Bug 156763 - Web Inspector: Hide ["PrivateSymbol.concatSlowPath"] and other PrivateSymbols from memory edges
Summary: Web Inspector: Hide ["PrivateSymbol.concatSlowPath"] and other PrivateSymbols...
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: Web Inspector (show other bugs)
Version: WebKit Nightly Build
Hardware: All All
: P2 Normal
Assignee: Devin Rousso
URL:
Keywords: InRadar
Depends on:
Blocks:
 
Reported: 2016-04-19 15:32 PDT by Timothy Hatcher
Modified: 2016-09-12 14:48 PDT (History)
9 users (show)

See Also:

Attachments
Screenshot (82.34 KB, image/png)
2016-04-19 15:32 PDT, Timothy Hatcher 		no flags Details
Patch (2.75 KB, patch)
2016-09-10 15:00 PDT, Devin Rousso 		joepeck: review+
Details | Formatted Diff | Diff
Patch (2.64 KB, patch)
2016-09-12 14:18 PDT, Devin Rousso 		no flags Details | Formatted Diff | Diff
Show Obsolete (1) View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Timothy Hatcher 2016-04-19 15:32:11 PDT 
These should be considered internal / private.
Comment 1 Timothy Hatcher 2016-04-19 15:32:24 PDT 
Created attachment 276763 [details]
Screenshot
Comment 2 Alexey Proskuryakov 2016-09-08 13:59:51 PDT 
<rdar://problem/28213642>
Comment 3 Devin Rousso 2016-09-10 15:00:00 PDT 
Created attachment 288502 [details]
Patch
Comment 4 Joseph Pecoraro 2016-09-12 13:18:47 PDT 
Comment on attachment 288502 [details]
Patch

I've always hesitated doing this quick fix because what if someone has a string property that actually starts with the string "PrivateSymbol". That said, I think this handles the 99% case. We should have a FIXME or file a follow-up to at some point actually distinguish between a private symbol property (an internal implementation detail of JavaScriptCore) and a regular Symbol property.
Comment 5 Devin Rousso 2016-09-12 13:40:20 PDT 
(In reply to comment #4)
> I've always hesitated doing this quick fix because what if someone has a
> string property that actually starts with the string "PrivateSymbol". That
> said, I think this handles the 99% case. We should have a FIXME or file a
> follow-up to at some point actually distinguish between a private symbol
> property (an internal implementation detail of JavaScriptCore) and a regular
> Symbol property.

Just so I understand correctly, where does the "PrivateSymbol" actually originate from?  Does it just represent a Symbol(), or is it something different?
Comment 6 Devin Rousso 2016-09-12 14:18:49 PDT 
Created attachment 288609 [details]
Patch
Comment 7 WebKit Commit Bot 2016-09-12 14:48:54 PDT 
Comment on attachment 288609 [details]
Patch

Clearing flags on attachment: 288609

Committed r205827: <http://trac.webkit.org/changeset/205827>
Comment 8 WebKit Commit Bot 2016-09-12 14:48:59 PDT 
All reviewed patches have been landed.  Closing bug.