Bug 123111 – Upstream ENABLE(REMOTE_INSPECTOR) and enable on iOS and Mac

• Component: WebKit Misc.
• Status: RESOLVED FIXED

[Joseph Pecoraro]

ENABLE(REMOTE_INSPECTOR) is an alternative to ENABLE(INSPECTOR_SERVER) for remotely debugging via the inspector protocol.

REMOTE_INSPECTOR on Mac platforms exposes an XPC interface to receive page listings, setup/disconnect a remote debugging session, and send/receive inspector protocol socket messages. This is an alternative to the Web Socket approach enabled by INSPECTOR_SERVER.

[Joseph Pecoraro]

Created attachment 214782 [details]
[PATCH] Add ENABLE(REMOTE_INSPECTOR) Implementation

This is a working WebKit1 Mac / iOS implementation. There are extra iOS bits that I'll put in a separate patch for node highlighting.

[WebKit Commit Bot]

Attachment 214782 [details] did not pass style-queue:

Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'Source/JavaScriptCore/ChangeLog', u'Source/JavaScriptCore/Configurations/FeatureDefines.xcconfig', u'Source/WTF/ChangeLog', u'Source/WTF/wtf/FeatureDefines.h', u'Source/WebCore/ChangeLog', u'Source/WebCore/Configurations/FeatureDefines.xcconfig', u'Source/WebCore/WebCore.exp.in', u'Source/WebKit/ChangeLog', u'Source/WebKit/WebKit.xcodeproj/project.pbxproj', u'Source/WebKit/cf/ChangeLog', u'Source/WebKit/cf/WebCoreSupport/WebInspectorClientCF.cpp', u'Source/WebKit/ios/ChangeLog', u'Source/WebKit/ios/WebCoreSupport/WebInspectorClientIOS.mm', u'Source/WebKit/mac/ChangeLog', u'Source/WebKit/mac/Configurations/FeatureDefines.xcconfig', u'Source/WebKit/mac/Misc/WebKitLogging.h', u'Source/WebKit/mac/WebCoreSupport/WebFrameLoaderClient.mm', u'Source/WebKit/mac/WebCoreSupport/WebInspectorClient.h', u'Source/WebKit/mac/WebCoreSupport/WebInspectorClient.mm', u'Source/WebKit/mac/WebInspector/remote/WebInspectorClientRegistry.h', u'Source/WebKit/mac/WebInspector/remote/WebInspectorClientRegistry.mm', u'Source/WebKit/mac/WebInspector/remote/WebInspectorRelayDefinitions.h', u'Source/WebKit/mac/WebInspector/remote/WebInspectorRemoteChannel.h', u'Source/WebKit/mac/WebInspector/remote/WebInspectorRemoteChannel.mm', u'Source/WebKit/mac/WebInspector/remote/WebInspectorServer.h', u'Source/WebKit/mac/WebInspector/remote/WebInspectorServer.mm', u'Source/WebKit/mac/WebInspector/remote/WebInspectorServerWebViewConnection.h', u'Source/WebKit/mac/WebInspector/remote/WebInspectorServerWebViewConnection.mm', u'Source/WebKit/mac/WebInspector/remote/WebInspectorServerWebViewConnectionController.h', u'Source/WebKit/mac/WebInspector/remote/WebInspectorServerWebViewConnectionController.mm', u'Source/WebKit/mac/WebInspector/remote/WebInspectorXPCWrapper.h', u'Source/WebKit/mac/WebInspector/remote/WebInspectorXPCWrapper.m', u'Source/WebKit/mac/WebKit.exp', u'Source/WebKit/mac/WebView/WebView.mm', u'Source/WebKit/mac/WebView/WebViewData.h', u'Source/WebKit/mac/WebView/WebViewData.mm', u'Source/WebKit/mac/WebView/WebViewInternal.h', u'Source/WebKit/mac/WebView/WebViewPrivate.h', u'Source/WebKit2/ChangeLog', u'Source/WebKit2/Configurations/FeatureDefines.xcconfig']" exit_code: 1
Source/WebKit/mac/WebCoreSupport/WebInspectorClient.h:89:  The parameter name "remoteChannel" adds no information, so it should be removed.  [readability/parameter_name] [5]
Source/WebKit/mac/WebInspector/remote/WebInspectorXPCWrapper.h:39:  This { should be at the end of the previous line  [whitespace/braces] [4]
Source/WebKit/mac/WebInspector/remote/WebInspectorXPCWrapper.h:49:  Extra space before ( in function call  [whitespace/parens] [4]
Source/WebKit/mac/WebInspector/remote/WebInspectorXPCWrapper.h:50:  Extra space before ( in function call  [whitespace/parens] [4]
Source/WebKit/mac/WebInspector/remote/WebInspectorXPCWrapper.h:51:  Extra space before ( in function call  [whitespace/parens] [4]
Source/WebKit/mac/WebInspector/remote/WebInspectorXPCWrapper.h:52:  Extra space before ( in function call  [whitespace/parens] [4]
Source/WebKit/mac/WebInspector/remote/WebInspectorClientRegistry.h:42:  Extra space before ( in function call  [whitespace/parens] [4]
Source/WebKit/mac/ChangeLog:83:  ChangeLog entry has no bug number  [changelog/bugnumber] [5]
Source/WTF/wtf/FeatureDefines.h:873:  Multi-line string ("...") found.  This lint script doesn't do well with such strings, and may give bogus warnings.  They're ugly and unnecessary, and you should use concatenation instead".  [readability/multiline_string] [5]
Total errors found: 9 in 40 files


If any of these errors are false positives, please file a bug against check-webkit-style.

