=?gb2312?q?=BB=D8=B8=B4=A3=BA=20Re:=20=BB=D8=B8=B4=A3=BA=20Re:=20[antlr-i?= =?gb2312?q?nterest]=20token=20position=20detection=20problem?=

[=?gb2312?q?=CB=EF=BC=CD=B8=D5=20Jigang=20(Robert)=20Sun?=]

Thanks a lot.

In lexer grammar, I changed "  '\t' {setTabSize(4);} " to " |  {setTabSize(4);} '\t' ". Works ok.
I am sure, in this way, TAB width had changed from 8 to 4 before the first TAB met. 


Jigang


--- Martin Probst <mail at martin-probst.com>дµÀ:

> 
> > As showed on the attached picture, in the 4th line "	 
> > System.out.println(8);", the position of '8'
> > is detected as line = 4, column=28,the only TAB character before  
> > "System" is conted as 8 spaces; I
> > made an exact  same text by copy and paste,this time the position  
> > of 8 inside method add is
> > line=17, column=24, the TAB inside is counted as 4 correctly.
> 
> I might be mistaken (no ANTLR source tree at hands) but aren't you  
> supposed to do:
> "setTabWidth(4);" before starting to parse? E.g.
> 
> Parser myParser = new Parser();
> myParser.setTabWidth(4);
> myParser.axiom();
> 
> I think you're setting it every time you encounter a tab, but only  
> after actually consuming the tab. This way, the first tab will always  
> be mistaken, subsequent ones will be read correctly.
> 
> Regards,
> Martin
> 
> 



	

	
		
___________________________________________________________ 
ÑÅ»¢1GÃâ·ÑÓÊÏä°Ù·Ö°Ù·ÀÀ¬»øÐÅ 
http://cn.mail.yahoo.com/