Propose v6.2.1 #6934
Closed
Propose v6.2.1 #6934
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
I often want to run a test many times to see if a failure can be recreated and I believe this is a common use case. We even have this job in the CI https://ci.nodejs.org/job/node-stress-single-test/configure but often you want to run it on a specific machine. This patch adds the --repeat option so that you can repeat the selected set of tests a number of times. Given existing options in test.py this will allow you to run one or more tests for the number of repeats specified. For example: tools/test.py -j8 --repeat 1000 parallel/test-process-exec-argv runs the test-process-exec-argv test 1000 times, running 8 copies in parallel tools/test.py --repeat 2 would run the entire test suite twice. PR-URL: #6700 Reviewed-By: Ben Noorhduis <[email protected]> Reviewed-By: James M Snell <[email protected]> Reviewed-By: Fedor Indutny <[email protected]> Reviewed-By: thefourtheye - Sakthipriyan Vairamani <[email protected]> Reviewed-By: joaocgreis - João Reis <[email protected]> Reviewed-By: Johan Bergström <[email protected]>
Avoid using deprecated getter syntax plus other miscellaneous updates. PR-URL: #6766 Reviewed-By: Colin Ihrig <[email protected]>
The hash link format has changed from #HASH_LINK to #HASH-LINK. PR-URL: #6817 Reviewed-By: Roman Klauke <[email protected]> Reviewed-By: James M Snell <[email protected]>
Refs: #6578 PR-URL: #6783 Reviewed-By: James M Snell <[email protected]>
PR-URL: #6812 Reviewed-By: Brian White <[email protected]> Reviewed-By: James M Snell <[email protected]> Reviewed-By: Roman Reiss <[email protected]>
Ref: #6578 PR-URL: #6717 Reviewed-By: Robert Lindstaedt <[email protected]> Reviewed-By: James M Snell <[email protected]>
This is purely cleanup and carries no visible behavioural changes. Up to now, `this._closed` was used in zlib.js as a synonym of `!this._handle`. This change makes this connection explicit and removes the `_closed` property from zlib streams, as the previous duplication has been the cause of subtle errors like #6034. This also makes zlib errors lead to an explicit `_close()` call rather than waiting for garbage collection to clean up the handle, thus returning memory resources earlier in the case of an error. Add a getter for `_closed` so that the property remains accessible by legacy code. PR-URL: #6574 Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: James M Snell <[email protected]>
PR-URL: #6762 Reviewed-By: James M Snell <[email protected]> Reviewed-By: Brian White <[email protected]> Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: Fedor Indutny <[email protected]> Reviewed-By: Johan Bergström <[email protected]>
Previously a `checks.workers` boolean was conditionally set, but never checked. Additionally, it was never actually set because `cluster.onlineWorkers` is always undefined. PR-URL: #6535 Reviewed-By: Anna Henningsen <[email protected]>
Currently we are overwriting the entire env object of the child-process spawned in `npm-test-install`. This commit alternatively clones the `process.env` object and modifies it with the neccessary changes before passing it the the spawned process. Fixes: #6736 PR-URL: #6797 Reviewed-By: Michael Dawson <[email protected]> Reviewed-By: James M Snell <[email protected]>
Per: #6142 Clarify the prefix argument. Fixes: #6142 PR-URL: #6800 Reviewed-By: Ben Noordhuis <[email protected]> Reviewed-By: Roman Klauke <[email protected]>
This commit fixes a regression in basename() for the case when a supplied extension name completely matches the resulting basename. Fixes: #6587 PR-URL: #6590 Reviewed-By: James M Snell <[email protected]> Reviewed-By: Evan Lucas <[email protected]>
The TypeError checks are already done later on in the test file for all path functions. PR-URL: #6590 Reviewed-By: James M Snell <[email protected]> Reviewed-By: Evan Lucas <[email protected]>
PR-URL: #6590 Reviewed-By: James M Snell <[email protected]> Reviewed-By: Evan Lucas <[email protected]>
