Release proposal: v1.5.0

[rvagg]

Lots of stuff has landed, we have the new Buffer#indexOf() and I see a bunch of stuff on the verge of landing too--which may make it in to this release if we give it a couple of days. I'm calling 48 hours (approx) on this release.

Again it'd be nice to have attention on #1005 if we can have some @iojs/platform-windows people (or non-Windows people even! It shouldn't be that hard) to focus on some fixes to reduce the error count.

• [b27931b0fe] - benchmark: fix wrk check (Brian White) #1076
• [2b79052494] - benchmark: check for wrk ahead of running benchmarks (Johan Bergström) #982
• [31421afe89] - buffer: reword Buffer.concat error message (Chris Dickinson) joyent/node#8723
• [78581c8d90] - (SEMVER-MINOR) buffer: add indexOf() method (Trevor Norris) #561
• [37bb1df7c4] - build: remove mdb from io.js (Johan Bergström) #1023
• [726671cb0e] - build: add basic mips/mipsel support (Ben Noordhuis) #1045
• [a45d4f8fd6] - build: remove tools/wrk from the tree (Johan Bergström) #982
• [dee07e2983] - deps: make node-gyp work with io.js (cjihrig) #990
• [fe14802fb7] - deps: upgrade npm to 2.7.0 (Forrest L Norvell) #1080
• [31142415de] - doc: add TC meeting 2015-02-18 minutes (Rod Vagg) #1051
• [6190a2236b] - doc: remove cjihrig from TC (cjihrig) #1056
• [9741291fe9] - doc: fix child_process heading depth (Sam Roberts) #1038
• [c8110692a5] - doc: add explanations for querystring (Robert Kowalski) joyent/node#9259
• [8fb711e06c] - doc: fix default value of opts.decodeURIComponent (h7lin) joyent/node#9259
• [6433ad1eef] - doc: add missing newline in CHANGELOG (Rod Vagg)
• [555a7c48cf] - events: optimize listener array cloning (Brian White) #1050
• [4d0329ebeb] - (SEMVER-MINOR) fs: remove unnecessary usage of .hasOwnProperty() (Jonathan Ong) #635
• [4874182065] - http: send Content-Length when possible (Christian Tellnes) #1062
• [08133f45c7] - http: optimize outgoing requests (Brendan Ashworth) #605
• [dccb69a21a] - js_stream: fix leak of instances (Fedor Indutny) #1078
• [4ddd6406ce] - lib: avoid .toLowerCase() call in Buffer#write() (Ben Noordhuis) #1048
• [bbf54a554a] - lib: hand-optimize Buffer constructor (Ben Noordhuis) #1048
• [9d2b89d06c] - net: allow port 0 in connect() (cjihrig) joyent/node#9268
• [e0835c9cda] - node: improve performance of nextTick (Trevor Norris) #985
• [8f5f12bb48] - smalloc: export constants from C++ (Vladimir Kurchatkin) #920
• [0697f8b44d] - smalloc: validate arguments in js (Vladimir Kurchatkin) #920
• [1640dedb3b] - src: fix ucs-2 buffer encoding regression (Ben Noordhuis) #1042
• [2eda2d6096] - src: fix external string length calculation (Ben Noordhuis) #1042
• [4aea16f214] - src: rename confusingly named local variable (Ben Noordhuis) #1042
• [c9ee654290] - src: simplify node::Utf8Value() (Ben Noordhuis) #1042
• [364cc7e08a] - src: remove NODE_INVALID_UTF8 environment variable (Ben Noordhuis) #1042
• [826cde8661] - src: fix gc heuristic for external twobyte strings (Ben Noordhuis) #1042
• [f5b7e18243] - src: remove unused code (Ben Noordhuis) #1042
• [4ae64b2626] - src: extract node env init out of process init (Petka Antonov) #980
• [b150c9839e] - src: fix -Wempty-body compiler warnings (Ben Noordhuis) #974
• [fb284e2e4d] - src: fix compiler warning in smalloc.cc (Ben Noordhuis) #1055
• [583a868bcd] - stream_wrap: add HandleScope's in uv callbacks (Fedor Indutny) #1078
• [e2fb733a95] - test: simplify parallel/test-stringbytes-external (Ben Noordhuis) #1042
• [7b554b1a8f] - test: don't spawn child processes in domain test (Ben Noordhuis) #974
• [b72fa03057] - test: adds a test for undefined value in setHeader (Ken Perkins) #970
• [563771d8b1] - test: split parts out of host-headers test into its own test (Johan Bergström) #1049
• [671fbd5a9d] - test: refactor all tests that depends on crypto (Johan Bergström) #1049
• [c7ad320472] - test: check for openssl cli and provide path if it exists (Johan Bergström) #1049
• [71776f9057] - test: remove unused https imports (Johan Bergström) #1049
• [3d5726c4ad] - test: introduce a helper that checks if crypto is available (Johan Bergström) #1049
• [d0e7c359a7] - test: don't assume process.versions.openssl always is available (Johan Bergström) #1049
• [e1bf6709dc] - test: fix racey-ness in tls-inception (Fedor Indutny) #1040
• [fd3ea29902] - test: fix test-fs-access when uid is 0 (Johan Bergström) #1037
• [5abfa930b8] - test: make destroyed-socket-write2.js more robust (Michael Dawson) joyent/node#9270
• [1009130495] - tests: fix race in test-http-curl-chunk-problem (Julien Gilli)
• [bd1bd7e38d] - timer: Improve performance of callbacks (Ruben Verborgh) #406
• [7b3b8acfa6] - tls: accept empty net.Sockets (Fedor Indutny) #1046
• [c09c90c1a9] - tls_wrap: do not hold persistent ref to parent (Fedor Indutny) #1078
• [3446ff417b] - tty: do not add shutdown method to handle (Fedor Indutny) #1073
• [abb00cc915] - url: throw for invalid values to url.format (Christian Tellnes) #1036
• [abd3ecfbd1] - win,test: fix test-stdin-from-file (Bert Belder)

