WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
197161
Remove DeprecatedOptional
https://bugs.webkit.org/show_bug.cgi?id=197161
Summary
Remove DeprecatedOptional
Alex Christensen
Reported
2019-04-22 09:07:59 PDT
Remove DeprecatedOptional
Attachments
Patch
(9.68 KB, patch)
2019-04-22 09:09 PDT
,
Alex Christensen
no flags
Details
Formatted Diff
Diff
Archive of layout-test-results from ews121 for ios-simulator-wk2
(4.91 MB, application/zip)
2019-04-22 14:17 PDT
,
EWS Watchlist
no flags
Details
Patch
(17.98 KB, patch)
2019-04-24 11:58 PDT
,
Alex Christensen
no flags
Details
Formatted Diff
Diff
Patch
(17.98 KB, patch)
2019-04-24 15:33 PDT
,
Alex Christensen
no flags
Details
Formatted Diff
Diff
Patch
(16.33 KB, patch)
2019-04-25 10:35 PDT
,
Alex Christensen
no flags
Details
Formatted Diff
Diff
Show Obsolete
(4)
View All
Add attachment
proposed patch, testcase, etc.
Alex Christensen
Comment 1
2019-04-22 09:09:46 PDT
Created
attachment 367942
[details]
Patch
Chris Dumez
Comment 2
2019-04-22 09:14:24 PDT
Did you check with Joe? I broke some builds (non-ews) when I tried to do it last because we need to maintain some binary compatibility is some system frameworks.
Darin Adler
Comment 3
2019-04-22 12:45:44 PDT
Comment on
attachment 367942
[details]
Patch Oops, I didn’t see the comment from Chris when marking this as reviewed
Alex Christensen
Comment 4
2019-04-22 13:36:31 PDT
I think this is OK since it's been several years, this would only affect open source developers working on super old operating systems, and I think it's not needed any more because we're back to using WTF::Optional (see
http://trac.webkit.org/r209326
) but I'll wait for Joe or Devin to agree.
Devin Rousso
Comment 5
2019-04-22 13:41:56 PDT
(In reply to Alex Christensen from
comment #4
)
> I think this is OK since it's been several years, this would only affect open source developers working on super old operating systems, and I think it's not needed any more because we're back to using WTF::Optional (see
http://trac.webkit.org/r209326
) but I'll wait for Joe or Devin to agree.
That's correct. If none of the bots complain, then I think this is fine to remove.
Chris Dumez
Comment 6
2019-04-22 13:48:10 PDT
(In reply to Devin Rousso from
comment #5
)
> (In reply to Alex Christensen from
comment #4
) > > I think this is OK since it's been several years, this would only affect open source developers working on super old operating systems, and I think it's not needed any more because we're back to using WTF::Optional (see
http://trac.webkit.org/r209326
) but I'll wait for Joe or Devin to agree. > That's correct. If none of the bots complain, then I think this is fine to > remove.
You will be proven wrong very soon I fear. It tried to remove this file like 1 month ago when I renamed Optional and it passed EWS but failed on internal bots.
Chris Dumez
Comment 7
2019-04-22 13:49:27 PDT
Anyway, I do not mind trying but please monitor the bots after landing. Worse case scenario we roll it out.
EWS Watchlist
Comment 8
2019-04-22 14:17:17 PDT
Comment on
attachment 367942
[details]
Patch
Attachment 367942
[details]
did not pass ios-sim-ews (ios-simulator-wk2): Output:
https://webkit-queues.webkit.org/results/11963367
New failing tests: http/tests/paymentrequest/payment-response-payerName-attribute.https.html http/tests/paymentrequest/payment-response-payerPhone-attribute.https.html http/tests/paymentrequest/updateWith-method-pmi-handling.https.html http/tests/paymentrequest/payment-request-change-shipping-address.https.html http/tests/paymentrequest/payment-response-complete-method.https.html http/tests/paymentrequest/payment-address-attributes-and-toJSON-method.https.html http/tests/paymentrequest/payment-request-change-shipping-option.https.html http/tests/paymentrequest/payment-response-payerEmail-attribute.https.html http/tests/paymentrequest/payment-response-methodName-attribute.https.html http/tests/paymentrequest/payment-response-retry-method.https.html http/tests/paymentrequest/payment-request-merchant-validation.https.html http/tests/paymentrequest/payment-request-show-method.https.html
EWS Watchlist
Comment 9
2019-04-22 14:17:19 PDT
Created
attachment 367972
[details]
Archive of layout-test-results from ews121 for ios-simulator-wk2 The attached test failures were seen while running run-webkit-tests on the ios-sim-ews. Bot: ews121 Port: ios-simulator-wk2 Platform: Mac OS X 10.13.6
Alex Christensen
Comment 10
2019-04-23 12:58:29 PDT
Comment on
attachment 367942
[details]
Patch Joe said "Let's do this" in more elegant words. I believe the test failures are unrelated. I'll check the bots after this lands.
WebKit Commit Bot
Comment 11
2019-04-23 13:26:05 PDT
Comment on
attachment 367942
[details]
Patch Clearing flags on attachment: 367942 Committed
r244558
: <
https://trac.webkit.org/changeset/244558
>
WebKit Commit Bot
Comment 12
2019-04-23 13:26:06 PDT
All reviewed patches have been landed. Closing bug.
