RpcServer, RpcClient: strip HF_ prefix from getversion response

[AnnaShaleva]

It's nice to have clear hardfork ame without hardfork prefix so that the resulting hardfork JSON-ized name can be applicable by both C# and NeoGo nodes.

Ref. #822.

[Jim8y]

weird to make C# aligned to neogo, instead of neogo aligned to C#. If it is not a bug or improvement proposal, i dont have any motivation to stay aligned to neogo.

@shargon Any thought on this?

[roman-khimov]

It's obviously an improvement for the RPC protocol. These prefixes are OK for some code-level constants in a big module, but they're not really interesting for RPC reply that has hardforks section.

Note also, that their introduction into the protocol was accidental, not intentional (as well as inclusion into 3.6.1).

[cschuchardt88]

you have to remove

neo-modules/tests/Neo.Network.RPC.Tests/RpcTestCases.json

Lines 2495 to 2500 in 6ead780

	"hardforks": [
	{
	"name": "HF_Aspidochelone",
	"blockheight": 0
	}
	]

for the tests to work again.

Or Change to reflect your changes.

HF_HF_Aspidochelone is your name that you are using. No need to add prefix. the enum name as it already.

Prefix should be removed. and should be like this Aspidochelone.

Its in section hardfork already. so no need for prefix of HF that's redundant

May need to fix #825 as well...

[Jim8y]

Changes to the interface should be very careful, we dont know if someone is already using it or not, and it is not a good experience to change the inteface from time to time. Yes, without HF prefix is much reasonable, but since it has HF and has being in this way for a well, then it become a historical problem.

I wont stand my opinion if @shargon agrees to update, but before that, i dont think it is necessary or should be done in another compatible way without breaking existing interface.

[cschuchardt88]

why?

[AnnaShaleva]

See the #838 (comment) and updated version.

[cschuchardt88]

you are adding it again. why?

[cschuchardt88]

Test fails with System.ArgumentException: Requested value 'HF_HF_Aspidochelone' was not found.

[AnnaShaleva]

I accidentally forgot to commit the test file changes, fixed.

[shargon]

I prefer to use a method to check if the string startWith, instead of replace always the first 3 chars.

[AnnaShaleva]

As Roman said in #838 (comment), the initial idea for hardforks naming was to use "Aspidochelone", "Basilisk" and etc. for user-facing outputs (see the neo-project/neo#2749 (comment)), and HF_Aspidochelone is just a code-level enum implementation that is expected to stay at the code level. So it would be nice to keep human-readable names in the JSON response output, as we already have them under hardforks section.

[AnnaShaleva]

I prefer to use a method to check if the string startWith, instead of replace always the first 3 chars.

Fixed, see the updated version, please.