lib: fix error name #26738

BridgeAR · 2019-03-18T13:20:57Z

This makes sure Node.js errors have a "regular" error name while keeping the error code in the stack trace. Please have a look at the individual commits for a more detailed description.

Fixes: #26669
Fixes: #20253

Checklist

make -j4 test (UNIX), or vcbuild test (Windows) passes
tests and/or benchmarks are included
documentation is changed or added
commit message follows commit guidelines

nodejs-github-bot · 2019-03-18T13:20:59Z

@BridgeAR build started: https://ci.nodejs.org/blue/organizations/jenkins/node-test-pull-request-lite-pipeline/detail/node-test-pull-request-lite-pipeline/2963/pipeline

BridgeAR · 2019-03-18T15:08:28Z

@gabrielschulhof I am not familar with the n-api code and I went ahead to remove the error code from the n-api error names. What I do in JS is to add the code only to the stack trace instead of the name property. Is there a chance to do the same for n-api as well?

BridgeAR · 2019-03-18T16:25:21Z

CI https://ci.nodejs.org/job/node-test-pull-request/21633/ ✔️

joyeecheung · 2019-03-18T19:24:36Z

Note that Error.prototype.toString() is specified

Return the string-concatenation of name, the code unit 0x003A (COLON), the code unit 0x0020 (SPACE), and msg.

If we are going semver-major anyway, I'd prefer we just move the code to the message, and follow the behavior of native errors, that way we don't need to hack around Error.captureStackTrace.

BridgeAR · 2019-03-18T20:28:34Z

@joyeecheung I am aware of the spec but my goal here was to keep the current behavior how the stack is visible while making the sure the name is set to the expected one.

I personally like that the code is not part of the error message.

we don't need to hack around Error.captureStackTrace

I personally would not call it a hack. To me it seems like the hideStackFrame function is easy to use and understand and it has the benefit that the code is a) smaller b) more straight forward (we are able to throw immediately) and c) we always make sure the stack is not calculated twice.

jasnell · 2019-03-18T20:35:03Z

I'm -1 on moving the code to the message

BridgeAR · 2019-03-18T20:44:13Z

@jasnell with my implementation the output is exactly as it was before (the code is not in the message). The only difference is that the name property does not contain the code anymore.

jasnell · 2019-03-18T20:47:05Z

That's fine for me.

…

On Mon, Mar 18, 2019, 15:44 Ruben Bridgewater ***@***.***> wrote: @jasnell <https://github.com/jasnell> with my implementation the output is exactly as it was before (the code is *not* in the message). The only difference is that the name property does not contain the code anymore. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#26738 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAa2eQbNsn2SUj0R7KsQLmF4K095uXeSks5vX_qlgaJpZM4b5wyo> .

Trott · 2019-03-19T01:46:03Z

lib/.eslintrc.yaml

@@ -15,6 +15,8 @@ rules:
    # Config specific to lib
    - selector: "NewExpression[callee.name=/Error$/]:not([callee.name=/^(AssertionError|NghttpError)$/])"
      message: "Use an error exported by the internal/errors module."
+    - selector: "CallExpression[callee.object.name='Error'][callee.property.name='captureStackTrace']"


Given that there are 17 exceptions to this rule in lib, are we sure this is a good rule to have?

It might be difficult to identify the cases which should not use Error.captureStackTrace without this rule.

I count 16 exceptions while 7 of those are all inside internal/errors and I would normally just add a file wide exception explicitly for this case but that would require to write an individual rule instead of using the no-restricted-syntax rule.

Three more cases would ideally also switch to using hideStackFrames() but C++ errors still work quite differently than our JS errors (besides N-API errors which try to mirror the JS side) and handling errors inside callbacks is not yet possible. I already have an idea to improve those further if this PR lands.

That leaves six other cases of which five seem to be exceptions to the rule in a way that they are neither directly related to our errors nor should they ever use hideStackFrames().

I updated the hideStackFrames() function and now there ~~is one~~ are two exception less than before.

Trott · 2019-03-19T01:54:23Z

@joyeecheung I am aware of the spec but my goal here was to keep the current behavior how the stack is visible while making the sure the name is set to the expected one.

