Codecov Report

Merging #236 into master will not change coverage.
The diff coverage is n/a.

@@          Coverage Diff          @@
##           master   #236   +/-   ##
=====================================
  Coverage      79%    79%           
=====================================
  Files          19     19           
  Lines        3701   3701           
=====================================
  Hits         2924   2924           
  Misses        567    567           
  Partials      210    210

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 8048762...4278c1c. Read the comment docs.

I'd say this functionality is already covered with Count(), which returns zero if no records are found. It looks a bit prettier tho, you have a point there.

I haven't looked at the code, so maybe I'm missing the point, but sql's count is not efficient way of performing "exists" check.

AFAIK, modern postgres versions use indices to perform simple counts. This should be quite fast (O(n), but in memory).

If I read your code correctly, you perform a query and call rs.Next(), which actually copies values to memory. Only the PK is copied tho (which I missed when I looked at it before), so your method is probably faster.

This seems good to me.

Now that I maintain this project actively, I'm interested on getting this merged. Please add some documentation and usage examples to README.md, as well as some tests to the suite. Then, unless anyone has anything to object, I'll merge it.

Hey @roobre , unfortunately for kallax I switched to sqlboiler. For now I don't have a time for finishing this, maybe later if it's not urgent. Otherwise example usage is provided in my first comment above, and testing this should be straight-forward, so you can get it merged.