1
$\begingroup$

Extracting the points from a line in a ListPlot in this answer uses

points = Cases[Normal@plot, Line[pts_] :> pts, Infinity];

This works in DateListPlot (provided Joined->True) (example below, using Last to get the y-axis data only)

Clear[test,plot,points];
test={{"2010",1},{"2011",2},{"2012",3},{"2014",4},{"2015",5},{"2016",6}};
plot=DateListPlot[test,Joined->True]
points=Last/@Flatten[Cases[Normal@plot,Line[pts_]:>pts,Infinity],1]

enter image description here

(*OUTPUT
{1.,2.,3.,4.,5.,6.}

However, when the plot is limited by PlotRange using date values (lower plot), the function returns the entire list of points.

plot2=DateListPlot[test,Joined->True,PlotRange->{{"2012","2015"},{0,6}}]
points2=Last/@Flatten[Cases[Normal@plot2,Line[pts_]:>pts,Infinity],1]

enter image description here

(*OUTPUT
{1.,2.,3.,4.,5.,6.}

Is there a function, like points which can return only the points displayed on the plot in DateListPlot under 'PlotRange` ?

Improve this question
$\endgroup$
4
  • 1
    $\begingroup$ You can just use Select on the result with conditions similar to those in PlotRange. $\endgroup$
    – Kuba ♦
    Jun 9, 2014 at 23:27
  • 1
    $\begingroup$ Related $\endgroup$
    – ciao
    Jun 9, 2014 at 23:33
  • $\begingroup$ @Kuba Appreciated. Pickett is along those lines too and delves into the plot function to retrieve the bounds $\endgroup$
    – PlaysDice
    Jun 9, 2014 at 23:41
  • $\begingroup$ @rasher Thanks. The linked answer is a serious Module indeed! $\endgroup$
    – PlaysDice
    Jun 9, 2014 at 23:51

2 Answers 2

Reset to default
3
$\begingroup$

This should do it:

points2 = 
 With[{range = Part[PlotRange /. Options[plot2], 1]}, 
  Select[Flatten[Cases[Normal@plot2, Line[pts_] :> pts, Infinity], 1],
     First@range <= First@# && First@# <= Last@range &][[All, 2]]]

With plot2 as in your example:

{3., 4., 5.}

Improve this answer
$\endgroup$
3
  • $\begingroup$ Yes! I was hoping someone could retrieve the values with Options magic. $\endgroup$
    – PlaysDice
    Jun 9, 2014 at 23:40
  • 1
    $\begingroup$ I worked through what you did. Using the Options to make a set of replace rules and PlotRange as the "target of those replace rules" is very neat. I learned something new (again). $\endgroup$
    – PlaysDice
    Jun 10, 2014 at 2:20
  • $\begingroup$ @PlaysDice I'm glad you liked it! :) $\endgroup$
    – C. E.
    Jun 10, 2014 at 2:27
2
$\begingroup$

A variant of Pickett's answer:

  • using new-in-10 RegionMember
  • with a terse way to extract the plot range
  • extended to either Line or Point data
  • written as a reusable function

Code:

inrangeData[gr_Graphics] := 
  Select[Rectangle @@ (PlotRange[gr]\[Transpose]) // RegionMember] /@ 
    Cases[Normal @ gr, _Point | _Line, -1][[All, 1]]

Test:

inrangeData[plot2]

{{{3.53436*10^9, 3.}, {3.59752*10^9, 4.}, {3.62906*10^9, 5.}}}

ListPlot[Prime @ Range @ 25, PlotRange -> {20, 40}]

% // inrangeData

enter image description here

{{{9., 23.}, {10., 29.}, {11., 31.}, {12., 37.}}}

The extra bracket depth is there to allow the function to handle plots with multiple lines or sets:

ListLinePlot[
  Table[{k, PDF[BinomialDistribution[50, p], k]}, {p, {0.3, 0.5, 0.8}}, {k, 0, 50}], 
  PlotRange -> {0.08, 0.15},
  Filling -> Axis
]

% // inrangeData

enter image description here

{
 {
  {11.838, 0.08}, {11.8528, 0.08035}, {12., 0.0838297}, {13., 0.105017},
  {14., 0.118948}, {15., 0.122347}, {16., 0.1147}, {17., 0.0983144},
  {17.8527, 0.08035}, {17.8693, 0.08}
 },
 {
  {22.0685, 0.08}, {22.089, 0.08035}, {23., 0.0959617}, {24., 0.107957},
  {25., 0.112275}, {26., 0.107957}, {27., 0.0959617}, {27.911, 0.08035},
  {27.9315, 0.08}
 },
 {
  {37.1629, 0.08}, {37.1755, 0.08035}, {38., 0.103275}, {39., 0.127108},
  {40., 0.139819}, {41., 0.136409}, {42., 0.116922}, {43., 0.0870116},
  {43.2105, 0.08035}, {43.2216, 0.08}
 }
}
Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.