Re: Memoization in compile-time
On Friday, 13 March 2015 at 13:51:43 UTC, Dennis Ritchie wrote: And you can somehow memoization stuff at compile time? If you want memoization at compile time I would suggest using the template version of Factorial as template instantiations are cached by the compiler. However, std.functional.memoize may work as well at compile time, I don't know.
Re: is eC alot like D?
On Wednesday, 11 March 2015 at 03:16:50 UTC, Taylor Hillegeist wrote: So I found http://ec-lang.org/ it seems alot like D, But it has a company backing it. It just seems interesting. heh they choose a lot worse name for a language than D :D. https://eclang.codeplex.com/wikipage?title=Intro%20to%20Ec
Re: Memoization in compile-time
On 03/13/2015 11:28 AM, Ali Çehreli wrote:
enum int[] factorials = memoizeFactorials(N);
Oops! That's generally a trap! The array better be 'static' because a
manifest constant like 'enum factorials' would be inserted everywhere it
is used. (Similar to a C macro.)
I've been scratching my head why the assertion below was failing.
It sould be 'static':
static int[] factorials = memoizeFactorials(N);
If it's an enum, the following assert will fail:
foreach (i; 0 .. N) {
assert(factorials.ptr + i == (factorials[i]));
}
Make it a 'static', it will pass because then there will be just one
factorials array.
Ali
Re: embedding Pyd in Windows program
On Friday, 13 March 2015 at 16:30:07 UTC, Matt wrote: Thank you, adding the subConfigurations section to dub.json seems to allow the program to compile and run successfully. However, to test the code, I first tried in the main program: --- py_eval!string(import sys\nprint(sys.path)); --- and it didn't print anything. No error, the program just ran and closed. Then, I removed that and added print(sys.path) to my site.py as you suggested (the sys module is already imported), and still the program runs without printing anything, so I'm a little confused by that. I'll probably try running some of the example code, see if I can figure it out, but if you can advise, that would be fantastic. Thank you for all the help so far, it's really been appreciated I've tried the hello.d example from GitHub, and get the error message: Error: template instance py_eval!string template 'py_eval' is not defined Unfortunately, that's all I seem to get. Is py_eval not deprecated in 0.9.4?
Re: Memoization in compile-time
On 03/13/2015 06:16 AM, weaselcat wrote: confusingly, D uses enum for named compile-time constants. http://ddili.org/ders/d.en/enum.html Actually, 'static' works as well. I have enumerated the cases here: http://ddili.org/ders/d.en/functions_more.html#ix_functions_more.CTFE quote For a function to be executed at compile time, it must appear in an expression that in fact is needed at compile time: - Initializing a static variable - Initializing an enum variable - Calculating the length of a fixed-length array - Calculating a template value argument /quote Are there other cases that is worth adding to that list? Ali
Re: is eC alot like D?
On 03/13/2015 10:18 AM, Dude wrote: On Wednesday, 11 March 2015 at 03:16:50 UTC, Taylor Hillegeist wrote: So I found http://ec-lang.org/ it seems alot like D, But it has a company backing it. It just seems interesting. heh they choose a lot worse name for a language than D :D. https://eclang.codeplex.com/wikipage?title=Intro%20to%20Ec Following Go's lead: :D http://en.wikipedia.org/wiki/Go!_%28programming_language%29 Ali
Re: Memoization in compile-time
On Friday, 13 March 2015 at 18:38:16 UTC, Ali Çehreli wrote:
On 03/13/2015 11:28 AM, Ali Çehreli wrote:
enum int[] factorials = memoizeFactorials(N);
Oops! That's generally a trap! The array better be 'static'
because a manifest constant like 'enum factorials' would be
inserted everywhere it is used. (Similar to a C macro.)
I've been scratching my head why the assertion below was
failing.
