Hello,
The following works for me:
library(Hmisc)
timeseries<-c(1950,2000,2050,2100)
dataseries<-seq(1:4)
dataseries1<-c(1,2,3,4)
dataseries2<-c(1.5,2.5,3.5,4)
plot(timeseries,dataseries,type="n")
lines(timeseries,dataseries1,col="red",type="l")
lines(timeseries,dataseries2,col="blue",type="l")
curves<-list()
curves[[1]]<-list(x=timeseries,y=dataseries1)
curves[[2]]<-list(x=timeseries,y=dataseries2)
dataSeriesLabels<-c("this legend should have pch=19","this legend should have
pch=5")
labcurve(curves, dataSeriesLabels, type=rep("l",2), lty=rep("solid",2), lwd=1,
col=c("red","blue"), keys=c(19,5), cex=1, adj="auto", keyloc=c("auto"),
transparent=FALSE)
Another way is to use legend:
library(Hmisc)
timeseries<-c(1950,2000,2050,2100)
dataseries<-seq(1:4)
dataseries1<-c(1,2,3,4)
dataseries2<-c(1.5,2.5,3.5,4)
plot(timeseries,dataseries,type="n")
lines(timeseries,dataseries1,col="red",type="o",pch=19)
lines(timeseries,dataseries2,col="blue",type="o",pch=5)
curves<-list()
curves[[1]]<-list(x=timeseries,y=dataseries1)
curves[[2]]<-list(x=timeseries,y=dataseries2)
dataSeriesLabels<-c("this legend should have pch=19","this legend should
have pch=5")
legend(1950,3.6,dataSeriesLabels,lty=1, pch=c(19,5), col=c('red','blue'))
HTH,
Pascal
2013/8/28 Igor Ribeiro <[email protected]>
> Hi Pascal,
> Still... there is something wrong. I changed the two lines as you
> suggested, and now I'm using the keys parameter. But the output is not as
> expected. That's why I was controlling the points using points function,
> because that works well.
> Let me try to show you with the code:
>
> # Example 1 - the points are not over the lines and are not distributed
> the way I want (see Example 2)
> library(Hmisc)
> rm(list=ls(all=TRUE))
> timeseries<-c(1950,2000,2050,2100)
> dataseries<-seq(1:4)
> dataseries1<-c(1,2,3,4)
> dataseries2<-c(1.5,2.5,3.5,4)
> plot(timeseries,dataseries,type="n")
> lines(timeseries,dataseries1,col="red",type="l")
> lines(timeseries,dataseries2,col="blue",type="l")
> curves<-list()
> curves[[1]]<-list(x=timeseries,y=dataseries1)
> curves[[2]]<-list(x=timeseries,y=dataseries2)
> dataSeriesLabels<-c("this legend should have pch=19","this legend should
> have pch=5")
> labcurve(curves, dataSeriesLabels, type=rep("s",2), lty=rep("solid",2),
> lwd=1, col=c("red","blue"), keys=c(19,5), cex=1, adj="auto",
> keyloc=c("auto"), transparent=FALSE)
>
> # Example 2 - this is the way I want the lines, but then I can't put the
> points (pch) in the legend...
> library(Hmisc)
> rm(list=ls(all=TRUE))
> timeseries<-c(1950,2000,2050,2100)
> dataseries<-seq(1:4)
> dataseries1<-c(1,2,3,4)
> dataseries2<-c(1.5,2.5,3.5,4)
> plot(timeseries,dataseries,type="n")
> points (timeseries, dataseries1, pch=19, col="red", cex=1.3)
> lines(timeseries,dataseries1,col="red",type="l")
> points (timeseries, dataseries2, pch=5, col="blue", cex=1.3)
> lines(timeseries,dataseries2,col="blue",type="l")
> curves<-list()
> curves[[1]]<-list(x=timeseries,y=dataseries1)
> curves[[2]]<-list(x=timeseries,y=dataseries2)
> dataSeriesLabels<-c("this legend should have pch=19","this legend should
> have pch=5")
> labcurve(curves, dataSeriesLabels, type=rep("s",2), lty=rep("solid",2),
> lwd=1, col=c("red","blue"), keys="lines", cex=1, adj="auto",
> keyloc=c("auto"), transparent=FALSE)
>
> You see the difference? I need to put the legend generated in Example 1 in
> the plot generated in Example 2. So far I was unable to mix them!
> I appreciate all your help, thank you.
> Igor.
>
>
> On Mon, Aug 26, 2013 at 9:44 PM, Pascal Oettli <[email protected]> wrote:
>
>> Hello,
>>
>> That is why I changed 2 lines in the code. Because points are misplaced
>> if you keep points with the lines.
