xPath coding help

[#1]

I am attempting to learn/use xPath in retrieving elements from the JMP Display Output.  For my learning, I have a very simple Bivariate output, and I am attempting to retrieve the Prediction formula, directly from the display output.

The JSL

names default to here(1);
dt = Open("$Sample_Data/big class.jmp");

ow = Bivariate(
	Y( :weight ),
	X( :height ),
	Fit Line( {Line Color( "Medium Dark Green" )} )
);

The Issue:

I can easily retrieve the formula using the Display Tree call of

zap = report(ow)["Linear Fit"][TextBox(1)] << get text;

I can also access it by retrieving the text from the 4th TextBox from an xPath retrieval

zip = (report(ow) << XPath("//TextEditBox"))[4] << get text ;

But since the value of 4 for the subscript, may change if additional fits are added etc.  I would like to use xPath to retrieve the 1st textbox after an outlinebox that has the value of "Linear Fit"

Do any of you xPath experts have a solution.

[#2]

You should be able to do it all with XPath. The first example gives you all the texts. The second example tries to use an XPath function, but I can't get the syntax to work, though.

Names Default to Here( 1 );

dt = Open( "$SAMPLE_DATA/Big Class.jmp" );

biv = dt << Bivariate( Y( :weight ), X( :height ), Fit Line( 1 ) );

biv rep = biv << Report;

// get all texts in a list
biv rep << XPath( "//TextEditBox/text()" );

// get specific text
target = "=";
biv rep << XPath( "//TextEditBox/[contains(text(),target)]/text()" );

[#3]

Aha! This syntax works!

biv rep << XPath( "//TextEditBox[contains(text(),\!"=\!")]/text()" );

[#4]

Super......

I am still thinking there must be a way to return the value from a TextEditBox after detecting a specific OutlineBox, i.e.return the first text box after a specified Outline Box

[#5 · Accepted solution]

rpt=ow<<report;
actText=(rpt << xpath("//OutlineBox[text()='Linear Fit']//TextEditBox"))[1]<<get text;

[#6]

Thank you very much!

[#7]

Here are some other xpath expressions that I've found helpful for navigating the tree structure.

// Note: the * in these examples can also be replaced by the class of
// the element you are looking for e.g. TableBox, OutlineBox

rpt = report(ow);

// Gets immediate children of the Outlinebox (ItsTableBox)
rpt << xpath("//OutlineBox[text()='Summary of Fit']/*");

// Gets all descendants of the Outlinebox (TableBox AND all of its ColBoxes)
rpt << xpath("//OutlineBox[text()='Summary of Fit']//*"); 

// Gets following elements at the same level of the tree
//(Lack of Fit, Analysis of Variances, Parameter Estimates OutlineBoxes)
rpt << xpath("//OutlineBox[text()='Summary of Fit']/following-sibling::*")

// Gets preceding elements at the same level of the tree (TextEditBox with formula)
rpt << xpath("//OutlineBox[text()='Summary of Fit']/preceding-sibling::*")

// Gets ancestors of the box (ListBox, OutlineBox["Linear Fit"], IfBox, OutlineBox["Bivariate Fit..."])
rpt << xpath("//OutlineBox[text()='Summary of Fit']/ancestor::*")

This is the tree structure being navigated in the examples.

This poster has even more examples https://community.jmp.com/t5/Discovery-Summit-2015/JMP-Discovery-2015-Mining-JMP-Reports-v10-pdf/ta-...

[#8]

You CAN approach it that way, Jim, but that way subverts the point of XPath language. It is supposed to do all the work for you.