[R] How to import timestamps from emails into R

Gabor Grothendieck ggrothendieck at gmail.com
Wed Jun 24 16:15:56 CEST 2009


See the example in plot.zoo labelled Fancy X Axis:

library(zoo)
example(plot.zoo)
?plot.zoo

On Wed, Jun 24, 2009 at 10:11 AM, Thomas Levine<thomas.levine at gmail.com> wrote:
> One last tiny problem: How do I add months to the scale? It currently just
> has years
> http://school.thomaslevine.org/mywall.png
>
> Thanks again
>
> Tom
>
> On Sat, Jun 20, 2009 at 12:14 PM, Thomas Levine <thomas.levine at gmail.com>
> wrote:
>>
>> I wasn't really thinking that far ahead; plot tries to do something, so I
>> figured I'd try that as I had little other idea of what to do.
>>
>> The plot(tt) actually does what I want, though; the scales are just very
>> messed-up.
>>
>> Tom
>>
>> On Sat, Jun 20, 2009 at 11:58 AM, Gabor Grothendieck
>> <ggrothendieck at gmail.com> wrote:
>>>
>>> If that is the situation then plot(tt) in your post could not have been
>>> what you wanted in any case, e.g. plot(10:20)
>>>
>>> On Sat, Jun 20, 2009 at 11:49 AM, Thomas Levine<thomas.levine at gmail.com>
>>> wrote:
>>> > This produces the x-axis is the index, and the y-axis is time. It has
>>> > all of
>>> > the time information on the same axis, allowing me to plot cumulative
>>> > occurrences by time (my original plan) if the times are sorted, which
>>> > they
>>> > should be.
>>> >
>>> > I think I'll end up using some variant of plot(tt,seq_along(tt)),
>>> > putting
>>> > the time axis along the bottom.
>>> >
>>> > Thanks
>>> >
>>> > Tom
>>> >
>>> > On Sat, Jun 20, 2009 at 11:15 AM, Gabor Grothendieck
>>> > <ggrothendieck at gmail.com> wrote:
>>> >>
>>> >> Try this:
>>> >>
>>> >> plot(seq_along(tt), tt)
>>> >>
>>> >>
>>> >> On Sat, Jun 20, 2009 at 10:55 AM, Thomas
>>> >> Levine<thomas.levine at gmail.com>
>>> >> wrote:
>>> >> > Here's what I get
>>> >> >> head(tt)
>>> >> > [1] "2008-02-20 03:09:51 EST" "2008-02-20 12:12:57 EST"
>>> >> > [3] "2008-03-05 09:11:28 EST" "2008-03-05 17:59:40 EST"
>>> >> > [5] "2008-03-09 09:00:09 EDT" "2008-03-29 15:57:16 EDT"
>>> >> >
>>> >> > But I can't figure out how to plot this now. plot(tt) does not
>>> >> > appear to
>>> >> > be
>>> >> > univariate. I get the same plot with plot(as.Date(tt)), which would
>>> >> > make
>>> >> > sense if time is used because of the range of the dates and the
>>> >> > insignificance of the times of day.
>>> >> >> head(as.Date(tt))
>>> >> > [1] "2008-02-20" "2008-02-20" "2008-03-05" "2008-03-05" "2008-03-09"
>>> >> > [6] "2008-03-29"
>>> >> >
>>> >> > plot(tt) and plot(as.Date(tt)) give something like year as a
>>> >> > function of
>>> >> > the
>>> >> > rest of the date. Here they are
>>> >> >
>>> >> >
>>> >> > Here are the addresses
>>> >> > http://thomaslevine.org/time/tt.png
>>> >> > http://thomaslevine.org/time/as.Date.tt.png
>>> >> >
>>> >> > Tom
>>> >> >
>>> >> > On Fri, Jun 19, 2009 at 6:21 PM, Gabor Grothendieck
>>> >> > <ggrothendieck at gmail.com> wrote:
>>> >> >>
>>> >> >> Try this:
>>> >> >>
>>> >> >>
>>> >> >> Lines <- "Sun, 14 Jun 2009 07:33:00 -0700
>>> >> >> Sun, 14 Jun 2009 08:35:10 -0700
>>> >> >> Sun, 14 Jun 2009 21:26:34 -0700
>>> >> >> Mon, 15 Jun 2009 19:47:47 -0700
>>> >> >> Wed, 17 Jun 2009 21:50:41 -0700"
>>> >> >>
>>> >> >> # L <- readLines("myfile.txt")
>>> >> >> L <- readLines(textConnection(Lines))
>>> >> >> tt <- as.POSIXct(L, format = "%a, %d %b %Y %H:%M:%S")
>>> >> >>
>>> >> >>
>>> >> >>
>>> >> >> On Fri, Jun 19, 2009 at 6:06 PM, Thomas
>>> >> >> Levine<thomas.levine at gmail.com>
>>> >> >> wrote:
>>> >> >> > I am analysing occurrences of a phenomenon by time, and each of
>>> >> >> > these
>>> >> >> > timestamps taken from email headers represents one occurrence.
>>> >> >> > (The
>>> >> >> > last
>>> >> >> > number is the time zone.) I can easily change the format.
>>> >> >> >
>>> >> >> > Sun, 14 Jun 2009 07:33:00 -0700
>>> >> >> > Sun, 14 Jun 2009 08:35:10 -0700
>>> >> >> > Sun, 14 Jun 2009 21:26:34 -0700
>>> >> >> > Mon, 15 Jun 2009 19:47:47 -0700
>>> >> >> > Wed, 17 Jun 2009 21:50:41 -0700
>>> >> >> >
>>> >> >> > I've found documentation for a plethora of ways of importing time
>>> >> >> > data,
>>> >> >> > but
>>> >> >> > I can't decide how to approach it. Any ideas on what may be the
>>> >> >> > cleanest
>>> >> >> > way? The only special concern is that I'll want to plot these
>>> >> >> > data by
>>> >> >> > date
>>> >> >> > and time, meaning that I would rather not bin all of the
>>> >> >> > occurrences
>>> >> >> > from
>>> >> >> > one day.
>>> >> >> >
>>> >> >> > The time zone isn't important as these are all local times; the
>>> >> >> > time
>>> >> >> > zone
>>> >> >> > only changes as a function of daylight savings time, so I
>>> >> >> > probably
>>> >> >> > shouldn't
>>> >> >> > use it at all.
>>> >> >> >
>>> >> >> > Tom
>>> >> >> >
>>> >> >> >        [[alternative HTML version deleted]]
>>> >> >> >
>>> >> >> > ______________________________________________
>>> >> >> > R-help at r-project.org mailing list
>>> >> >> > https://stat.ethz.ch/mailman/listinfo/r-help
>>> >> >> > PLEASE do read the posting guide
>>> >> >> > http://www.R-project.org/posting-guide.html
>>> >> >> > and provide commented, minimal, self-contained, reproducible
>>> >> >> > code.
>>> >> >> >
>>> >> >
>>> >> >
>>> >
>>> >
>>
>
>




More information about the R-help mailing list