Prepare release 1.7 #2834

wz1000 · 2022-04-18T08:16:02Z

Waiting for haskell.org to come back...

konn

LGTM as for Splice Plugin!

fendor

LGTM, awesome!

ChangeLog.md

GenChangelogs.hs

fendor · 2022-04-18T09:12:07Z

Actually one question, don't we have to bump the versions of all plugins? Since we are modifying their dependency requirements and have to re-upload them to hackage?

wz1000 · 2022-04-18T09:21:48Z

Actually one question, don't we have to bump the versions of all plugins? Since we are modifying their dependency requirements and have to re-upload them to hackage?

No, I believe hackage revisions should be sufficient for the rest of the plugins.

July541 · 2022-04-18T10:56:27Z

It would be great if we can bundle hls-stylish-haskell-plugin in ghc 9.2 with the new release. #2836

And, should we enabling it in ghc9.0 with a compromise solution? #2820

hasufell · 2022-04-18T11:01:01Z

@wz1000 try

ghcup config set url-source '{"OwnSource": "https://mirror.sjtu.edu.cn/ghcup/yaml/ghcup/data/ghcup-0.0.6.yaml"}'

to enable a mirror.

wz1000 · 2022-04-18T11:06:45Z

ghcup config set url-source '{"OwnSource": "https://mirror.sjtu.edu.cn/ghcup/yaml/ghcup/data/ghcup-0.0.6.yaml"}'

Thanks. See https://gitlab.haskell.org/haskell/haskell-language-server/-/pipelines/50432

wz1000 · 2022-04-18T11:08:04Z

It would be great if we can bundle hls-stylish-haskell-plugin in ghc 9.2 with the new release. #2836

Sure , we can include this PR.

And, should we enabling it in ghc9.0 with a compromise solution? #2820

Its not exactly clear to me what the compromise is.

July541 · 2022-04-18T12:26:54Z

Its not exactly clear to me what the compromise is.

The stylish-haskell 9.0 branch is on my fork, migrate to the official repo is not cheap.

It is worth waiting if we are not eager. #2820 (comment)

pepeiborra · 2022-04-18T19:28:59Z

Actually one question, don't we have to bump the versions of all plugins? Since we are modifying their dependency requirements and have to re-upload them to hackage?

No, I believe hackage revisions should be sufficient for the rest of the plugins.

policeman (build from HEAD) is great for reviewing version bumps, but sadly I have found it doesn't work very well for HLS. If you find a way to make it work nicely, please share!

michaelpj · 2022-04-18T19:31:38Z

No, I believe hackage revisions should be sufficient for the rest of the plugins.

Is this a good idea? Won't that make the previous version unbuildable? Since the versions of the plugins which worked with it before will now not work with it since you'll have mutated the version bounds.

Ailrun

LGTM. Here's my minor comment:

Ailrun · 2022-04-18T19:46:36Z

ChangeLog.md

+- Distribute dynamically linked binaries for HLS to avoid statically linking against GLIBC
+  and system libraries, and to avoid unpredictable failures due to subtle differences
+  between the GHC used to compile HLS and the GHC installed on the users machine
+  (@hasufell, #2675, #2431)


We may need to change https://github.com/haskell/haskell-language-server/blob/9fca35d7277ff4e459b39e0c503931c9339fe394/docs/troubleshooting.md#static-binaries this doc as well

I don't think that is needed, since the same caveats about static binaries apply. The difference is that this time we aren't distributing static binaries but are distributing dynamic binaries. If someone tries to compile static binaries on their own then the caveat still applies (and they will get a warning popup).

@wz1000 I mean, the document mentions "local building" as the easiest way to get a dynamic binary, but now it's probably easier to just download the binary.

pepeiborra · 2022-04-18T20:07:55Z

No, I believe hackage revisions should be sufficient for the rest of the plugins.

Is this a good idea? Won't that make the previous version unbuildable? Since the versions of the plugins which worked with it before will now not work with it since you'll have mutated the version bounds.

I second this, previous versions would not become unbuildable if the change was only increasing the upper bound. I question the choice of clamped ^> version bounds.

wz1000 · 2022-04-18T20:11:39Z

Is this a good idea? Won't that make the previous version unbuildable? Since the versions of the plugins which worked with it before will now not work with it since you'll have mutated the version bounds.

I second this, previous versions would not become unbuildable if the change was only increasing the upper bound. I question the choice of clamped ^> version bounds.

