Remove set_unconditionally & BeValue
#1131
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This apparently only existed to get around the orphan rules for `Persistable`, but we can just do that by moving the implementation into the same crate that defines that trait (`fontir`). In the process of this work I discovered that avar needs special handling, because it is the one in this context that is allowed to explicitly be `None`. This had previously been handled by `BeValue`; to support this use case I also added an explicit `PersistableOption` type, but thinking about it more now I think this is unnecesary? Shouldn't it be fine to just serialize an empty avar? But my quick attempt to do this ran into some confusing test failures, so this seems fine for now.
This method had a TODO saying it shouldn't be necessary, which sounded like a challange.
|
Oh nice, if we now have partialeq everywhere we should be able to tear down a bunch of that indeed. Memory is failing me wrt avar weirdness. |
rsheeter
reviewed
Nov 18, 2024
fontbe/src/avar.rs
Outdated
| @@ -17,6 +18,42 @@ use crate::{ | |||
| orchestration::{AnyWorkId, BeWork, Context, WorkId}, | |||
| }; | |||
|
|
|||
| /// Avar is a special case where sometimes we want to explicitly have an empty one | |||
There was a problem hiding this comment.
Note in comment why this exists, e.g. why not just Option<Avar>
rsheeter
approved these changes
Nov 18, 2024
There was a problem hiding this comment.
Definitely looks like an improvement and makes sense if write-fonts now has partialeq we shouldn't need to work around it not having it.
As you suggest, I don't think we need to special case avar so. Maybe as a follow-on we can delete that enum and let avar be like everything else?
|
I tried just deleting the option but it broke some tests that did not seem immediately related, so I settled on this. |
Instead of having a general purpose 'maybe missing' type we can just treat the avar as a singular special case, which is simpler.
cmyr
force-pushed
the
remove-set-unconditionally
branch
from
bbb0b19 to
3c79244
Compare
8 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
How this happened:
set_unconditionally, instead of justsetwrite-fontstypes don't implPartialEqBeValuePersistabletrait for the write-fonts types in the same crate where the trait is defined (fontir)avarwe want to be able to store an explicitOption<T>(the only case like this) and we were repurposing theBeValueto also do that jobso I added aPersistableOptiontype to explicitly handle this case (which I still do not fully understand)MaybeEmptyAvartype.Bonus:
Why do we have this special case for avar? why do we need an explicit option? Why can't we just serialize an empty table in the case where we want to have an empty table in the final font? cc @rsheeter