Radar WebKit Bug Importer
Comment 13
2019-04-23 13:27:22 PDT
<
rdar://problem/50141786
>
Shawn Roberts
Comment 14
2019-04-23 17:25:21 PDT
The tests listed in EWS failure are actually crashing on iOS Simulator and Debug.
https://build.webkit.org/builders/Apple%20iOS%2012%20Simulator%20Release%20WK2%20%28Tests%29/builds/3750
https://build.webkit.org/builders/Apple%20iOS%2012%20Simulator%20Debug%20WK2%20%28Tests%29/builds/3419
Debug is hitting assertion failures and causing 13 crashes ASSERTION FAILED: dlopen(/System/Library/Frameworks/PassKit.framework/PassKit, 2): Symbol not found: __ZN9Inspector17BackendDispatcher19reportProtocolErrorESt8optionalIlENS0_15CommonErrorCodeERKN3WTF6StringE
https://build.webkit.org/results/Apple%20iOS%2012%20Simulator%20Debug%20WK2%20(Tests)/r244565%20(3417)/fast/css/appearance-apple-pay-button-crash-log.txt
WebKit Commit Bot
Comment 15
2019-04-23 17:26:39 PDT
Re-opened since this is blocked by
bug 197219
Alex Christensen
Comment 16
2019-04-24 11:58:24 PDT
Created
attachment 368154
[details]
Patch
EWS Watchlist
Comment 17
2019-04-24 12:01:23 PDT
Attachment 368154
[details]
did not pass style-queue: ERROR: Source/JavaScriptCore/inspector/InspectorBackendDispatcherCompatibility.h:38: Use 'WTF::Optional<>' instead of 'std::optional<>'. [runtime/wtf_optional] [4] ERROR: Source/JavaScriptCore/inspector/InspectorBackendDispatcherCompatibility.cpp:26: Found header this file implements before WebCore config.h. Should be: config.h, primary header, blank line, and then alphabetically sorted. [build/include_order] [4] ERROR: Source/JavaScriptCore/inspector/InspectorBackendDispatcherCompatibility.cpp:31: Use 'WTF::Optional<>' instead of 'std::optional<>'. [runtime/wtf_optional] [4] Total errors found: 3 in 6 files If any of these errors are false positives, please file a bug against check-webkit-style.
Alex Christensen
Comment 18
2019-04-24 15:33:04 PDT
Created
attachment 368192
[details]
Patch
EWS Watchlist
Comment 19
2019-04-24 15:34:20 PDT
Attachment 368192
[details]
did not pass style-queue: ERROR: Source/JavaScriptCore/inspector/InspectorBackendDispatcherCompatibility.h:38: Use 'WTF::Optional<>' instead of 'std::optional<>'. [runtime/wtf_optional] [4] ERROR: Source/JavaScriptCore/inspector/InspectorBackendDispatcherCompatibility.cpp:26: Found header this file implements before WebCore config.h. Should be: config.h, primary header, blank line, and then alphabetically sorted. [build/include_order] [4] ERROR: Source/JavaScriptCore/inspector/InspectorBackendDispatcherCompatibility.cpp:31: Use 'WTF::Optional<>' instead of 'std::optional<>'. [runtime/wtf_optional] [4] Total errors found: 3 in 6 files If any of these errors are false positives, please file a bug against check-webkit-style.
Anders Carlsson
Comment 20
2019-04-25 08:01:41 PDT
"iOS12" should say "iOS 12".
Darin Adler
Comment 21
2019-04-25 08:59:11 PDT
Comment on
attachment 368192
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=368192&action=review
> Source/JavaScriptCore/ChangeLog:19 > + * inspector/InspectorBackendDispatcherCompatibility.cpp: Added. > + (Inspector::BackendDispatcher::reportProtocolError): > + * inspector/InspectorBackendDispatcherCompatibility.h: Added.
I think the contents of InspectorBackendDispatcherCompatibility.h could all be entirely within InspectorBackendDispatcherCompatibility.cpp; we don’t need two files.
> Source/JavaScriptCore/inspector/InspectorBackendDispatcherCompatibility.cpp:30 > +// This symbol must be exported from JavaScriptCore to maintain binary compatibility with iOS12.
I’d like the other implicit half of this to be explicitly stated. "We can, and should, remove this entire file once we no longer need binary compatibility with iOS 12."
Alex Christensen
Comment 22
2019-04-25 10:35:34 PDT
(In reply to Darin Adler from
comment #21
)
> Comment on
attachment 368192
[details]
> Patch > > View in context: >
https://bugs.webkit.org/attachment.cgi?id=368192&action=review
> > > Source/JavaScriptCore/ChangeLog:19 > > + * inspector/InspectorBackendDispatcherCompatibility.cpp: Added. > > + (Inspector::BackendDispatcher::reportProtocolError): > > + * inspector/InspectorBackendDispatcherCompatibility.h: Added. > > I think the contents of InspectorBackendDispatcherCompatibility.h could all > be entirely within InspectorBackendDispatcherCompatibility.cpp; we don’t > need two files.
Aha! I initially had it as an inline function that wasn't exporting a symbol and I thought I needed a header. I actually just needed the definition separate from the declaration, even if in the same file.
> > Source/JavaScriptCore/inspector/InspectorBackendDispatcherCompatibility.cpp:30 > > +// This symbol must be exported from JavaScriptCore to maintain binary compatibility with iOS12. > > I’d like the other implicit half of this to be explicitly stated. "We can, > and should, remove this entire file once we no longer need binary > compatibility with iOS 12."
Ok!
Alex Christensen
Comment 23
2019-04-25 10:35:53 PDT
Created
attachment 368246
[details]
Patch
Alex Christensen
Comment 24
2019-04-25 10:37:10 PDT
http://trac.webkit.org/r244652
Note
You need to
log in
before you can comment on or make changes to this bug.
Top of Page
Format For Printing
XML
Clone This Bug