<div dir="ltr">No typo, just poor reading comprehension. I guess I'll make amends and take a look at #1. I'll leave someone else to look at #4 though.<br></div><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Dec 3, 2014 at 2:59 PM, Sean McBride <span dir="ltr"><<a href="mailto:sean@rogue-research.com" target="_blank">sean@rogue-research.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Andy,<br>
<br>
Maybe you have a typo there, but #3 (the kd tree one) was fixed by Dan Lipsa already. #1 and #4 need someone...<br>
<span class="HOEnZb"><font color="#888888"><br>
Sean<br>
</font></span><div class="HOEnZb"><div class="h5"><br>
<br>
On Wed, 3 Dec 2014 14:57:14 -0500, Andy Bauer said:<br>
<br>
>Hi Sean,<br>
><br>
>I volunteer to look at #3. For some reason I missed this email thread --<br>
>thanks for your persistence!<br>
><br>
>Best,<br>
>Andy<br>
><br>
>On Wed, Dec 3, 2014 at 12:08 PM, Sean McBride <<a href="mailto:sean@rogue-research.com">sean@rogue-research.com</a>><br>
>wrote:<br>
><br>
>> Thanks David for fixing #2, and thanks Dan Lipsa for doing #3.<br>
>><br>
>> Anyone know vtkPolyLine.cxx to look at #1?<br>
>><br>
>> And to add a #4:<br>
>><br>
>> Common/ExecutionModel/vtkCompositeDataPipeline.cxx:1047:5: Value stored to<br>
>> 'outputPort' is never read<br>
>><br>
>> The whole block of code seems pointless because nothing actually uses<br>
>> 'outputPort' after setting it.<br>
>><br>
>> // Save the information about COMPOSITE_INDICES() as needed in the data<br>
>> // object.<br>
>> int outputPort = 0;<br>
>> if(request->Has(FROM_OUTPUT_PORT()))<br>
>> {<br>
>> outputPort = request->Get(FROM_OUTPUT_PORT());<br>
>> outputPort = (outputPort >= 0 ? outputPort : 0);<br>
>> }<br>
>><br>
>> Thanks,<br>
>><br>
>> Sean<br>
>><br>
>><br>
>> On Fri, 21 Nov 2014 10:20:54 -0700, David Gobbi said:<br>
>><br>
>> >Hi Sean,<br>
>> ><br>
>> >I can fix vtkCellPicker.cxx by making the code able to handle the<br>
>> >case that property is null.<br>
>> ><br>
>> > - David<br>
>> ><br>
>> ><br>
>> >On Fri, Nov 21, 2014 at 9:46 AM, Sean McBride <<a href="mailto:sean@rogue-research.com">sean@rogue-research.com</a>><br>
>> >wrote:<br>
>> ><br>
>> >> Hi all,<br>
>> >><br>
>> >> Occasionally, I try to fix some of the hundreds of clang static analyzer<br>
>> >> warnings in VTK. Some are easy, some are false positives, and some are<br>
>> >> real problems that I don't know how to fix. I'd like to appeal to you<br>
>> all<br>
>> >> to help fix these 3:<br>
>> >><br>
>> >> 1)<br>
>> >> file: Common/DataModel/vtkPolyLine.cxx<br>
>> >> method: vtkPolyLine::GenerateSlidingNormals<br>
>> >> lines: 227-237<br>
>> >> problem: dead code. The "if ( largeRotation )" branch sets "theta" then<br>
>> >> nothing uses it.<br>
>> >><br>
>> >> 2)<br>
>> >> file: Rendering/Core/vtkCellPicker.cxx<br>
>> >> method: vtkCellPicker::IntersectVolumeWithLine<br>
>> >> line: 691<br>
>> >> problem: the 'property' pointer is initialized to null, and if the<br>
>> >> subsequent 'if' is not entered, it's certainly still null at line 691<br>
>> and<br>
>> >> dereferenced.<br>
>> >><br>
>> >> 3)<br>
>> >> file: Common/DataModel/vtkKdTree.cxx<br>
>> >> line: 3188<br>
>> >> method: vtkKdTree::GenerateRepresentationWholeSpace<br>
>> >> problem: the first time through the 'for' loop always does "1 << -1".<br>
>> You<br>
>> >> can't left shift by a negative count.<br>
>> >><br>
>> >> I don't want to blindly delete dead code that maybe shouldn't be dead,<br>
>> or<br>
>> >> just throw in null checks, so hopefully someone knows this code!<br>
<br>
<br>
</div></div></blockquote></div><br></div>