Threaded Mode | Linear Mode

ggaliens · (This post was last modified: 09-25-2014 06:05 PM by ggaliens.)

Not at all sure what is actually calling math:log10.
But this happens when applying tools/scene-info for a flat object (flatten a cube).

Code:

Dump written 2014-9-25_13-52

Version: 1.5.3

Window: geom

Reason: badarith

Short stack trace:

[]

Long stack trace:

[{math,log10,[0.0],[]},

 {wings,'-get_object_info/2-fun-0-',1,[{file,"wings.erl"},{line,1855}]},

 {lists,map,2,[{file,"lists.erl"},{line,1237}]},

 {lists,map,2,[{file,"lists.erl"},{line,1237}]},

 {wings,get_object_info,2,[{file,"wings.erl"},{line,1867}]},

 {wings,'-area_volume_info/1-lc$^0/1-0-',2,[{file,"wings.erl"},{line,1835}]},

 {wings,area_volume_info,1,[{file,"wings.erl"},{line,1835}]},

 {wings,command_1,2,[{file,"wings.erl"},{line,857}]}]

Is it possible for one row in call-stack shown above to be "missing" for some reason ?
Or is the call stack in the crash_dump always fairly complete ?

Micheus ... I'm thinking you might have change the formatting logic ?

math:log10 function will get really nasty for small numbers near zero.

Micheus ... I'm thinking you might have change the formatting logic ?

math:log10 function will get really nasty for small numbers near zero.

Maybe we could just make this a wings_prefs item and leave the fancy logic out ?

ggaliens · 09-25-2014, 09:14 PM

OK ... Micheus ... I have made a bad assumption here ... maybe you didn't put in there the math:log10
Nevertheless ... that code needs a special fix near ZERO.

Anybody ? I'd advocate for a dialog item for the precision ... maybe not even a prefs ... maybe just an dialog item right inside the existing dialog ? The oddball decimal crafty code should be reverted.

Dan ... toss down a better idea and I'll try coding, branch, pull request.

ggaliens · 09-27-2014, 03:31 PM

OK . I have something to submit here via github / pullrequest.
I have created a preference.

Will submit after a few more tests.