It sould be 'static':
static int[] factorials = memoizeFactorials(N);
If it's an enum, the following assert will fail:
foreach (i; 0 .. N) {
assert(factorials.ptr + i == (factorials[i]));
}
Make it a 'static', it will pass because then there will be
just one factorials array.
Ali
Thanks.
And you can make the same memoized variable was available at
compile time and at run time? :)
Re: Compilation changes
On 03/13/2015 05:26 AM, bearophile wrote: Now I think it doesn't work any more, giving errors like this: That may be related to the recent changes in the build system. Have you been following the following threads? (I haven't been; so, I am sure whether they apply.) [dmd-internals] DMD now requires a working D compiler to be build Proposal : aggregated dlang git repository dmd 2.066.1 cannot build phobos 2.066.1 Ali
Re: Memoization in compile-time
On 03/13/2015 06:51 AM, Dennis Ritchie wrote:
And you can somehow memoization stuff at compile time?
Scarily simple. :D
import std.stdio;
enum N = 15;
enum int[] factorials = memoizeFactorials(N);
int[] memoizeFactorials(int n)
{
if (!__ctfe) {
// Make sure that this function is never called at run time
assert(false);
}
int[] result = new int[n];
result[0] = 1;
foreach (i; 1 .. n) {
result[i] = result[i - 1] * i;
}
return result;
}
int fact(int n)
{
return factorials[n];
}
void main()
{
foreach (i; 0 .. N) {
writeln(fact(i));
}
}
Ali
Re: OutputDebugString()
On Friday, 13 March 2015 at 21:12:52 UTC, Robert M. Münch wrote: How do I declare such missing Windows API functions myself? In the file you want to use it, you can just write extern(Windows) void OutputDebugStringA(in char*); and it should work... or whatever the signature is, check msdn, and remember the ones that take strings tend to need A or W for the ascii or Unicode variants. dmd comes with a bunch of windows lib link files but not all of them. This one should work though. Just add the file, e.g. user32.lib, to the compile command line. You can also do pragma(lib, user32); and such in the main D file itself you are building and it will add automatically.
Re: Memoization in compile-time
On 03/13/2015 12:09 PM, Dennis Ritchie wrote: And you can make the same memoized variable was available at compile time and at run time? :) Yes, run-time is always possible and if it can be computed at compile time, compile-time is also possible. Ali
Re: Compilation changes
Ali Çehreli: That may be related to the recent changes in the build system. Right. Have you been following the following threads? (I haven't been; so, I am sure whether they apply.) I have not. [dmd-internals] DMD now requires a working D compiler to be build Proposal : aggregated dlang git repository dmd 2.066.1 cannot build phobos 2.066.1 I will try to get something out of those threads, thank you :-) Bye, bearophile
Re: OutputDebugString()
On Friday, 13 March 2015 at 21:12:52 UTC, Robert M. Münch wrote: Hi, I want to use the Windows OutputDebugString() which is not defined anywhere. How do I declare such missing Windows API functions myself? And with which libaries do I then need to link? Does DMD contain all necessary Windows link libs or am I supposed to get them myself? So, it's not clear what to do if a Windows API function is missing. And, how can I contribute new declared ones to get them integrated into the standard distribution? I'm pretty sure the correct way to do it is as follows: --- import core.sys.windows.windows; extern (Windows) OutputDebugString( LPCTSTR ); /// Add other externs here ---
Re: Memoization in compile-time
On Friday, 13 March 2015 at 21:16:24 UTC, Ali Çehreli wrote: On 03/13/2015 12:09 PM, Dennis Ritchie wrote: And you can make the same memoized variable was available at compile time and at run time? :) Yes, run-time is always possible and if it can be computed at compile time, compile-time is also possible. Thanks.
