Re: [R] [ggplot2] Wind rose orientation

2009-12-07 Thread Thomas S. Dye
On Dec 7, 2009, at 7:00 AM, hadley wickham wrote: >> The idea of plotting a wind rose must be fairly common. I wonder >> if it >> would make sense to have a switch that would wrap data around the >> ends of a >> continuous scale? > > Probably - but it requires a lot of work, because ggplot2

Re: [R] [ggplot2] Wind rose orientation

2009-12-07 Thread hadley wickham
> The idea of plotting a wind rose must be fairly common.  I wonder if it > would make sense to have a switch that would wrap data around the ends of a > continuous scale? Probably - but it requires a lot of work, because ggplot2 doesn't currently support circular scales, which is what you really

Re: [R] [ggplot2] Wind rose orientation

2009-12-07 Thread Thomas S. Dye
ose with another package and wait >> patiently >> until ggplot2 plots the full compass. >> Thanks again for a terrific software package. >> All the best, >> Tom >> Begin forwarded message: >> >> From: "Thomas S. Dye" >> Date: Decembe

Re: [R] [ggplot2] Wind rose orientation

2009-12-06 Thread hadley wickham
n forwarded message: > > From: "Thomas S. Dye" > Date: December 3, 2009 9:42:27 PM HST > To: hadley wickham > Cc: "r-help@r-project.org" > Subject: Re: [R] [ggplot2] Wind rose orientation > Hi Hadley, > > That solved a lot of problems.  Thanks! >

Re: [R] [ggplot2] Wind rose orientation

2009-12-03 Thread Thomas S. Dye
Hi Hadley, That solved a lot of problems. Thanks! Do you get a vertically oriented bar? Here I get 15 bars with a space where the north bar, the 16th bar, should be. All the best, Tom On Dec 3, 2009, at 8:16 PM, hadley wickham wrote: Hi Thomas, Ok, the key thing that you were missing w

Re: [R] [ggplot2] Wind rose orientation

2009-12-03 Thread hadley wickham
Hi Thomas, Ok, the key thing that you were missing was: scale_x_continuous(limits = c(0, 360)) Since you don't have any data at 0, and because ggplot2 doesn't know that your variable had intrinsic meaning as a degree, it was starting zero degrees at 22.5. A few other tweaks below: wind.data$wi

Re: [R] [ggplot2] Wind rose orientation

2009-12-03 Thread Thomas S. Dye
Aloha Hadley, Thanks very much for ggplot. It's a terrific piece of work. Specifying width = 1 in the call to geom_bar didn't change the orientation of the coordinates. If you run the example, you'll see that 100 is horizontal, where 90 would be on the compass. Here is a reproducible e

Re: [R] [ggplot2] Wind rose orientation

2009-12-03 Thread hadley wickham
Hi Thomas, I suspect you want geom_bar(stat = "identity", width = 1), but it's hard to be sure without a reproducible example. Hadley On Thu, Dec 3, 2009 at 8:18 PM, Thomas S. Dye wrote: > Aloha all, > > I love using ggplot.  It took a while to get used to the grammar of > graphics, but it is

[R] [ggplot2] Wind rose orientation

2009-12-03 Thread Thomas S. Dye
Aloha all, I love using ggplot. It took a while to get used to the grammar of graphics, but it is starting to get easy now that I am thinking in a more structured way. A question. I'm making a wind rose that I'd like to be oriented with due north straight up. I've discovered that the ori