https://issues.apache.org/ooo/show_bug.cgi?id=124091
--- Comment #7 from Ted Ralphs <[email protected]> --- I'm not sure I understand your patch, Pedro. To me, it seems as though both lines are part of the linearity test. I can't easily find where the approxEqual function is defined but from the comment in the code, I guess it involves taking the ratio of the two numbers so that it if one of them is zero, this causes a problem. Aside from that, the two lines are testing the same thing: fTwo = fInitial + 2.0 * fCoeff fInitial = fTwo - 2.0 * fCoeff I'm not sure I see why the two tests should be separate. It still just looks like a tolerance issue to me. The approxEqual function is saying that two values that should actually be considered equal (for this purpose) are not equal. Can you say more? -- You are receiving this mail because: You are on the CC list for the issue. You are the assignee for the issue. You are watching all issue changes.
