Documentation link to date_breaks is broken

[jwhendy]

Not sure this was the proper place to report, but the fix should be easy.

On the documentation for scale_x_datetime, the breaks documentation is like so:

breaks
A vector of breaks, a function that given the scale limits returns a vector of breaks, or a character vector, specifying the width between breaks. For more information about the first two, see continuous_scale, for more information about the last, see date_breaks`.

I think inside-r has redone their URL scheme, so the link for date_breaks is dead.

It should point here (at least, it appears that's the analogous site on inside-r).

---

As an aside, this still seems to be an issue since inside-r features the 0.8.9 documentation, and I'm using time data so I thought to myself, "I wonder if there's an equivalent, like time_breaks?" Guess what, there is, or at least used to be, but it's not featured in the 0.9.3 documentation on CRAN.

And as a last aside, why does the ggplot2 documentation feature usage information for date_break, but never actually give any formal documentation of what the argument(s) need to be? Clicking the hyperref in the PDF for date_breaks with in a browser or the downloaded file just takes me to the top of the file.

[hadley]

Should be fixed in the next release. There aren't really any arguments to date_breaks since the whole point is that it picks the breaks automatically without any user intervention.

[jwhendy]

Thanks for fixing. Correct in that it mostly picks the right breaks, but sometimes you want, say, 5 minute breaks, not 15 minute breaks as picked automatically. That's what I meant.

I also found out that date_breaks() takes pretty much any time unit argument; I thought date_breaks() was for things like days, weeks, months, etc. and time_breaks() was for hours, minutes, and seconds. Turns out date_breaks() takes all of them, so I'm a happy camper on both fronts :)

[hadley]

Hmmm, weird, the documentation from inside-r isn't actually right. ?scales::date_breaks is somewhat more informative.

[jwhendy]

Completely agree. I guess I should start using actual R help instead of always googling for the answer...