I'm inclined to agree with Joyee that moving the code to the message property would be best. First, message is clearly intended to be something human-readable and mutable over time. So that's the place to put custom stuff. Second, the custom error stuff is new enough that a semver major change like that probably won't be catastrophic on the ecosystem. But if we wait a few more releases, it is likely to become impossible to make such a change. So my inclination is: Let's do it now while there's a chance that we still can.

Trott · 2019-03-19T02:19:27Z

Thanks for doing this. It's way more modified files than I would have expected. Yikes!

lib/internal/errors.js

lib/util.js

BridgeAR · 2019-03-19T09:38:47Z

lib/.eslintrc.yaml

@@ -15,6 +15,8 @@ rules:
    # Config specific to lib
    - selector: "NewExpression[callee.name=/Error$/]:not([callee.name=/^(AssertionError|NghttpError)$/])"
      message: "Use an error exported by the internal/errors module."
+    - selector: "CallExpression[callee.object.name='Error'][callee.property.name='captureStackTrace']"


It might be difficult to identify the cases which should not use Error.captureStackTrace without this rule.

I count 16 exceptions while 7 of those are all inside internal/errors and I would normally just add a file wide exception explicitly for this case but that would require to write an individual rule instead of using the no-restricted-syntax rule.

Three more cases would ideally also switch to using hideStackFrames() but C++ errors still work quite differently than our JS errors (besides N-API errors which try to mirror the JS side) and handling errors inside callbacks is not yet possible. I already have an idea to improve those further if this PR lands.

That leaves six other cases of which five seem to be exceptions to the rule in a way that they are neither directly related to our errors nor should they ever use hideStackFrames().

BridgeAR · 2019-03-19T10:07:50Z

@joyeecheung @Trott even if we plan on moving the code inside the error message, I would still like to land this as is instead of working on an completely different approach which would have to go to the TSC as @jasnell pointed out that he's against that approach. This does seem to improve the situation for everyone already by at least removing the code property from the name. Is there a reason not to accept this as is and then to look further into this?

Trott · 2019-03-20T05:37:29Z

Is there a reason not to accept this as is and then to look further into this?

No objection from me on that approach. It's not my first choice, but yeah, sometimes you have to do things in phases and hope it doesn't stall.

BridgeAR · 2019-03-20T12:54:37Z

Just as a side note: using the hideStackFrames() function has another benefit over the former implementation: if such functions are stacked (e.g. some validation functions call other validation functions inside of them), then the most outer function is the one used to hide the stack frame instead of the most inner one as it is the case so far.

BridgeAR · 2019-03-20T14:57:05Z

I just updated this PR with two more commits: the first one improves the hideStackFrames functions (now only Node.js core error frames will be hidden. E.g., programming errors have the full stack trace; if more than a single error is created inside of the wrapped function, all errors hide the internal frames; the function will now work on arbitrary code inside of the wrapped function and does not rely anymore on the return value or thrown errors). The other commit improves the performance to create AssertionErrors (I could move that into a separate PR but it felt natural to this one).

CI https://ci.nodejs.org/job/node-test-pull-request/21687/

BridgeAR · 2019-03-21T18:02:07Z

If we want to remove the code from the name, we should likely do that for v12. Otherwise more people might rely on it and it becomes a more significant breaking change.

BridgeAR · 2019-03-22T03:27:45Z

@joyeecheung

does this affect #23926

AFAIC this has no other impact than the code before. I just remove the name property before calling Error.prepareStackTrace.

BridgeAR · 2019-03-22T16:24:34Z

It would be nice if someone would just confirm their LG right quick before landing.

This updates all Node.js errors by removing the `code` being part of the `name` property. Instead, the name is just changed once on instantiation, the stack is accessed to create the stack as expected and then the `name` property is set back to it's original form. PR-URL: nodejs#26738 Fixes: nodejs#26669 Fixes: nodejs#20253 Reviewed-By: Gus Caplan <me@gus.host> Reviewed-By: Matteo Collina <matteo.collina@gmail.com> Reviewed-By: Michaël Zasso <targos@protonmail.com> Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>

