You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I want to start releasing alpha versions while I work on this release, so we can see how much we’ve improved in terms of compatibility with the specification.
Hey, we no longer need to have backerns. That was necessary when this version was being developed (it targeted Node >4, and TextDecoder support didn’t exist yet). Now the idea is to migrate everything to avoid using Node.js buffers and instead use things like Uint8Array
Great, I've read almost all of the comments about the Buffer being inconsistent, and since the decision has already been made, I would love to work on it.
edit: Switching to Uint8Array seems harder than I thought 😄
haha yeah, don’t try to do everything in a single PR, do it in separate ones. That also makes it easier for me to review. As for UTF-16, I think we could probably rely only on TextDecoder for the decoder; I haven’t checked yet whether the inconsistencies that existed before are still there.
Activity
GroophyLifefor commentedon Feb 6, 2026
Hi Sebastian, is #235 not solved with 7a64f21 ? Or might I'm missing some parts.
bjohansebas commentedon Feb 6, 2026
Hey, we no longer need to have backerns. That was necessary when this version was being developed (it targeted Node >4, and TextDecoder support didn’t exist yet). Now the idea is to migrate everything to avoid using Node.js buffers and instead use things like Uint8Array
GroophyLifefor commentedon Feb 7, 2026
Great, I've read almost all of the comments about the Buffer being inconsistent, and since the decision has already been made, I would love to work on it.
edit: Switching to Uint8Array seems harder than I thought 😄
bjohansebas commentedon Feb 7, 2026
haha yeah, don’t try to do everything in a single PR, do it in separate ones. That also makes it easier for me to review. As for UTF-16, I think we could probably rely only on TextDecoder for the decoder; I haven’t checked yet whether the inconsistencies that existed before are still there.
bjohansebas commentedon Oct 31, 2026
P.S. I’m almost done making the changes to the public API so we can release a beta