unicodia
unicodia copied to clipboard
Mercury13
Flag emojis are unsearchable
Searching for any flag emoji only provides results for the "regional indicator symbol" letters.
Example: Searching for 🇺🇸 only returns 🇺 and 🇸.
This is both a problem and a feature.
- Emoji are unsearchable, but how to search them? Could you suggest?
- This is a feature, split into characters. And 🇺🇸 is actually 🇺+🇸.
P.S. When I deleted characters between U and S, they ligated into USA flag.
P.S. When I deleted characters between U and S, they ligated into USA flag.
Deleted what???
@curya , maybe you have some suggestions?
I'm just a lowly end user so I'm probably not the right person to ask. But it seems to me that if I type "us flag", it should bring up the correct emoji. (The normal U and S, not the regional indicator symbols.) This is also the way emojipedia.org handles it. https://emojipedia.org/search?q=us%20flag
@curya You have written “🇺 and 🇸”. When I delete “␣and␣”, I instantly get 🇺🇸. Right now emoji search is not implemented, and what problems I have:
- How to tell multi-character emoji from a normal single character.
- How to return to search list if we moved to wrong place.
- How to treat “us flag” if I have “flag: United States”
This is not just emoji library, it’s the entire Unicode.
I instantly get 🇺🇸 in Firefox+W10, it has Twemoji on board. If you have Chromelike+W10, you won’t get, as W10 has no flag emoji, and Chromelikes completely rely on system emoji. They just display some strange U + some strange S.
This is what I get when typing 🇺🇸 after deleting " and ".
And like I said, I'm probably the wrong person to ask about how to solve this problem. I'm not a programmer. Maybe you can contact the people at emojipedia and ask how they did it?
Maybe a separate search box for only emojis?
Not from programmer’s perspective, but from user’s. I see W10 + really high DPI (glad that hiDPI works somehow, it was ha-ard).
W10 does not have US emoji, so US in strange font is everything it can. And there’s no conceptual difference between blonde 👱🏻♀️ (which W10 can) and US flag 🇺🇸 (which W10 cannot).
So what’s the difference? My scope is the entire Unicode, they cover emoji sequences only. I can insert some temporary feature (e.g. US flag) when you paste THE EMOJI ITSELF. Not “US flag” text, but emoji itself. But it breaks a bit the feature “break into codepoints”, so I need some interface justification, e.g.
Another interface breakup. When you click that US flag, you go somewhere, and there’s hard to return to search results. What to do? These are purely interface troubles, I can program somehow of course, but they will be incomplete.
It shouldn't matter what OS I'm on. Unicodia should display the emoji correctly. In this case, the flags. They display just fine in the Library tab:
Pasting the emoji itself is a step in the right direction. But I feel like the program should also return the correct emoji when searching for "us flag" or "uk flag" or whatever. (Not regional indicator letters).
Just a suggestion. Maybe move the emoji search to the Library tab?
UA, US and so on are actually two codepoints and do not fit straightly into my “break into codepoints” feature. That’s why I’m asking to devise a solution. Right now I’m thinking about data structures that will find finished multi-character emoji inside text, but that’s another thing.
@curya Just by tags you see, I took your request seriously, and thinking what to do and how to help you. I know that Unicodia has these problems.
I appreciate it! Unfortunately, I'm not sure I can help you. Maybe there should be a separate "emoji" tab so we can search flags (🇺🇸) and all variations of emojis (like 👱🏻♀️)?
@curya Recently I’ve read a well-known story about Pontiac that started when the owner bought chocolate ice-cream, and refused when he bought vanilla ice-cream. It was hard to translate from owner’s language to engineer’s one. Vanilla ice-cream (default in the USA, thus term “vanilla” for original unmodded software) was in a separate fridge, and buying it was quicker. Because of carburettor deficiencies, the car failed to start in a specific time range after stopping. (VAZ owners will come: cover it with a wet rag.)
I will not release until Noto runs through all their fonts. And I’ll write more.
@curya Recently I’ve read a well-known story about Pontiac that started when the owner bought chocolate ice-cream, and refused when he bought vanilla ice-cream. It was hard to translate from owner’s language to engineer’s one. Vanilla ice-cream (default in the USA, thus term “vanilla” for original unmodded software) was in a separate fridge, and buying it was quicker. Because of carburettor deficiencies, the car failed to start in a specific time range after stopping. (VAZ owners will come: cover it with a wet rag.)
I think something is lost in translation here...
@curya Is it OK?
Looks good to me. I'll have to test it when the new version is released.
@curya Google “pontiac vanilla ice cream” About VAZ (Volga Automobile Plant): to own a car/moped in the USSR, you needed to do most of maintenance for yourself, so Soviet car-owners knew lots of lifehacks.
Noto did nothing today, waiting for monthly release.
Does this support the other multi-character emojis as well? 👱🏻♀️ 🧑🧑🧒 🐦🔥 🍋🟩
Words “flag” and “of” are flag-related keywords. “Of” is dubious (the last counts if several), and “flag” is just a keyword. So “of flag ua” and “ua flag” do the same. More than 3 words, digits → bad string, no search
@curya Normal emoji search isn’t done yet, mostly because of special words like “person” and “flag”. But test what I made.
@curya Making an emergency fix right now, I’ll think later about true emoji search.
@curya @MAZ06 This isn’t exactly your task, but did today…
- You can find a character by its emoji name: e.g. both “savoring” (emoji) and “savouring” (character)