Core API refinements and efficiency improvements #3493

ghost · 2016-12-09T15:56:43Z

commit e1c190a
Author: Lars Gierth larsg@systemli.org
Date: Wed Nov 16 06:21:15 2016 +0100

coreapi: smarter way of dealing with the different APIs

License: MIT
Signed-off-by: Lars Gierth <larsg@systemli.org>

commit 2e15ba9
Author: Lars Gierth larsg@systemli.org
Date: Fri Dec 9 16:35:28 2016 +0100

coreapi: be more flexible about inputs

While coreapi functions previously only accepted path strings,
they now accept anything from Path structs and strings,
Multihash structs and strings, and most notably: Cid structs,
strings and byte arrays. See go-cid.Parse() for more information.

If the argument is a string containing at least one forward-slash,
it's treated as a path and resolved accordingly, i.e. including
IPNS resolution and link traversal.

Anything else will be parsed as a Cid and fetched directly using DAG.Get().

This change makes various use cases of the Core API more efficient,
since there's no more need for expensive string/Path/Cid conversions.

License: MIT
Signed-off-by: Lars Gierth <larsg@systemli.org>

commit 7241c02
Author: Lars Gierth larsg@systemli.org
Date: Fri Dec 9 16:49:24 2016 +0100

gateway: make serving index.html a little more efficient

Prevents unneccessary conversions and lookups :):)

License: MIT
Signed-off-by: Lars Gierth <larsg@systemli.org>

hsanjuan · 2016-12-09T20:15:15Z

core/coreapi/interface/interface.go

-// 	ID() CoreID
-// 	Version() CoreVersion
-// }
+type Ref interface{}


I object to such type for API methods parameters. See my comment in #3490 .

whyrusleeping · 2016-12-14T19:47:10Z

blocked on #3490

whyrusleeping · 2017-02-14T23:02:40Z

@lgierth progress here? I think its probably safe to move this to the 0.4.7 milestone for now?

ghost · 2017-02-17T00:44:57Z

progress here? I think its probably safe to move this to the 0.4.7 milestone for now?

Yes good to move, it needs the two changes we discussed but which I haven't posted here yet (will dig them up).

whyrusleeping · 2017-03-11T01:25:21Z

@lgierth any progress here since last update?

ghost · 2017-03-11T01:50:25Z

will take care of this first thing in the docs sprint, so we can get it into 0.4.8

License: MIT Signed-off-by: Lars Gierth <larsg@systemli.org>

The new coreiface.Path maps a path to the cid.Cid resulting from a full path resolution. The path is internally represented as a go-ipfs/path.Path, but that doesn't matter to the outside. Apart from the path-to-CID mapping, it also aims to hold all resolved segment CIDs of the path. Right now it only exposes Root(), and only for flat paths a la /ipfs/Qmfoo. In other cases, the root is nil. In the future, resolution will internally use go-ipfs/path.Resolver.ResolvePathComponents and thus always return the proper resolved segments, via Root(), or a future Segments() func. - Add coreiface.Path with Cid() and Root(). - Add CoreAPI.ResolvePath() for getting a coreiface.Path. - All functions now expect and return coreiface.Path. - Add ParsePath() and ParseCid() for constructing a coreiface.Path. - Add coreiface.Node and Link which are simply go-ipld-node.Node and Link. - Add CoreAPI.ResolveNode() for getting a Node from a Path. License: MIT Signed-off-by: Lars Gierth <larsg@systemli.org>

ghost · 2017-03-17T02:59:33Z

coreapi: make the interfaces path centric

The new coreiface.Path maps a path to the cid.Cid
resulting from a full path resolution.

The path is internally represented as a go-ipfs/path.Path,
but that doesn't matter to the outside.

Apart from the path-to-CID mapping, it also aims to hold all
resolved segment CIDs of the path. Right now it only exposes
Root(), and only for flat paths a la /ipfs/Qmfoo. In other cases,
the root is nil.

In the future, resolution will internally use
go-ipfs/path.Resolver.ResolvePathComponents and thus always return
the proper resolved segments, via Root(), or a future Segments() func.