[rvagg]

commit log doubled overnight

[Fishrock123]

win,test: fix test-stdin-from-file (Bert Belder)

Guessing the commit tool does not parse multiple subsystems?

[rvagg]

Guessing the commit tool does not parse multiple subsystems?

Yeah, it does as of changelog-maker@1.2.1, fixed it just after updating here, I've updated again with the new output.

[rvagg]

https://jenkins-iojs.nodesource.com/job/iojs+any-pr+multi/246/

Current v1.x, with expected failures and still ~8 failures for Windows.

[Starefossen]

Excited!

[rvagg]

https://iojs.org/download/nightly/v1.4.4-nightly20150305b27931b0fe/

consider that an RC, we'll roll another one tomorrow before release build

[qfox]

I need nighty build to make test cases for win. Not sure how to install the latest version from sources.

Oops. Sorry. https://iojs.org/download/nightly/

[Fishrock123]

@zxqfox there are both .msi installers there and windows binaries in the folders at the top. :)

[Fishrock123]

If possible, #1080 (npm@2.7.0) and #1078 (tls mem leak fix) (both landed) would be good to make it into this.

[qfox]

Just fyi: I can't repeat errors with the current v1.4.4-nightly20150305b27931b0fe, that's great ;-)

@Fishrock123 Yeah, by mad. ;-)

[Fishrock123]

https://jenkins-iojs.nodesource.com/view/iojs/job/iojs+any-pr+multi/250/ -- looks like the test for #1068 is (still / again) failing on windows.

[rvagg]

at this stage I'm not going to let existing errors on Windows hold up on a release but we should definitely be on a path towards reducing the number of errors

[rvagg]

https://iojs.org/download/nightly/v1.4.4-nightly20150306dee07e2983/

RC just now with a bunch of new stuff in it including the (hopeful) TLSWrap leak fixes reported in #1075. Please test.

[snostorm]

Ping @iojs/website that we're close to a new release. I'll stalk this thread myself for the next few hours.