When using `Errors.captureStackFrames` the error's stack property is set again. This adds a helper function that wraps this functionality in a simple API that does not only set the stack including the `code` property but it also improves the performance to create the error. The helper works for thrown errors and errors returned from wrapped functions in case they are Node.js core errors. PR-URL: nodejs#26738 Fixes: nodejs#26669 Fixes: nodejs#20253 Reviewed-By: Gus Caplan <me@gus.host> Reviewed-By: Matteo Collina <matteo.collina@gmail.com> Reviewed-By: Michaël Zasso <targos@protonmail.com> Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>

This adds a custom eslint rule to verify that `Error.captureStackTrace()` is only called if necessary. In most cases the helper function should be used instead. PR-URL: nodejs#26738 Reviewed-By: Gus Caplan <me@gus.host> Reviewed-By: Matteo Collina <matteo.collina@gmail.com> Reviewed-By: Michaël Zasso <targos@protonmail.com> Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>

All calls verify the cb is called, thus using `common.mustCall` is not required inside of this function. PR-URL: nodejs#26738 Reviewed-By: Gus Caplan <me@gus.host> Reviewed-By: Matteo Collina <matteo.collina@gmail.com> Reviewed-By: Michaël Zasso <targos@protonmail.com> Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>

This is a first step to align the n-api errors towards errors created in JS. The stack still has to be updated to add the error code. PR-URL: nodejs#26738 Fixes: nodejs#26669 Fixes: nodejs#20253 Reviewed-By: Gus Caplan <me@gus.host> Reviewed-By: Matteo Collina <matteo.collina@gmail.com> Reviewed-By: Michaël Zasso <targos@protonmail.com> Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>

This improves the performance for AssertionError by deactivating duplicated stack frame creation. PR-URL: nodejs#26738 Reviewed-By: Gus Caplan <me@gus.host> Reviewed-By: Matteo Collina <matteo.collina@gmail.com> Reviewed-By: Michaël Zasso <targos@protonmail.com> Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>

BridgeAR · 2019-03-23T01:58:12Z

Thanks a lot for the reviews.