[EFL EWS Bot]

Comment on attachment 214782 [details]
[PATCH] Add ENABLE(REMOTE_INSPECTOR) Implementation

Attachment 214782 [details] did not pass efl-ews (efl):
Output: http://webkit-queues.appspot.com/results/8848137

[EFL EWS Bot]

Comment on attachment 214782 [details]
[PATCH] Add ENABLE(REMOTE_INSPECTOR) Implementation

Attachment 214782 [details] did not pass efl-wk2-ews (efl-wk2):
Output: http://webkit-queues.appspot.com/results/8828113

[Joseph Pecoraro]

Comment on attachment 214782 [details]
[PATCH] Add ENABLE(REMOTE_INSPECTOR) Implementation

View in context: https://bugs.webkit.org/attachment.cgi?id=214782&action=review

>> Source/WTF/wtf/FeatureDefines.h:873
>> +#error "ENABLE(REMOTE_INSPECTOR) requires ENABLE(INSPECTOR)
> 
> Multi-line string ("...") found.  This lint script doesn't do well with such strings, and may give bogus warnings.  They're ugly and unnecessary, and you should use concatenation instead".  [readability/multiline_string] [5]

lol, yes it looks like I'm missing an ending quote.

[Timothy Hatcher]

Comment on attachment 214782 [details]
[PATCH] Add ENABLE(REMOTE_INSPECTOR) Implementation

View in context: https://bugs.webkit.org/attachment.cgi?id=214782&action=review

> Source/WebKit/mac/WebCoreSupport/WebInspectorClient.mm:256
> +    Page* page = core(m_webView);
> +    if (page) {
> +        page->inspectorController()->disconnectFrontend();
> +    }

Could put Page* page in the if() and drop the { }.

> Source/WebKit/mac/WebInspector/remote/WebInspectorRelayDefinitions.h:38
> +#define WIRSimulatorTCPPortNumber               27753
> +#define WIRXPCMachPortName                      "com.apple.webinspector"
> +#define WIRServiceAvailableNotification         "com.apple.webinspectord.available"
> +#define WIRServiceAvailabilityCheckNotification "com.apple.webinspectord.availability_check"
> +#define WIRServiceEnabledNotification           "com.apple.webinspectord.enabled"
> +#define WIRServiceDisabledNotification          "com.apple.webinspectord.disabled"

These could be const NSString * instead. Also there isn't any reason for the symbol to keep the "WIR" prefix, it could differ even if the string content needs to keep "WIR".

> Source/WebKit/mac/WebInspector/remote/WebInspectorServerWebViewConnectionController.mm:281
> +    NSString *connectionIdentifier = [dictionary objectForKey:WIRConnectionIdentifierKey];
> +    [self pushListing:connectionIdentifier];

One line?

[Joseph Pecoraro]

Comment on attachment 214782 [details]
[PATCH] Add ENABLE(REMOTE_INSPECTOR) Implementation

View in context: https://bugs.webkit.org/attachment.cgi?id=214782&action=review

>> Source/WebKit/mac/WebInspector/remote/WebInspectorRelayDefinitions.h:38
>> +#define WIRServiceDisabledNotification          "com.apple.webinspectord.disabled"
> 
> These could be const NSString * instead. Also there isn't any reason for the symbol to keep the "WIR" prefix, it could differ even if the string content needs to keep "WIR".

These are used by XPC/notify APIs which take C strings, so I just kept them C String constants. Would there be an advantage to making them NSStrings? Just seems like more work unless they are actually used as NSStrings.

I agree that there is no need to keep the defines WIR prefixed. The consistency, and prefix makes it obvious to me that these are shared constants. If I changed them, I'd like to change them to "RWI", but of course the values need to stay "WIR" for backwards compatibility.

[Joseph Pecoraro]

(In reply to comment #7)
> (From update of attachment 214782 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=214782&action=review
> 
> >> Source/WebKit/mac/WebInspector/remote/WebInspectorRelayDefinitions.h:38
> >> +#define WIRServiceDisabledNotification          "com.apple.webinspectord.disabled"
> > 
> > These could be const NSString * instead. Also there isn't any reason for the symbol to keep the "WIR" prefix, it could differ even if the string content needs to keep "WIR".
> 
> These are used by XPC/notify APIs which take C strings, so I just kept them C String constants. Would there be an advantage to making them NSStrings? Just seems like more work unless they are actually used as NSStrings.
> 
> I agree that there is no need to keep the defines WIR prefixed. The consistency, and prefix makes it obvious to me that these are shared constants. If I changed them, I'd like to change them to "RWI", but of course the values need to stay "WIR" for backwards compatibility.

You mentioned this could be "const char*" types. I'll address this in a follow-up.

[Joseph Pecoraro]

Landed in: <http://trac.webkit.org/changeset/158050>