157153 - Bug in layout/html/table/src/nsTableRowGroupFrame.cpp

Status: RESOLVED FIXED

(Core :: Layout: Tables, defect)

Description

[tenthumbs]

In layout/html/table/src/nsTableRowGroupFrame.cpp, gcc says

nsTableRowGroupFrame.cpp:820: warning: unsigned value < 0 is always 0

and the code is

for (rowFrame = startRowFrame, rowIndex = 0; rowFrame && (extra > 0); 
rowFrame = rowFrame->GetNextRow(), rowIndex++) {
    RowInfo& rInfo = rowInfo[rowIndex];
    if (rInfo.hasPctHeight) {
      nscoord rowExtra = PR_MAX(0, rInfo.pctHeight - rInfo.height); <****
      rowExtra = PR_MIN(rowExtra, extra);
      UpdateHeights(rInfo, rowExtra, heightOfRows, heightOfUnStyledRows);
      extra -= rowExtra;
    }
  }

This all means that rInfo.pctHeight - rInfo.height is an unsigned quantity 
so it is always > 0 so rowExtra = rInfo.pctHeight - rInfo.height _always_.

That doesn't seem to be the intent.

Comment 1

[Kevin McCluskey (gone)]

->Karnaze

Comment 2

[Bernd]

thats the right level for me :-)

Comment 3

[Bernd]

*** Bug 166359 has been marked as a duplicate of this bug. ***

Comment 4

[Bernd]

Attachment: patch + QueryInterface Cleanup

Comment 5

[karnaze (gone)]

Comment on attachment 109034 [details] [diff] [review]
patch + QueryInterface Cleanup

+  if(NS_FAILED(result)) 
+    return result;
+  if(!cellFrame)
+    return NS_ERROR_FAILURE;
+

Can't this be simplified to

+  if(!cellFrame)
+    return result;

or

+  if(NS_FAILED(result))
+    return result;

Comment 6

[Bernd]

Chris, yes you are right 

+  if(NS_FAILED(result))
+    return result;

is already enough.

Comment 7

[Robert O'Callahan (:roc) (email my personal email if necessary)]

Comment on attachment 109034 [details] [diff] [review]
patch + QueryInterface Cleanup

sr=roc+moz

Comment 8

[Bernd]

fix checked in on dec 13th 2002