I think there is another memory leak when result != nil AND PQresultStatus(result) != PGRES_COMMAND_OK, I would fixed it but I don't know anything about your coding conventions and result isn't accessible inside the guard statement

Edit: never mind solved it like in executeTransaction

Codecov Report

Merging #40 into master will decrease coverage by 0.04%.
The diff coverage is 60%.

@@            Coverage Diff             @@
##           master      #40      +/-   ##
==========================================
- Coverage   73.42%   73.37%   -0.05%     
==========================================
  Files           3        3              
  Lines         444      447       +3     
==========================================
+ Hits          326      328       +2     
- Misses        118      119       +1

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 79ff943...47200d0. Read the comment docs.

@tunniclm Should we have a look at this?

@EnriqueL8 Missed this notification, sorry! Yes, I think this looks important and we should investigate and merge if it looks good.

LGTM, @ianpartridge what do you think?

LGTM, merge when you're ready. @EnriqueL8 can you also audit the rest of the codebase for missing calls to PQclear()?

Could do

I believe the three cases (Query, PreparedStatement and Transaction) are now covered:

• Query uses a helper function to clear the result
• Transaction calls PQClear directly
• PrepareStatement is this PR which mimics the Transaction case

Cool, thanks. I think this should be tagged as 1.1.1 after merge, can you do this?

okay, merging