[6.x] n-api: backport to v6.x#19447
[6.x] n-api: backport to v6.x#19447gabrielschulhof wants to merge 183 commits intonodejs:v6.x-stagingfrom
Conversation
95aece3 to
17f4970
Compare
|
Fixes nodejs/abi-stable-node#298. |
|
@gabrielschulhof great work. Is this landed with the experimental flag in place? We may want to consider doing a semver minor for this... not 100%. We were not planning another minor for 6.x before it went into maintenance mode though /cc @nodejs/lts, thoughts? |
|
The experimental flag is not in place in this patch. It captures the full
circle of having it in, and then having it removed.
…On Mon, Mar 19, 2018 at 9:30 PM, Myles Borins ***@***.***> wrote:
@gabrielschulhof <https://github.com/gabrielschulhof> great work. Is this
landed with the experimental flag in place? We may want to consider doing a
semver minor for this... not 100%. We were not planning another minor for
6.x before it went into maintenance mode though
/cc @nodejs/lts <https://github.com/orgs/nodejs/teams/lts>, thoughts?
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#19447 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AA7k0TXggTNRk5Bk0u7kKq6HSKoLYgNbks5tgFufgaJpZM4SwaB6>
.
|
|
Our hope is to be able to claim, when 10.0.0 arrives, that N-API is now
stable in the latest of each of 6.x, 8.x, and 10.x.
On Mon, Mar 19, 2018 at 9:54 PM, Schulhof, Gabriel <
gabriel.schulhof@intel.com> wrote:
… The experimental flag is not in place in this patch. It captures the full
circle of having it in, and then having it removed.
On Mon, Mar 19, 2018 at 9:30 PM, Myles Borins ***@***.***>
wrote:
> @gabrielschulhof <https://github.com/gabrielschulhof> great work. Is
> this landed with the experimental flag in place? We may want to consider
> doing a semver minor for this... not 100%. We were not planning another
> minor for 6.x before it went into maintenance mode though
>
> /cc @nodejs/lts <https://github.com/orgs/nodejs/teams/lts>, thoughts?
>
> —
> You are receiving this because you were mentioned.
> Reply to this email directly, view it on GitHub
> <#19447 (comment)>, or mute
> the thread
> <https://github.com/notifications/unsubscribe-auth/AA7k0TXggTNRk5Bk0u7kKq6HSKoLYgNbks5tgFufgaJpZM4SwaB6>
> .
>
|
|
... and binary compatible in all three, of course!
On Mon, Mar 19, 2018 at 9:56 PM, Schulhof, Gabriel <
gabriel.schulhof@intel.com> wrote:
… Our hope is to be able to claim, when 10.0.0 arrives, that N-API is now
stable in the latest of each of 6.x, 8.x, and 10.x.
On Mon, Mar 19, 2018 at 9:54 PM, Schulhof, Gabriel <
***@***.***> wrote:
> The experimental flag is not in place in this patch. It captures the full
> circle of having it in, and then having it removed.
>
> On Mon, Mar 19, 2018 at 9:30 PM, Myles Borins ***@***.***>
> wrote:
>
>> @gabrielschulhof <https://github.com/gabrielschulhof> great work. Is
>> this landed with the experimental flag in place? We may want to consider
>> doing a semver minor for this... not 100%. We were not planning another
>> minor for 6.x before it went into maintenance mode though
>>
>> /cc @nodejs/lts <https://github.com/orgs/nodejs/teams/lts>, thoughts?
>>
>> —
>> You are receiving this because you were mentioned.
>> Reply to this email directly, view it on GitHub
>> <#19447 (comment)>, or mute
>> the thread
>> <https://github.com/notifications/unsubscribe-auth/AA7k0TXggTNRk5Bk0u7kKq6HSKoLYgNbks5tgFufgaJpZM4SwaB6>
>> .
>>
>
>
|
|
There is another PR that was just opened that would be semver minor against 6.x as well I'll open an issue in the release repo tomorrow to discuss it |
|
To that end I've actually tested the following sequence:
```
git checkout v6.x-backport-n-api
git clean -x -d -f -f
./configure
make -j4
make -j4 test-addons-napi
git checkout v8.x-backport-n-api-updated-squashed
./configure
make -j4
/usr/bin/python2.7 tools/test.py --mode=release addons-napi
git checkout master
./configure
make -j4
/usr/bin/python2.7 tools/test.py --mode=release addons-napi
```
... and I've found it to work. That is, test/addons-napi can be run without
rebuilding.
