CppCon 2014: James McNellis "Unicode in C++"

[@2000YG]

since 2017 the german uppercase for ß is ẞ (its own letter) Before that there was no uppercase form of ß SS seems like a workaround to make a program produce a desireble result in most cases.

[@christo161]

To write portable c++ code with non-ascii characters in string literals was always a forbidden mystic black magic :D

[@2002budokan]

If you know what Unicode is and interested only how all these are implemented in C++ skip to 30:07

[@JimBalter]

"Any time there are numbers on the slide they're hexadecimal numbers" next slide: "32 control characters, 95 printable characters" I literally laughed out loud! No harm done but you have to admit it's funny.

[@hl2mukkel]

This guy makes great talks!

[@Dereline]

4:48 Алло Йоба, это ты? :D

[@tymscar]

There is an error at 17:25. U+0065 is e. Lovely talk regardless!

[@JivanPal]

The codepoint should read "U+0069", then the slide is correct 🙂

[@3oder7]

Nice "Easteregg" at 0:11, i see what you did there :D

[@xarcaz]

I'd be surprised if there were anyone that didn't notice it...

[@jonskunator]

Gosh...everything in computer science is getting so complicated. Even storing a piece of text is already quite involved. On the other hand, C++ seems to quite nicely abstract away some of the burden.

[@llothar68]

No it's much easier now. Before we had hunderts of encoding, now we have UTF-8

[@shizeeque]

C++ is not a perfect choice for text processing. Go from Google has native UTF-8 support for example. Even "Hello World" in Go prints text in Chinese golang.org/

[@mihainita5325]

And actually what was explained here is not quite accurate. The C++ standard is messier than that. The video says that char16_t and char32_t are UTF-16 / UTF-32. But that is not what the standard says. That is only true if the STDC_UTF_16 (respectively STDC_UTF_32) macros are defined.

[@nwodon8481]

Visual C++ fucks up existing code at every new Release. GCC doesnt has this sort of Problem at all.

[@Dereline]

And russian "Hello" -> "Привет" ;)

[@shizeeque]

Алло is a phone response in Russian so Hello -> Алло works fine too. Sort of. :)

[@tetramaximum]

это он гугл-транслэйт так подколол ;)

[@tetramaximum]

long story short: use Boost

[@nexusclarum8000]

In the year 2020 there shouldn't be a programming language without full unicode support. At least UTF-8. I think unicode is stupid anyways. Unicode should have been a specialist format. They should have had 2-byte/4-byte code point for all human languages without all the normalization/grapheme cluster non-sense. Oh two symbols look identical but are different? Who cares. Make them same. I just don't get it.... I'm just pissed off how annoying unicode is in C++.

[@JimBalter]

I think the developers of unicode were an order of magnitude smarter than you.

[@abobe5572]

@@JimBalter over-engineering things is not about being smarter

[@avtem]

65,536 characters won't be enough if you put useless symbols like that rabbit. Just alphabets and signs like -,.;:*. For that use images or dedicated fonts like windlings

[@JimBalter]

There are 1,114,112 code points, not only 65,536 characters. Did you watch the video?