You can save XML like this
dt = Open( "$SAMPLE_DATA/Big Class.jmp" );
bv = dt << Bivariate( Y( :weight ), X( :height ), Fit Line( {Line Color( {212, 73, 88} )} ) );
xml = bv<<getxml;
file = savetextfile("$temp/teport.xml", xml);
XPATH can search the xml. The xml is a representation of the report's display box tree. It will not include pictures of the report, but you could gather those pictures from the <PictureBox> nodes that XPATH could locate. I think all the numbers in the report will be there, and they will require a bit of study to grab them:
<OutlineBox leftOffset=\!"0\!" topOffset=\!"19\!" width=\!"251\!" height=\!"124\!" helpKey=\!"Bivar Summary\!" isOpen=\!"true\!">Summary of Fit<TableBox leftOffset=\!"14\!" topOffset=\!"29\!" width=\!"237\!" height=\!"97\!">
<StringColBox width=\!"172\!" height=\!"95\!">
<StringColBoxHeader></StringColBoxHeader>
<StringColBoxItem>RSquare</StringColBoxItem>
<StringColBoxItem>RSquare Adj</StringColBoxItem>
<StringColBoxItem>Root Mean Square Error</StringColBoxItem>
<StringColBoxItem>Mean of Response</StringColBoxItem>
<StringColBoxItem>Observations (or Sum Wgts)</StringColBoxItem>
</StringColBox>
<NumberColBox leftOffset=\!"172\!" topOffset=\!"0\!" width=\!"64\!" height=\!"95\!">
<NumberColBoxHeader></NumberColBoxHeader>
<NumberColBoxItem>0.502917</NumberColBoxItem>
<NumberColBoxItem>0.489836</NumberColBoxItem>
<NumberColBoxItem>15.85786</NumberColBoxItem>
<NumberColBoxItem>105</NumberColBoxItem>
<NumberColBoxItem>40</NumberColBoxItem>
</NumberColBox>
</TableBox>
</OutlineBox>
pairing up Root Mean Square Error with 15.85786
Table Box with two columns
will require loading both columns of the table. There won't be a key for Root Mean Square Error, but there is a way to find it.
This is an example of how XML and JSON are nearly identical languages (tree structured, tags ) but in practice XML tends to represent reports while JSON tends to represent data.
If you build something cool that you can share, maybe you can do a presentation at a JMP conference!
Craige