On Mon, Mar 19, 2018 at 9:57 PM, Schulhof, Gabriel <
gabriel.schulhof@intel.com> wrote:
… ... and binary compatible in all three, of course!
On Mon, Mar 19, 2018 at 9:56 PM, Schulhof, Gabriel <
***@***.***> wrote:
> Our hope is to be able to claim, when 10.0.0 arrives, that N-API is now
> stable in the latest of each of 6.x, 8.x, and 10.x.
>
> On Mon, Mar 19, 2018 at 9:54 PM, Schulhof, Gabriel <
> ***@***.***> wrote:
>
>> The experimental flag is not in place in this patch. It captures the
>> full circle of having it in, and then having it removed.
>>
>> On Mon, Mar 19, 2018 at 9:30 PM, Myles Borins ***@***.***>
>> wrote:
>>
>>> @gabrielschulhof <https://github.com/gabrielschulhof> great work. Is
>>> this landed with the experimental flag in place? We may want to consider
>>> doing a semver minor for this... not 100%. We were not planning another
>>> minor for 6.x before it went into maintenance mode though
>>>
>>> /cc @nodejs/lts <https://github.com/orgs/nodejs/teams/lts>, thoughts?
>>>
>>> —
>>> You are receiving this because you were mentioned.
>>> Reply to this email directly, view it on GitHub
>>> <#19447 (comment)>, or mute
>>> the thread
>>> <https://github.com/notifications/unsubscribe-auth/AA7k0TXggTNRk5Bk0u7kKq6HSKoLYgNbks5tgFufgaJpZM4SwaB6>
>>> .
>>>
>>
>>
>
|
|
@gabrielschulhof thanks for all the hard work on this. |
|
@mhdawson NP! Happy to help! |
17f4970 to
f0872f3
Compare
f0872f3 to
a80ec42
Compare
|
@MylesBorins I have now re-done this backport with complete commits as much as possible, much like the v8.x backport. |
a80ec42 to
2966812
Compare
|
I think I'm in favor. It would be nice to have N-API in all of the supported release lines. |
|
I’m in favor! |
|
204 commits on 569 files with less than a month before maintenance... we don't even have time to follow our process for testing a semver-minor. I'm floating between -0 and -1. Looking forward to more people chiming in |
|
@gabrielschulhof I think several commits got pulled in that have absolutely no relation to n-api. As an example, the whole reformatting of the docs to 80-chars width has been pulled in, and a change in the eslint. Those would have to be removed IMHO. @MylesBorins those are the the vast majority of the changed and the touched files. |
|
Specifically 66e6d0bdcd346af488aaf3c3d383c5fe204d6c77 and 0b2d5bc3fd1025d9bef2e9934d4b898826b30236 (this one is 243 files!) |
|
wow, lots of work here, +1 from me as long as you think it's fairly low risk .. I haven't reviewed the changeset but if there are any major departures to how it's been applied to 8.x+ could you let us know so we can have some idea of areas of risk? |
The vast majority of the touched files are just for linting changes, so it's very hard to review. IMHO the changes in itself is relatively low risk (as long as it passes CITGM and we can verify that n-api modules works here and in 8), as it's a new feature. |
This adds RegExp or error constructor arguments to the remaining places where it is missing in preparation for the commit that will enforce the presence of at least two arguments. Backport-PR-URL: #19447 PR-URL: #12270 Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: Rich Trott <rtrott@gmail.com>
Missed in ca786c3. This does not actually affect the outcome because returning `nullptr` or `this` from a constructor has the same effect. Backport-PR-URL: #19447 PR-URL: #12318 Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
