merkledag

package module
v0.11.0 Latest Latest
Warning

This package is not in the latest version of its module.

Go to latest
Published: Jun 9, 2023 License: MIT Imports: 24 Imported by: 471

README

go-merkledag

Coverage Status

go-merkledag implements the 'DAGService' interface and adds two ipld node types, Protobuf and Raw

Status

❗ This library is maintained, but not actively developed. It will continue to receive fixes and security updates for users that depend on it. However, it may be deprecated in the future and it is recommended that you use alternatives to the functionality in go-merkledag, including:

License

MIT © Juan Batiz-Benet

Documentation

Overview

Package merkledag implements the IPFS Merkle DAG data structures.

Index

Constants

This section is empty.

Variables

View Source
var (
	ErrNotProtobuf  = fmt.Errorf("expected protobuf dag node")
	ErrNotRawNode   = fmt.Errorf("expected raw bytes node")
	ErrLinkNotFound = fmt.Errorf("no link by that name")
)

Common errors

View Source
var ErrReadOnly = fmt.Errorf("cannot write to readonly DAGService")

ErrReadOnly is used when a read-only datastructure is written to.

Functions

func DecodeProtobufBlock

func DecodeProtobufBlock(b blocks.Block) (format.Node, error)

DecodeProtobufBlock is a block decoder for protobuf IPLD nodes conforming to node.DecodeBlockFunc

func DecodeRawBlock

func DecodeRawBlock(block blocks.Block) (format.Node, error)

DecodeRawBlock is a block decoder for raw IPLD nodes conforming to `node.DecodeBlockFunc`.

func FetchGraph

func FetchGraph(ctx context.Context, root cid.Cid, serv format.DAGService) error

FetchGraph fetches all nodes that are children of the given node

func FetchGraphWithDepthLimit

func FetchGraphWithDepthLimit(ctx context.Context, root cid.Cid, depthLim int, serv format.DAGService) error

FetchGraphWithDepthLimit fetches all nodes that are children to the given node down to the given depth. maxDepth=0 means "only fetch root", maxDepth=1 means "fetch root and its direct children" and so on... maxDepth=-1 means unlimited.

func NewDAGService

func NewDAGService(bs bserv.BlockService) *dagService

NewDAGService constructs a new DAGService (using the default implementation). Note that the default implementation is also an ipld.LinkGetter.

func NewReadOnlyDagService

func NewReadOnlyDagService(ng ipld.NodeGetter) ipld.DAGService

NewReadOnlyDagService takes a NodeGetter, and returns a full DAGService implementation that returns ErrReadOnly when its 'write' methods are invoked.

func NewSession

func NewSession(ctx context.Context, g ipld.NodeGetter) ipld.NodeGetter

NewSession returns a session backed NodeGetter if the given NodeGetter implements SessionMaker.

func PrefixForCidVersion

func PrefixForCidVersion(version int) (cid.Prefix, error)

PrefixForCidVersion returns the Protobuf prefix for a given CID version

func ProtoNodeConverter added in v0.4.0

func ProtoNodeConverter(b blocks.Block, nd ipld.Node) (legacy.UniversalNode, error)

func RawNodeConverter added in v0.4.0

func RawNodeConverter(b blocks.Block, nd ipld.Node) (legacy.UniversalNode, error)

func V0CidPrefix

func V0CidPrefix() cid.Prefix

V0CidPrefix returns a prefix for CIDv0

func V1CidPrefix

func V1CidPrefix() cid.Prefix

V1CidPrefix returns a prefix for CIDv1 with the default settings

func Walk added in v0.2.0

func Walk(ctx context.Context, getLinks GetLinks, c cid.Cid, visit func(cid.Cid) bool, options ...WalkOption) error

WalkGraph will walk the dag in order (depth first) starting at the given root.

func WalkDepth added in v0.2.0

func WalkDepth(ctx context.Context, getLinks GetLinks, c cid.Cid, visit func(cid.Cid, int) bool, options ...WalkOption) error

WalkDepth walks the dag starting at the given root and passes the current depth to a given visit function. The visit function can be used to limit DAG exploration.

func WrapSession added in v0.5.0

func WrapSession(s *bserv.Session) format.NodeGetter

WrapSession wraps a blockservice session to satisfy the format.NodeGetter interface