Landed in c757cb1...afce912 🎉

felixfbecker · 2019-03-25T15:47:59Z

Thank you so much for fixing this

Notable changes: * assert: * improve performance to instantiate errors (Ruben Bridgewater) [#26738](#26738) * validate required arguments (Ruben Bridgewater) [#26641](#26641) * adjust loose assertions (Ruben Bridgewater) [#25008](#25008) * async_hooks: * remove deprecated emitBefore and emitAfter (Matteo Collina) [#26530](#26530) * remove promise object from resource (Andreas Madsen) [#23443](#23443) * bootstrap * make Buffer and process non-enumerable (Ruben Bridgewater) [#24874](#24874) * buffer: * use stricter range checks (Ruben Bridgewater) [#27045](#27045) * harden SlowBuffer creation (ZYSzys) [#26272](#26272) * harden validation of buffer allocation size (ZYSzys) [#26162](#26162) * do proper error propagation in addon methods (Anna Henningsen) [#23939](#23939) * child_process: * change the defaults maxBuffer size (kohta ito) [#27179](#27179) * harden fork arguments validation (ZYSzys) [#27039](#27039) * use non-infinite maxBuffer defaults (kohta ito) [#23027](#23027) * console: * don't use ANSI escape codes when TERM=dumb (Vladislav Kaminsky) [#26261](#26261) * crypto: * remove legacy native handles (Tobias Nießen) [#27011](#27011) * decode missing passphrase errors (Tobias Nießen) [#25208](#25208) * move DEP0113 to End-of-Life (Tobias Nießen) [#26249](#26249) * remove deprecated crypto.\_toBuf (Tobias Nießen) [#25338](#25338) * set `DEFAULT\_ENCODING` property to non-enumerable (Antoine du Hamel) [#23222](#23222) * deps: * silence irrelevant V8 warning (Michaël Zasso) [#26685](#26685) * update postmortem metadata generation script (cjihrig) [#26685](#26685) * V8: un-cherry-pick bd019bd (Refael Ackermann) [#26685](#26685) * V8: cherry-pick 6 commits (Michaël Zasso) [#26685](#26685) * V8: cherry-pick d82c9af (Anna Henningsen) [#26685](#26685) * V8: cherry-pick e5f01ba (Anna Henningsen) [#26685](#26685) * V8: cherry-pick d5f08e4 (Anna Henningsen) [#26685](#26685) * V8: cherry-pick 6b09d21 (Anna Henningsen) [#26685](#26685) * V8: cherry-pick f0bb5d2 (Anna Henningsen) [#26685](#26685) * V8: cherry-pick 5b0510d (Anna Henningsen) [#26685](#26685) * V8: cherry-pick 91f0cd0 (Anna Henningsen) [#26685](#26685) * V8: cherry-pick 392316d (Anna Henningsen) [#26685](#26685) * V8: cherry-pick 2f79d68 (Anna Henningsen) [#26685](#26685) * sync V8 gypfiles with 7.4 (Ujjwal Sharma) [#26685](#26685) * update V8 to 7.4.288.13 (Ujjwal Sharma) [#26685](#26685) * bump minimum icu version to 63 (Ujjwal Sharma) [#25852](#25852) * silence irrelevant V8 warnings (Michaël Zasso) [#25852](#25852) * V8: cherry-pick 7803fa6 (Jon Kunkee) [#25852](#25852) * V8: cherry-pick 58cefed (Jon Kunkee) [#25852](#25852) * V8: cherry-pick d3308d0 (Michaël Zasso) [#25852](#25852) * V8: cherry-pick 74571c8 (Michaël Zasso) [#25852](#25852) * cherry-pick fc0ddf5 from upstream V8 (Anna Henningsen) [#25852](#25852) * sync V8 gypfiles with 7.3 (Ujjwal Sharma) [#25852](#25852) * sync V8 gypfiles with 7.2 (Michaël Zasso) [#25852](#25852) * update V8 to 7.3.492.25 (Michaël Zasso) [#25852](#25852) * add s390 asm rules for OpenSSL-1.1.1 (Shigeki Ohtsu) [#19794](#19794) * sync V8 gypfiles with 7.1 (Refael Ackermann) [#23423](#23423) * update V8 to 7.1.302.28 (Michaël Zasso) [#23423](#23423) * doc: * update behaviour of fs.writeFile (Sakthipriyan Vairamani (thefourtheye)) [#25080](#25080) * add internal functionality details of util.inherits (Ruben Bridgewater) [#24755](#24755) * errors: * update error name (Ruben Bridgewater) [#26738](#26738) * fs: * use proper .destroy() implementation for SyncWriteStream (Matteo Collina) [#26690](#26690) * improve mode validation (Ruben Bridgewater) [#26575](#26575) * harden validation of start option in createWriteStream (ZYSzys) [#25579](#25579) * make writeFile consistent with readFile wrt fd (Sakthipriyan Vairamani (thefourtheye)) [#23709](#23709) * http: * validate timeout in ClientRequest() (cjihrig) [#26214](#26214) * return HTTP 431 on HPE\_HEADER\_OVERFLOW error (Albert Still) [#25605](#25605) * switch default parser to llhttp (Anna Henningsen) [#24870](#24870) * change DEP0066 to a runtime deprecation (Morgan Roderick) [#24167](#24167) * else case is not reachable (szabolcsit) [#24176](#24176) * lib: * move DEP0021 to end of life (cjihrig) [#27127](#27127) * remove Atomics.wake (Gus Caplan) [#27033](#27033) * validate Error.captureStackTrace() calls (Ruben Bridgewater) [#26738](#26738) * refactor Error.captureStackTrace() usage (Ruben Bridgewater) [#26738](#26738) * move DTRACE\_\* probes out of global scope (James M Snell) [#26541](#26541) * deprecate \_stream\_wrap (Sam Roberts) [#26245] (#26245) * don't use `util.inspect()` internals (Ruben Bridgewater) [#24971](#24971) * improve error message for MODULE\_NOT\_FOUND (Ali Ijaz Sheikh) [#25690](#25690) * requireStack property for MODULE\_NOT\_FOUND (Ali Ijaz Sheikh) [#25690](#25690) * move DEP0029 to end of life (cjihrig) [#25377](#25377) * move DEP0028 to end of life (cjihrig) [#25377](#25377) * move DEP0027 to end of life (cjihrig) [#25377](#25377) * move DEP0026 to end of life (cjihrig) [#25377](#25377) * move DEP0023 to end of life (cjihrig) [#25280](#25280) * move DEP0006 to end of life (cjihrig) [#25279](#25279) * remove unintended access to deps/ (Anna Henningsen) [#25138](#25138) * move DEP0120 to end of life (cjihrig) [#24862](#24862) * use ES6 class inheritance style (Ruben Bridgewater) [#24755](#24755) * remove `inherits()` usage (Ruben Bridgewater) [#24755](#24755) * module: * remove dead code (Ruben Bridgewater) [#26983](#26983) * mark DEP0019 as End-of-Life (Ruben Bridgewater) [#26973](#26973) * throw an error for invalid package.json main entries (Ruben Bridgewater) [#26823](#26823) * don't search in require.resolve.paths (cjihrig) [#23683](#23683) * n-api: * remove code from error name (Ruben Bridgewater) [#26738](#26738) * net: * do not manipulate potential user code (Ruben Bridgewater) [#26751](#26751) * emit "write after end" errors in the next tick (Ouyang Yadong) [#24457](#24457) * deprecate \_setSimultaneousAccepts() undocumented function (James M Snell) [#23760](#23760) * net,http2: * merge setTimeout code (ZYSzys) [#25084](#25084) * os: * implement os.type() using uv\_os\_uname() (cjihrig) [#25659](#25659) * process: * global.process, global.Buffer getters (Guy Bedford) [#26882](#26882) * move DEP0062 (node --debug) to end-of-life (Joyee Cheung) [#25828](#25828) * exit on --debug and --debug-brk after option parsing (Joyee Cheung) [#25828](#25828) * improve `--redirect-warnings` handling (Ruben Bridgewater) [#24965](#24965) * readline: * support TERM=dumb (Vladislav Kaminsky) [#26261](#26261) * repl: * add welcome message (gengjiawen) [#25947](#25947) * fix terminal default setting (Ruben Bridgewater) [#26518](#26518) * check colors with .getColorDepth() (Vladislav Kaminsky) [#26261](#26261) * deprecate REPLServer.rli (Ruben Bridgewater) [#26260](#26260) * src: * remove unused INT\_MAX constant (Sam Roberts) [#27078](#27078) * update NODE\_MODULE\_VERSION to 72 (Ujjwal Sharma) [#26685](#26685) * remove `AddPromiseHook()` (Anna Henningsen) [#26574](#26574) * update NODE\_MODULE\_VERSION to 71 (Michaël Zasso) [#25852](#25852) * clean up MultiIsolatePlatform interface (Anna Henningsen) [#26384](#26384) * properly configure default heap limits (Ali Ijaz Sheikh) [#25576](#25576) * remove icuDataDir from node config (GauthamBanasandra) [#24780](#24780) * explicitly allow JS in ReadHostObject (Yang Guo) [#23423](#23423) * update postmortem constant (cjihrig) [#23423](#23423) * update NODE\_MODULE\_VERSION to 68 (Michaël Zasso) [#23423](#23423) * tls: * support TLSv1.3 (Sam Roberts) [#26209](#26209) * return correct version from getCipher() (Sam Roberts) [#26625](#26625) * check arg types of renegotiate() (Sam Roberts) [#25876](#25876) * add code for ERR\_TLS\_INVALID\_PROTOCOL\_METHOD (Sam Roberts) [#24729](#24729) * emit a warning when servername is an IP address (Rodger Combs) [#23329](#23329) * disable TLS v1.0 and v1.1 by default (Ben Noordhuis) [#23814](#23814) * remove unused arg to createSecureContext() (Sam Roberts) [#24241](#24241) * deprecate Server.prototype.setOptions() (cjihrig)[ #23820](#23820) * load NODE\_EXTRA\_CA\_CERTS at startup (Ouyang Yadong) [#23354](#23354) * util: * change inspect compact and breakLength default (Ruben Bridgewater) [#27109](#27109) * improve inspect edge cases (Ruben Bridgewater) [#27109](#27109) * only the first line of the error message (Simon Zünd) [#26685](#26685) * don't set the prototype of callbackified functions (Ruben Bridgewater) [#26893](#26893) * rename callbackified function (Ruben Bridgewater) [#26893](#26893) * increase function length when using `callbackify()` (Ruben Bridgewater) [#26893](#26893) * prevent tampering with internals in `inspect()` (Ruben Bridgewater) [#26577](#26577) * fix proxy inspection (Ruben Bridgewater) [#26241](#26241) * prevent leaking internal properties (Ruben Bridgewater) [#24971](#24971) * protect against monkeypatched Object prototype for inspect() (Rich Trott) [#25953](#25953) * treat format arguments equally (Roman Reiss) [#23162](#23162) * win, fs: * detect if symlink target is a directory (Bartosz Sosnowski) [#23724](#23724) * zlib: * throw TypeError if callback is missing (Anna Henningsen)[ #24929](#24929) * make “bare” constants un-enumerable (Anna Henningsen) [#24824](#24824) PR-URL: #26930