This API doesn't serve much purpose, and is only likely to cause confusion and bugs. The intention was that this would return the number of characters in a string independent of encoding, but that's not generally useful. In almost all cases, one of the encoding-specific napi_get_value_string_* APIs is more correct. (Pass a null buffer if only the encoded length is desired.) Anyway the current implementation of napi_get_value_string_length() is technically wrong: it returns the number of 2-byte code units of the UTF-16 encoding, but there are actually some characters that are encoded as two UTF-16 code units. Note the JavaScript String.prototype.length property returns the number of UTF-16 code units, which may be different from the number of characters. So, getting the true character count is not common with JavaScript, and is probably best left to specialized internationalization libraries. Backport-PR-URL: #19447 PR-URL: #12496 Fixes: nodejs/abi-stable-node#226 Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com> Reviewed-By: Jeremiah Senkpiel <fishrock123@rocketmail.com> Reviewed-By: James M Snell <jasnell@gmail.com>
We left out null-checks for many of the parameters passed to our APIs. In particular, arguments of type `napi_value` were often accepted without a null-check, even though they should never be null. Additionally, many APIs simply returned `napi_ok` on success. This leaves in place an error that may have occurred in a previous N-API call. Others (those which perform `NAPI_PREAMBLE(env)` at the top) OTOH explicitly clear the last error before proceeding. With this modification all APIs explicitly clear the last error on success. Fixes: nodejs/abi-stable-node#227 Backport-PR-URL: #19447 PR-URL: #12539 Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com> Reviewed-By: Anna Henningsen <anna@addaleax.net> Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
N-API is somewhat strict about blocking calls to many APIs while there is a pending exception. The NAPI_PREAMBLE macro at the beginning of many API implementations checks for a pending exception. However, a subset of the APIs (which don't call back into JavaScript) still need to work while in a pending-exception state. This changes the reference APIs (equivalent to v8::Persistent) and handle scope APIs so that they can be used for cleanup up while an exception is pending. We may decide to similarly enable a few other APIs later, (which would be a non-breaking change) but we know at least these are needed now to unblock some specific scenarios. Fixes: nodejs/abi-stable-node#122 Fixes: nodejs/abi-stable-node#228 Backport-PR-URL: #19447 PR-URL: #12524 Reviewed-By: Anna Henningsen <anna@addaleax.net> Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com> Reviewed-By: Gabriel Schulhof <gabriel.schulhof@intel.com>
- Add a test project to addons-napi that covers the N-API reference and external APIs - Fix a bug in napi_typeof that was found by the new tests Backport-PR-URL: #19447 PR-URL: #12551 Reviewed-By: Anna Henningsen <anna@addaleax.net> Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Background: To enable N-API support for node versions back to v4, the
N-API code can also be built as an external addon. To make maintenance
easier, a single codebase needs to support both built-in and external
scenarios, along with Node versions >= 4 (and corresponding V8
versions).
This change includes several minor fixes to avoid using node
internal APIs and support older V8 versions:
- Expand node::arraysize
- In the CHECK_ENV() macro, return an error code instead of calling
node::FatalError(). This is more consistent with how other invalid
arguments to N-API functions are handled.
- In v8impl::SetterCallbackWrapper::SetReturnValue(), do nothing
instead of calling node::FatalError(). This is more consistent
with JavaScript setter callbacks, where any returned value is
silently ignored.
- When queueing async work items, get the uv default loop instead of
getting the loop from node::Environment::GetCurrent(). Currently
that returns the same loop anyway. If/when node supports multiple
environments, it should have a public API for getting the
environment & event loop, and we can update this implementation
then.
