fix: copy value in cv.map

[r3v4s]

package mapindex

var mapus map[uint64]string = make(map[uint64]string)

func init() {
	mapus[3] = "three"
	mapus[5] = "five"
	mapus[9] = "nine"
}

func FindMapWithKey(k uint64) string {
	return mapus[k]
}

println(mapus[5]) will return 'five' in test cases run by gno test which is expected

HOWEVER, with deployed package

$ gnokey maketx addpkg \
-remote 127.0.0.1:26657 \
-broadcast=true \
-chainid dev \
-gas-fee 1ugnot \
-gas-wanted 9000000 \
-memo "" \
-pkgdir ./ \
-pkgpath gno.land/r/demo/mapindex \
test1

calling

$ gnokey maketx call \
-remote 127.0.0.1:26657 \
-broadcast=true \
-chainid dev \
-gas-fee 1ugnot \
-gas-wanted 9000000 \
-memo "" \
-pkgpath gno.land/r/demo/mapindex \
-func FindMapWithKey \
-args "3" \
test1

will return nothing for value, and 'string' for type

$ gnokey maketx call \
-remote 127.0.0.1:26657 \
-broadcast=true \
-chainid dev \
-gas-fee 1ugnot \
-gas-wanted 9000000 \
-memo "" \
-pkgpath gno.land/r/demo/mapindex \
-func FindMapWithKey \
-args "3" \
test1
Enter password.

( string)
OK!
GAS WANTED: 9000000
GAS USED:   73301

Related PR

• fix(gnovm): load map from state #932 @tbruyelle

Contributors' checklist...

• Added new tests, or not needed, or not feasible
• Provided an example (e.g. screenshot) to aid review or the PR is self-explanatory
• Updated the official documentation or not needed
• No breaking changes were made, or a BREAKING CHANGE: xxx message was included in the description
• Added references to related issues and PRs
• Provided any useful hints for running manual tests
• Added new benchmarks to generated graphs, if any. More info here.

[moul]

Looks good.

It would be beneficial if you could incorporate a unit test to prevent potential regressions in the future. Thank you.

[r3v4s]

@moul This bug only occurs in actual environment.
which means you have to run your own gnoland local node then publish package and call it

both of running test file(_test.gno) with gno test or something like this file doesn't reproduce this bug.

So, I'm not sure about what you meant by unit test (at least for this case), could be more specific?

[moul]

@gfanton is developing a new txtar-based integration system (https://github.com/gnolang/gno/compare/master...gfanton:feat/gnoland-txtar-driver?expand=1) designed for crafting such tests.

However, for straightforward tests like these, a simpler alternative might be more fitting.

Regardless, your PR is solid. I'll append a label for enhanced QA in the future. Edit: #1115. Edit: blocked until we finish a more straightforward integration test system to use this PR as a good use case.

[moul]

Blocked until we have a better integration test system.

i.e., #1117, cc @gfanton

[thehowl]

@r3v4s master now has txtar integration tests; you're welcome to update this PR and add reproducible tests for what you outlined in the OP :)

[codecov]

Codecov Report

Attention: 2 lines in your changes are missing coverage. Please review.

Comparison is base (a2b5f29) 56.17% compared to head (121dcb6) 56.16%.

Files	Patch %	Lines
gnovm/pkg/gnolang/realm.go	0.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1112      +/-   ##
==========================================
- Coverage   56.17%   56.16%   -0.01%     
==========================================
  Files         439      439              
  Lines       66242    66244       +2     
==========================================
- Hits        37209    37204       -5     
- Misses      26143    26149       +6     
- Partials     2890     2891       +1     

Flag	Coverage Δ
go-1.21.x	∅ <ø> (∅)
misc	∅ <ø> (∅)
misc-_test.genstd	∅ <ø> (∅)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

Reason for blocking has been resolved; @moul had previously approved the PR.