On Sunday 2013-09-08 00:10 -0700, Dirk Schulze wrote: > I need help to interpret the following sentence on interpolation of length, percentage, or calc: > > "" > [..] both values are converted into a βcalc()β function that is the sum of a length and a percentage (each possibly zero), and these βcalc()β functions have each half interpolated as real numbers. > "" > > Just looking at the length to percentage case: If I have an interpolation off 20px to 20%, I have to create a calc() function for each of it of the following form? > > calc(<length> + <percentage>) > > What would be the value for the both calc() functions? 10% of the way from 20px to 20% would be calc(0.9 * 20px + 0.1 * 20%). 50% of the way from 20px to 20% would be calc(0.5 * 20px + 0.5 * 20%). 80% of the way from 20px to 20% would be calc(0.2 * 20px + 0.8 * 20%). > What does need to be half interpolated? I must assume that the order (<length> + <percentage> vs <percentage> + <length>) matters in this case, right? No, the order doesn't matter. > Looking at length to calc: If I have 20px to calc(50% - 20px), how do the interpolation candidates look like? 10% of the way from 20px to calc(50% - 20px) would be calc(0.9 * 20px + 0.1 * (50% - 20px)) which is the same as calc(16px + 5%). 50% of the way from 20px to calc(50% - 20px) would be calc(0.5 * 20px + 0.5 * (50% - 20px)) which is the same as calc(0px + 25%). 80% of the way from 20px to calc(50% - 20px) would be calc(0.2 * 20px + 0.8 * (50% - 20px)) which is the same as calc(-12px + 40%). -David -- π L. David Baron http://dbaron.org/ π π’ Mozilla https://www.mozilla.org/ π Before I built a wall I'd ask to know What I was walling in or walling out, And to whom I was like to give offense. - Robert Frost, Mending Wall (1914)Received on Sunday, 8 September 2013 07:46:58 UTC
This archive was generated by hypermail 2.4.0 : Monday, 23 January 2023 02:14:32 UTC