Closed Bug 763223 Opened 11 years ago Closed 11 years ago

"ASSERTION: parentSize should be valid" with huge padding

Categories

(Core :: Layout, defect)

Product:
Core
Component:
Layout
Platform:
x86_64
macOS
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

()

Status:
RESOLVED FIXED
Milestone:
mozilla16

People

(Reporter: jruderman, Assigned: ehsan.akhgari)

References

Details

(Keywords: assertion, regression, testcase)

Attachments

(3 files)

testcase
286 bytes, text/html
Details
stack trace
1.31 KB, text/plain
Details
Patch (v1)
2.13 KB, patch
dbaron
: review+
Details | Diff | Splinter Review
Attached file testcase — 
###!!! ASSERTION: parentSize should be valid: 'parentSize.width != NS_INTRINSICSIZE && parentSize.height != NS_INTRINSICSIZE', file layout/base/nsCSSFrameConstructor.cpp, line 11994

This assertion was added in bug 157681.
Attached file stack trace — 

    
    

    
  
So the right padding here is a large value, which causes tons of these types of warnings: "have unconstrained width; this should only result from very large sizes, not attempts at intrinsic width calculation: 'aContainingBlockWidth != NS_UNCONSTRAINEDSIZE'", and also causes the width value to be unconstrained size, which triggers this assertion.

How are we supposed to handle cases like this with huge values in layout?

    
    

    
  
We don't really have a plan for these.  :(

    
    

    
  
So there isn't too much we can do about this assertion, except for taking it out?

    
    

    
  
I have an idea in bug 765861.  For now, this should probably be downgraded to a warning.

    
    

    
  

      
      
      
      

        Attached patch
          
          
        Patch (v1)
        — 
          — Splinter Review
      

    


  
This patch converts the assertion into a warning.
Assignee: nobody → ehsan
Status: NEW → ASSIGNED

        Attachment #634258 -
        Flags: review?(dbaron)

    
    

    
  
Comment on attachment 634258 [details] [diff] [review]
Patch (v1)

Please use NS_WARN_IF_FALSE instead of the #ifdef and NS_WARNING.  r=dbaron with that

        Attachment #634258 -
        Flags: review?(dbaron) → review+

    
    

    
  
https://hg.mozilla.org/mozilla-central/rev/a7f575fe03e0
Status: ASSIGNED → RESOLVED
Closed: 11 years ago
Resolution: --- → FIXED

          You need to log in
          before you can comment on or make changes to this bug.