>>
>> Regards,
>> Pascal
>>
>>
>> 2013/8/27 Igor Ribeiro <[email protected]>
>>
>>> Because I'm already controlling points using points function. So I don't
>>> want labcurve to change anything on the lines... Just draw the legend the
>>> way I need.
>>> On Aug 26, 2013 8:35 PM, "Pascal Oettli" <[email protected]> wrote:
>>>
>>>> Hello,
>>>>
>>>> Please keep the r-list included when you reply.
>>>>
>>>> Why do you want to add points to lines only in the legend? If so, the
>>>> legend would be incorrect.
>>>>
>>>> Regards,
>>>> Pascal
>>>>
>>>>
>>>> 2013/8/26 Igor Ribeiro <[email protected]>
>>>>
>>>>> Hi Pascal,
>>>>> Thank you very much - your solution works partially - it will include
>>>>> the pch in the legend, but also in the curve, which I don't want. I don't
>>>>> see a way to remove the pch from the curve and leave it only in the
>>>>> legend.
>>>>> Do you have any idea? method="none" doesn't work because then pch is not
>>>>> drawn in the legend :(
>>>>> Thanks!
>>>>> Igor.
>>>>>
>>>>>
>>>>> On Fri, Aug 23, 2013 at 3:09 AM, Pascal Oettli <[email protected]>wrote:
>>>>>
>>>>>> Hello,
>>>>>>
>>>>>> In labcurve, use "keys=c(19,5)". It is said in examples provided in
>>>>>> the help page.
>>>>>>
>>>>>> Regards,
>>>>>> Pascal
>>>>>>
>>>>>>
>>>>>> 2013/8/23 Igor Ribeiro <[email protected]>
>>>>>>
>>>>>>> Dear all,
>>>>>>> I have a plot with two lines and I'm using labcurbe (package Hmisc)
>>>>>>> to show
>>>>>>> the legend.
>>>>>>> Everything works well, except that the line displayed in legend box
>>>>>>> should
>>>>>>> have same style (plotting character) as the line in the plot. This is
>>>>>>> achieved by pch parameter and looks like that it's only supported via
>>>>>>> putKey function (according to the documention here:
>>>>>>> http://www.inside-r.org/packages/cran/Hmisc/docs/bezier).
>>>>>>>
>>>>>>> However, I've been unsucessful to correctly use this function
>>>>>>> (putKey).
>>>>>>> Could someone help me fix the code below?
>>>>>>>
>>>>>>> library(Hmisc)
>>>>>>>
>>>>>>> rm(list=ls(all=TRUE))
>>>>>>> timeseries<-c(1950,2000,2050,2100)
>>>>>>> dataseries<-seq(1:4)
>>>>>>> dataseries1<-c(1,2,3,4)
>>>>>>> dataseries2<-c(1.5,2.5,3.5,4)
>>>>>>> plot(timeseries,dataseries,type="n")
>>>>>>> lines(timeseries,dataseries1,col="red",type="o",pch=19)
>>>>>>> lines(timeseries,dataseries2,col="blue",type="o",pch=5)
>>>>>>> curves<-list()
>>>>>>> curves[[1]]<-list(x=timeseries,y=dataseries1)
>>>>>>> curves[[2]]<-list(x=timeseries,y=dataseries2)
>>>>>>> dataSeriesLabels<-c("this legend should have pch=19","this legend
>>>>>>> should
>>>>>>> have pch=5")
>>>>>>>
>>>>>>> # PROBLEM BELOW
>>>>>>> #putKey(z=curves,labels=dataSeriesLabels,pch=19)
>>>>>>>
>>>>>>> labcurve(curves, dataSeriesLabels, type=rep("s",2),
>>>>>>> lty=rep("solid",2),
>>>>>>> lwd=1, col=c("red","blue"), keys="lines", cex=1, adj="auto",
>>>>>>> keyloc=c("auto"), transparent=FALSE)
>>>>>>>
>>>>>>> Thank you,
>>>>>>> --
>>>>>>> Igor.
>>>>>>> [email protected]
>>>>>>>
>>>>>>> [[alternative HTML version deleted]]
>>>>>>>
>>>>>>> ______________________________________________
>>>>>>> [email protected] 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.
>>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Igor.
>>>>> [email protected]
>>>>>
>>>>>
>>>>
>>
>
>
> --
> Igor.
> [email protected]
>
>
[[alternative HTML version deleted]]
______________________________________________
[email protected] 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.
