A couple of years ago, I posted about a technique I picked up from Marie de Blois that allows conflict-checking two- and four-part field-only armory with the O&A complex search form.
It entails running a search for the line type, plus each of the tinctures, and for the tinctures together in reverse order, and the codes for field-only and peripheral-only.
So, to filter the O&A for possible conflicts with “Per fess argent and sable” you could search for:
- PFESS:pl
- PFESS:argent
- PFESS:~and sable
- PFESS:sable:~and argent
- FO — give this line a weight of 2
- PO
Any results that appear with a score of 4 or higher are a potential conflict.
It turns out that the same approach can also be used for fields divided into more than four parts, but the coding for the tinctures requires a slightly different approach.
To filter the O&A for possible conflicts with “Barry wavy Or and gules” you could search for:
- FIELD DIV.-BARRY:wavy
- FIELD DIV.-BARRY:or
- FIELD DIV.-BARRY:~and or
- FIELD DIV.-BARRY:gules
- FIELD DIV.-BARRY:~and gules
- FO — give this line a weight of 2
- PO
Any results that appear with a score of 4 or higher are a potential conflict.
The extra tincture lines are there to handle cases in which a change of one tincture causes the other to be coded differently.
Because tinctures in multiple-part divisions are always coded in alphabetical order, “barry wavy or and gules” and “barry wavy gules and or” will both be encoded in the database as “FIELD DIV.-BARRY:gules:~and or”.
However, we also want our search to find “barry wavy argent and gules”, which would be a conflict as it has only one DC for changing the tincture of half the field — and that armory would be encoded as “FIELD DIV.-BARRY:argent:~and gules”.
Therefore our search needs to include both “:gules” and “~and gules” to ensure that we’ll match the tincture in either position.
Some of these search terms will never match anything and can be excluded — because argent is the first tincture alphabetically, we’ll never have a multi-part field coded as “~and argent”, and similarly because vert is last alphabetically, we’ll never see it appear in this context as “vert” — but they do no harm and are safe to include for purposes of consistency.
[Update, Mar 2021:] I realized recently that both of these recipes have a fatal flaw — there is only a DC between per-fess and barry, not an SC, and thus the above searches could overlook a conflicting registration. I’ll need to put some additional effort into an updated recipe that covers these cases.