OutputDebugString()
Hi, I want to use the Windows OutputDebugString() which is not defined anywhere. How do I declare such missing Windows API functions myself? And with which libaries do I then need to link? Does DMD contain all necessary Windows link libs or am I supposed to get them myself? So, it's not clear what to do if a Windows API function is missing. And, how can I contribute new declared ones to get them integrated into the standard distribution? -- Robert M. Münch http://www.saphirion.com smarter | better | faster
Re: OutputDebugString()
On Friday, 13 March 2015 at 21:12:52 UTC, Robert M. Münch wrote: Hi, I want to use the Windows OutputDebugString() which is not defined anywhere. The declaration can be already part of WindowsAPI project: https://github.com/AndrejMitrovic/WindowsAPI just see it there. How do I declare such missing Windows API functions myself? And with which libaries do I then need to link? This function has rather unique implementation too. Are you sure, you don't want to see its MSDN entry?
Re: embedding Pyd in Windows program
On Friday, 13 March 2015 at 19:05:59 UTC, Matt wrote: example code, see if I can figure it out, but if you can advise, that would be fantastic. Thank you for all the help so far, it's really been appreciated My penitence for not putting this information on readthedocs. I've tried the hello.d example from GitHub, and get the error message: Error: template instance py_eval!string template 'py_eval' is not defined Unfortunately, that's all I seem to get. Is py_eval not deprecated in 0.9.4? hmm. that's odd. are you trying to build with dub? the only examples that build with dub are in examples/dub. the rest use distutils.
Re: how to pass a ubyte[] to c interface?
On 13/03/2015 7:47 p.m., zhmt wrote:
On Friday, 13 March 2015 at 06:39:31 UTC, Rikki Cattermole wrote:
On 13/03/2015 7:35 p.m., zhmt wrote:
ubyte[] arr ;
I pass the arr.ptr to c program, it fails silently.
Is there any way to cast a ubyte[] to a clang pointer?
Theoretically this should work.
D:
void func(ubyte[] value) {
func(value.length, value.ptr);
}
extern(C) void func(size_t length, ubyte*);
C:
void func(size_t length, ubyte*) {
// ...
}
Thank you for your confirm,I will try again.
Just be careful about GC pointers. They could be free'd and the c
library thinks it is still live.
Either keep a copy of the pointer alive on the heap somewhere or tell
the GC to ignore it.
Re: how to pass a ubyte[] to c interface?
On Friday, 13 March 2015 at 06:39:31 UTC, Rikki Cattermole wrote:
On 13/03/2015 7:35 p.m., zhmt wrote:
ubyte[] arr ;
I pass the arr.ptr to c program, it fails silently.
Is there any way to cast a ubyte[] to a clang pointer?
Theoretically this should work.
D:
void func(ubyte[] value) {
func(value.length, value.ptr);
}
extern(C) void func(size_t length, ubyte*);
C:
void func(size_t length, ubyte*) {
// ...
}
Thank you for your confirm,I will try again.
how to pass a ubyte[] to c interface?
ubyte[] arr ; I pass the arr.ptr to c program, it fails silently. Is there any way to cast a ubyte[] to a clang pointer?
Re: how to pass a ubyte[] to c interface?
On 13/03/2015 7:35 p.m., zhmt wrote:
ubyte[] arr ;
I pass the arr.ptr to c program, it fails silently.
Is there any way to cast a ubyte[] to a clang pointer?
Theoretically this should work.
D:
void func(ubyte[] value) {
func(value.length, value.ptr);
}
extern(C) void func(size_t length, ubyte*);
C:
void func(size_t length, ubyte*) {
// ...
}
Re: how to pass a ubyte[] to c interface?
I have some source code of vibe.d, it does this in the same way,
and it works .
void read(ubyte[] dst)
{
checkConnected(false);
acquireReader();
scope(exit) releaseReader();
while (dst.length 0) {
checkConnected(false);
logTrace(evbuffer_read %d bytes (fd %d), dst.length,
m_ctx.socketfd);
auto nbytes = bufferevent_read(m_ctx.event, dst.ptr,
dst.length);
logTrace( .. got %d bytes, nbytes);
dst = dst[nbytes .. $];
if( dst.length == 0 ) break;
checkConnected(false);
m_ctx.core.yieldForEvent();
}
logTrace(read data);
}
Re: how to pass a ubyte[] to c interface?