Thanks for noticing this, I will relax the bounds on the plugins which haven't been bumped.

eddiemundo

LGTM!

isovector

lgtm for tactics

wz1000 · 2022-04-19T07:35:15Z

I have relaxed the bounds on all the packages for which the version wasn't bumped.

It has been changed to `cabal-ghc92.project`

9.2.2 is broken (https://gitlab.haskell.org/ghc/ghc/-/issues/21196)

wz1000 requested review from jneira, hasufell, pepeiborra, isovector, Ailrun, konn, kokobd, eddiemundo, berberman, georgefst, fendor and drsooch as code owners April 18, 2022 08:16

konn approved these changes Apr 18, 2022

View reviewed changes

fendor approved these changes Apr 18, 2022

View reviewed changes

ChangeLog.md Outdated Show resolved Hide resolved

GenChangelogs.hs Outdated Show resolved Hide resolved

wz1000 force-pushed the wip/1.7-release-prep branch from 3ae38b3 to f31ea85 Compare April 18, 2022 10:53

wz1000 force-pushed the wip/1.7-release-prep branch from f31ea85 to 9fca35d Compare April 18, 2022 11:05

xsebek mentioned this pull request Apr 18, 2022

Completions broken for pre-qualified import #2824

Closed

Ailrun approved these changes Apr 18, 2022

View reviewed changes

eddiemundo approved these changes Apr 18, 2022

View reviewed changes

isovector approved these changes Apr 18, 2022

View reviewed changes

wz1000 force-pushed the wip/1.7-release-prep branch from 9fca35d to fb983b8 Compare April 19, 2022 07:34

wz1000 force-pushed the wip/1.7-release-prep branch from fb983b8 to 3ac3c85 Compare April 19, 2022 07:36

wz1000 requested review from michaelpj, guibou and Anton-Latukha as code owners April 19, 2022 07:58

michaelpj approved these changes Apr 19, 2022

View reviewed changes

wz1000 force-pushed the wip/1.7-release-prep branch 5 times, most recently from 9337f58 to 7e1c11e Compare April 20, 2022 07:38

wz1000 added 3 commits April 20, 2022 14:49

Prepare 1.7.0.0

f01cff0

Remove references to cabal-ghc921.project

8d0550b

It has been changed to `cabal-ghc92.project`

gitlab-ci: Add 9.2.1 job for windows

032f192

9.2.2 is broken (https://gitlab.haskell.org/ghc/ghc/-/issues/21196)

wz1000 force-pushed the wip/1.7-release-prep branch from 7e1c11e to 5845c7d Compare April 20, 2022 09:19

Fix build on windows 9.2.2

4592162

wz1000 force-pushed the wip/1.7-release-prep branch from 5845c7d to 4592162 Compare April 20, 2022 11:06

wz1000 merged commit 641eed4 into master Apr 21, 2022

Uh oh!

Prepare release 1.7 #2834

Prepare release 1.7 #2834

Uh oh!

Conversation

wz1000 commented Apr 18, 2022 • edited by gitpod-io bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

konn left a comment

Choose a reason for hiding this comment

Uh oh!

fendor left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

fendor commented Apr 18, 2022

Uh oh!

wz1000 commented Apr 18, 2022

Uh oh!

July541 commented Apr 18, 2022

Uh oh!

hasufell commented Apr 18, 2022

Uh oh!

wz1000 commented Apr 18, 2022

Uh oh!

wz1000 commented Apr 18, 2022

Uh oh!

July541 commented Apr 18, 2022

Uh oh!

pepeiborra commented Apr 18, 2022

Uh oh!

michaelpj commented Apr 18, 2022

Uh oh!

Ailrun left a comment

Choose a reason for hiding this comment

Uh oh!

Ailrun Apr 18, 2022

Choose a reason for hiding this comment

Uh oh!

wz1000 Apr 20, 2022

Choose a reason for hiding this comment

Uh oh!

Ailrun Apr 20, 2022

Choose a reason for hiding this comment

Uh oh!

pepeiborra commented Apr 18, 2022

Uh oh!

wz1000 commented Apr 18, 2022

Uh oh!

eddiemundo left a comment

Choose a reason for hiding this comment

Uh oh!

isovector left a comment

Choose a reason for hiding this comment

Uh oh!

wz1000 commented Apr 19, 2022

Uh oh!

Uh oh!

wz1000 commented Apr 18, 2022 •

edited by gitpod-io bot

Loading