Re: Should Inline be private in the private part of a package spec?

[Shark8 <onewingedshark@gmail.com>]

On Tuesday, August 14, 2012 9:19:02 AM UTC-6, Vasiliy Molostov wrote:
> Shark8 <onewingedshark@gmail.com> писал(а) в своём письме Tue, 14 Aug 2012  
> 
> 18:44:42 +0400:
> 
> > These problems can make "diff"ing the XML source very difficult." --  
> > http://c2.com/cgi/wiki?XmlSucks
> 
> Probably if you have unclear view it is better to use xmldiff tool. Also,  
> try to sucks against
> Scala/Java/Groovy and friends - you might want to wonder where you stay  
> with you opinion while others moving forward.

Sometimes you simply don't have access to such tools, even if you're aware of them. But thank you for telling me about XMLdiff, I'll have to check it out next time I have to muck around with XML. (If I'm lucky that'll be "never again".)

> 
> This is a description landuage rooted from its grandfather EDIF, being  
> born in AutoCAD field as an interchange text file format. So the purpose  
> is clear.

Oh, I'm not saying it has an unclear purpose. I'm saying that it's a hard format to use nicely, to keep things correct with. Though I do pity those who have been forced to work projects where regexes are used on XML. (That seems to be a surprisingly common attempt, though it's not possible as it requires something more powerful; the same as between an FSM and PDA, map to those problems, IIRC.)

> 
> To get many tools capable to use compiler project/compilation/database it  
> is the best choice. And a tremendous asis related interfaces can go to the  
> bookshelf.

I don't think that's true. (That XML is the best choice for compilers to use, I mean; I haven't had cause to use ASIS so I can't comment on it.)