News for hackers. Stuff that matters.
Re:How a comment on Hacker News led to 4½ new Unicode charac... by smegel on Thursday June 23, 2016 @12:01AM (#11958775)
> Ask the people behind your Operating System and those who design your favourite fonts to start supporting Unicode 9!

Surely not every font has to create glyphs for every Unicode character...how does that work? Is there a kind of "fall-back" font for characters not implemented?

Re:How a comment on Hacker News led to 4½ new Unicode charac... by medgno on Thursday June 23, 2016 @12:13AM (#11958803)
Systems have a way of finding fall-back fonts. At least on OS X (err, macOS) they will satisfy glyph requests in order from the font you specify, some built-in fonts, and then any fonts that have the required glyphs. Finally, a "last resort" font is used to fill in any remaining glyphs[1].

A result of this is that if you ask for a character such as "𓀴" (U+13034 EGYPTIAN HIEROGLYPH A044) in a monospace font, the symbol you get back can be variable width.

[1]: https://en.wikipedia.org/wiki/Fallback_font

Re:How a comment on Hacker News led to 4½ new Unicode charac... by the8472 on Thursday June 23, 2016 @01:18AM (#11958993)
I have a font for that, but the character is unreadably small while it's perfectly fine for latin characters.

Many other unicode symbols also suffer from this problem. E.g. ␀ is the printable version of the unprintable NUL (\0) control character, but it's so small at 13.3px / 10pt CSS font size that it's difficult to distinguish from the other control pictures.

␀ ␁ ␂ ␃ ␄ ␅ ␆ ␇ ␈ ␉ ␊ ␋ ␌ ␍ ␎ ␏ ␐ ␑ ␒ ␓ ␔ ␕ ␖ ␗ ␘ ␙ ␚ ␛ ␜ ␝ ␞ ␟ ␠ ␡

How those look on display w/o pixel scaling: https://i.imgur.com/lAAyjXu.png

Re:How a comment on Hacker News led to 4½ new Unicode charac... by sscotth on Thursday June 23, 2016 @02:41AM (#11959257)
Imagine if ﷽ was sized to a monospace font.
Re:How a comment on Hacker News led to 4½ new Unicode charac... by chme on Thursday June 23, 2016 @07:21AM (#11960164)
For someone who can't see this glyph, here some different renders:

http://i.imgur.com/g7igcwP.png

Re:How a comment on Hacker News led to 4½ new Unicode charac... by chme on Thursday June 23, 2016 @02:55AM (#11959304)
In the name of god, that has to be the most complex glyph I have ever seen.
Re:How a comment on Hacker News led to 4½ new Unicode charac... by acqq on Thursday June 23, 2016 @06:16AM (#11959897)
> In the name of god, that has to be the most complex glyph

And actually it is in the Arabic "name of God," thanks!

https://en.wiktionary.org/wiki/%D8%A8%D8%B3%D9%85_%D8%A7%D9%...

I didn't know there's a whole ligature as a single code point U+FDFD available, thanks sscotth. The code point is also easy to remember.

(It, accidentally, also most probably represents the words first said by the Orlando shooter when calling 911, when the translation in the transcription is compared.)

Is this the biggest glyph in Unicode?

Re:How a comment on Hacker News led to 4½ new Unicode charac... by legulere on Thursday June 23, 2016 @03:04AM (#11959336)
https://en.m.wikipedia.org/wiki/Biangbiang_noodles#/media/Fi...

Biang from Biang biang noodles is also fairly complex (though not supported by Unicode)

Re:How a comment on Hacker News led to 4½ new Unicode charac... by ojii on Thursday June 23, 2016 @04:32AM (#11959590)
Re:How a comment on Hacker News led to 4½ new Unicode charac... by outworlder on Friday June 24, 2016 @07:18PM (#11974231)
Just tried it in Emacs, renders just fine

(M-x ucs-insert FDFD)

Re:How a comment on Hacker News led to 4½ new Unicode charac... by lucaspiller on Thursday June 23, 2016 @02:29AM (#11959215)
They are also very small on my computer (Chrome on OS X 10.11.5, external non-'Retina' monitor), but they are vertically aligned instead of diagonal:

http://i.imgur.com/kKfLYgx.jpg

Interestingly in Safari the hieroglyph is shown (still small though), I wonder why Chrome can't find it?

Re:How a comment on Hacker News led to 4½ new Unicode charac... by rbanffy on Thursday June 23, 2016 @04:54AM (#11959662)
Funny how the OS won't even try some of them...

http://imgur.com/oQEHbA7

Re:How a comment on Hacker News led to 4½ new Unicode charac... by heinrich5991 on Thursday June 23, 2016 @02:37AM (#11959247)
FWIW on my screen, they're really distinguishable: https://i.imgur.com/IdUK8MV.png.
Re:How a comment on Hacker News led to 4½ new Unicode charac... by j4_james on Thursday June 23, 2016 @08:15AM (#11960342)
That looks like your browser is somehow misinterpreting the character set of the page. What you're seeing appears to be something like code page 437 (https://en.wikipedia.org/wiki/Code_page_437) rather than unicode.

What browser/os are you using? Is it possible you're going through a proxy that is altering the HTTP headers?

Re:How a comment on Hacker News led to 4½ new Unicode charac... by yrro on Thursday June 23, 2016 @02:51PM (#11963477)
Could be that he has a font that provides code page 437-like outlines for those codepoints.
Re:How a comment on Hacker News led to 4½ new Unicode charac... by the8472 on Thursday June 23, 2016 @02:52AM (#11959295)
The spade symbol[0] is codepoint 9824, the acknowledge control picture[1] (␆) is codepoint 9222. They shouldn't be using the same glyph.

[0] http://www.fileformat.info/info/unicode/char/2660/index.htm [1] http://www.fileformat.info/info/unicode/char/2406/index.htm

Re:How a comment on Hacker News led to 4½ new Unicode charac... by chme on Thursday June 23, 2016 @03:42AM (#11959423)
With DejaVu they are distinguishable enough:

http://i.imgur.com/WemOI28.png

Re:How a comment on Hacker News led to 4½ new Unicode charac... by petepete on Thursday June 23, 2016 @02:46AM (#11959272)
On Android N with Chrome 51

http://m.imgur.com/tKnyPad

Re:How a comment on Hacker News led to 4½ new Unicode charac... by umanwizard on Thursday June 23, 2016 @01:09AM (#11958964)
Ah, so is that why emojis don't display correctly in vim? They spill out of the character's bounding box...
Re:How a comment on Hacker News led to 4½ new Unicode charac... by legulere on Thursday June 23, 2016 @02:54AM (#11959299)
That sounds more like a problem of your terminal.

The basic problem is that Unicode characters (which may consist of as many codepoints as you like) have varying width. For instance Chinese has characters that are displayed over a width of two normal monospace characters.

See this hacker news post: https://news.ycombinator.com/item?id=10206380

So the problem with emojis is that they aren't really specified to be full width or half width characters: https://ianrenton.com/blog/adventures-in-emoji/

Re:How a comment on Hacker News led to 4½ new Unicode charac... by legulere on Thursday June 23, 2016 @10:20AM (#11961030)
Apparently emoji representations are defined to be wide by unicode 9 now: http://www.unicode.org/reports/tr11/tr11-31.html#Modificatio...
Re:How a comment on Hacker News led to 4½ new Unicode charac... by pavanky on Thursday June 23, 2016 @12:39AM (#11958877)
And I see a boxes (on linux)
Re:How a comment on Hacker News led to 4½ new Unicode charac... by benkuykendall on Thursday June 23, 2016 @01:03AM (#11958943)
That's not an inherent problem with GNU/Linux. Shows up fine for me because I have "ttf-ancient-scripts" (https://aur.archlinux.org/packages/ttf-ancient-fonts/) installed.
Re:How a comment on Hacker News led to 4½ new Unicode charac... by kalleboo on Thursday June 23, 2016 @12:49AM (#11958909)
On my Mac it gets displayed with "Noto Sans Egyptian Hieroglyhs" which appears to be included in the system (/Library/Application Support/Apple/Fonts/Language Support) but can be downloaded for other platforms from here https://www.google.com/get/noto/
Re:How a comment on Hacker News led to 4½ new Unicode charac... by aarongolliver on Thursday June 23, 2016 @12:48AM (#11958904)
Same on win10/chrome. Works on edge though.
Re:How a comment on Hacker News led to 4½ new Unicode charac... by pcr0 on Thursday June 23, 2016 @01:06AM (#11958952)
I see a box with these two lines inside it:

013

034

OSX 10.11.4

Re:How a comment on Hacker News led to 4½ new Unicode charac... by lifthrasiir on Thursday June 23, 2016 @01:47AM (#11959088)
There are some, and I have tried to make one myself [1] to learn OpenType and scripts. It is hard and takes much time and effort to get a good one (Noto fonts [2] are probably the closest font that covers the entire Unicode and looks great). I admit that it is still hard even when you don't have the looks-good requirement.

[1] https://github.com/lifthrasiir/unison

[2] https://www.google.com/get/noto/

Re:How a comment on Hacker News led to 4½ new Unicode charac... by johncolanduoni on Thursday June 23, 2016 @12:18AM (#11958820)
This isn't even possible for most font formats. There are over 100k assigned code points, but TTF and OTF both use unsigned 16-bit glyph indices, giving a maximum of 65,536 glyphs per font. Not all code points map to glyphs, but some glyphs are also not directly associated to code points (for example, some Unicode combining marks require separate glyphs in certain combinations).
Re:How a comment on Hacker News led to 4½ new Unicode charac... by Gracana on Thursday June 23, 2016 @12:12AM (#11958796)
Re:How a comment on Hacker News led to 4½ new Unicode charac... by abritinthebay on Thursday June 23, 2016 @12:13AM (#11958801)
Sort of. Font substitution is a thing in most operating systems. As long as one font on the system has those characters it should work.

Not always true though, and can depend on the application using the right system calls for rendering text (for example).