vm: pretty-print remaining opcodes

[fyrchik]

There are a few opcodes which have parameter but are not yet pretty-printed. Fix this.

[codecov]

Codecov Report

Merging #1231 into master will decrease coverage by 0.01%.
The diff coverage is 66.66%.

@@            Coverage Diff             @@
##           master    #1231      +/-   ##
==========================================
- Coverage   68.07%   68.05%   -0.02%     
==========================================
  Files         215      216       +1     
  Lines       18508    18524      +16     
==========================================
+ Hits        12599    12607       +8     
- Misses       5227     5235       +8     
  Partials      682      682              

Impacted Files	Coverage Δ
pkg/core/interop/crypto/interop.go	100.00% <ø> (ø)
pkg/rpc/request/txBuilder.go	83.33% <0.00%> (ø)
pkg/vm/contract_checks.go	82.19% <ø> (ø)
pkg/vm/interop.go	78.26% <ø> (ø)
pkg/vm/vm.go	90.68% <25.00%> (-0.65%)	⬇️
pkg/compiler/codegen.go	89.67% <100.00%> (ø)
pkg/core/interop/callback/callback.go	100.00% <100.00%> (ø)
pkg/core/interop/context.go	96.72% <100.00%> (ø)
pkg/core/interop/interopnames/convert.go	100.00% <100.00%> (ø)
pkg/core/interops.go	100.00% <100.00%> (ø)
... and 5 more

---

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 95d86b6...b2e53fe. Read the comment docs.

[fyrchik]

Looks nice (from the test contract):

528      LDLOC          10 (0a)
530      CONVERT        Integer (21)
532      STLOC          9 (09)
534      LDLOC          9 (09)
536      LDLOC          8 (08)
538      LT
539      JMPIFNOT_L     575 (36/24000000)
544      PUSHDATA1      696e73756666696369656e742066756e6473 ("insufficient funds")
564      SYSCALL        System.Runtime.Log (cfe74796)

[fyrchik]

I haven't implemented maps from id-to-name and reverse, because FromID is used only in PrintOps(). Not hard to add though.

[roman-khimov]

neo-vm update (not strictly needed as the contents we care about didn't really change) should be explicit, not hidden in som a1b9706.

Have you tested 2136011 performance impact? It's being used in VM, so we shouldn't waste time there.

[roman-khimov]

This is what I wanted to avoid --- yet another name registry duplicating declarations made by interops.

[roman-khimov]

This is a perfect case for some code-generation tool, but Go isn't very friendly to those, so as a compromise I think we can make the following:

• add interopnames (not names, please) package that would just contain string constants for all interop functions.
• use its constants for interop registrations
• use its constants for id-to-name in vm/cli

At least that would remove string duplication that is very easy to make error in. It still leaves some potential for constant usage to get out of sync, but that's unavoidable given the constraints we have.

outdated