This notebook is the same idea as Links to Example Code. 

D Programming Language tips

 (Adam D. Ruppe)

The Two-Towers Problem

The ability to create escape strings is a waste of the backslash character.

benchmark small sums among multiprecision integers (MPI)

Profiling with Digital Mars D Compiler on Windows 

Batch files for running multiple DMD versions

request a url

 htmlget.d

readln()

can D check, whether a class A an Interface B supports?

One nice thing that opApply capable of is it can avoid heap activity by stack-allocating data during iteration.

line fflush (stdout); must be written before getch()

I'm gearing up for changing std.algorithm, and I am thinking of making
the major change of favoring lazy evaluation throughout.

efficient sorting routine

Re: .bat file to help compile easier - dmd/build

Re: Operator overloading -- lets collect some use cases

Composite Oriented Programming

template AttributeGetSet

Using DMD's -v to build dependencies

longArithmetic

 benchmark

toString ugliness

A lightweight module for class extensions in D

Re: Is __traits(derivedMembers) supposed to return mangled names?

Re: Scope storage class

 stack allocated vs. heap allocated

std.algorithm

benchmarks

 matrix mul

D: A Newbie-Oriented Tutorial

import std.thread;

Forward declarations of template specilizations.

private import

make sure that ESP points to first
parameter on the stack

 asm

how asm blocks mixed with normal D code

Roles of variables in novice-level procedural programming

backtrace handler

Re: RegExp

Re: Templated interface

Re: Templated interface

Re: random number within a specified range?

Actually, when it comes to string processing, D is decidedly *not* a
"performance language".

random k-sample of a file

"foo(int, long)" is not supported

 oops, this is my mistake. it's works with mysterious way.

Re: Overloading templates

template Const(T)

ArrayBuilder is quite useful

#line decoder

optimized version of array operations, using SSE2 instructions

Revised CPUID/cache info

UTF-8 issues

I know this has been solved but I'd like to point out that windows
offers GetPrivateProfileString.  Its not portable and not that efficient
(although that would only be a concern if your reading 10000's of entries).

http://msdn.microsoft.com/en-us/library/ms724353(VS.85).aspx

Re: Windows PSAPI

 Using coffimplib with a psapi.lib taken from the M$ platform SDK will 

probably work.  The stdcall mangling is present in the .lib file, but 
not the .dll.  I didn't test the resulting OMF lib file, but the names 
are correctly mangled, "_GetModuleFileNameExA@16" and so on.

http://ftp.digitalmars.com/coffimplib.zip

If there's any constructor defined for S, then S(args) is a constructor
call.

If there's any opCall defined for S, then S(args) is an opCall call.

Otherwise, it's a struct literal.

To construct an object of struct S:

auto s = S(arguments);

Palindromes

is it expected that this does an infinite loop?

recent CPUs' branch predictions really are as good as people claim

yield, C# etc

random getstate/setstate

The SSE2 is being used

 Array Operators

Recursive Templates

First benchmark (DMD 1.034 and 2.018), just D against itself, not used GCC yet, the results show that vector ops are generally slower

xflatten, flatten in D
In the last versions of my D language libs I have re-written maybe for the 4-th time the function and the iterator to flatten a given iterable in an eager/lazy way, and this time I think the code may be as flexible and fast as it gets.

In D you can also solve the same problem without gotos, shortening the code, using a named 'continue' statement (but the code with goto is faster)

Travelling salesman problem

Re: random number generator

http://dsource.org/projects/meta

Alpha release of meta, a utility library for compile time functions and template metaprogramming

template isFloat(T)

I have updated the solutions to the Google problems (improved puzzle 6):
http://www.fantascienza.net/leonardo/js/interview_puzzles.zip

multisort function

std.getopt

module googlehunt;

Python dictionary inspired hash table implementation

third-party win32 bindings

I have then implemented the same algoritm in D language, it needs about 7.7 seconds to run.

getopt() sees only one short option in bundled options - the one that is given to getopt() first.

X[]==Y[] is 7X slower than it should be

Re: Functions with package protection

Re: chmod and chown in D

float f2;
f2 !<>= f2; // true

enum : bool

test the allocation speed

application that maps numbers to words from a list of words

TooltipLabelListener

uses DWT

import d.func: map, range;
import d.time: clock;
import std.traits: ReturnType;
import d.extra: NewVoidCArray, delVoidC, NewVoidGCArray;

mixin creating an enum

import std.algorithm;

struct A
{
    int Merge(int) { return 0; }
}

void main(string[] args)
{
    A a;
    auto fun = &a.Merge;
    map!(fun)([1, 2, 3]);
}

> Is it possible to map! or reduce! a member function or delegate against a
> range?

test if a parameter is a struct

GDC does inline ref-arg functions.