On 03/12/2015 11:35 PM, zhmt wrote: ubyte[] arr ; I pass the arr.ptr to c program Unless there is sentinel value at the end of the array, you must also pass the number of elements (as Rikki Cattermole has shown). However, if the C function holds on to that pointer for later use, you must also keep the array alive. You can ensure that by two general ways: a) The D-side slice must not be a local slice. For example, it can be a member of a long-living object or it is a module-scope variable. b) Tell the garbage collector that there is indeed a reference to that memory block somewhere else (even though there is no D-side reference to it). You can do this with GC.addRoot. (The opposite function is GC.removeRoot.) , it fails silently. That's not good. Is there any way to cast a ubyte[] to a clang pointer? Going off-topic, I think you mean the C language when you say clang, which may be confusing because there is also the compiler called clang. :) Ali
Re: how to pass a ubyte[] to c interface?
On Friday, 13 March 2015 at 06:56:33 UTC, Ali Çehreli wrote:
On 03/12/2015 11:35 PM, zhmt wrote:
ubyte[] arr ;
I pass the arr.ptr to c program
Unless there is sentinel value at the end of the array, you
must also pass the number of elements (as Rikki Cattermole has
shown).
However, if the C function holds on to that pointer for later
use, you must also keep the array alive. You can ensure that by
two general ways:
a) The D-side slice must not be a local slice. For example, it
can be a member of a long-living object or it is a module-scope
variable.
b) Tell the garbage collector that there is indeed a reference
to that memory block somewhere else (even though there is no
D-side reference to it). You can do this with GC.addRoot. (The
opposite function is GC.removeRoot.)
, it fails silently.
That's not good.
Is there any way to cast a ubyte[] to a clang pointer?
Going off-topic, I think you mean the C language when you say
clang, which may be confusing because there is also the
compiler called clang. :)
Ali
sorry for clang, I mean c , not clang of mac.
Thanks for you advice, I use coroutine, so the array will hold in
stack, will not be gc before return.
And I successfully passed the arr.ptr to c program. I found the
root of problem:
The order of method declaration in c++ class and d interface must
be exactly the same:
this is d code:
interface CConn
{
void free();
int connect(ubyte* ip,int port,CFiberCtx fiberctx);
int read(ubyte* buf,int offset, int len,CFiberCtx fiberctx);
int readSome(ubyte* buf,int offset, int len,CFiberCtx fiberctx);
int write(ubyte* buf,int offset, int len,CFiberCtx fiberctx);
}
this is c++ class:
class CConn
{
public:
// dlang interfaces;
virtual void free();
virtual int connect(char* ip,int port,CFiberCtx *fiberctx);
virtual int read(char* buf,int offset, int len,CFiberCtx
*fiberctx);
virtual int write(char* buf,int offset, int len,CFiberCtx
*fiberctx); //ERROR : swap this line with below, everything will
be ok
virtual int readSome(char* buf,int offset, int len,CFiberCtx
*fiberctx);
}
After swapping the last two lines, everything goes well.
The tutorial of d dont mention this.
Thanks for all replies.
