equinox
|
|
Posted on: May 30, 2013, 10:44:44 am |
|
|
Joined: May 2013
Posts: 1
|
Hi, how can I fix ? tnk1000
|
|
|
Logged
|
|
|
|
Josh @ Dreamland
|
|
Reply #1 Posted on: May 30, 2013, 11:19:43 am |
|
|
Prince of all Goldfish
Location: Pittsburgh, PA, USA Joined: Feb 2008
Posts: 2950
|
It's possible you're trying to run a 32-bit ENIGMA from a 64-bit JVM. Do you have a 32-bit Java installed?
|
|
|
Logged
|
"That is the single most cryptic piece of code I have ever seen." -Master PobbleWobble "I disapprove of what you say, but I will defend to the death your right to say it." -Evelyn Beatrice Hall, Friends of Voltaire
|
|
|
|
|
polygone
|
|
Reply #4 Posted on: May 30, 2013, 03:25:41 pm |
|
|
Location: England Joined: Mar 2009
Posts: 794
|
malkierian were you originally using 64 bit Java though? You might need to fully un-install it if you were.
|
|
|
Logged
|
I honestly don't know wtf I'm talking about but hopefully I can muddle my way through.
|
|
|
|
polygone
|
|
Reply #6 Posted on: May 30, 2013, 03:34:17 pm |
|
|
Location: England Joined: Mar 2009
Posts: 794
|
My system defaults to 64-bit, yes. I'm on Windows 7 after all. Why does it matter? Nothing else I've ever run has specified 32-bit OR 64-bit.
EDIT: Tried the older installer, and it actually compiled stuff, but I was still getting errors in compile, and the compileEGMf one too.
Errors are probably normal. But what errors were they? The fact that it compiles is making me question Robert's stuff now though.
|
|
|
Logged
|
I honestly don't know wtf I'm talking about but hopefully I can muddle my way through.
|
|
|
malkierian
|
|
Reply #7 Posted on: May 30, 2013, 03:43:08 pm |
|
|
Joined: May 2013
Posts: 18
|
Here's the log for a clean start: http://pastebin.com/3hnUDjCFThis time, after extracting and before running ENIGMA.exe for the first time, I changed "run" to use the 32-bit java from the start. Still got the compile errors, still got the compileEGMf error at the end. But yes, at least it's compiling now, unlike with the latest installer which starts Lateral and goes straight to the compileEGMf error.
|
|
|
Logged
|
|
|
|
polygone
|
|
Reply #8 Posted on: May 30, 2013, 03:45:17 pm |
|
|
Location: England Joined: Mar 2009
Posts: 794
|
Oh no, that's the same problem. It was just compiling the objects that come already pre-compiled in the newer Zip.
It's still most likely the java then. I can't wait until NaturalGM is up and running with ENIGMA :P
|
|
« Last Edit: May 30, 2013, 03:47:56 pm by polygone »
|
Logged
|
I honestly don't know wtf I'm talking about but hopefully I can muddle my way through.
|
|
|
|
polygone
|
|
Reply #10 Posted on: May 30, 2013, 04:06:40 pm |
|
|
Location: England Joined: Mar 2009
Posts: 794
|
Josh has apparently seen the problem, to do with MinGW updating a type and making it incompatible. He doesn't seem happy about it :p But then he's rarely happy when MinGW update something and screw him over.
|
|
|
Logged
|
I honestly don't know wtf I'm talking about but hopefully I can muddle my way through.
|
|
|
|
Josh @ Dreamland
|
|
Reply #12 Posted on: May 30, 2013, 04:25:11 pm |
|
|
Prince of all Goldfish
Location: Pittsburgh, PA, USA Joined: Feb 2008
Posts: 2950
|
Relatively soon. I'm deciding what to do. The assholes used [snip]long long unsigned int[/snip] as their size_type, which is NOT valid in the previous ISO. I get warnings for an attempt to use long long in my own code. Still, I suppose our only option is to overload it now that the alternative is an error. The intelligent thing to do would be to overload it for size_type, but that would conflict with the other overloads on systems wherein it was just long. Their stupid change puts me in an awkward position.
EDIT: Good news; this was an ISO decision, not a MinGW decision. The type [snip]long long int[/snip] is now guaranteed to exist, and be no less than 64 bits and at least as large as a regular [snip]long int[/snip].
I've asked polygone to replace the [snip]USE_LONG_LONG[/snip] macro in the compiler with [snip]__cplusplus >= 201100[/snip], which should let us filter out compilers that support the new standard a little better.
How did you end up with a compiler that uses the new standard by default?
|
|
« Last Edit: May 30, 2013, 04:56:05 pm by Josh @ Dreamland »
|
Logged
|
"That is the single most cryptic piece of code I have ever seen." -Master PobbleWobble "I disapprove of what you say, but I will defend to the death your right to say it." -Evelyn Beatrice Hall, Friends of Voltaire
|
|
|
polygone
|
|
Reply #13 Posted on: May 30, 2013, 04:53:58 pm |
|
|
Location: England Joined: Mar 2009
Posts: 794
|
Ok I've committed update. Hopefully it will work.
|
|
|
Logged
|
I honestly don't know wtf I'm talking about but hopefully I can muddle my way through.
|
|
|
malkierian
|
|
Reply #14 Posted on: May 30, 2013, 05:10:25 pm |
|
|
Joined: May 2013
Posts: 18
|
How did you end up with a compiler that uses the new standard by default?
Not exactly sure, except that maybe it hails back to when I was trying to get C11 working for some feature therein I can't even remember what it was. But both my 32-bit and 64-bit installations were installed to be C11-compatible, so they may have defaulted to it. I'll download the latest stuff and try it again. Do I need to compile from source, or will one of the previous installers work? (Or maybe the installer's been updated?)
|
|
|
Logged
|
|
|
|
|