Static libraries are just collections of *optional* additional
object files.

Building Libraries

isDerivedFromA

newCtfeArray

This should extend to any semi-reserved name.  The compiler used to allow
redefinition of 'sizeof' and 'offsetof' but checks were added to disallow
that.  That mechanism should just be extended.

 semi-reserved names, such as init

import std.stdio, std.stream, std.string, std.ctype, std.gc;

Re: U++ Core vs D programming language

import std.file;
import std.stdio;
import std.date;

hyperlink regular expression pattern (std.regexp)

interface IFruit

Re: This needs a different approach

function table:

auto functionArray = [APPLE:&groupModule.appleModule.eat,
PEAR:&groupModule.pearModule.eat, .., PLUM:&groupModule.plumModule.eat];

functionArray[data.type]();

import std.stdio;
import std.string;

int main( string[] args) {
    char[] a = cast(char[])args[0].toupper();    
    a[2..5] = "XXX";
    a = cast(char[])tolower( cast(string)a );
    writefln(a);
    return 0;
}

John Conway's Game of Life in D v1.0

 See also: Web-News v.1.6.3

What is nothrow for?

A few lines of code to clean the trace.log produced by DMD

Are private functions that aren't declared final in the vtable?

 Apparently not.

import std.string;
import std.algorithm;

void main(char[][] args)
{
    string[] words = ["c", "ab", "bb"];
    sort!("a < b", SwapStrategy.stable)(words);
    assert(words == ["ab", "bb", "c"]);
}

 (with suggested correction)

The template with non-tuple parameter is preferred if that parameter is a value (Case 1). The template with tuple parameter is preferred if the non-tuple parameter is a type (Case 2).

Completely at compile time, it parses a text file describing a mapping between an 8-bit encoding and Unicode, and builds data structures and functions to convert between that encoding and Unicode.

Function template that takes as types one or more exception types, and
as argument a (lazy) expression. It returns true if once called the
expression throws one of the specified exception(s).

I'm still flabbergasted by the cool things you can do inside
fixed-length matrix/vector templates.

For instance (this is all from my matrix/vector classes in OpenMesh/D
which are basically a complete rewrite of the original C++ versions)

struct MatrixT(T, int M, int N)
{
     alias T Scalar;

     union {
         Scalar[M*N] values_ = void;
         static if(M<=10 && N<=10) {
             mixin(_gen_elements!(M,N)("Scalar"));
         }
     }
}

// Generate the elements m00,m01,m02... elements
// For matrices with dimensions < 10 x 10 only.
private string _gen_elements(int M, int N)(string type) {
     static assert(M<=10);
     static assert(N<=10);
     char[] Num = "0123456789";
     string ret;
     for(int col=0; col<N; ++col) {
         ret ~= type ~ " ";
         for (int row=0; row<M; row++) {
             ret ~= "m" ~ Num[row] ~ Num[col];
             if (row!=M-1) ret ~=  ", ";
         }
         ret ~= ";\n";
     }
     return ret;
}

This creates aliases to each element of the matrix of the form m00, m01,
m02, but only if the matrix is small enough.  I have no idea how you'd
do that in C++.  I don't think it's really possible.


Here's another -- generating the unrolled code for multiplying two
matrices together:

/// Multiply MxN matrix time NxP matrix
void mat_mult_ret(T,int M,int N,int P)(
     /*const*/ref MatrixT!(T,M,N) A,
     /*const*/ref MatrixT!(T,N,P) B,
     inout MatrixT!(T,M,P) ret)
{
     with(ret) {
         mixin(_gen_mat_mult_body!(M,N,P)("A","B"));
     }
}

private string _gen_mat_mult_body(int M, int N, int P)(string A, string B)
{
     string ret;
     for (int row=0; row<M; ++row) {
         for(int col=0; col<P; ++col) {
             ret ~= "        ";
             ret ~= "values_["~ctfe_itoa(col*M+row)~"] = ";
             for(int jj; jj<N; ++jj) {
                 string JJ = ctfe_itoa(jj);
                 string iA = ctfe_itoa( row+jj *M );
                 string iB = ctfe_itoa( jj +col*N );
                 ret ~= A ~ ".values_["~iA~"]*"~B~".values_["~iB~"]";
                 if (jj != N-1) { ret ~= " + "; }
                 else { ret ~= ";\n"; }
             }
         }
         ret ~= "\n";
     }
     return ret;
}


Wow.  The code generating function basically just looks like a standard
triply-nested matrix multiply, but instead of doing multiplies it just
splits out code that implements multiplies.