Re: embedding Pyd in Windows program
On Friday, 13 March 2015 at 01:40:34 UTC, Ellery Newcomer wrote: On 03/11/2015 07:59 PM, Matt wrote: Right, copying site.py into my program's working dir sorts out the missing module error, but I now get a syntax error: file=sys.stderr) ^ SyntaxError: invalid syntax Error executing command run: Program exited with code 1 I googled this, and apparently it's caused by an older interpreter. The only files I have transferred over are: python3.DLL python3.LIB python34.LIB Does anyone have any idea what I might be doing wrong, or how I can fix this? I'm guessing your python path is screwed up. In your embedded python, import sys print (sys.path) make sure everything is pointing where it should. Failing that, are you using the correct dub configuration? without checking, I believe it defaults to python 2.7. Failing that, you transferred python3.dll from somewhere to somewhere? I believe pyd links to \windows\system32\python3.dll or some such. At least, I think that's where those lib files point to. not sure. If that is an issue and you want to use your own lib files, you will need to generate OMF files from them. I used the ~0.9.4 branch in dub, and I'm not sure how to change configuration. Do you mean I should be using ~master or ~develop? as for the transferring the python3.dll, I downloaded and installed a fresh copy of the latest Python build, version 3.4, and copied the libraries from there. Even so, I had to add PYD_PACKAGE_DIR manually to get it to work. I'm not sure I'm going to be able to run the python snippet you sent, since my program, which currently opens a window for three seconds before closing again, doesn't even get that far, and chokes when I run py_init().
How to reference another dub project?
I have two dub projects, one is library, the other one is app. The library is referenced by the app. How to reference it in dub.json? I found the subPackage, It does not meet my needs exactly.
Re: How to reference another dub project?
On Friday, 13 March 2015 at 08:24:11 UTC, zhmt wrote:
I have two dub projects, one is library, the other one is app.
The library is referenced by the app. How to reference it in
dub.json?
I found the subPackage, It does not meet my needs exactly.
I got it:
dependencies: {
gamelibd:{path:../gamelibd/}
},
Re: Memoization in compile-time
On Friday, 13 March 2015 at 02:38:18 UTC, Rikki Cattermole wrote:
You could assign it to e.g. an enum. Or force it over using
meta-programming.
And this code can be rewritten to D?
template int n
struct Factorial
{
enum { value = n * Factorialn - 1::value };
};
template
struct Factorial0
{
enum { value = 1 };
};
int main()
{
constexpr auto x = Factorial5::value;
constexpr auto y = Factorial7::value;
}
Compilation changes
Time ago Dmitry Olshansky gave me a nice script to compile dmd+Phobos from github, starting like this: My recipe on Windows, by Dmitry Olshansky: 0. Make sure there are no other DMD in the Windows path. 1. Get the latest DMD release zip you can find, unzip to some drive root(! otherwise get's trickier) 2. Delete all within src subfolder 3. Get a Git console in \dmd2\src, then the usual: ... Now I think it doesn't work any more, giving errors like this: make -fwin32.mak C=backend TK=tk ROOT=root HOST_DC= reldmd make -fwin32.mak C=backend TK=tk ROOT=root HOST_DC= OPT=-o DEBUG= LFLAGS=-L/delexe/la dmd.exe run idgen Error: 'run' not found --- errorlevel 1 --- errorlevel 1 So is Dmitry (or someone else) willing and able to tell me how to fix my compilation script and what to do? Thank you, bye, bearophile
Re: is eC alot like D?
On Wednesday, 11 March 2015 at 03:16:50 UTC, Taylor Hillegeist wrote: So I found http://ec-lang.org/ it seems alot like D, But it has a company backing it. It just seems interesting. Seems to me that structs are not the same as in D, and structs in D are very powerful. I don't like the fact that they mention object orientation right from the start, as if it was the be all end all of programming - which it isn't. Hm. I have to admit that I'm biased towards D, but D has come a long way and is the result of serious input from loads of users and experts. Any new language has a lot of catching up to do. In a way it's funny that every new language builds up an infrastructure first, toolchains etc. Like someone who's trying to learn how to play guitar buys expensive amps and flashy effect devices before s/he can even play a single chord. However, language features aside, how is D doing in the mobile sector? Most new languages boast Android/iOS support. Even if the language is inferior to D, people will opt for any language that can be ported to mobile platforms, regardless of its features.
