Thread 105916458

>Adjacent string literals are concatenated in C/C++
const char* str1 = "This is" "a" "single" "string";
const char* strt2 = "So"
"is"
"this";

literally for what reason? just to make programatically finding string literals a little bit more annoying?

"probably so \
that \
you don't have to \
do \
this"

What I find worse in C++ is the fact that the comma is an overloadable operator.

>>105916458 (OP)
>just to make programatically finding string literals a little bit more annoying?
Yeah it's to fuck with you personally and not so people can generate strings with the preprocessor.

>>105916458 (OP)
> string literals concatenated
It’s not guaranteed.
I just did this, and my second string was a “guard” pattern.
I wrote over the first string bounds, printed out my guard… looked the exact same.
Disassembled with /Fa and lo and behold, the compiler put some extra C++ junk in there between the strings. Not even alignment padding (you’d expect the second string might start on a 32, 64, or 128-bit boundary for simd memxxx operations.

>>105916517
Why are you splitting it up between lines to begin with?
>for formatting reasons!
damn that's crazy, maybe this isn't the right way to go about the problem, and #embed should have been in since fuckin C89??

>>105916745
Got a godbolt?

>>105916551
lol? what's the fucking usecase?

what does 結城晴 have to do with this?

>>105916745
Got a pic? genuinely curious about this behaviour

>>105916458 (OP)
Its a feature, one that I have used multiple times. Maybe try making something other than a script next time?

>>105916551
so long as ; is not overloadable everythings fine

>>105916845
I just used a prolog dcg over a token stream to do:
string_literals([Literal|Literals]) --> string_literal(Literal), string_literals(Literals).
string_literals([]) --> [].


where string_literal is another dcg that handles a single string literal. it's trivial, but annoying to have to handle additional cases because people like you are too retarded to read a single-line string with whitespace characters like a normal fucking programmer.

>>105917058
I just used a prolog dcg over a token stream to do:
string_literals([Literal|Literals]) --> string_literal(Literal), string_literals(Literals).
string_literals([]) --> [].


where string_literal is another dcg that handles a single string literal. it's trivial, but annoying to have to handle additional cases because people like you are too retarded to read a single-line string with whitespace characters like a normal fucking programmer.

>>105916458 (OP)
It simplifies preprocessor fuckery.
Wait until you learn about multi-character literals in C.

>>105916829
>>105917058
The two strings are defined consecutively and looks like OP’s example. They weren’t const though.

>>105916745
>>105916829
>>105917058
Oops I was wrong, it’s not doing that.
The extra junk was between the first string and the second string. Not the concatenated string.
Picrel is the extra junk you might get between OPs str1 and strt2, which was still surprising