REST API method to get path to latest available blockchain state

[dmugtasimov]

TODO:

• Use thenewboston_node.business_logic.blockchain.file_blockchain.get_blockchain_state_filename_meta
• GET /api/v1/blockchain-states-meta/?limit=&offset=&ordering=
• ordering by block_number in both directions must be supported. do not forget to handle special case of blockchain genesis state ( implement by yielding blockchain states in either direction)
• GET /api/v1/blockchain-states-meta/{block_number}/ (if leading zeros are provided then they should be striped from block_number
• Use blockchain application
• Base "urls" on "network_addresses" of the declared node

Response example:

[
        {
		"last_block_number": null,
                "url_path": "/blockchain/account-root-files/0/0/0/0/0/0/0/0/000000000!-arf.msgpack.xz",
		"urls": ["http://3.143.205.184:8555/blockchain/account-root-files/0/0/0/0/0/0/0/0/000000000!-arf.msgpack.xz"]
	},
	{
		"last_block_number": 10,
                "url_path": "/blockchain/account-root-files/0/0/0/0/0/0/0/0/0000000010-arf.msgpack.xz",
		"urls": ["http://3.143.205.184:8555/blockchain/account-root-files/0/0/0/0/0/0/0/0/0000000010-arf.msgpack.xz"]
	}
]

Use this code for pagination:

from collections import OrderedDict

from rest_framework.pagination import LimitOffsetPagination
from rest_framework.response import Response


class CustomLimitOffsetPaginationBase(LimitOffsetPagination):
    default_limit = 5
    max_limit = 20

    def get_paginated_dict(self, data):
        raise NotImplementedError('Must be implemented in a child class')

    def get_paginated_response(self, data):
        return Response(self.get_paginated_dict(data))


class CustomLimitOffsetPagination(CustomLimitOffsetPaginationBase):

    def get_paginated_dict(self, data):
        return OrderedDict((('count', self.count), ('results', data)))

[dmugtasimov]

@andbortnik I rename path to url_path

[andbortnik]

@dmugtasimov Please review #291

What do we do if the requested node is unregistered in the blockchain? In the PR urls is empty in this case.

[dmugtasimov]

@andbortnik let's serve as usual. Just put a HIGH priority TODO about it. Because normally if a node unregisters itself it should also become offline and also even if it does not since it is unregistered other nodes should not consider it as available and therefore do not make API requests to it.

[dmugtasimov]

@andbortnik I added some comments (maybe I will have more later). The main issue is that representation logic was put into business logic. Please, address the comments.

[dmugtasimov]

@andbortnik also if you wish you may submit changes in smaller PRs. I do it sometimes myself: just abandon the entire large PR and make many more smaller PRs untill I cover everything I had in the large PR originally. This up to you, just a suggestion.

[andbortnik]

@dmugtasimov Fixed

[dmugtasimov]

@andbortnik is this PR #329 ready for merge?

[andbortnik]

@dmugtasimov Ready for review:

• Blockchain states meta API: return 404 if block number not found/accept null and genesis special values #329
• Blockchain state meta API: return 400 if node undeclared #350
• Blockchain state meta API: support memory blockchain #351

[dmugtasimov]

@andbortnik #329 merged, thank you

[dmugtasimov]

@andbortnik please, address the comment for #350

[dmugtasimov]

@andbortnik rebase required: #351

[andbortnik]

@dmugtasimov Ready for review

• Blockchain state meta API: return 400 if node undeclared #350
• Blockchain state meta API: support memory blockchain #351
• Blockchain state meta API: use custom ordering filter #364

[dmugtasimov]

@andbortnik #350 merged, thank you

[dmugtasimov]

Fixup: #366

[dmugtasimov]

@andbortnik Thank you #351 is merged

[dmugtasimov]

@andbortnik thank you, #364 merged

[dmugtasimov]

@andbortnik is there anything left to do in this task?

[andbortnik]

@andbortnik is there anything left to do in this task?

Nothing. I'll close the task