// Sort data using keys as the order

uint[] keys, data;

assert(keys.length == data.length);

ulong[] arr;
arr.length = keys.length;

foreach(int i, uint _; keys) arr[i] = (cast(uint)keys[i])<<32 | data[i];

arr.sort;

foreach(int i, uint _; keys) data[i] = cast(uint)(arr & 0x0ffff_ffff);

might be called ClassToTupleExample

requires proposed "funcptr" syntax

auto, this, ~this

for, continue, break, try, catch

Lazy evaluation, delegate, cast

get/set speed, final

struct extending with a mixin

foreach(int i, dchar c; char[] str)

weak pointer

switch simulation

subject: Can't declare "template parameter which takes value of any type"

can't use import in functions

benchmark

use ubyte[] to store binary data,
use void[] for typed stuff that the type doesn't matter for

add empty main() function in D code

Re: Inner classes - More expressiveness needed

The D Programming Language - Bugzilla Issue 1594 - version not honored for invarient declaration

The D Programming Language - Bugzilla Issue 1583

compatibility with D1 and D2

can't use import in a function and must have main

Implicit Function Template Instantiation (It's D jargon because D's template instantiation used to always be explicit)

compiles in DMD 2.007

benchmark

with "Proof of Concept"

Re: Full closures for D

Re: Full closures for D

Walter: 'D 2.007 brings full closures to D. I was tired of D being denigrated for not having "real" closures.'

byte[] bytesof = tobytes((&structvar)[0..1]);

Bitfield, _SumLens, Getter, Setter

Bitfield, BitfieldImpl, etc.

compiles and works with dmd 1.022

for DMD 2.012

for std2?

Either you use a Subversion client (like tortoisesvn.tigris.org) to download the trunk or you use the following URL to download a Zip file with the latest revision:

http://www.dsource.org/projects/bindings/changeset/251/trunk/win32?old_path=%2F&format=zip

"251" in this URL is the revision and should be changed to the latest version.

(or instead of a number, just use "head" for the latest version)

based on C code: http://nickmudge.info/c/hashtest.c.txt

"scrapple.tools' ThreadPool class has a futures implementation"

response to "If a command-line argument to a D program contains a non-ascii character, that argument doesn't get into main(). This happens even if console code page is 65001. This is most annoying because it cannot be worked around."

const (D 1.x) = enum (D 2.009)

DMD 2.009

DMD 2.009

invariant(int) can be assigned to an invariant(int)*

DMD 2.007, but not DMD 2.008

DMD 2.008

"contains" and "find" templates

by BCS

Subject: Re: function returning a tuple

subject: Can't declare a "template parameter which takes anything"

See also: 67884, 67862, 67858

http://www.digitalmars.com/webnews/newsgroups.php?art_group=digitalmars.D&article_id=67884

http://www.digitalmars.com/webnews/newsgroups.php?art_group=digitalmars.D&article_id=67862

http://www.digitalmars.com/webnews/newsgroups.php?art_group=digitalmars.D&article_id=67858

error without line number

"real a = 5/2;" set a to 2.

you can't put import in a function

can't use "import" in a function

demangle to get fully qualified name (FQN)

how to create non-rectangular, non-uniform, non-whatever windows with HTMLayout (flowerd)

import std.algorithm;

struct A
{
    int Merge(int) { return 0; }
}

void main(string[] args)
{
    A a;
    map!((int x) { return a.Merge(x); })([1, 2, 3]);
}

This is because the enum declaration creates its own namespace.

Use anymous enumerate, aliases, or namespace.

used 'sleep' from the Linux shell

template Const(T)

Re: Overloading templates

random k-sample of a file

Appender

use temporary because of a bug

You can use this query to get a sorted list of bugs on which there are at
least one vote:

  http://d.puremagic.com/issues/buglist.cgi?query_format=advanced&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&votes=1&order=bugs.votes,bugs.bug_id

To construct the query yourself: on the advanced search page, put a 1 in
the "Only issues with at least ___ votes" box.

uint len

small function template

assert (!__traits (compiles ...

Unification and extension of compile-time reflection

Scope storage class

 "Man or boy" test

Re: sscanf() equivalent for D using Phobos library?

"nbody" benchmarks of the Shootout

complex objects using mixins

Himeno test

You may also want to take a look at xchan() and the template mixin Chainable:

http://www.fantascienza.net/leonardo/so/dlibs/func.html

(And there's lot of other lazy stuff there that you may appreciate.)

Foreach Range Properties (D 2.x)

D_Version2

Guideline for D1/2 compatible code

D_Version2

multiprecision integers (long)

Immutable Things, where do they all come from?