|
From: | Michael D Godfrey |
Subject: | Re: Should interp1.m function allow jumps in X-values? |
Date: | Tue, 21 Aug 2012 19:57:23 -0400 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:14.0) Gecko/20120717 Thunderbird/14.0 |
Ben, I just took a quick look at interp1q which is described as as a simplified version of interp1 which provides only linear interpolation. The documentation also says that it "performs little error checking". Then it says that it provides "no error checking." This later is the correct statement. I doubt if the saving in execution speed is worth the lack of error checking. But, this points out that there is more to do in order to bring the interpolation code into better shape. This adds a bit to my argument that first the documentation should be made to match the code, then someone with some significant time should work through the capabilities and bring them into better shape. In the meantime, the fewer code patches which modify functionality the better. Do you agree? Michael |
[Prev in Thread] | Current Thread | [Next in Thread] |