Add coreiface.Path with Cid() and Root().
Add CoreAPI.ResolvePath() for getting a coreiface.Path.
All functions now expect and return coreiface.Path.
Add ParsePath() and ParseCid() for constructing a coreiface.Path.
Add coreiface.Node and Link which are simply go-ipld-node.Node and Link.
Add CoreAPI.ResolveNode() for getting a Node from a Path.

whyrusleeping · 2017-03-22T01:47:50Z

core/coreapi/interface/interface.go

-	Add(context.Context, io.Reader) (*cid.Cid, error)
-	Cat(context.Context, string) (Reader, error)
-	Ls(context.Context, string) ([]*Link, error)
+	Add(context.Context, io.Reader) (Path, error)


Hrm... this one feels weird. I like that Cat takes a path and returns a reader, and that Add is the inverse operation, but it just feels weird that Add returns a path...

I think i'll get over it. The Path in question can be thought of as /ipfs/Cid

whyrusleeping

This LGTM, thanks @lgierth :)

We're this much closer to having the gateway extracted!

ghost added topic/core-api Topic core-api topic/gateway Topic gateway labels Dec 9, 2016

ghost added this to the ipfs 0.4.5 milestone Dec 9, 2016

ghost added status/in-progress In progress need/review Needs a review labels Dec 9, 2016

hsanjuan reviewed Dec 9, 2016

View reviewed changes

ghost mentioned this pull request Dec 10, 2016

Core API needs to take flexible inputs #3490

Closed

whyrusleeping added the status/blocked Unable to be worked further until needs are met label Dec 14, 2016

ghost force-pushed the feat/coreapi-refinements branch from 7241c02 to c6dc74b Compare December 16, 2016 21:36

Kubuxu modified the milestones: ipfs 0.4.6, ipfs 0.4.5 Jan 10, 2017

whyrusleeping modified the milestones: Ipfs 0.4.7, ipfs 0.4.6 Feb 14, 2017

whyrusleeping modified the milestones: Ipfs 0.4.8, Ipfs 0.4.7 Mar 11, 2017

ghost mentioned this pull request Mar 14, 2017

Create interface-ipfs-core for go-ipfs ipfs-inactive/docs#24

Closed

ghost force-pushed the feat/coreapi-refinements branch from c6dc74b to 963a812 Compare March 16, 2017 15:47

Lars Gierth added 2 commits March 17, 2017 02:35

coreapi: smarter way of dealing with the different APIs

e260d2f

License: MIT Signed-off-by: Lars Gierth <larsg@systemli.org>

ghost force-pushed the feat/coreapi-refinements branch from 963a812 to ee45b8d Compare March 17, 2017 02:58

ghost requested review from whyrusleeping and Kubuxu March 20, 2017 18:16

whyrusleeping reviewed Mar 22, 2017

View reviewed changes

whyrusleeping approved these changes Mar 22, 2017

View reviewed changes

whyrusleeping merged commit 7dee31b into master Mar 22, 2017

whyrusleeping deleted the feat/coreapi-refinements branch March 22, 2017 01:57

whyrusleeping removed the status/in-progress In progress label Mar 22, 2017

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Core API refinements and efficiency improvements #3493

Core API refinements and efficiency improvements #3493

ghost commented Dec 9, 2016 •

edited by Kubuxu

Loading

hsanjuan Dec 9, 2016

whyrusleeping commented Dec 14, 2016

whyrusleeping commented Feb 14, 2017

ghost commented Feb 17, 2017

whyrusleeping commented Mar 11, 2017

ghost commented Mar 11, 2017

ghost commented Mar 17, 2017

whyrusleeping Mar 22, 2017

whyrusleeping left a comment

Core API refinements and efficiency improvements #3493

Core API refinements and efficiency improvements #3493

Conversation

ghost commented Dec 9, 2016 • edited by Kubuxu Loading

hsanjuan Dec 9, 2016

Choose a reason for hiding this comment

whyrusleeping commented Dec 14, 2016

whyrusleeping commented Feb 14, 2017

ghost commented Feb 17, 2017

whyrusleeping commented Mar 11, 2017

ghost commented Mar 11, 2017

ghost commented Mar 17, 2017

whyrusleeping Mar 22, 2017

Choose a reason for hiding this comment

whyrusleeping left a comment

Choose a reason for hiding this comment

ghost commented Dec 9, 2016 •

edited by Kubuxu

Loading