Make sure that `catch-stdout-error` has written data before the destination process exits. Fixes: #6791 PR-URL: #6808 Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: James M Snell <[email protected]> Reviewed-By: Roman Klauke <[email protected]>
In 5d38d54, an additional property in node_config.cc was added whose definition depends on having the local `env` variable declared, which in turn depended on `NODE_HAVE_I18N_SUPPORT` being defined. Moving `env = ...` out of the `#ifdef` block allows building via `./configure --without-intl` again. PR-URL: #6820 Reviewed-By: James M Snell <[email protected]> Reviewed-By: Myles Borins <[email protected]> Reviewed-By: Ben Noordhuis <[email protected]>
The test is currently flaky and CI provides no real information because the test times out rather than failing on an assertion. Add logging to gather more information about the failure. Refs: #6754 PR-URL: #6769 Reviewed-By: Colin Ihrig <[email protected]>
Remove port increment by `1337` which appears to sometimes result in a port that is already in use. There is no reason not to use `common.PORT`. PR-URL: #6769 Fixes: #6754 Reviewed-By: Colin Ihrig <[email protected]>
Since I was doing the necessary git archaeology anyway, I took the time to add YAML information to the docs about when `addMembership()` and `dropMembership()` first appeared in their current forms. PR-URL: #6753 Ref: #6578 Reviewed-By: James M Snell <[email protected]> Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: Ben Noordhuis <[email protected]>
The only tests for `addMembership()` and `dropMembership()` (from the `dgram` module) were in `test/internet` which means they almost never get run. This adds checks in `test/parallel`. PR-URL: #6753 Reviewed-By: James M Snell <[email protected]> Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: Ben Noordhuis <[email protected]>
On OSX it's possible that the fd is replaced, so use the proper libuv API to get the correct fd. PR-URL: #6753 Reviewed-By: Ben Noordhuis <[email protected]> Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: James M Snell <[email protected]>
PR-URL: #6805 Reviewed-By: Anna Henningsen <[email protected]> Reviewed-By: Ben Noordhuis <[email protected]> Reviewed-By: James M Snell <[email protected]>
Fix `options` usage on `lib/_http_agent.js` for the Legacy API. Fixes: #5051 PR-URL: #5190 Reviewed-By: Ben Noordhuis <[email protected]> Reviewed-By: Brian White <[email protected]> Reviewed-By: James M Snell <[email protected]>
Fixes: #6765 PR-URL: #6809 Reviewed-By: Ben Noordhuis <[email protected]> Reviewed-By: James M Snell <[email protected]> Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: Sakthipriyan Vairamani <[email protected]> Reviewed-By: Michael Dawson <[email protected]>
This makes it so you can see why the check fails if it does. Typically that sort of thing can happen if you are modifying bootstrapping or `process`. PR-URL: #6786 Reviewed-By: Ben Noordhuis <[email protected]> Reviewed-By: Anna Henningsen <[email protected]>
Git logs print my full name Robert Jefe Lindstaedt. When I did #6389 I forgot simply forgot it. PR-URL: #6880 Reviewed-By: Ben Noordhuis <[email protected]>
Updates the dns module documentation to include documentation on the resolveNaptr method, and also adds the option NAPTR to the list of valid values for rrtype in dns.resolve(hostname[, rrtype], callback). PR-URL: #6586 Fixes: #6507 Reviewed-By: Anna Henningsen <[email protected]> Reviewed-By: James M Snell <[email protected]> Reviewed-By: Roman Reiss <[email protected]>
Replace booleans with `common.mustCall()`, migrate from `var` to `const`, and apply minor formatting changes. PR-URL: #6756 Reviewed-By: Fedor Indutny <[email protected]> Reviewed-By: Santiago Gimeno <[email protected]> Reviewed-By: James M Snell <[email protected]>
General improvements to vm module documentation PR-URL: #6827 Reviewed-By: Ben Noordhuis <[email protected]>
In preparation for a lint rule to flag `__defineGetter__`, refactor the one remaining instance in the code base. PR-URL: #6774 Reviewed-By: Ben Noordhuis <[email protected]> Reviewed-By: James M Snell <[email protected]>
