According to 1.4.3.1 of the P5010X222 & I5010X223 guides, numerically equivalent subloops do not need to be sent in alpha-order.
Should Numerically equivalent subloops like the one below pass syntax check in a translator or must they be in order?
2300 CLM CLM*479-1-9*40***11:B:1*Y*A*Y*Y
2300 DTP DTP*454*D8*20110323
2300 DTP DTP*431*D8*20110530
2300 REF REF*D9*123456789
2300 CR2 CR2********A
2300 HI HI*BK:7231
2310C NM1 NM1*77*2*Smith CHIROPRACTIC INC*****XX*1649257239
2310C N3 N3*4817 NE 2ND LOOP
2310C N4 N4*OCALA*FL*344701524
2310B NM1 NM1*82*1*ZANETTI DEBOLT*DINA****XX*1649257239
2310B PRV PRV*PE*PXC*111N00000X
2400 LX LX*1
Numerically equivalent loops can be sent in any order. In the scenario presented it is valid to have a 2310C loop appear in the data before a 2310B loop.
This issue is explicitly addressed in X12.59 (IMPLEMENTATION OF EDI STRUCTURES-SEMANTIC IMPACT), section 6.2, example number 2 (Loop Independence).
5010X223