remove old hardfork checks

[oxarbitrage]

There are certain checks in the codebase that are safeguards added before a consensus upgrade is triggered and that are not needed after that date pass.

This checks can only be removed at the next hardfork as they are part of the consensus code.

This ticket is to identify and remove the most of this checks we can for the next consensus release.

Some of this checks are easily identfied in the code with Todo comments just as:
https://github.com/bitshares/bitshares-core/blob/master/libraries/chain/market_evaluator.cpp#L160-L163

[jmjatlanta]

Question 1: In the process of reviewing the code, I have found 4 HARDFORK #defines (so far) that are only #defines. There is no code related to them. I can only assume there used to be, and it was removed when it was no longer needed. Can I remove such xxx.hf files from the repository?

[jmjatlanta]

I am researching each one, collecting information. The summary is being collected here: https://docs.google.com/spreadsheets/d/1l5H3WpkxVQpyI4HzfLMaqfNl1TYRrFBBcEGtXpB7VUE

I realize Google Docs isn't the best place to post such information, but once the data becomes more static, I will post the results here for posterity.

[pmconrad]

Yes, unused hardfork constants can be removed.

[jmjatlanta]

I am coming to the end of my research. One is confusing me. HARDFORK_CORE_583_TIME. Comments say it can be removed, but it seems dangerous to do so.

https://github.com/bitshares/bitshares-core/blob/319d7b96a4d29902db80631022bb021ca79ac950/libraries/chain/market_evaluator.cpp#L302:L334

My synopsis: That GRAPHENE_ASSERT is in the log, therefore it was executed at least once. Changing the behavior to the "else" below includes an extra check, which (in my mind) could cause a change in consensus. Therefore I believe the "if" should remain. Am I missing something?

Update: I am running both the "if" and "else" logic to see if the results would change should the "if" be removed.

[pmconrad]

Hm. Operations that fail on this assertion shouldn't make it into the blockchain.

The only exception that comes to mind is if the operation is contained in a proposal. Can you check? I agree that this cannot be removed if the assertion is triggered.
The comment may have overlooked the possibility of proposals. (Or maybe the comment means that the comment should be removed, not the if. :-)

[abitmore]

I admit that I overlooked the possibility of proposals. If the data has been included in the chain, we should keep the code, and perhaps update the comment explaining why we kept it.

[jmjatlanta]

Reopening

[jmjatlanta]

Fixed in part by #1718. Outstanding issues moved to #1743