-
Notifications
You must be signed in to change notification settings - Fork 29.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
build: remove cares header from tarball #10283
Conversation
cc/ @richardlau |
@bnoordhuis Is the rationale for removing zlib the same as that for c-ares? |
I'm not so sure about removing the zlib headers. From nodejs/node-gyp#1055 (comment):
|
Also the important thing here is consistency. At the moment it is possible to write a native addon that |
This doesn't sound like a good idea to me. |
Which part? Removing zlib or c-ares? |
c-ares sounds like a good idea to remove - not exported on Windows and not kept stable means its not a useable API/ABI zlib sounds like people are using, and its useable |
7ca0fee
to
f9e4e5e
Compare
Okay, I've left zlib in and just removed c-ares. Thoughts @Fishrock123 @sam-github @richardlau ? |
Still semver-major, idk... we might need some ecosystem stats? I'm fairly certain I've heard of addons using c-ares before. |
From my point of view I just want some consistency (hence the referenced pull request in Whatever we decide, we should also tidy up the addons docs. |
It's pretty broken, though. I receive the occasional bug report and I always end up steering people into bundling c-ares with their add-on. |
@gibfahn Maybe remove |
@jasnell Is this something that the CTC should look at? EDIT: It's |
@ChALkeR Is there an easy way to find out whether modules are using the c-ares headers (rather than bundling c-ares with their addon)? |
@ChALkeR ping |
cc/ @nodejs/ctc , does anyone object to this? I think we should either remove the headers or support them. If using the bundled version doesn't work properly then I'd vote to remove. cc/ @Fishrock123 any further thoughts? It'd be good to resolve this one way or the other before Node 8. |
No objection from me. If necessary we can put this on this weeks CTC call agenda to make sure folks look at it. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'd agree that unless we are willing to commit to support/keep consistent we should remove LGTM.
I'm +1 on this in semver-major. I think we have a fairly high degree of confidence it's not being used and if it is then it's not even being maintained as a stable API. Not a great reason to do so, but removing it will at least smoke out anyone using it and we can reassess if we actually find anyone. The less API we have to maintain the better! |
@gibfahn Are you satisfied at this point that this has received sufficient CTC review that you are comfortable landing it? @Fishrock123 It's unclear (to me, at least) if you are objecting in a "don't do this" kind of way or in a "-0 but I won't stop it if everyone else thinks it's a good idea" kind of way. Can you clarify? |
@Trott yeah, I think everyone should at least have been made aware of it by now. |
f9e4e5e
to
ef57d09
Compare
The bundled c-ares isn't very suitable for consumption by addons, isn't kept stable, and isn't exported on windows. PR-URL: nodejs#10283 Refs: nodejs/node-gyp#1055 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com> Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
ef57d09
to
a1028d5
Compare
Checklist
make -j4 test
(UNIX), orvcbuild test nosign
(Windows) passesAffected core subsystem(s)
build
Description of change
The bundled c-ares isn't very suitable for consumption by addons,
isn't kept stable, and isn't exported on windows.
Ref: nodejs/node-gyp#1055
CI: https://ci.nodejs.org/job/node-test-commit/6660/