Bug 235093 - Build WebKitSwift when building with make or build-webkit
Summary: Build WebKitSwift when building with make or build-webkit
Status: REOPENED
Alias: None
Product: WebKit
Classification: Unclassified
Component: New Bugs (show other bugs)
Version: WebKit Nightly Build
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Alex Christensen
URL:
Keywords: InRadar
Depends on:
Blocks:
 
Reported: 2022-01-11 20:03 PST by Alex Christensen
Modified: 2022-01-12 17:17 PST (History)
4 users (show)

See Also:


Attachments
Patch (3.64 KB, patch)
2022-01-11 20:04 PST, Alex Christensen
no flags Details | Formatted Diff | Diff
Patch (2.60 KB, patch)
2022-01-12 11:10 PST, Alex Christensen
no flags Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Alex Christensen 2022-01-11 20:03:17 PST
Build WebKitSwift when building with make or build-webkit
Comment 1 Alex Christensen 2022-01-11 20:04:52 PST
Created attachment 448900 [details]
Patch
Comment 2 Alexey Proskuryakov 2022-01-12 09:26:41 PST
Comment on attachment 448900 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=448900&action=review

> Source/WebKit/ChangeLog:9
> +        For various internal reasons, we have to have no depency between the two,
> +        but we command line builders want to build and test it just like everyone else.

Not sure if I know all of the internal reasons, but typically, we add new Xcode projects, not add independent targets to existing ones. Is there a reason why this case has to be different?

> Makefile.shared:113
> +		$1 xcodebuild $2 $(OTHER_OPTIONS) $(XCODE_TARGET) $(XCODE_OPTIONS) $3 $(WEBKIT_ADDITIONAL_XCODE_MAKEFILE_ARGS) | $(OUTPUT_FILTER) && exit $${PIPESTATUS[0]} \

This new variable is vaguely named, but looks like we can just use XCODE_TARGET?
Comment 3 Alex Christensen 2022-01-12 11:10:40 PST
Created attachment 448970 [details]
Patch
Comment 4 Alexey Proskuryakov 2022-01-12 11:22:18 PST
Comment on attachment 448970 [details]
Patch

Seems fine as a temporary approach, as discussed on Slack.
Comment 5 Eric Carlson 2022-01-12 11:22:38 PST
Comment on attachment 448970 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=448970&action=review

> Source/WebKit/ChangeLog:9
> +        but we command line builders want to build and test it just like everyone else.

Do you mean "but *the* command line builders"?
Comment 6 Alex Christensen 2022-01-12 11:24:26 PST
I am a command line builder because I use the command line to build WebKit more often than building from within Xcode.  I believe there are several of us.
Comment 7 EWS 2022-01-12 12:19:10 PST
Committed r287944 (245973@main): <https://commits.webkit.org/245973@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 448970 [details].
Comment 8 Radar WebKit Bug Importer 2022-01-12 12:20:16 PST
<rdar://problem/87476125>
Comment 9 Ryan Haddad 2022-01-12 16:28:08 PST
Monterey builds appear to be failing with errors in WKGroupSession.swift after this change: https://build.webkit.org/#/builders/368/builds/1643/steps/8/logs/stdio
Comment 10 Ryan Haddad 2022-01-12 17:17:36 PST
Reverted r287944 for reason:

Breaks Monterey build.

Committed r287964 (245993@trunk): <https://commits.webkit.org/245993@trunk>