Conversation
hannahhoward
left a comment
There was a problem hiding this comment.
LGTM and on the the two connected PRs.. but I don't consider myself a core API expert by any means.
I have two quick questions:
- Why move unixfile to go-unixfs & why put tar-writer in go-ipfs-files instead of go-unixfs?
- Am I correct in understanding that neither of these two changes are strickly essential to delivering the refactor to make get use CoreAPI?
Non-blocking, just want to understand.
Because However: ipfs/go-unixfs#59 (comment)
We need some way to go from a |
As a note - I decided to drop it as the combo of |
License: MIT Signed-off-by: Łukasz Magiera <magik6k@gmail.com>
License: MIT Signed-off-by: Łukasz Magiera <magik6k@gmail.com>
License: MIT Signed-off-by: Łukasz Magiera <magik6k@gmail.com>
License: MIT Signed-off-by: Łukasz Magiera <magik6k@gmail.com>
7601f27 to
21a05dd
Compare
| return err | ||
| } | ||
|
|
||
| res.SetLength(uint64(size)) |
There was a problem hiding this comment.
This is actually incorrect but it's an existing bug... #5690
There was a problem hiding this comment.
The alternative is not setting this at all, but it will break progress bar... (We could pass this in a separate header)
There was a problem hiding this comment.
I think we have to. Unfortunately, that'll require a change to the commands library.
Stebalien
left a comment
There was a problem hiding this comment.
Looks like this doesn't work with sharding. t0260-sharding.sh is failing.
|
Error is: "Error: cannot write to readonly DAGService" |
|
The issue is: https://github.com/ipfs/go-unixfs/blob/0eb1ef8f33371b1b9f74f74655b584144919ce91/file/unixfile.go#L121. This is causing the hamt to serialize and save itself. |
|
It should probably just do |
|
Ah, right, it breaks the progress bar ._. |
|
Maybe we should have a separate hamt reader? |
License: MIT Signed-off-by: Łukasz Magiera <magik6k@gmail.com>
21a05dd to
7c2aa0e
Compare
Refactor ipfs get to use CoreAPI This commit was moved from ipfs/kubo@35aaa98
Depends on: