Make most emojis full-width #5795

leonMSFT · 2020-05-07T18:38:55Z

The table that we refer to in CodepointWidthDetector.cpp to determine
whether or not a codepoint should be rendered as Wide vs Narrow was
based off EastAsianWidth[1]. If a codepoint wasn't included in this
table, they're considered Narrow. Many emojis aren't specified in the
EAW list, so this PR supplements our table with emoji codepoints from
emoji-data[2] in order to render most, if not all, emojis as full-width.

There are certain codepoints I've added to the comments (in case we want
to add them officially to the table in the future) that Microsoft
decided to give an emoji presentation even if it's specified as
Narrow/Ambiguous in the EAW list and are not specified in the Unicode
emoji list. These include all of the Mahjong Tiles block, different
direction pencils (✎✐), different pointing index fingers (☜, ☞) among
others. I have no idea if I've captured all of them, as I don't know of
an easy way to detect which are Microsoft specific emojis.

Validation Steps Performed

I have looked at so many emojis that I dream emoji.

These screenshots aren't encompassing all emoji but I've tried to grab
a couple from all across the codepoint ranges:

Before:

After:

[1] http://www.unicode.org/Public/UCD/latest/ucd/EastAsianWidth.txt
[2] https://www.unicode.org/Public/13.0.0/ucd/emoji/emoji-data.txt

Closes #900

zadjii-msft

I think I'm fine with this, but let's get the expert in on this.

Also: Add some screenshots 😄

zadjii-msft · 2020-05-07T20:32:18Z

@msftbot make sure @miniksa signs off on this

ghost · 2020-05-07T20:32:23Z

Hello @zadjii-msft!

Because you've given me some instructions on how to help merge this pull request, I'll be modifying my merge approach. Here's how I understand your requirements for merging this pull request:

I'll only merge this pull request if it's approved by @miniksa

If this doesn't seem right to you, you can tell me to cancel these instructions and use the auto-merge policy that has been configured for this repository. Try telling me "forget everything I just told you".

DHowett-MSFT · 2020-05-07T20:34:01Z

src/types/CodepointWidthDetector.cpp

- UnicodeRange{ 0x23f0, 0x23f0, CodepointWidth::Wide },
- UnicodeRange{ 0x23f3, 0x23f3, CodepointWidth::Wide },
- UnicodeRange{ 0x2460, 0x24e9, CodepointWidth::Ambiguous },
+ UnicodeRange{ 0x2328, 0x232a, CodepointWidth::Wide },


If possible -- and I know this is a lot of work -- i'd like a comment on each line that refers to a footnote about us overriding these. Just so we can come find it later.

DHowett-MSFT

(Changes requested in comment)

j4james · 2020-05-08T12:20:45Z

I'm not an expert on this, but I've followed a few other discussions on this subject, and my understanding was that it was vital that we exactly follow whatever standard everyone else uses for determining character width. Applications like editors need to be able to predicate exactly how much of a line of text they can fit on the screen before it wraps. They can't do that if the standard says a character should be narrow, but we're rendering it as wide.

I don't know exactly what the standards are that we should be following, so maybe everything in this PR is good, but the mention of "Microsoft specific emojis" makes it sound like we're making up own rules for some characters, and I suspect that's going to cause problems.

miniksa · 2020-05-08T17:03:25Z

I'm not an expert on this, but I've followed a few other discussions on this subject, and my understanding was that it was vital that we exactly follow whatever standard everyone else uses for determining character width. Applications like editors need to be able to predicate exactly how much of a line of text they can fit on the screen before it wraps. They can't do that if the standard says a character should be narrow, but we're rendering it as wide.

I don't know exactly what the standards are that we should be following, so maybe everything in this PR is good, but the mention of "Microsoft specific emojis" makes it sound like we're making up own rules for some characters, and I suspect that's going to cause problems.

I agree with James here. If Microsoft are the only one emojifying a particular Unicode codepoint, we should just center that glyph inside the 1 cell that it's supposed to be taking up.

EDIT: Or we should put just those ones behind a setting called experimental.wideMicrosoftEmoji = true

leonMSFT · 2020-05-08T17:56:45Z

That's fair, I was unsure about putting the Microsoft specific emojis in, because if I didn't, the lower_right_pencil u270e emoji that was noticed to be tiny in #900 would stay tiny since it isn't in the standard.

miniksa · 2020-05-08T18:17:57Z

That's fair, I was unsure about putting the Microsoft specific emojis in, because if I didn't, the lower_right_pencil u270e emoji that was noticed to be tiny in #900 would stay tiny since it isn't in the standard.

Yeah that's why I said maybe we make it an experimental toggle that is off by default. So purists can have what Unicode dictates and Microsoft fanboys can have their full width pencils.

DHowett-MSFT · 2020-05-08T18:52:46Z

Yeah that's why I said maybe we make it an experimental toggle that is off by default. So purists can have what Unicode dictates and Microsoft fanboys can have their full width pencils.

I am deeply concerned with us plugging a configuration option into this global singleton function (since right now we don't have a good place to store state fo codepoint width detection, and we need conpty to agree on the widths with terminal, and there's no good way to communicate that state down to conpty)

leonMSFT · 2020-05-08T19:33:23Z

I'll leave the MS codepoints I've found in the comments so we can come back to implement them as we wish in the future then?

miniksa · 2020-05-08T19:46:20Z

Yeah that's why I said maybe we make it an experimental toggle that is off by default. So purists can have what Unicode dictates and Microsoft fanboys can have their full width pencils.

I am deeply concerned with us plugging a configuration option into this global singleton function (since right now we don't have a good place to store state fo codepoint width detection, and we need conpty to agree on the widths with terminal, and there's no good way to communicate that state down to conpty)

Oof. OK. Yeah you're right.

I'll leave the MS codepoints I've found in the comments so we can come back to implement them as we wish in the future then?

Yeah that's probably best.

miniksa

🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉

DHowett-MSFT · 2020-05-08T21:51:17Z

James is right, but right now the best thing we have is "vague consensus of various wcwidth implementations" and we're only making things slightly more inconsistent by adding another one. I'm not happy that we have to do this, but I think that applications are going to have a rough time regardless since the landscape here is so bad. wcwidth operates on a wchar_t, and there's no practical way to handle things like the emoji variation selector or other composing characters.

Unfortunately, emoji are just one place where column-counting breaks down for applications. ☹️

The table that we refer to in `CodepointWidthDetector.cpp` to determine whether or not a codepoint should be rendered as Wide vs Narrow was based off EastAsianWidth[1]. If a codepoint wasn't included in this table, they're considered Narrow. Many emojis aren't specified in the EAW list, so this PR supplements our table with emoji codepoints from emoji-data[2] in order to render most, if not all, emojis as full-width. There are certain codepoints I've added to the comments (in case we want to add them officially to the table in the future) that Microsoft decided to give an emoji presentation even if it's specified as Narrow/Ambiguous in the EAW list and are _not_ specified in the Unicode emoji list. These include all of the Mahjong Tiles block, different direction pencils (✎✐), different pointing index fingers (☜, ☞) among others. I have no idea if I've captured all of them, as I don't know of an easy way to detect which are Microsoft specific emojis. ## Validation Steps Performed I have looked at so many emojis that I dream emoji. These screenshots aren't encompassing _all_ emoji but I've tried to grab a couple from all across the codepoint ranges: Before: ![before](https://user-images.githubusercontent.com/57155886/81445092-2051a980-912d-11ea-9739-c9f588da407d.png) After: ![after](https://user-images.githubusercontent.com/57155886/81445107-2778b780-912d-11ea-9615-676c2150e798.png) [1] http://www.unicode.org/Public/UCD/latest/ucd/EastAsianWidth.txt [2] https://www.unicode.org/Public/13.0.0/ucd/emoji/emoji-data.txt Closes #900 (cherry picked from commit 7ae3433)

A couple of codepoints, namely the card suites, male and female signs, and white and black smiling faces were changed to have a two-column width as part of #5795 since they were specified as emoji in Unicode's emoji list v13.0[1]. These particular glyphs also show up in some of the most fundamental code pages, such as CP437[2] and WGL4[3]. We should not be touching the width of the glyphs in these codepages, as suddenly changing a long-time-running narrow glyph to use two-columns all of a sudden will surely break (and has already broken) things. [1] https://www.unicode.org/Public/13.0.0/ucd/emoji/emoji-data.txt [2] https://en.wikipedia.org/wiki/Code_page_437 [3] https://en.wikipedia.org/wiki/Windows_Glyph_List_4 Closes #5822

A couple of codepoints, namely the card suites, male and female signs, and white and black smiling faces were changed to have a two-column width as part of #5795 since they were specified as emoji in Unicode's emoji list v13.0[1]. These particular glyphs also show up in some of the most fundamental code pages, such as CP437[2] and WGL4[3]. We should not be touching the width of the glyphs in these codepages, as suddenly changing a long-time-running narrow glyph to use two-columns all of a sudden will surely break (and has already broken) things. [1] https://www.unicode.org/Public/13.0.0/ucd/emoji/emoji-data.txt [2] https://en.wikipedia.org/wiki/Code_page_437 [3] https://en.wikipedia.org/wiki/Windows_Glyph_List_4 Closes #5822 (cherry picked from commit cf62922)

ghost · 2020-05-13T17:12:17Z

🎉Windows Terminal Release Candidate v0.11.1333.0 (1.0rc2) has been released which incorporates this pull request.:tada:

Handy links:

This reverts commit 7ae3433.

TapeWerm · 2020-05-18T03:23:35Z

When I type a ☢ emoji and backspace it in Ubuntu 18.04 I need to Ctrl-L to make it disappear. This does not happen on cmd. Should I open a separate issue for this?

DHowett-MSFT · 2020-05-18T03:31:19Z

Nope. We fixed this with #5934 😄 we got some of the widths wrong

The table that we refer to in `CodepointWidthDetector.cpp` to determine whether or not a codepoint should be rendered as Wide vs Narrow was based off EastAsianWidth[1]. If a codepoint wasn't included in this table, they're considered Narrow. Many emojis aren't specified in the EAW list, so this PR supplements our table with emoji codepoints from emoji-data[2] in order to render most, if not all, emojis as full-width. There are certain codepoints I've added to the comments (in case we want to add them officially to the table in the future) that Microsoft decided to give an emoji presentation even if it's specified as Narrow/Ambiguous in the EAW list and are _not_ specified in the Unicode emoji list. These include all of the Mahjong Tiles block, different direction pencils (✎✐), different pointing index fingers (☜, ☞) among others. I have no idea if I've captured all of them, as I don't know of an easy way to detect which are Microsoft specific emojis. ## Validation Steps Performed I have looked at so many emojis that I dream emoji. These screenshots aren't encompassing _all_ emoji but I've tried to grab a couple from all across the codepoint ranges: Before: ![before](https://user-images.githubusercontent.com/57155886/81445092-2051a980-912d-11ea-9739-c9f588da407d.png) After: ![after](https://user-images.githubusercontent.com/57155886/81445107-2778b780-912d-11ea-9615-676c2150e798.png) [1] http://www.unicode.org/Public/UCD/latest/ucd/EastAsianWidth.txt [2] https://www.unicode.org/Public/13.0.0/ucd/emoji/emoji-data.txt Closes microsoft#900

A couple of codepoints, namely the card suites, male and female signs, and white and black smiling faces were changed to have a two-column width as part of microsoft#5795 since they were specified as emoji in Unicode's emoji list v13.0[1]. These particular glyphs also show up in some of the most fundamental code pages, such as CP437[2] and WGL4[3]. We should not be touching the width of the glyphs in these codepages, as suddenly changing a long-time-running narrow glyph to use two-columns all of a sudden will surely break (and has already broken) things. [1] https://www.unicode.org/Public/13.0.0/ucd/emoji/emoji-data.txt [2] https://en.wikipedia.org/wiki/Code_page_437 [3] https://en.wikipedia.org/wiki/Windows_Glyph_List_4 Closes microsoft#5822

leonMSFT added 3 commits May 5, 2020 12:35

adding the rest of the emoji codepoints in to see

19e4c53

comment

e0eac63

comments and microsoft emoji additions

e8e2beb

ghost added Area-Rendering Text rendering, emoji, complex glyph & font-fallback issues Issue-Bug It either shouldn't be doing this or needs an investigation. Priority-2 A description (P2) Product-Terminal The new Windows Terminal. labels May 7, 2020

zadjii-msft approved these changes May 7, 2020

View reviewed changes

ghost added the AutoMerge Marked for automatic merge by the bot when requirements are met label May 7, 2020

DHowett-MSFT reviewed May 7, 2020

View reviewed changes

DHowett-MSFT suggested changes May 8, 2020

View reviewed changes

ghost added the Needs-Author-Feedback The original author of the issue/PR needs to come back and respond to something label May 8, 2020

marking the overridden ones

7ae5d20

ghost removed the Needs-Author-Feedback The original author of the issue/PR needs to come back and respond to something label May 8, 2020

removed MS emoji codepoints

ee2b59f

leonMSFT added 4 commits May 8, 2020 12:52

added a few missing

224f880

hot springs, female sign

56000d4

spell check

fc0fe03

wow i can't do math

9e2ee25

miniksa approved these changes May 8, 2020

View reviewed changes

DHowett-MSFT approved these changes May 8, 2020

View reviewed changes

ghost merged commit 7ae3433 into master May 8, 2020

ghost deleted the dev/lelian/emojibig branch May 8, 2020 22:31

j4james mentioned this pull request May 9, 2020

Card suite Unicode characters are the wrong width #5822

Closed

leonMSFT mentioned this pull request May 12, 2020

Revert some emoji back to narrow width #5866

Merged

ghost mentioned this pull request May 13, 2020

Certain "emoji" are still half-sized #900

Closed

DHowett-MSFT pushed a commit that referenced this pull request May 16, 2020

Revert "Make most emojis full-width (#5795)"

b2bd271

This reverts commit 7ae3433.

pagerc mentioned this pull request Dec 5, 2020

Feature: Explicit option for rendering emojis as Wide mintty/mintty#1065

Closed

This pull request was closed.

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Make most emojis full-width #5795

Make most emojis full-width #5795

leonMSFT commented May 7, 2020 •

edited by DHowett-MSFT

Loading

zadjii-msft left a comment

zadjii-msft commented May 7, 2020

ghost commented May 7, 2020

DHowett-MSFT May 7, 2020

DHowett-MSFT left a comment

j4james commented May 8, 2020

miniksa commented May 8, 2020 •

edited

Loading

leonMSFT commented May 8, 2020

miniksa commented May 8, 2020

DHowett-MSFT commented May 8, 2020

leonMSFT commented May 8, 2020

miniksa commented May 8, 2020

miniksa left a comment

DHowett-MSFT commented May 8, 2020

ghost commented May 13, 2020

TapeWerm commented May 18, 2020

DHowett-MSFT commented May 18, 2020

Make most emojis full-width #5795

Make most emojis full-width #5795

Conversation

leonMSFT commented May 7, 2020 • edited by DHowett-MSFT Loading

Validation Steps Performed

zadjii-msft left a comment

Choose a reason for hiding this comment

zadjii-msft commented May 7, 2020

ghost commented May 7, 2020

DHowett-MSFT May 7, 2020

Choose a reason for hiding this comment

DHowett-MSFT left a comment

Choose a reason for hiding this comment

j4james commented May 8, 2020

miniksa commented May 8, 2020 • edited Loading

leonMSFT commented May 8, 2020

miniksa commented May 8, 2020

DHowett-MSFT commented May 8, 2020

leonMSFT commented May 8, 2020

miniksa commented May 8, 2020

miniksa left a comment

Choose a reason for hiding this comment

DHowett-MSFT commented May 8, 2020

ghost commented May 13, 2020

TapeWerm commented May 18, 2020

DHowett-MSFT commented May 18, 2020

leonMSFT commented May 7, 2020 •

edited by DHowett-MSFT

Loading

miniksa commented May 8, 2020 •

edited

Loading