How can we get sample hits of auto features?

 https://nexus.hexagon.com/community/public/pc-dmis/f/pc-dmis-for-cmms/117085/extract-sample-points 

For sample hit #2, I could change 1 to 2 ?

Yes, that is correct.

Yes, that is correct.

Thank you.

We have some programs which use datum DEF based on Datum ABC. The datum D is a surface with high offset up to 5 mm in 50 mm long area.  I could use surface check to get datum E and Datum F surface locations, which are circle or round slot. If I could get both datum E and datum F sample hits, I could use them as temp datum D and use relative small  prehit and retract as 2mm than 5 mm. 

If you also want to use less variables, you can throw them into an array, and use something like this. So instead of 27 variables per feature, you can cut it down to 9, if using 3 sample hits. V1[1] is first parameter of the array, so Sample Hit 1 Theoretical X.

$$ NO,
            Sample Hit 1
            ASSIGN/V1=ARRAY(GETTEXT("SAMPLE_HIT_THEO_X", 1, {CIR2}),GETTEXT("SAMPLE_HIT_THEO_Y", 1, {CIR2}),GETTEXT("SAMPLE_HIT_THEO_Z", 1, {CIR2}))
            ASSIGN/V2=ARRAY(GETTEXT("SAMPLE_HIT_THEO_I", 1, {CIR2}),GETTEXT("SAMPLE_HIT_THEO_J", 1, {CIR2}),GETTEXT("SAMPLE_HIT_THEO_K", 1, {CIR2}))
            ASSIGN/V3=ARRAY(GETTEXT("SAMPLE_HIT_MEAS_X", 1, {CIR2}),GETTEXT("SAMPLE_HIT_MEAS_Y", 1, {CIR2}),GETTEXT("SAMPLE_HIT_MEAS_Z", 1, {CIR2}))
F1         =GENERIC/POINT,DEPENDENT,CARTESIAN,$
            NOM/XYZ,<V1[1],V1[2],V1[3]>,$
            MEAS/XYZ,<V3[1],V3[2],V3[3]>,$
            NOM/IJK,<V2[1],V2[2],V2[3]>,$
            MEAS/IJK,<V2[1],V2[2],V2[3]>
$$ NO,
            Sample Hit 2
            ASSIGN/V4=ARRAY(GETTEXT("SAMPLE_HIT_THEO_X", 2, {CIR2}),GETTEXT("SAMPLE_HIT_THEO_Y", 2, {CIR2}),GETTEXT("SAMPLE_HIT_THEO_Z", 2, {CIR2}))
            ASSIGN/V5=ARRAY(GETTEXT("SAMPLE_HIT_THEO_I", 2, {CIR2}),GETTEXT("SAMPLE_HIT_THEO_J", 2, {CIR2}),GETTEXT("SAMPLE_HIT_THEO_K", 2, {CIR2}))
            ASSIGN/V6=ARRAY(GETTEXT("SAMPLE_HIT_MEAS_X", 2, {CIR2}),GETTEXT("SAMPLE_HIT_MEAS_Y", 2, {CIR2}),GETTEXT("SAMPLE_HIT_MEAS_Z", 2, {CIR2}))
F2         =GENERIC/POINT,DEPENDENT,CARTESIAN,$
            NOM/XYZ,<V4[1],V4[2],V4[3]>,$
            MEAS/XYZ,<V6[1],V6[2],V6[3]>,$
            NOM/IJK,<V5[1],V5[2],V5[3]>,$
            MEAS/IJK,<V5[1],V5[2],V5[3]>
$$ NO,
            Sample Hit 3
            ASSIGN/V7=ARRAY(GETTEXT("SAMPLE_HIT_THEO_X", 3, {CIR2}),GETTEXT("SAMPLE_HIT_THEO_Y", 3, {CIR2}),GETTEXT("SAMPLE_HIT_THEO_Z", 3, {CIR2}))
            ASSIGN/V8=ARRAY(GETTEXT("SAMPLE_HIT_THEO_I", 3, {CIR2}),GETTEXT("SAMPLE_HIT_THEO_J", 3, {CIR2}),GETTEXT("SAMPLE_HIT_THEO_K", 3, {CIR2}))
            ASSIGN/V9=ARRAY(GETTEXT("SAMPLE_HIT_MEAS_X", 3, {CIR2}),GETTEXT("SAMPLE_HIT_MEAS_Y", 3, {CIR2}),GETTEXT("SAMPLE_HIT_MEAS_Z", 3, {CIR2}))
F3         =GENERIC/POINT,DEPENDENT,CARTESIAN,$
            NOM/XYZ,<V7[1],V7[2],V7[3]>,$
            MEAS/XYZ,<V9[1],V9[2],V9[3]>,$
            NOM/IJK,<V8[1],V8[2],V8[3]>,$
            MEAS/IJK,<V8[1],V8[2],V8[3]>

Thank you. Much better 

have you thought about doing it the other way around?

measure an auto plane around the hole/slot, then use that as the reference feature for the hole/slot and that way you do not need to use sample hits for them.

I used a wrong number, the Datum D and Datum E were  around 400mm apart, not 50 mm.  I use temp Datum D to measure datum D with 2mm prehit/retract instead of 5 mm, and better surface vector too.