Awesome IPLD Endeavour #89
Conversation
daviddias
commented
Oct 3, 2016
daviddias
commented
- remove extension support (no need with CID)
- blockstore starts getting blockBlobs instead of blocks (so that it can know the right key)
|
Why not add cid support directly to Block, The current |
|
I scratched my head a lot thinking about this :). Although having a .cid() on the block would be super sweet, you can't really tell which hash function to use, otherwise, it breaks the Block abstraction. Only the Block Service, hinted by the caller with the CID, can tell which data type and hash function should it use for the data it is looking for. |
| @@ -2,7 +2,7 @@ | |||
| "name": "ipfs-repo", | |||
| "version": "0.9.1", | |||
| "description": "IPFS Repo implementation", | |||
| "main": "lib/index.js", | |||
| "main": "src/index.js", | |||
There was a problem hiding this comment.
TODO: change before merge
| @@ -5,16 +5,16 @@ const pull = require('pull-stream') | |||
| const Lock = require('lock') | |||
| const base32 = require('base32.js') | |||
| const path = require('path') | |||
| const write = require('pull-write') | |||
| const pullWrite = require('pull-write') | |||
| function multihashToPath (multihash, extension) { | ||
| extension = extension || 'data' | ||
| function multihashToPath (multihash) { | ||
| const extension = 'data' |
There was a problem hiding this comment.
if this is fixed, why not make it a constant like PREFIX_LENGTH
| const tasks = blockBlobs.map((blockBlob) => { | ||
| return (cb) => { | ||
| writeBlock(blockBlob, (err, meta) => { | ||
| if (err) { |
There was a problem hiding this comment.
the nesting is strong in this one, maybe simplify this somehow?
|
|
||
| cb() | ||
| }) | ||
| } |
There was a problem hiding this comment.
👎 this change makes it harder to read, and is a matter of personal preference
There was a problem hiding this comment.
I can agree on that for line 44, however, for line 44, the left (previous) is way harder to read because the way the 3 functions are nested is not entirely clear.
| }) | ||
|
|
||
| // TODO ??Why does a putStream need to be a source as well?? |
There was a problem hiding this comment.
@dignifiedquire could you clarify on this one?
daviddias
force-pushed
the
awesome-ipld
branch
from
16c86c1
to
2796f53
Compare
| }) | ||
|
|
||
| // TODO ?? Why does a putStream need to be a source as well?? |
daviddias
force-pushed
the
awesome-ipld
branch
from
2796f53
to
145bc4d
Compare
… that now it receives the key in which it should store it
daviddias
force-pushed
the
awesome-ipld
branch
from
145bc4d
to
f7e4047
Compare
|
for record, @dignifiedquire approved |