Notable changes: * assert: * validate required arguments (Ruben Bridgewater) [#26641](#26641) * adjust loose assertions (Ruben Bridgewater) [#25008](#25008) * async_hooks: * remove deprecated `emitBefore` and `emitAfter` (Matteo Collina) [#26530](#26530) * remove promise object from resource (Andreas Madsen) [#23443](#23443) * bootstrap: make Buffer and process non-enumerable (Ruben Bridgewater) [#24874](#24874) * buffer: * use stricter range checks (Ruben Bridgewater) [#27045](#27045) * harden `SlowBuffer` creation (ZYSzys) [#26272](#26272) * harden validation of buffer allocation size (ZYSzys) [#26162](#26162) * do proper error propagation in addon methods (Anna Henningsen) [#23939](#23939) * child_process: * remove `options.customFds` (cjihrig) [#25279](#25279) * harden fork arguments validation (ZYSzys) [#27039](#27039) * use non-infinite `maxBuffer` defaults (kohta ito) [#23027](#23027) * console: * don't use ANSI escape codes when `TERM=dumb` (Vladislav Kaminsky) [#26261](#26261) * crypto: * remove legacy native handles (Tobias Nießen) [#27011](#27011) * decode missing passphrase errors (Tobias Nießen) [#25208](#25208) * remove `Cipher.setAuthTag()` and `Decipher.getAuthTag()` (Tobias Nießen) [#26249](#26249) * remove deprecated `crypto._toBuf()` (Tobias Nießen) [#25338](#25338) * set `DEFAULT\_ENCODING` property to non-enumerable (Antoine du Hamel) [#23222](#23222) * deps: * update V8 to 7.4.288.13 (Michaël Zasso, cjihrig, Refael Ackermann) (Anna Henningsen, Ujjwal Sharma) [#26685](#26685) * bump minimum icu version to 63 (Ujjwal Sharma) [#25852](#25852) * update OpenSSL to 1.1.1b (Sam Roberts, Shigeki Ohtsu) [#26327](#26327) * errors: * update error name (Ruben Bridgewater) [#26738](#26738) * fs: * use proper .destroy() implementation for SyncWriteStream (Matteo Collina) [#26690](#26690) * improve mode validation (Ruben Bridgewater) [#26575](#26575) * harden validation of start option in `createWriteStream()` (ZYSzys) [#25579](#25579) * make writeFile consistent with readFile wrt fd (Sakthipriyan Vairamani (thefourtheye)) [#23709](#23709) * http: * validate timeout in `ClientRequest()` (cjihrig) [#26214](#26214) * return HTTP 431 on `HPE_HEADER_OVERFLOW` error (Albert Still) [#25605](#25605) * switch default parser to llhttp (Anna Henningsen) [#24870](#24870) * Runtime-deprecate `outgoingMessage._headers` and `outgoingMessage._headerNames` (Morgan Roderick) [#24167](#24167) * lib: * remove `Atomics.wake()` (Gus Caplan) [#27033](#27033) * move DTRACE\_\* probes out of global scope (James M Snell) [#26541](#26541) * deprecate `_stream_wrap` (Sam Roberts) [#26245](#26245) * use ES6 class inheritance style (Ruben Bridgewater) [#24755](#24755) * module: * remove unintended access to deps/ (Anna Henningsen) [#25138](#25138) * improve error message for MODULE\_NOT\_FOUND (Ali Ijaz Sheikh) [#25690](#25690) * requireStack property for MODULE\_NOT\_FOUND (Ali Ijaz Sheikh) [#25690](#25690) * remove dead code (Ruben Bridgewater) [#26983](#26983) * make `require('.')