Re: Memoization in compile-time
On Friday, 13 March 2015 at 12:49:48 UTC, Dennis Ritchie wrote:
On Friday, 13 March 2015 at 02:38:18 UTC, Rikki Cattermole
wrote:
You could assign it to e.g. an enum. Or force it over using
meta-programming.
And this code can be rewritten to D?
template int n
struct Factorial
{
enum { value = n * Factorialn - 1::value };
};
template
struct Factorial0
{
enum { value = 1 };
};
int main()
{
constexpr auto x = Factorial5::value;
constexpr auto y = Factorial7::value;
}
You can translate it directly:
template Factorial(int n)
{
static if (n == 0)
{
enum Factorial = 1;
}
else static if (n 0)
{
enum Factorial = n * Factorial!(n - 1);
}
else
{
static assert(false, n cannot be negative);
}
}
int main()
{
//Calculated at compile time
auto x = Factorial!5;
auto y = Factorial!7;
}
However, it's much easier to just write a function and decide
whether you want to calculate its value at runtime or compile
time.
int factorial(int n)
in
{
assert(n = 0, n cannot be 0);
}
body
{
return n == 0 ? 1 : n * factorial(n - 1);
}
void main()
{
//Evaluated at compile time
enum x = factorial(5);
//Evaluated at runtime
auto y = factorial(7);
}
Re: Memoization in compile-time
On Friday, 13 March 2015 at 12:58:13 UTC, Meta wrote:
On Friday, 13 March 2015 at 12:49:48 UTC, Dennis Ritchie wrote:
On Friday, 13 March 2015 at 02:38:18 UTC, Rikki Cattermole
wrote:
You could assign it to e.g. an enum. Or force it over using
meta-programming.
And this code can be rewritten to D?
template int n
struct Factorial
{
enum { value = n * Factorialn - 1::value };
};
template
struct Factorial0
{
enum { value = 1 };
};
int main()
{
constexpr auto x = Factorial5::value;
constexpr auto y = Factorial7::value;
}
You can translate it directly:
template Factorial(int n)
{
static if (n == 0)
{
enum Factorial = 1;
}
else static if (n 0)
{
enum Factorial = n * Factorial!(n - 1);
}
else
{
static assert(false, n cannot be negative);
}
}
int main()
{
//Calculated at compile time
auto x = Factorial!5;
auto y = Factorial!7;
}
However, it's much easier to just write a function and decide
whether you want to calculate its value at runtime or compile
time.
int factorial(int n)
in
{
assert(n = 0, n cannot be 0);
}
body
{
return n == 0 ? 1 : n * factorial(n - 1);
}
void main()
{
//Evaluated at compile time
enum x = factorial(5);
//Evaluated at runtime
auto y = factorial(7);
}
Thanks.
Re: Memoization in compile-time
On Friday, 13 March 2015 at 13:16:27 UTC, weaselcat wrote:
On Friday, 13 March 2015 at 12:49:48 UTC, Dennis Ritchie wrote:
On Friday, 13 March 2015 at 02:38:18 UTC, Rikki Cattermole
wrote:
You could assign it to e.g. an enum. Or force it over using
meta-programming.
And this code can be rewritten to D?
template int n
struct Factorial
{
enum { value = n * Factorialn - 1::value };
};
template
struct Factorial0
{
enum { value = 1 };
};
int main()
{
constexpr auto x = Factorial5::value;
constexpr auto y = Factorial7::value;
}
confusingly, D uses enum for named compile-time constants.
http://ddili.org/ders/d.en/enum.html
If you take Rikki's example and apply it to an enum(i.e,
enum x = factorial(5);
)
the program will fail to compile if it can't be computed at
compile-time.
woops, walked away to get coffee before I submitted and got
beaten to the punch by a better answer : )
Re: Memoization in compile-time
On Friday, 13 March 2015 at 12:49:48 UTC, Dennis Ritchie wrote:
On Friday, 13 March 2015 at 02:38:18 UTC, Rikki Cattermole
wrote:
You could assign it to e.g. an enum. Or force it over using
meta-programming.
