Ah, I just realized I've been using gcc 4.3.0 for a long time! I thought I was using my trusted 3.4.x version!
When I switch to the 3.4 version I get tuple errors too. I'm pretty useless with templates, so I'm not sure how to fix that...
--
Mats
Ah, I just realized I've been using gcc 4.3.0 for a long time! I thought I was using my trusted 3.4.x version!
When I switch to the 3.4 version I get tuple errors too. I'm pretty useless with templates, so I'm not sure how to fix that...
--
Mats
Compilers can produce warnings - make the compiler programmers happy: Use them!
Please don't PM me for help - and no, I don't do help over instant messengers.
Really, 4.3.0? Man, I guess it's time for an update.
Code:#include <cmath> #include <complex> bool euler_flip(bool value) { return std::pow ( std::complex<float>(std::exp(1.0)), std::complex<float>(0, 1) * std::complex<float>(std::atan(1.0) *(1 << (value + 2))) ).real() < 0; }
I'm on "4.5-Something". ^_^
Mats: I got the email regarding the issue (#2). I've suggested before that you close them both.
Sebastian: Try changing the expansion--the crappy implementation macro--to specify the target template type of the nested template for every use. (The example explains the suggested modifications.)
Soma
Code:template <> template <> \ inline const tuple<2,type>& tuple<2,type>::operator = (const tuple<2,otype>& v)Code:template <> template <> \ inline const tuple<2,type>& tuple<2,type>::operator= <otype> (const tuple<2,otype>& v)
No go. And unfortunately, mingw's latest stable version of g++ is 3.4.5, and it won't compile it, either. I'll see if I can figure out a workaround.
Code:#include <cmath> #include <complex> bool euler_flip(bool value) { return std::pow ( std::complex<float>(std::exp(1.0)), std::complex<float>(0, 1) * std::complex<float>(std::atan(1.0) *(1 << (value + 2))) ).real() < 0; }
Well, it finally compiled after commenting out these lines:
...although I have no idea if that might introduce any bugs. Also added a copy of Borlands touch.exe, since mingw doesn't come with one. And of course the configure and makefile changes.Code://UINT32_TUPLE_SPECS(int16_t,uint16_t) //UINT32_TUPLE_SPECS(uint16_t,int16_t)
Code:#include <cmath> #include <complex> bool euler_flip(bool value) { return std::pow ( std::complex<float>(std::exp(1.0)), std::complex<float>(0, 1) * std::complex<float>(std::atan(1.0) *(1 << (value + 2))) ).real() < 0; }
I'd recommend getting the GnuWin32 package, instead of using a 16-bit touch.
But we could probably remove the need for reference to ranlib altogether.
--
Mats
Compilers can produce warnings - make the compiler programmers happy: Use them!
Please don't PM me for help - and no, I don't do help over instant messengers.
I have fixed several minor issues
- Saving to text-file (.BAS) (caused by calling fstream::write() with a zero length which set the "fail" bit).
- Compile errors with older compiler.
- Makefile problems.
I still have a problem LOADING .BAS files.
--
Mats
Compilers can produce warnings - make the compiler programmers happy: Use them!
Please don't PM me for help - and no, I don't do help over instant messengers.
I have made a bit of progress with the loading .BAS files, but not quite there yet. Now it simply needs to stop when the file is consumed...
--
Mats
Compilers can produce warnings - make the compiler programmers happy: Use them!
Please don't PM me for help - and no, I don't do help over instant messengers.
I just pushed another bit of updates.
However, I think I've broken the "save" functionality, so I need to unbreak that. Nothing obvious shows up in the past updates, so I'm not sure what it is. Still works to save with "tsave".
--
Mats
Compilers can produce warnings - make the compiler programmers happy: Use them!
Please don't PM me for help - and no, I don't do help over instant messengers.
I just started on "string variables". It's nowhere near complete.
Also done a whole lot more code cleanup.
--
Mats
Compilers can produce warnings - make the compiler programmers happy: Use them!
Please don't PM me for help - and no, I don't do help over instant messengers.
I've been pretty busy lately, so I'm afraid I'm going to have to retire from the project (for now, at least). Is everything working well at the moment?
Code:#include <cmath> #include <complex> bool euler_flip(bool value) { return std::pow ( std::complex<float>(std::exp(1.0)), std::complex<float>(0, 1) * std::complex<float>(std::atan(1.0) *(1 << (value + 2))) ).real() < 0; }
As far as I know, yes.Originally Posted by Sebastiani
--
Mats
Compilers can produce warnings - make the compiler programmers happy: Use them!
Please don't PM me for help - and no, I don't do help over instant messengers.
Just spent some more time optimizing (more fun than adding new features!)
Now I have a file that is 192KB and it takes between 50% to 70% of the time it took 7 changesets ago to run the BM2 benchmarks.
--
Mats
Compilers can produce warnings - make the compiler programmers happy: Use them!
Please don't PM me for help - and no, I don't do help over instant messengers.
Mats, I tried pulling the most recent snapshot, but it seems to be missing the include file mbstate.h?
Code:#include <cmath> #include <complex> bool euler_flip(bool value) { return std::pow ( std::complex<float>(std::exp(1.0)), std::complex<float>(0, 1) * std::complex<float>(std::atan(1.0) *(1 << (value + 2))) ).real() < 0; }
Yes, I did notice that shortly after I submittied, but I had a problem uploading it (my wireless network sometimes decides that "Hey, I'm on a break right now, I'm not passing any packets to the rest of the world. Come back later"). As it was way too late in the evening anyways, I decided that was a good sign for "go to bed".
I've posted it now.
--
Mats
Compilers can produce warnings - make the compiler programmers happy: Use them!
Please don't PM me for help - and no, I don't do help over instant messengers.