- Use v8::Maybe::FromJust() instead of the newer alias ToChecked()
Backport-PR-URL: #19447
PR-URL: #12674
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Add the initial documentation for the N-API This PR is a result of work in the abi-stable-node repo: https://github.com/nodejs/abi-stable-node/tree/doc, with this PR being the cumulative work on the documentation sections in that repo with the following contributors in alphabetical order: Author: Arunesh Chandra <arunesh.chandra@microsoft.com> Author: Gabriel Schulhof <gabriel.schulhof@intel.com> Author: Hitesh Kanwathirtha <hiteshk@microsoft.com> Author: Jason Ginchereau <jasongin@microsoft.com> Author: Michael Dawson <michael_dawson@ca.ibm.com> Author: Sampson Gao <sampsong@ca.ibm.com> Author: Taylor Woll <taylor.woll@microsoft.com> Backport-PR-URL: #19447 PR-URL: #12549 Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Jeremiah Senkpiel <fishrock123@rocketmail.com> Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Start the transition to Array.prototype.includes() and String.prototype.includes(). This commit refactors most of the comparisons of Array.prototype.indexOf() and String.prototype.indexOf() return values with -1 to the former methods in tests. Backport-PR-URL: #19447 PR-URL: #12604 Refs: #12586 Reviewed-By: Alexey Orlenko <eaglexrlnk@gmail.com> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Gibson Fahnestock <gibfahn@gmail.com>
These APIs do not need a try-catch around their body, because no exceptions are thrown in their implementation: - `napi_is_array()` - `napi_get_value_string_latin1()` - `napi_get_value_string_utf8()` - `napi_get_value_string_utf16()` - `napi_get_value_external()` - `napi_is_buffer()` - `napi_is_arraybuffer()` - `napi_get_arraybuffer_info()` - `napi_is_typedarray()` - `napi_get_typedarray_info()` Fixes: nodejs/abi-stable-node#238 Backport-PR-URL: #19447 PR-URL: #12705 Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com> Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: Jason Ginchereau <jasongin@microsoft.com> Reviewed-By: Anna Henningsen <anna@addaleax.net>
Add coverage for N-API functions related to throwing and creating errors. A number of these are currently showing as not having any coverage in the nightly code coverage reports. Backport-PR-URL: #19447 PR-URL: #12729 Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: James M Snell <jasnell@gmail.com>
Improved test coverage for napi_make_callback by porting the existing addons/make_callback test to n-api Backport-PR-URL: #19447 PR-URL: #12409 Reviewed-By: Anna Henningsen <anna@addaleax.net> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Refael Ackermann <refack@gmail.com>
Backport-PR-URL: #19447 PR-URL: #12864 Ref: #12551 (comment) Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
When the number of args requested is greater than the actual number of args supplied to the function call, the remainder of the args array should be filled in with `undefined` values. Because of this bug, the remainder of the array was left uninitialized, which could cause a crash. Refer to the documentation for the `argv` parameter at https://github.com/nodejs/node/blob/master/doc/api/n-api.md#napi_get_cb_info Backport-PR-URL: #19447 PR-URL: #12863 Reviewed-By: Anna Henningsen <anna@addaleax.net> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
- Create a handle scope before invoking the async completion callback, because it is basically always needed, easy for user code to forget, and this makes it more consistent with ordinary N-API function callbacks. - Check for an unhandled JS exception after invoking an async completion callback, and report it via `node::FatalException()`. - Add a corresponding test case for an exception in async callback. Previously, any unhandled JS exception thrown from a `napi_async_complete_callback` would be silently ignored. Among other things this meant assertions in some test cases could be undetected. Backport-PR-URL: #19447 PR-URL: #12838 Reviewed-By: Anna Henningsen <anna@addaleax.net> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com> Reviewed-By: James M Snell <jasnell@gmail.com>
- fix 2 broken links - fix capitalization in description of napi_create_array-with-length Backport-PR-URL: #19447 PR-URL: #12889 Reviewed-By: Sakthipriyan Vairamani <thechargingvolcano@gmail.com> Reviewed-By: Refael Ackermann <refack@gmail.com> Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: Luigi Pinca <luigipinca@gmail.com> Reviewed-By: Daniel Bevenius <daniel.bevenius@gmail.com> Reviewed-By: MichaëZasso <targos@protonmail.com> Reviewed-By: Gibson Fahnestock <gibfahn@gmail.com> Reviewed-By: Hitesh Kanwathirtha <digitalinfinity@gmail.com>
Backport-PR-URL: #19447 PR-URL: #12898 Fixes: #7129 Reviewed-By: Sam Roberts <vieuxtech@gmail.com> Reviewed-By: Sakthipriyan Vairamani <thechargingvolcano@gmail.com> Reviewed-By: Refael Ackermann <refack@gmail.com> Reviewed-By: Anna Henningsen <anna@addaleax.net> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Rich Trott <rtrott@gmail.com> Reviewed-By: Gibson Fahnestock <gibfahn@gmail.com> Reviewed-By: Timothy Gu <timothygu99@gmail.com> Reviewed-By: Daniel Bevenius <daniel.bevenius@gmail.com> Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Use `common.mustCall()` to confirm that function is invoked. Backport-PR-URL: #19447 PR-URL: #12959 Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: Sakthipriyan Vairamani <thechargingvolcano@gmail.com> Reviewed-By: Gibson Fahnestock <gibfahn@gmail.com> Reviewed-By: Anna Henningsen <anna@addaleax.net> Reviewed-By: James M Snell <jasnell@gmail.com>
Backport-PR-URL: #19447 PR-URL: #12974 Reviewed-By: Gibson Fahnestock <gibfahn@gmail.com> Reviewed-By: Sakthipriyan Vairamani <thechargingvolcano@gmail.com> Reviewed-By: Colin Ihrig <cjihrig@gmail.com> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Jason Ginchereau <jasongin@microsoft.com>
|
landed in e54b8e8...3aa5b7d If anything is out of the ordinary we can find it in the release tests and open separate PRs to help fix. Was able to automate applying the Backport-PR-URL meta data to all the commits using the following command |
src/node_api_backport.cc and src/node_api_backport.h are added
containing stubs for internal APIs used by N-API.
