http://d.puremagic.com/bugzilla/show_bug.cgi?id=111

           Summary: appending a dchar to a char[]
           Product: D
           Version: 0.154
          Platform: PC
        OS/Version: Windows
            Status: NEW
          Severity: normal
          Priority: P2
         Component: DMD
        AssignedTo: bugzilla@digitalmars.com
        ReportedBy: lio@lunesu.com


#void main() {
#  char[] t;
#//  t ~= '我'; // Error: Can only append to dynamic arrays, not char[]
~= wchar
#  t ~= "我";
#  char[] r;
#  foreach(dchar d; t) {
#    int rel = r.length;
#    r ~= d;
#    if (d>127) assert(r.length-rel>1); // fails
#  }
#  assert(r.length == t.length); // fails
#}

First of all: since I'm allowed to write "r ~= dchar", I should also be allowed to do the same with a constant, '\u6211'. This probably got posted before.

The second one seems similar: it compiles without complaint, but it's not correctly appending the 2-byte dchar to the char[].


-- 

http://d.puremagic.com/bugzilla/show_bug.cgi?id=111


smjg@iname.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |smjg@iname.com




------- Comment #1 from smjg@iname.com  2006-04-21 08:20 -------
Nothing like the error I get:

bugz111.d(3): unterminated character constant
bugz111.d(3): #line integer ["filespec"]\n expected
bugz111.d(3): unterminated character constant
bugz111.d(4): found 't' when expecting ';' following 'statement'
bugz111.d(4): found '~=' instead of statement

This is more understandable, since after the lexer has found one character, namely the '&', it expects the closing quote.  However, it appears that the compiler is poor at recovering from this error.


-- 

http://d.puremagic.com/bugzilla/show_bug.cgi?id=111





------- Comment #2 from smjg@iname.com  2006-04-21 09:01 -------
Having looked at the post on digitalmars.D.bugs from just before it was filed here, it now appears that either Bugzilla or the reporter's browser mangled the code.  It was actually an oriental character embedded directly in the code. Let's see what happens if I try doing it in Mozilla....

#  t ~= '我';
#  t ~= "我";


-- 

http://d.puremagic.com/bugzilla/show_bug.cgi?id=111





------- Comment #3 from lio@lunesu.com  2006-04-21 09:15 -------
Use '\u6211' resp. "\u6211" instead for the same effect.


-- 

d-bugmail@puremagic.com schrieb am 2006-04-20:
> First of all: since I'm allowed to write "r ~= dchar", I should also be allowed to do the same with a constant, '\u6211'. This probably got posted before.
>
> The second one seems similar: it compiles without complaint, but it's not correctly appending the 2-byte dchar to the char[].

Added to DStress as http://dstress.kuehne.cn/run/o/opCatAssign_19_A.d http://dstress.kuehne.cn/run/o/opCatAssign_19_B.d http://dstress.kuehne.cn/run/o/opCatAssign_19_C.d http://dstress.kuehne.cn/run/o/opCatAssign_19_D.d http://dstress.kuehne.cn/run/o/opCatAssign_19_E.d http://dstress.kuehne.cn/run/o/opCatAssign_19_F.d

Thomas