Re: V4.1 Carriage Returns w/ ReplaceRepeated in Windows?
- To: mathgroup at smc.vnet.net
- Subject: [mg27632] Re: V4.1 Carriage Returns w/ ReplaceRepeated in Windows?
- From: Jens-Peer Kuska <kuska at informatik.uni-leipzig.de>
- Date: Fri, 9 Mar 2001 02:35:45 -0500 (EST)
- Organization: Universitaet Leipzig
- References: <9872dr$8u9@smc.vnet.net>
- Sender: owner-wri-mathgroup at wolfram.com
Hi,
it is not a small bug, it's perfectly correct.
The parser must know on a line end,
that the expression continues. If the expression is
only syntactical complete on a line end you must add
the operator at the end to force the parser to read
the next line.
Thats the reason why Mathematica break sums a+b+c+d+e into
a+b+
c+d+e
and not in the matematical usual continuation notation
a+b
+c+d+e
Regards
Jens
"John A. Gunnels" wrote:
>
> I have run into a rather odd "problem" (easily worked around, but ...)
> in my recent upgrade to Version 4.1.
>
> Under version 4.0, I could type and evaluate something
> like the following:
> x //. {a -> b}
> //. {c -> d}
> [where I hit the return between "b}" and the second "//."
>
> Now, under 4.1, this yields a syntax error:
> Syntax::tsntxi:
> "//. {c->d}" is incomplete; more input is needed
>
> while
> x //. {a -> b} //. {c -> d}
> works as expected (well, it's a simplified example
> which does nothing, but ... no syntax error and the
> analog in my application works).
>
> So I tried FullForm[Hold[<expression>]] with and without
> the line return (thinking that, perhaps, some odd character
> was inserted or something), only to see that the two were
> identical. Further, copying and pasting from inside the
> evaluated FullForm[Hold[]], WITH a carriage return, gives
> me a cell that looks like the one that gives me a syntax
> error (above), yet no error is reported upon evaluation.
> FURTHER, if I copy and paste the "]" (the right brace
> corresponding to the "Hold["), then erase it (otherwise, of
> course, it's not syntactically correct), evaluating the cell
> gives me the same syntax error is above.
>
> I realize that this is fairly trivial, but I would like to
> know if anyone else has run into this. It _seems_ like a
> _small_ bug to me, but ...
>
> Thanks,
> John A. Gunnels
> gunnels at cs.utexas.edu