Remove non-ascii characters from internal code comments #53608
Conversation
typescript-bot
added
the
For Uncommitted Bug
|
This PR doesn't have any linked issues. Please open an issue that references this PR. From there we can discuss and prioritise. |
|
This seems okay, but, why is it a requirement for Deno? Can you exempt vendored code? How can we make sure to not regress? What if we add special characters to document something? |
Don't worry about it on your side. Our code just fails to build and it's easy to workaround. Only about 4 of these comments actually ended up in the output so I think it's really rare.
At that point I'll just add something to transform them.
No, it allows us to use v8 external one-byte strings which drastically reduces our snapshot size. Once one of these characters ends up in the code then it goes to utf-16. |
That's super interesting! |
| @@ -129,7 +129,7 @@ function getDeclarationSiteFix(context: CodeFixContext | CodeFixAllContext, expr | |||
| } | |||
| }); | |||
| // No fix-all because it will already be included once with the use site fix, | |||
| // and for simplicity the fix-all doesn‘t let the user choose between use-site and declaration-site fixes. | |||
| // and for simplicity the fix-all doesn't let the user choose between use-site and declaration-site fixes. | |||
There was a problem hiding this comment.
Can't stop looking at this ‘ psuedoapostrophe
In Deno we now require 7-bit ascii for our internal code. When TypeScript builds, some of these characters end up in the build output. It seems like there weren't that many instances of non-ascii characters in TypeScript's comments so is it ok to remove them? It makes my life a little easier and makes the code a bit more consistent.
I only did a search for the characters that ended up in the build output in the "src" directory and nothing else.