tryashtar

This was not fixed in 25w36a. It affects that version and is still present through 1.21.11-pre5.

When horses were first added to Java in 13w16a, they grew gradually. In 14w28b, a bug was introduced (MC-61535) that caused baby horses to look like adults. It was fixed in 1.8.1-pre1, but baby horses no longer grew gradually. Indeed, the code responsible for this was removed in that exact version. It also introduced another bug that caused horses to visually appear as adults when given food. That bug was in turn fixed in 15w47a. Gradually growing horses was never re-added. When I reported MCPE-121628, I was under the mistaken impression that horses gradually growing was removed from Java in 17w45a, a later snapshot that simplified the model. Still, horses only gradually grew in Java for the first year of their existence. In fact, by the time they were ported to Pocket Edition, this feature was already gone, and probably should not have been ported. Bedrock only just now removed it.

There is no bug here. Thank you for providing the video, it was helpful. In your video, you interact with the books in a different order just before the command starts detecting the blocks as different. The last-interacted with slot is a relevant part of chiseled bookshelves' data. You can see that a chiseled bookshelf outputs redstone strength to a comparator equal to the number of its last interacted-with slot. Therefore, the command is correctly determining that two chiseled bookshelves with a different last interacted-with slot are not identical.

Also, it’s interesting that /replaceitem and /loot replace block don't work on cauldrons (reporting that it’s not a container), but /loot insert does.

Which loot table would that be?

I’d like to mention that my preferred fix for this issue involves removing /title times altogether and moving the times to each /title command. The statefulness of times is not useful, it’s just confusing.

Java does not have facing in summon, and also does not have skippable command arguments, so I don’t know what you mean.

This is a misunderstanding about the meaning of “optional argument”

It does not mean “you can skip this argument and write the next one instead”

It means “if you stop writing the command before this argument, the command is valid”

Therefore spawnPos is correctly marked as optional, because /summon allay is a complete command.

Thanks for the info! I’ll go ahead and mark as plausible for now so Mojang will take a look

attack_range is not an attribute, because it’s more than just one value (it has a lower and upper bound). entity_interaction_range is the attribute version, which spears don’t touch.

I can’t reproduce this; the world loads for me

There's no semicolon; what you're seeing is the pixel at the top-right of the lowercase letter K

Apparently, water breathing preventing water mobs from suffocating in air is a bedrock feature, so this may be valid

I found an even worse nitpick: “Player teleports” appears when any mob holding this item dies:

diamond[death_protection={death_effects:[{type:"teleport_randomly"}]}]

I would really prefer if the statistics actually tracked using an item for its specific purpose, and wasn’t confounded with interactions that merely moved it. MC-298275 is very unpleasant to me. I want to see how many totems I’ve activated and apples I’ve eaten, not how many times I’ve moved either of those things in and out of a pot or shelf or item frame.