This commit provides a rewrite of StringDecoder that both improves performance (for non-single-byte encodings) and understandability. Additionally, StringDecoder instantiation performance has increased considerably due to inlinability and more efficient encoding name checking. PR-URL: #6777 Reviewed-By: James M Snell <[email protected]> Reviewed-By: Anna Henningsen <[email protected]> Reviewed-By: Ben Noordhuis <[email protected]>
A possibly-buggy fixture server uses `common.PORT+1000` for its port rather than `common.PORT`. That could result in it clashing with other ports if tests are run in parallel. The test runner increments `common.PORT` by 100 for each running instance for tests. Change to use common.PORT and have the tests that use the fixture start with common.PORT+1 for anything they need. PR-URL: #6990 Refs: #6989 Reviewed-By: Myles Borins <[email protected]> Reviewed-By: Ben Noordhuis <[email protected]> Reviewed-By: Santiago Gimeno <[email protected]> Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: Michael Dawson <[email protected]>
If a socket is sent to a child, any data that is buffered in the socket will not be sent to the child. The child will only receive data from the socket that is sent after the child has the socket. PR-URL: #6951 Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: Anna Henningsen <[email protected]>
PR-URL: #7032 Reviewed-By: Anna Henningsen <[email protected]> Reviewed-By: Colin Ihrig <[email protected]>
PR-URL: #7006 Reviewed-By: Robert Jefe Lindstaedt <[email protected]> Reviewed-By: James M Snell <[email protected]> Reviewed-By: Jeremiah Senkpiel <[email protected]>
PR-URL: #7007 Reviewed-By: Ben Noordhuis <[email protected]> Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: James M Snell <[email protected]>
According to kill(2), kill returns `EPERM` error if when signalling a process group any of the members could not be signalled. PR-URL: #7046 Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: Ben Noordhuis <[email protected]>
PR-URL: #6984 Reviewed-By: Anna Henningsen <[email protected]> Reviewed-By: Rich Trott <[email protected]> Reviewed-By: Yorkie Liu <[email protected]>
Fishrock123
force-pushed
the
propose-v6.2.1
branch
from
b1a05eb
to
ba67f77
Compare
|
Will probably happen tomorrow (Wednesday) as I'm afk for most of today. |
Add a step in vcbuild.bat to create a minimal package including node and npm that can be used as an alternative to the MSI. Compress the node.pdb file as zip and 7z. All files are uploaded as part of build-release. Reviewed-By: Joao Reis <[email protected]> Reviewed-By: Rod Vagg <[email protected]> Reviewed-By: Alexis Campailla <[email protected]> PR-URL: #5995 Fixes: nodejs/build#299 Fixes: #5696
When debug flags are passed to clustered applications, the debug port is rewritten for each worker process to avoid collisions. Prior to this commit, each debug flag would get a unique value. This commit reworks the logic to assign the same port value to all debug flags for a single worker. PR-URL: #7050 Reviewed-By: Ben Noordhuis <[email protected]> Reviewed-By: Santiago Gimeno <[email protected]>
There are several places in the cluster module where a version of process.send() is called, but the result is swallowed. Most of these cases are internal, but Worker.prototype.send(), which is publicly documented, also suffers from this problem. This commit exposes the return value to facilitate better error handling, and bring Worker.prototype.send() into compliance with the documentation. PR-URL: #6998 Reviewed-By: James M Snell <[email protected]> Reviewed-By: Ron Korving <[email protected]> Reviewed-By: Ben Noordhuis <[email protected]>
- Only `@@toStringTag` affects `util.isError()`, this is the reason why it uses `Object.prototype.toString.call(argument)` under the hood. - Shows an actual Euro symbol for reference. - Uses line-drawing characters for the URL chart & fixes the chart borders. PR-URL: #7026 Reviewed-By: James M Snell <[email protected]> Reviewed-By: Anna Henningsen <[email protected]> Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: Benjamin Gruenbaum <[email protected]>
