2.x: parseToAliases is missing but the README shipped at 2.2.0 still documents it

[yoobi]

Description of Bug

Affects 2.2.0 (upgrading from 1.9.7).

The README shipped at tag 2.2.0 documents EmojiParser#parseToAliases(String) and
EmojiParser#parseToAliases(String, FitzpatrickAction) (lines 231-251), but the function does not
exist in the code at that tag, and it is not on develop either.

EmojiParser.kt provides parseToUnicode, parseToHtmlDecimal, parseToHtmlHexadecimal,
parseFromUnicode, replaceAllEmojis, removeAllEmojis, removeEmojis, removeAllEmojisExcept
and extractEmojis — there is no emoji → short code transform under any name. So either the API was
dropped accidentally during the 2.x module refactor, or it was dropped deliberately and the
documentation was not updated.

Reproduction Steps

1. Depend on emojify + contract + moshi at 2.2.0.
2. Follow the README's "parse to aliases" section:

val emojiManager = EmojiManager.create(context, MoshiDeserializer())
val result = emojiManager.parseToAliases("An 😀awesome 😃string with a few 😉emojis!")

3. It fails to compile: unresolved reference parseToAliases. parseToUnicode, documented a few
   lines above it in the same README, resolves fine.

Additional Context

This is the 1.9.7 usage that no longer has an equivalent — outgoing text is stored as short codes,
incoming text is rendered back to emoji:

val payload = emojiManager.parseToAliases(userInput)    // emoji -> :short_code:
val display = emojiManager.parseToUnicode(serverText)   // :short_code: -> emoji  (still fine in 2.2.0)

If the removal was intentional, what is the recommended way to produce :short_code: output in 2.x?
Is parseFromUnicode with a custom EmojiTransformer the intended path, or is a helper planned
(something like parseToShortCodes, mirroring the new getForShortCode)? I am happy to open a PR
restoring it on top of parseFromUnicode if that would be welcome.

I searched the closed issues and did not find this reported.

[wax911]

Thanks for raising this. The missing parseToAliases(...) API is expected in 2.x, but the README was not updated correctly and still documents the old 1.x behaviour.

The change came from the 2.0 emoji data-model migration. One of the main goals of that migration was to stop maintaining the bundled emoji dataset manually and instead generate it from upstream emoji sources through the repository's Python generator. That allowed the library to keep pace with Unicode/Emojibase releases automatically, but it also meant aligning the Kotlin model with the data that the generator could reliably produce.

As part of that redesign, the old aliases field and the parser functionality built around it were removed. This was intentional and shipped as a major-version change.

Since then, shortcode support has been reintroduced in a different form. The current model exposes shortCodes, and EmojiManager.getForShortCode(...) provides shortcode lookup. What is still missing is the higher-level parsing functionality that existed previously:

• emoji → shortcode
• shortcode → emoji

So the underlying capability has started to return, but the parser implementation is incomplete.

Rather than restoring parseToAliases(...) and the old alias terminology, the direction should be to complete the current shortcode implementation with APIs such as parseToShortCodes(...) and the corresponding shortcode-to-Unicode conversion.

There are a few semantics we need to define first, particularly how a canonical shortcode is selected when an emoji has multiple shortcodes and how collisions should be handled in the reverse direction.

For this issue specifically, the README is incorrect and should be updated to reflect the actual 2.x API. I'll track the missing shortcode parsing functionality separately so it can be implemented against the current data model rather than reviving the legacy alias API.

Thanks for pointing this out and for offering to contribute.