Thread: Летс ми спик фром май харррт

The undeniable lack of Esperanto is the lack of natural (or native) literature and poetry or any expression of art which can transfer the idea of this art to native speakers of all other languages. While C++ has tons of texts, which are understandable for the people all over the world. Some examples of code are really the pieces of art. So C++ is the real candidate for the universal language

Originally Posted by CoffeeCup

... C++ has tons of texts, which are understandable for the people all over the world. Some examples of code are really the pieces of art. So C++ is the real candidate for the universal language

I suspect you are a fluent speaker, and therefore biased!

But actually, I read that someone made a language that contains a proportion of grammar and similarities to reflect the population of the world.... So it had lots of Chinese and Indian type words in it, obviously.
I do think Esperanto is nice and I like the "ideology" of Esperanto too.
(but I don't actually know the language... does anyone here speak it?)

Originally Posted by CoffeeCup

The undeniable lack of Esperanto is the lack of natural (or native) literature and poetry or any expression of art which can transfer the idea of this art to native speakers of all other languages. While C++ has tons of texts, which are understandable for the people all over the world. Some examples of code are really the pieces of art. So C++ is the real candidate for the universal language

Code:

Esperanto::Esperanto() { 
  char* dest=(char*) self; 
  idLang Lang=idEnglish;
  while(dest-(char*)self<=sizeof(self*)) {
    dest++*=(((char*)&::Languages[Lang])+rand()%sizeof(Languages[Lang]))*;
    if (++Lang>idChinese) Lang=idEnglish;
  }
  for(int i=0; i<nRules; i++) if (GrammarRules[i]->Size>3) GrammarRules[i]=NULL;
}

Originally Posted by it-ogo

Code:

Esperanto::Esperanto() { 
  char* dest=(char*) this; 
  idLang Lang=idEnglish;
  while(dest-(char*)this<=sizeof(*this)) {
    *dest++*= *(((char*)&::Languages[Lang])+rand()%sizeof(Languages[Lang]));
    if ((Lang = (idLang)((int)Lang + 1)) > idChinese) Lang=idEnglish;
  }
  for(int i=0; i<nRules; i++) if (GrammarRules[i]->Size>3) GrammarRules[i]=NULL;
}

A few corrections...

Originally Posted by Eric C.

A few corrections...

Code:

#define self this
idLang operator ++(idLang &, int);

//etc...

Originally Posted by it-ogo

Code:

#define self this
idLang operator ++(idLang &, int);

//etc...

Ok... then:

Code:

Esperanto::Esperanto() { 
  char* dest=(char*) self; 
  idLang Lang=idEnglish;
  while(dest-(char*)self<=sizeof(*self)) {
    *dest++*= *(((char*)&::Languages[Lang])+rand()%sizeof(Languages[Lang]));
    if (++Lang > idChinese) Lang=idEnglish;
  }
  for(int i=0; i<nRules; i++) if (GrammarRules[i]->Size>3) GrammarRules[i]=NULL;
}

But:

1. No code like "this*", in this case macros won't let you get away with a syntax error.
2. No code like "<pointer_name>*" when dereferencing the pointer. That's just wrong and gives you a syntax error as well.
3. It's generally recommended you define return types of such overloaded operators as "<class_name>&", ie:

Code:

idLang& operator ++(idLang &, int);

That allows you to avoid an unnecessary (and potentially expensive) call of the copy constructor.

Ви ар спикинг фром аувер харт.

Originally Posted by Eric C.

Ok... then:
1.
2.

Agreed. Mea culpa.

3. It's generally recommended you define return types of such overloaded operators as "<class_name>&", ie:

Code:

idLang& operator ++(idLang &, int);

That allows you to avoid an unnecessary (and potentially expensive) call of the copy constructor.

Hmmm... Actually I meant that idLang is enum rather than class. Otherwise we should invent something very special for "Languages[Lang]".

Originally Posted by it-ogo

Hmmm... Actually I meant that idLang is enum rather than class.

I see. Then it's quite ok.

Originally Posted by it-ogo

Otherwise we should invent something very special for "Languages[Lang]"

Like "operator int()"?