Treat negative length arguments to `Buffer()`/`allocUnsafe()` as if they were zero so the allocation does not affect the pool’s offset. Fixes: #7047 PR-URL: #7051 Reviewed-By: Sakthipriyan Vairamani <[email protected]> Reviewed-By: Ben Noordhuis <[email protected]> Reviewed-By: Сковорода Никита Андреевич <[email protected]> Reviewed-By: Trevor Norris <[email protected]> Reviewed-By: Rod Vagg <[email protected]>
Allow the operating system to provide an arbitrary available port rather than using `common.PORT`, as `common.PORT` makes it likely that a test will fail with `EADDRINUSE` as a side effect of an earlier test. PR-URL: #7013 Reviewed-By: Brian White <[email protected]> Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: James M Snell <[email protected]> Reviewed-By: Ben Noordhuis <[email protected]>
Previous version of weak used for gc tests emitted a warning on OS X. Updating to current version eliminates warning. PR-URL: #7014 Reviewed-By: Ben Noordhuis <[email protected]> Reviewed-By: James M Snell <[email protected]>
test-https-strict sometimes fails with EADDRINUSE in CI. Remove use of common.PORT to make the test resistant from side effects from other tests that may have not freed up the port. PR-URL: #7024 Reviewed-By: Brian White <[email protected]> Reviewed-By: Anna Henningsen <[email protected]> Reviewed-By: Colin Ihrig <[email protected]>
test-child-process-fork-net will sometimes fail in CI with EADDRINUSE because an earlier test failed to free common.PORT. Have the operating system provide an available port instead. PR-URL: #7033 Reviewed-By: Brian White <[email protected]> Reviewed-By: Ben Noordhuis <[email protected]> Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: Anna Henningsen <[email protected]>
Passing the uid via v8::Integer::New() converts it to a uint32_t. Which will trim the value early. Instead use v8::Number::New() to convert the int64_t to a double so that JS can see the full 2^53 range of uid's. PR-URL: #7096 Reviewed-By: Anna Henningsen <[email protected]> Reviewed-By: Colin Ihrig <[email protected]> Reviewed-By: Andreas Madsen <[email protected]>
OS X has a tiny 1kb hard-coded buffer size for stdout / stderr to TTYs (terminals). Output larger than that causes chunking, which ends up having some (very small but existent) delay past the first chunk. That causes two problems: 1. When output is written to stdout and stderr at similar times, the two can become mixed together (interleaved). This is especially problematic when using control characters, such as \r. With interleaving, chunked output will often have lines or characters erased unintentionally, or in the wrong spots, leading to broken output. CLI apps often extensively use such characters for things such as progress bars. 2. Output can be lost if the process is exited before chunked writes are finished flushing. This usually happens in applications that use `process.exit()`, which isn't infrequent. See #6980 for more info. This became an issue as result of the Libuv 1.9.0 upgrade. A fix to an unrelated issue broke a hack previously required for the OS X implementation. This resulted in an unexpected behavior change in node. The 1.9.0 upgrade was done in c3cec1e, which was included in v6.0.0. Full details of the Libuv issue that induced this are at #6456 (comment) Refs: #1771 Refs: #6456 Refs: #6773 Refs: #6816 PR-URL: #6895 Reviewed-By: Rod Vagg <[email protected]> Reviewed-By: Anna Henningsen <[email protected]>
Many thanks to thefourtheye and addaleax who helped make the python bits of this possible. See #6980 for more info regarding the related TTY issues. Refs: #6456 Refs: #6773 Refs: #6816 PR-URL: #6895 Reviewed-By: Rod Vagg <[email protected]> Reviewed-By: Anna Henningsen <[email protected]>
|
I'm sick and really don't have the energy to complete this today. I'm sorry. The most important things for the changelog are:
|
|
@Fishrock123 anything I can help with? |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
New plan is to do an RC with #6895 this week, and continue to work on that pull request's test before doing a release next week.
Proposing a patch this week.There are two minor commits but they are low priority. Above all, I'd like to get #6895 in (The OS X stdio fix), so this may be delayed a day or two.@nodejs/release