On Sunday, 30 July 2017 at 01:20:52 UTC, H. S. Teoh wrote:
On Sun, Jul 30, 2017 at 01:15:50AM +0000, Stefan Koch via
Digitalmars-d wrote:
Hi Guys,
I just ran the first moderately complex ctfe function
successfully through newCTFE!
This is the code that works now :
module bf_parser;
[...]
pragma(msg, parseBf("[,....]")[3].token);
It has been a year of work to get to this point.
And it might seem a little trivial.
But this is actually the a showcase of methods, slices,
strings and structs interacting.
Very nice indeed!
Out of curiosity, how does the performance of newCTFE compare
to the current CTFE with the above code? E.g., if you pass in
a BF program of non-trivial length, what's the difference in
performance?
T
newCTFE currently is about 3.65x faster.
(for a medium sized bf-program of 6467 bytes
which is parsed into 3441 RepeatedTokens);
