How are two strings compared?

Question

I am not sure how to form this question, since english isn't my primary language. . .

What is the process behind comparing two strings?

For example, how does a computer, on which logic, compare two strings?

#include <iostream> #include <string> int main() { std::string s1 {"b"}; std::string s2 {"abc"}; if(s1 > s2) { std::cout << s1 << " > " << s2; } else std::cout << s2 << " > " << s1; return 0; }

ouput: b > abc

How does a computer come up with this logic (even though it is correct).

I imagined computers logic to be converting chars into integers then comparing them by size, which is not the case here since if it was

b > abc would be treated like 98 > 97 + 98 + 99 which is incorrect.

98 > 97, that's it. Why would you add up the ASCII codes of the letters anyway? — ForceBru
– ForceBru, Commented Mar 25, 2017 at 18:28
The computer doesn't need to convert characters into integers - characters are already integer values — UnholySheep
– UnholySheep, Commented Mar 25, 2017 at 18:28
It compares the first pair of different characters. One that is larger belongs to a "larger" string. — HolyBlackCat
– HolyBlackCat, Commented Mar 25, 2017 at 18:29
On a side note, your else branch is wrong. If s1 is not greater than s2, it may also be that s1 is equal to s2. — zett42
– zett42, Commented Mar 25, 2017 at 18:31
When you say "a computer", this sounds like the CPU itself performed the comparison. That's a wrong picture. The lexicographical comparison performed for std::string is an operation on a much higher abstraction level. — Christian Hackl
– Christian Hackl, Commented Mar 25, 2017 at 18:31

Kerrek SB · Accepted Answer · 2017-03-25 18:28:16Z

3

The comparison logic is specified by the string's char traits, which for std::string is std::char_traits<char>::compare, which in turn specifies "lexicographic comparison". Each character is compared based on its numeric value, which is given by the encoding of the execution character set. On your platform, 'b' > 'a' is true, so s2 compares less than s1.

answered Mar 25, 2017 at 18:28

Kerrek SB

480k96 gold badges904 silver badges1.1k bronze badges

Sign up to request clarification or add additional context in comments.

6 Comments

zett42 Over a year ago

Where does the reference state anything about encoding and character sets being involved in string comparison?

463035818_is_not_an_ai Over a year ago

@zett42 it doesnt need to because the comparison only cares about the numerical value

zett42 Over a year ago

Just checked, the std::char_traits<char>::compare() implementation of MSVC++2017 does a simple memcmp(), so it doesn't really care about encodings and character sets.

Kerrek SB Over a year ago

@zett42: I think you're misunderstanding the point about encodings. The result of 'a' < 'b' depends on the numeric values of 'a' and 'b', which is precisely what the "encoding" of the execution character set is. If 'a' is 10 and 'b' is 20, the comparison is true, and if 'a' is 10 and 'b' is 5, the comparison is false.

zett42 Over a year ago

I just had to look up "execution character set", MSVC has a pretty good explanation: The execution character set is the encoding used for the text of your program that is input to the compilation phase after all preprocessing steps. ... still need to find a character set where 'a' < 'b' == false though ;-)

|

Collectives™ on Stack Overflow

How are two strings compared?

1 Answer 1

6 Comments

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

6 Comments

Linked

Related