Pull unique two column combinations and skip particular value

I need to create a summary table listing only unique entries from a source table. The key is split across two columns (Unit, Lesson). I've found an example online to do that much. The difficulty is that I also need to omit entries where 'Lesson' = 0, and I don't have a good enough grasp of array formulas to adapt the solution.

The formula I'm using is:

{=INDEX(Source[Unit],IFERROR(MATCH(TRUE(),ISNA(MATCH(Source[Unit]&Source[Lesson],Summary[[#Headers],[Unit]]:OFFSET([@Unit],-1,0)&Summary[[#Headers],[Lesson]]:OFFSET([@Lesson],-1,0),0)),0),1))}

Right now, it's returning:

Please Login or Register  to view this content.

What I need is:

Please Login or Register  to view this content.

I'm attaching a minimal working example. Any help you can offer on adding the additional criterion or making the whole thing more efficient would be greatly appreciated.

Figured it out.

{=INDEX(Source[Unit],IFERROR(MATCH(TRUE(),ISNA(MATCH(Source[Unit]&Source[Lesson],Summary[[#Headers],[Unit]]:D1&Summary[[#Headers],[Lesson]]:E1,0))*Source[Lesson]<>0,0),2))}

I do still wish there was a more elegant way of handling the #N/A produced by the first iteration, and that I could use structured references throughout without resorting to OFFSET, which slowed everything WAY down. I figured out a way to use INDEX instead that should have worked, but it produced circular reference errors. I'll settle for what I've got unless anyone has any strokes of genius.