| FazBrowse GitHub Viewer | Trending | | Home |
| Tools: [Download Repo ZIP] [Original HTTPS Page] |
|
@bnoordhuis build started: https://ci.nodejs.org/blue/organizations/jenkins/node-test-pull-request-lite-pipeline/detail/node-test-pull-request-lite-pipeline/2751/pipeline |
Sorry, something went wrong.
There was a problem hiding this comment.
LGTM with the doc comments being addressed.
Sorry, something went wrong.
There was a problem hiding this comment.
A suggestion about the naming and docs, but otherwise LGTM, thanks.
Sorry, something went wrong.
|
Thanks for the reviews, feedback incorporated. I also added a regression test (forgot to check it in.) |
Sorry, something went wrong.
|
@bnoordhuis no comment on calling it tls.DEFAULT_CA like the other defaults? |
Sorry, something went wrong.
No strong opinion anyway. :-) I was waiting to see if anyone else commented on that. FWIW, if I had to pick something in all caps, I'd probably opt for tls.CA_ROOTS or something like that. |
Sorry, something went wrong.
|
To be clear, I'm not opionated about the capsiness of the name, they both look like fine names, it's this story that seem strange to me: TLS exposes default values of some of the options as module properties:
and now:
Perhaps I'm unreasonably perturbed by pattern breaks? |
Sorry, something went wrong.
|
The tls.DEFAULT_FOO exports are reassignable (and the new value gets picked up the next time a connection is created) but this one isn't, that's why I picked a different style for the name. I suppose making the default CA roots fully overridable by assigning to tls.DEFAULT_CA or whatever isn't completely out of the question but I don't know. |
Sorry, something went wrong.
|
I'd really prefer if we could move away from that DEFAULT_FOO setter pattern on the module as it's just not going to work for the ESM side of things. That's not blocking for this PR but we need to consider an alternative approach at some point here. |
Sorry, something went wrong.
|
@jasnell That is a bit of a side-track, but what would it have to look like to be esm-ok? @bnoordhuis I see your point about not being able to be used to change the default, so being a bit different from the other DEFAULT_ definitions. I suspect reassignability might come as a feature request, and eventually we'll make it possible. It looks like you set the descriptor so that it will throw if anyone tries to set it, so that's good, nobody can complain that they set it and it didn't do anything. That was good enough for me, YMMV. |
Sorry, something went wrong.
|
I added one more sanity check to the test: https://ci.nodejs.org/job/node-test-pull-request/21294/ |
Sorry, something went wrong.
|
@sam-github @jasnell is this ready to land out of your perspective? (I did not check the comments fully) |
Sorry, something went wrong.
|
Would be good to get some feedback by @jasnell on esm-safety, but given this pattern is used for other top-level vars in tls, I can't see why it would be blocking. Would be good to have some @nodejs/collaborators weigh in on the naming. Its bike shedding, but we'll have to live with the shed for a long time, best be happy with it now. |
Sorry, something went wrong.
Since you're inviting the bike shed discussion: I'd prefer rootCertificates to roots. |
Sorry, something went wrong.
|
Is there consensus on the name? |
Sorry, something went wrong.
There was a problem hiding this comment.
LGTM I don't care about the names.
Sorry, something went wrong.
|
If bike shedding, I would also prefer rootCertificates (+0) to roots, but regardless the name am very happy to see this land. LGTM. |
Sorry, something went wrong.
|
I agree with @ronkorving and @Trott that rootCertificates is a nicer name but in the end I am fine with either. |
Sorry, something went wrong.
Sorry, something went wrong.
Sorry, something went wrong.
|
Rebased one more time. CI: https://ci.nodejs.org/job/node-test-pull-request/23139/ |
Sorry, something went wrong.
Sorry, something went wrong.
|
Again because 11:23:07 collect2: fatal error: ld terminated with signal 9 [Killed]: |
Sorry, something went wrong.
Sorry, something went wrong.
Sorry, something went wrong.
