@bomsy sorry, I don't have time to continue working on this. :)

• fixes lint errors

• rebased
• fixes lint error
• fixes react warning

I am sorry @bomsy, I've been busy lately. I'll get back to my patches next week. :)

• fixes lint errors

• rebased
• changes storing exceptions from sourceUrl to sourceId
• fixes lint errors

In D87444#2728900, @nchevobbe wrote:

I'm not sure how close I am to land D85398: Bug 1657310 - Attach target and thread from TargetList. r=jdescottes,ochameau. as I'm fighting some test failures, but maybe you want to wait until it lands? What do you think?

Sure no problem, let's wait to land that patch first and then continue with this one. :)

I realized that we can't use sourceId as a key for storing exceptions, because at the time when we get an exception from the resourceWatcher, the related source may not be initialized yet and we can not get sourceId at that moment.

• fixes lint errors

• fixes lint errors
• adds codeMirror.operation to ConsoleApiMessage component

I think it would be easier to handle the change of storing before we land this path. It could be confused to mix it with this patch. I'll open a new ticket a submit a new patch for it this weekend or on Monday.
We need to revert changes in /src/source-map.js and in firefox/create.js in this patch too.

@nchevobbe would be ok to open a new ticket for the change of storing exceptions from sourceActorId to sourceId?
I just realized that if we store them by sourceId we probably need to have an action for removing them from the store when the exception component umounts. Otherwise, the exception will be displayed after refreshing the page and pausing before the exception.
So far the sourceActorId changes after each refresh so we didn't need to handle this case.

• adds mochitest

thanks @nchevobbe and I am sorry for a late update.

• rebased
• adds error and debug to supported formats that should be displayed
• refactors ConsoleApiMessages component:
  • removes codeMirror.operation
  • removes selectedSource from mapStateToProps

@nchevobbe can we land this path? Then I can rebase inline console messages and submit changes. Thanks

• corrects assertion message

• adds corrections

Suggestions from D81678. Thanks
https://phabricator.services.mozilla.com/D81678#inline-485163

• rebased

I don't have access to try. Thanks

@nchevobbe could we land this patch? Thanks

@nchevobbe Thanks for the review. I am sorry for a late reply, I was away for a couple of days.

@loganfsmyth I am sorry for a late reply, I was away for a couple of days.

@nchevobbe it should work now, there was one merge conflict. Thanks

Thanks @nchevobbe!
I renamed most of the files from 'logMessage' to 'consoleApiMessage' so it removed most comments from them. I quoted the questions that have been removed to maintain the context. Thanks.

• adds mochitest browser_dbg-inline-console-messages.js
• adds Nicolas corrections
• changes all occurrences of 'logMessages' to 'consoleApiMessage'
• turns this feature off by default

Thanks @nchevobbe!

• swaps the order of functions in actions/exceptions.js

• changes clearLogMessage() and addLogMessage() to updateLogMessage() in LogMessage component
• adds requestIdleCallback to LogMessage component

@Harald I've made some changes to use`requestIdleCallback`, I'll upload it today. I can revert it back if I chose an incorrect approach so I don't block this path on it and leave it for follow-ups.

• removes changes in threadActor
• adds sourceActorId to frame.location
• adds sourceActorId = null to getGeneratedLocation() in source-map.js

@loganfsmyth I am sorry for the late reply.

• adds Itiel's corrections

• rebases the patch

Thanks @nchevobbe.

@davidwalsh was the stacktrace expanded? It should behave like the event tooltip in the Inspector's markup view. When the stacktrace is expanded, the tooltip gets closed by clicking outside, when it's closed it gets closed when mousing out. Not sure if it was expected behavior or not? Thanks.

Thanks @davidwalsh, I just updated the test plan with steps.
Here's the link https://elderly-beneficial-toucan.glitch.me/

• updates logMessages.js action to handle only console.log|info|warn
• adds an option to the overflow debugger settings for toggling the console messages
• fixes a lint error

@loganfsmyth could you please review this path? Thanks.

@jlast can we land this?

• adds a context menu item for hiding/showing the console messages in the editor
• adds an option to preferences for toggling the messages
• fixes lint errors

• adds a comment explaining why not use directly sourceActorId
• adds internalSourceId getter to sourceActor

ok, @nchevobbe let me know if I should continue working on this. I will need some guidance though. Thank you.

@jlast is the failure 'TEST-UNEXPECTED-FAIL | netwerk/test/unit/test_trr_case_sensitivity.js | xpcshell return code: 0' related to this patch? I can't replicate it locally. Thanks

• fixes mochitests failures - moves test functions from simple2.js to simple3.js

• removes addException from panel.js and adds addExceptionFromResource to actions/exceptions

• removes frameMenu util file and adds isValidRestartFrame() to Frames/FrameMenu.js
• adds frame.type
• updates all relevent files to pass flow test
• updates jest tests in SecondaryPanes/Frames/tests

A couple of comments. Thanks.

Hopefully, I managed to indicate the dependence of this patch on D78883 correctly. Thanks

• refactores frameMenu.js
• removes why and numFrames from frameMenu.js and updates all files where were used

• adds comments in frameMenu.js
• adds validation before setting this._requestedFrameRestart in _handleResumeLimit
• removes check frame.terminated from restartFrame

Thanks Logan!

Thanks Jason!

• fixes the mochitest browser_dbg-inline-exceptions.js

Thanks @loganfsmyth! I have a few questions:

• adds Logan's corrections:
• adds RESTARTED_FRAMES variable at the top of the thread.js
• adds this._requestedFrameRestart check at the top of the onEnterFrame hook
• refactors _makeOnPop to handle a potential restart request while inside of a pop handler
• adjusts _getNextStepFrame to skip frames that are restarted

• rebase
• adds Nicolas's corrections
• adds resourceWatcher to debugger/panel.js
• removes any changes in threadActor
• removes 'exception' event
• refactored addException() to work with 'resource' object

I searched through the getSymbols.spec.js and a similar issue can occur for all these examples where expression is set in the form of foo = value.
It seems that when there is a default value the parse fails to recognize it as an 'objectShorthand'.

As @nchevobbe suggested, we can use the resource API, which should allow us to completely remove the emitting 'exception' event in threadActor.
But using resource API in this patch depends on this patch https://phabricator.services.mozilla.com/D76447.
I'll make the next update as soon as it gets landed.

• makes changes in threadActor:
• adds support for restarting a frame on the last frame line
• removes the need for setting and then removing a breakpoint for pausing on the first breakable line of the restarted frame

• corrects the mochitest browser_dbg-restart-frame.js (forgot to simulate a right click on the frame)

Thanks Jason!

• adds corrections
• adds isValidRestartFrame util function in utils/pause/frames/frameMenu.js
• adds mochitest browser_dbg-restart-frame.js
• adds xpcshell test test_restartFrame-01.js
• hides restart frame context menu item for mutation, event, xhr breakpoints and for the oldest frame

Thanks @nchevobbe! I will add your corrections in the next update.