Skip to content
Search

C++:TrigraphSequence

lastmod 2018-08-17created 2015-03-264 edits

The C language has a set of substitution characters called trigraphs (TrigraphSequence) that take precedence over the preprocessor. This is because the character sets of some language regions cannot accommodate the entire C character set. (Most notably, ISO646 defines a character set of 82 characters, and it does not include some of the C character set.) In such cases C source code cannot be expressed, and trigraphs (Trigraph sequence) came into existence to make up for this. Trigraphs are substituted before any other preprocessing step, and there are 9 trigraphs in total. Each starts with two question marks (?) followed by one identifying character, and is replaced with some other character. The 9 are as follows. 

??=        #
??(        [
??<        {
??/        \
??)        ]
??>        }
??'        ^
??!        |
??-        ~

With code like the following, you will see it print the '#' character. 

printf(">>> ??= <<<\n");

Example

For reference, with C++ code like the following, the substitution only works correctly if you add the -ansi option. 

#include <iostream>
using namespace std;
int main () {
    cout << ">>> ??= <<<" << endl;
    return 0;
}

See also

Favorite site

sitemap