` never resolve outside the current directory (Ruben Bridgewater) [#26973](#26973) * throw an error for invalid package.json main entries (Ruben Bridgewater) [#26823](#26823) * don't search in `require.resolve.paths` (cjihrig) [#23683](#23683) * net: * remove `Server.listenFD()` (cjihrig) [#27127](#27127) * do not add `.host` and `.port` properties to DNS error (Ruben Bridgewater) [#26751](#26751) * emit "write after end" errors in the next tick (Ouyang Yadong) [#24457](#24457) * deprecate `_setSimultaneousAccepts()` undocumented function (James M Snell) [#23760](#23760) * os: * implement `os.type()` using `uv_os_uname()` (cjihrig) [#25659](#25659) * remove `os.getNetworkInterfaces()` (cjihrig) [#25280](#25280) * process: * make global.process, global.Buffer getters (Guy Bedford) [#26882](#26882) * move DEP0062 (node --debug) to end-of-life (Joyee Cheung) [#25828](#25828) * exit on --debug and --debug-brk after option parsing (Joyee Cheung) [#25828](#25828) * improve `--redirect-warnings` handling (Ruben Bridgewater) [#24965](#24965) * readline: * support TERM=dumb (Vladislav Kaminsky) [#26261](#26261) * repl: * add welcome message (gengjiawen) [#25947](#25947) * fix terminal default setting (Ruben Bridgewater) [#26518](#26518) * check colors with `.getColorDepth()` (Vladislav Kaminsky) [#26261](#26261) * deprecate REPLServer.rli (Ruben Bridgewater) [#26260](#26260) * src: * remove unused `INT_MAX` constant (Sam Roberts) [#27078](#27078) * update `NODE_MODULE_VERSION` to 72 (Ujjwal Sharma) [#26685](#26685) * remove `AddPromiseHook()` (Anna Henningsen) [#26574](#26574) * clean up `MultiIsolatePlatform` interface (Anna Henningsen) [#26384](#26384) * properly configure default heap limits (Ali Ijaz Sheikh) [#25576](#25576) * remove `icuDataDir` from node config (GauthamBanasandra) [#24780](#24780) * tls: * support TLSv1.3 (Sam Roberts) [#26209](#26209) * return correct version from `getCipher()` (Sam Roberts) [#26625](#26625) * check arg types of renegotiate() (Sam Roberts) [#25876](#25876) * add code for `ERR_TLS_INVALID_PROTOCOL_METHOD` (Sam Roberts) [#24729](#24729) * emit a warning when servername is an IP address (Rodger Combs) [#23329](#23329) * disable TLS v1.0 and v1.1 by default (Ben Noordhuis) [#23814](#23814) * remove unused arg to createSecureContext() (Sam Roberts) [#24241](#24241) * deprecate `Server.prototype.setOptions()` (cjihrig) [#23820](#23820) * load `NODE_EXTRA_CA_CERTS` at startup (Ouyang Yadong) [#23354](#23354) * util: * remove `util.print()`, `util.puts()`, `util.debug()` and `util.error()` (cjihrig) [#25377](#25377) * change inspect compact and breakLength default (Ruben Bridgewater) [#27109](#27109) * improve inspect edge cases (Ruben Bridgewater) [#27109](#27109) * only the first line of the error message (Simon Zünd) [#26685](#26685) * don't set the prototype of callbackified functions (Ruben Bridgewater) [#26893](#26893) * rename callbackified function (Ruben Bridgewater) [#26893](#26893) * increase function length when using `callbackify()` (Ruben Bridgewater) [#26893](#26893) * prevent tampering with internals in `inspect()` (Ruben Bridgewater) [#26577](#26577) * prevent Proxy traps being triggered by `.inspect()` (Ruben Bridgewater) [#26241](#26241) * prevent leaking internal properties (Ruben Bridgewater) [#24971](#24971) * protect against monkeypatched Object prototype for inspect() (Rich Trott) [#25953](#25953) * treat format arguments equally (Roman Reiss) [#23162](#23162) * win, fs: * detect if symlink target is a directory (Bartosz Sosnowski) [#23724](#23724) * zlib: * throw TypeError if callback is missing (Anna Henningsen) [#24929](#24929) * make “bare” constants un-enumerable (Anna Henningsen) [#24824](#24824) PR-URL: #26930

