     > However, it is possible it might have been a factor in another   
     > scenario. Also would clearly go against the common request to   
     > provide actual code and NOT a "sanitized" re-typing.   
      
   I intentionally wrote "unless this would be really necessary" to   
   indicate a situation where the problem seemingly was related to this   
   kind of code, therefore necessary to produce the problem.   
      
     > In addition, note there is nothing "non-standard" about this   
     > declaration. It is implementation-defined. Note the leading   
     > underscore.   
      
   You are incorrectly interpreting the standard. This kind of name is   
   ruled by [global.names] and there is nothing that says that it would   
   be implementation-defined what the effects are of using such a   
   name. Your argument holds only for implementations that give the name   
   _tmain the special meaning of a valid entry point name. A different   
   valid implementation might declare an object with the name _tmain in   
   the global namespace. In this case the function declaration provided   
   by the user would be ill-formed.   
      
     > Thus requesting posters to provide "portable" code for an   
     > *implementation* question is not reasonable and could in fact be   
     > obfuscating and misleading.   
      
   I was clearly not requesting it, when asking "Please do not use   
   non-portable "main" functions within this group" this was intended as   
   an advice. I recommend it. It also helps to raise interest for this   
   question to people that do not understand which effects a name like   
   "_tmain" are. It certainly make the problem more complex and steels   
   focus of the actual point of interest.   
      
   Greetings from Bremen,   
      
   Daniel Krügler   
      
      
   --   
         [ 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)