expectsError() is brought in from master and added to
tests/common/index.js.
N-API tests involving the JS async_hooks module are removed.
The following commits are partially or completely applied:
f24d0ec 6a5a9ad d20f1f0 5fb6f7f cd7d7b1 a03c90b
33e63fe 48b5c11 35c7238 d83f830 449d1c8 c698017
a29089d 9cb96ac 945eb1b 6e1c25c 463d1a4 caee112
755e07c a1bab82 2bead4b 1729af2 fe442f6 d3569b6
5ecd2e1 6101bd2 d379db3 a555397 c2b9048 629a447
47f664e 1286923 0c16b18 6e312c5 6c1906a c123467
f054855 f7c709f c90185c 316b5ef f75bc2c 3e06276
8938c4c 91c1ccd 8a86d9c 93acfe5 094d92b 94e2951
246aeac a893e79 ef49f55 d4cd8c2 50afd90 2475676
ff9a6bc 12c8b4d dc389bf 493340f 8f2c366 a4a9a3d
9cf3525 97ba69f 959c425 9b4ab14 51f92b6 e0b1394
3ee524b f2cb78c e04d23c 2336df1 0736ad4 201ecef
e07e708 7be4a84 b3f9b38 6bc82da a1c0804 f881789
3594223 17b818b 4957726 b33b3e1 84579b1 e6e58fd
cec6e21 05e4c1d 8b90250 a406a32 3070d53 a8c0a43
1976654 973c12f 0c258bd 92e5f5c 1a0727d 8c8c90b
290315a cb94905 a10856a 61e9ba1 7828698 1cdb41f
c77e6d3 7efb8f7 6c382de 82bad0b 70664bf e96b949
1fe0741 85d7d97 e6eb5c0 835c383 b72e702 af70c3b
ad664ea f3afe29 6968ead aa6fac6 77ca3cb 9926dfe
7849b52 e59987c 57a4ceb 4d1e086 bb29405 65eefa0
e3f7a54 73078d6 ac41db4 598a128 f52c707 d5b397c
8f3dab4 9c6804c 29df1a8 34cf8ad e36917b c9b6d95
ea927b3 732ae41 f803e77 0f1888f a71d205 ef28d85
3e18c49 d291338 ecf6a46 62e940d ca8a29c 2529119
01f4d9a c28418a df46fcb 7d9dfda ddba969 bd4b790
062071a 7a7ac1c 8ab8c33 fd54b10 bb91879 43e4efd
effeff1 d9ee297 1961900 2af49b6 260cd41 47919b3
f3ef971 bfade5a 4a7b7e8 a63b245 0dd8b9a 0083011
47c3c58 1b28022 9516aa1 abfd4bf 654afa2 147048a
2e3fef7 2bbabb1 0a734fe 73d9c0f 94a120c 8aca66a
cd32b77 a180259 0142276 deb9622 972bfe1 4271254
1d96803 4241577 b7a341d ba7bac5 468275a 6c60691
70b51c8 9d52222 46f2026 ad5f987 8bd26d3 9de2e15
0ec0272 affe0f2 9decfb1 ca786c3 afd5966 8fbace1
8460284 0a5bf4a 491d59d 4a21e39 56e881d
Checklist
make -j4 test(UNIX), orvcbuild test(Windows) passes