[buffer]: introduce the 'latin1' encoding string

[trevnorris]

• tests and code linting passes
• a test and/or benchmark is included
• documentation is changed or added
• the commit message follows commit guidelines

Affected core subsystem(s)

buffer, string_bytes

Description of change

When node began using the OneByte API (f150d56) it also switched to
officially supporting ISO-8859-1. Though at the time no new encoding
string was introduced.

Introduce the new encoding string 'latin1' to be more explicit. The
previous 'binary' is still around and only deprecated in documentation.
While many tests have switched to use 'latin1', there are still plenty
that do both 'binary' and 'latin1' checks side-by-side to ensure there
is no regression.

There's another minor cleanup commit included.

R=@bnoordhuis
R=@addaleax
R=@jasnell

CI: https://ci.nodejs.org/job/node-test-pull-request/2902/

[Fishrock123]

I think this is a good idea and a slow deprecation cost we should swallow so that things are make sense for the future.

[bnoordhuis]

LGTM

[jasnell]

semver-major because of the docs-only deprecation of binary

[addaleax]

@trevnorris Might be a good idea to separate out the deprecation to make this a semver-minor?

LGTM, and a big thanks for doing this.

[jasnell]

LGTM with green CI and CITGM. That said, not really sure there's a ton of benefit here as we'll likely never be able to get rid of the binary value.

[addaleax]

I don’t think anybody’s expecting to remove binary, but I’m pretty sure there’s a benefit for learners in this. People referring to “binary strings” just adds a lot of confusion to the topic of character encodings, and it’s already a confusing topic all by itself.

[trevnorris]

@jasnell didn't realize doc only deprecation would be semver-major. if that's the case can I just remove (deprecated) for a semver-minor patch and re-add it for next major.

edit: to clarify, reason for deprecation is as @addaleax pointed out. to simply point out that 'binary' encoding isn't actually a thing, and for clarity should use 'latin1'.

[jasnell]

All other doc only deprecations we've done have been treated as semver-majors as well.
Removing the deprecation would certainly bump this down to a minor.

[trevnorris]

@jasnell There you go. Changed documentation to state that 'binary' is an alias for 'latin1'.

[trevnorris]

@jasnell This mean I'm good to land?

CI after rebase: https://ci.nodejs.org/job/node-test-commit/3685/

[jasnell]

Lgtm

[trevnorris]

Thanks much! Landed in 54cc721 and e0b8dd5.

[bnoordhuis]

I'm changing this to semver-major again because the change to the encoding enum is an ABI change; the values of the enum change.

It could land in concert with #7284 as a semver-minor change but honestly, I'd just wait until v7.

[trevnorris]

I know it's been almost two years since this should have been introduced, but that means one more LTS of a confusing encoding name =/

If it can't be done then I'll deal with it, but think it'd be useful to land in v6.

[addaleax]

I’d be +1 for landing these together in a semver-minor, too.