BridgeAR added the semver-major PRs that contain breaking changes and should be released in the next major version. label Mar 18, 2019

BridgeAR requested review from mscdex, jasnell, Trott, addaleax and joyeecheung March 18, 2019 13:20

nodejs-github-bot added the lib / src Issues and PRs related to general changes in the lib or src directory. label Mar 18, 2019

BridgeAR force-pushed the fix-error-name branch from f61ffc0 to 455b8e6 Compare March 18, 2019 16:22

devsnek approved these changes Mar 18, 2019

View reviewed changes

Trott reviewed Mar 19, 2019

View reviewed changes

BridgeAR commented Mar 19, 2019

View reviewed changes

BridgeAR requested review from targos, apapirovski, ChALkeR and danbev March 21, 2019 18:03

BridgeAR mentioned this pull request Mar 22, 2019

lib: refactor validators #26809

Closed

4 tasks

joyeecheung mentioned this pull request Mar 22, 2019

CI failures: 20190322 nodejs/reliability#21

Closed

joyeecheung mentioned this pull request Mar 22, 2019

Investigate flaky parallel/test-https-connect-localport #26862

Closed

BridgeAR closed this Mar 23, 2019

BethGriggs mentioned this pull request Mar 26, 2019

v12.0.0 proposal #26930

Merged

jawa-the-hutt mentioned this pull request Apr 29, 2019

Error when installing plugin to project using vue-cli 3 nativescript-vue/vue-cli-plugin-nativescript-vue#11

Closed

BridgeAR deleted the fix-error-name branch January 20, 2020 11:56

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

lib: fix error name #26738

lib: fix error name #26738

BridgeAR commented Mar 18, 2019 •

edited

Loading

nodejs-github-bot commented Mar 18, 2019

BridgeAR commented Mar 18, 2019

BridgeAR commented Mar 18, 2019 •

edited

Loading

joyeecheung commented Mar 18, 2019

BridgeAR commented Mar 18, 2019

jasnell commented Mar 18, 2019

BridgeAR commented Mar 18, 2019

jasnell commented Mar 18, 2019 via email

Trott Mar 19, 2019

BridgeAR Mar 19, 2019

BridgeAR Mar 20, 2019 •

edited

Loading

Trott commented Mar 19, 2019

Trott commented Mar 19, 2019

BridgeAR Mar 19, 2019

BridgeAR commented Mar 19, 2019

Trott commented Mar 20, 2019

BridgeAR commented Mar 20, 2019

BridgeAR commented Mar 20, 2019

BridgeAR commented Mar 21, 2019

BridgeAR commented Mar 22, 2019

BridgeAR commented Mar 22, 2019

BridgeAR commented Mar 23, 2019

felixfbecker commented Mar 25, 2019

lib: fix error name #26738

lib: fix error name #26738

Conversation

BridgeAR commented Mar 18, 2019 • edited Loading

Checklist

nodejs-github-bot commented Mar 18, 2019

BridgeAR commented Mar 18, 2019

BridgeAR commented Mar 18, 2019 • edited Loading

joyeecheung commented Mar 18, 2019

BridgeAR commented Mar 18, 2019

jasnell commented Mar 18, 2019

BridgeAR commented Mar 18, 2019

jasnell commented Mar 18, 2019 via email

Trott Mar 19, 2019

Choose a reason for hiding this comment

BridgeAR Mar 19, 2019

Choose a reason for hiding this comment

BridgeAR Mar 20, 2019 • edited Loading

Choose a reason for hiding this comment

Trott commented Mar 19, 2019

Trott commented Mar 19, 2019

BridgeAR Mar 19, 2019

Choose a reason for hiding this comment

BridgeAR commented Mar 19, 2019

Trott commented Mar 20, 2019

BridgeAR commented Mar 20, 2019

BridgeAR commented Mar 20, 2019

BridgeAR commented Mar 21, 2019

BridgeAR commented Mar 22, 2019

BridgeAR commented Mar 22, 2019

BridgeAR commented Mar 23, 2019

felixfbecker commented Mar 25, 2019

BridgeAR commented Mar 18, 2019 •

edited

Loading

BridgeAR commented Mar 18, 2019 •

edited

Loading

BridgeAR Mar 20, 2019 •

edited

Loading