Fixes: nodejs#25824 PR-URL: nodejs#26415 Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Luigi Pinca <luigipinca@gmail.com> Reviewed-By: Matteo Collina <matteo.collina@gmail.com> Reviewed-By: Rich Trott <rtrott@gmail.com> Reviewed-By: Ron Korving <ron@ronkorving.nl> Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de> Reviewed-By: Sam Roberts <vieuxtech@gmail.com> Reviewed-By: Vse Mozhet Byt <vsemozhetbyt@gmail.com>
Fixes: #25824 PR-URL: #26415 Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Luigi Pinca <luigipinca@gmail.com> Reviewed-By: Matteo Collina <matteo.collina@gmail.com> Reviewed-By: Rich Trott <rtrott@gmail.com> Reviewed-By: Ron Korving <ron@ronkorving.nl> Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de> Reviewed-By: Sam Roberts <vieuxtech@gmail.com> Reviewed-By: Vse Mozhet Byt <vsemozhetbyt@gmail.com>
Notable changes:
* process:
* Log errors using `util.inspect` in case of fatal exceptions
(Ruben Bridgewater) #27243
* repl:
* Add `process.on('uncaughtException')` support (Ruben Bridgewater)
#27151
* stream:
* Implemented `Readable.from` async iterator utility (Guy Bedford)
#27660
* tls:
* Expose built-in root certificates (Ben Noordhuis)
#26415
* Support `net.Server` options (Luigi Pinca)
#27665
* Expose `keylog` event on TLSSocket (Alba Mendez)
#27654
* worker:
* Added the ability to unshift messages from the `MessagePort`
(Anna Henningsen) #27294
Notable changes:
* esm:
* Added the `--experimental-wasm-modules` flag to support
WebAssembly modules (Myles Borins & Guy Bedford)
#27659
* process:
* Log errors using `util.inspect` in case of fatal exceptions
(Ruben Bridgewater) #27243
* repl:
* Add `process.on('uncaughtException')` support (Ruben Bridgewater)
#27151
* stream:
* Implemented `Readable.from` async iterator utility (Guy Bedford)
#27660
* tls:
* Expose built-in root certificates (Ben Noordhuis)
#26415
* Support `net.Server` options (Luigi Pinca)
#27665
* Expose `keylog` event on TLSSocket (Alba Mendez)
#27654
* worker:
* Added the ability to unshift messages from the `MessagePort`
(Anna Henningsen) #27294
PR-URL: #27799
Notable changes:
* esm:
* Added the `--experimental-wasm-modules` flag to support
WebAssembly modules (Myles Borins & Guy Bedford)
#27659
* process:
* Log errors using `util.inspect` in case of fatal exceptions
(Ruben Bridgewater) #27243
* repl:
* Add `process.on('uncaughtException')` support (Ruben Bridgewater)
#27151
* stream:
* Implemented `Readable.from` async iterator utility (Guy Bedford)
#27660
* tls:
* Expose built-in root certificates (Ben Noordhuis)
#26415
* Support `net.Server` options (Luigi Pinca)
#27665
* Expose `keylog` event on TLSSocket (Alba Mendez)
#27654
* worker:
* Added the ability to unshift messages from the `MessagePort`
(Anna Henningsen) #27294
PR-URL: #27799
This is a partial backport of commit f1a3968 ("tls: expose built-in root certificates") from the master branch. The original commit adds a new API, this commit just backports the non-visible changes to ease backporting follow-up commits. Refs: nodejs#26415
This is a partial backport of commit f1a3968 ("tls: expose built-in root certificates") from the master branch. The original commit adds a new API, this commit just backports the non-visible changes to ease backporting follow-up commits. PR-URL: #26415 Backport-PR-URL: #29137 Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Luigi Pinca <luigipinca@gmail.com> Reviewed-By: Matteo Collina <matteo.collina@gmail.com> Reviewed-By: Rich Trott <rtrott@gmail.com> Reviewed-By: Ron Korving <ron@ronkorving.nl> Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de> Reviewed-By: Sam Roberts <vieuxtech@gmail.com> Reviewed-By: Vse Mozhet Byt <vsemozhetbyt@gmail.com>
| Back | FazBrowse Home | New Git URL |
Fixes: #25824