And this code can be rewritten to D?
template int n
struct Factorial
{
enum { value = n * Factorialn - 1::value };
};
template
struct Factorial0
{
enum { value = 1 };
};
int main()
{
constexpr auto x = Factorial5::value;
constexpr auto y = Factorial7::value;
}
confusingly, D uses enum for named compile-time constants.
http://ddili.org/ders/d.en/enum.html
If you take Rikki's example and apply it to an enum(i.e,
enum x = factorial(5);
)
the program will fail to compile if it can't be computed at
compile-time.
Re: Memoization in compile-time
And you can somehow memoization stuff at compile time?
Re: is eC alot like D?
On Fri, 13 Mar 2015 04:02:07 +, Jerome St-Louis wrote: To confirm guys eC does support C preprocessing. The C preprocessor is invoked before the eC compiler does its parsing. ah, i see. eC tries to be a superset of C as much possible, with only a few keyword clashes like 'class' as an exception. so they choose to throw away a possibility to include good type system into the language... ahem... looks bad for me. nice try, nevertheless. signature.asc Description: PGP signature
Re: embedding Pyd in Windows program
On Friday, 13 March 2015 at 09:38:45 UTC, Matt wrote:
I used the ~0.9.4 branch in dub, and I'm not sure how to
change configuration. Do you mean I should be using ~master
or ~develop?
nope. configurations are a different thing.
you can set them in your project's dub.json. example:
https://github.com/ariovistus/pyd/blob/master/examples/dub/simple_embedded/dub.json
for your project, I think you'd want
subConfigurations: {
pyd: python34,
}
as for the transferring the python3.dll, I downloaded and
installed a fresh copy of the latest Python build, version 3.4,
and copied the libraries from there. Even so, I had to add
PYD_PACKAGE_DIR manually to get it to work.
PYD_PACKAGE_DIR is necessary for linking the OMF files found in
infrastructure/windows. hmm. why did it not Just Work?
I'm not sure I'm going to be able to run the python snippet you
sent, since my program, which currently opens a window for
three seconds before closing again, doesn't even get that far,
and chokes when I run py_init().
you can put it in the top of site.py
Re: embedding Pyd in Windows program
On Friday, 13 March 2015 at 14:21:15 UTC, Ellery Newcomer wrote:
On Friday, 13 March 2015 at 09:38:45 UTC, Matt wrote:
I used the ~0.9.4 branch in dub, and I'm not sure how to
change configuration. Do you mean I should be using
~master or ~develop?
nope. configurations are a different thing.
you can set them in your project's dub.json. example:
https://github.com/ariovistus/pyd/blob/master/examples/dub/simple_embedded/dub.json
for your project, I think you'd want
subConfigurations: {
pyd: python34,
}
as for the transferring the python3.dll, I downloaded and
installed a fresh copy of the latest Python build, version
3.4, and copied the libraries from there. Even so, I had to
add PYD_PACKAGE_DIR manually to get it to work.
PYD_PACKAGE_DIR is necessary for linking the OMF files found in
infrastructure/windows. hmm. why did it not Just Work?
I'm not sure I'm going to be able to run the python snippet
you sent, since my program, which currently opens a window for
three seconds before closing again, doesn't even get that far,
and chokes when I run py_init().
you can put it in the top of site.py
Thank you, adding the subConfigurations section to dub.json
seems to allow the program to compile and run successfully.
However, to test the code, I first tried in the main program:
---
py_eval!string(import sys\nprint(sys.path));
---
and it didn't print anything. No error, the program just ran and
closed.
Then, I removed that and added print(sys.path) to my site.py as
you suggested (the sys module is already imported), and still the
program runs without printing anything, so I'm a little confused
by that. I'll probably try running some of the example code, see
if I can figure it out, but if you can advise, that would be
fantastic. Thank you for all the help so far, it's really been
appreciated