Types

type ComboService

type ComboService struct {
	Read  ipld.NodeGetter
	Write ipld.DAGService
}

ComboService implements ipld.DAGService, using 'Read' for all fetch methods, and 'Write' for all methods that add new objects.

func (*ComboService) Add

func (cs *ComboService) Add(ctx context.Context, nd ipld.Node) error

Add writes a new node using the Write DAGService.

func (*ComboService) AddMany

func (cs *ComboService) AddMany(ctx context.Context, nds []ipld.Node) error

AddMany adds nodes using the Write DAGService.

func (*ComboService) Get

func (cs *ComboService) Get(ctx context.Context, c cid.Cid) (ipld.Node, error)

Get fetches a node using the Read DAGService.

func (*ComboService) GetMany

func (cs *ComboService) GetMany(ctx context.Context, cids []cid.Cid) <-chan *ipld.NodeOption

GetMany fetches nodes using the Read DAGService.

func (*ComboService) Remove

func (cs *ComboService) Remove(ctx context.Context, c cid.Cid) error

Remove deletes a node using the Write DAGService.

func (*ComboService) RemoveMany

func (cs *ComboService) RemoveMany(ctx context.Context, cids []cid.Cid) error

RemoveMany deletes nodes using the Write DAGService.

type ErrorService

type ErrorService struct {
	Err error
}

ErrorService implements ipld.DAGService, returning 'Err' for every call.

func (*ErrorService) Add

func (cs *ErrorService) Add(ctx context.Context, nd ipld.Node) error

Add returns the cs.Err.

func (*ErrorService) AddMany

func (cs *ErrorService) AddMany(ctx context.Context, nds []ipld.Node) error

AddMany returns the cs.Err.

func (*ErrorService) Get

func (cs *ErrorService) Get(ctx context.Context, c cid.Cid) (ipld.Node, error)

Get returns the cs.Err.

func (*ErrorService) GetMany

func (cs *ErrorService) GetMany(ctx context.Context, cids []cid.Cid) <-chan *ipld.NodeOption

GetMany many returns the cs.Err.

func (*ErrorService) Remove

func (cs *ErrorService) Remove(ctx context.Context, c cid.Cid) error

Remove returns the cs.Err.

func (*ErrorService) RemoveMany

func (cs *ErrorService) RemoveMany(ctx context.Context, cids []cid.Cid) error

RemoveMany returns the cs.Err.

type GetLinks func(context.Context, cid.Cid) ([]*format.Link, error)

GetLinks is the type of function passed to the EnumerateChildren function(s) for getting the children of an IPLD node.

func GetLinksDirect

func GetLinksDirect(serv format.NodeGetter) GetLinks

GetLinksDirect creates a function to get the links for a node, from the node, bypassing the LinkService. If the node does not exist locally (and can not be retrieved) an error will be returned.

func GetLinksWithDAG

func GetLinksWithDAG(ng format.NodeGetter) GetLinks

GetLinksWithDAG returns a GetLinks function that tries to use the given NodeGetter as a LinkGetter to get the children of a given IPLD node. This may allow us to traverse the DAG without actually loading and parsing the node in question (if we already have the links cached).

type LinkSlice

type LinkSlice []*format.Link

LinkSlice is a slice of format.Links

func (LinkSlice) Len

func (ls LinkSlice) Len() int

func (LinkSlice) Less

func (ls LinkSlice) Less(a, b int) bool

func (LinkSlice) Swap

func (ls LinkSlice) Swap(a, b int)

type ProgressTracker

type ProgressTracker struct {
	Total int
	// contains filtered or unexported fields
}

ProgressTracker is used to show progress when fetching nodes.

func (*ProgressTracker) DeriveContext

func (p *ProgressTracker) DeriveContext(ctx context.Context) context.Context

DeriveContext returns a new context with value "progress" derived from the given one.

func (*ProgressTracker) Increment

func (p *ProgressTracker) Increment()

Increment adds one to the total progress.

func (*ProgressTracker) Value

func (p *ProgressTracker) Value() int

Value returns the current progress.

type ProtoNode

type ProtoNode struct {
	// contains filtered or unexported fields
}

ProtoNode represents a node in the IPFS Merkle DAG. nodes have opaque data and a set of navigable links. ProtoNode is a go-ipld-legacy.UniversalNode, meaning it is both a go-ipld-prime node and a go-ipld-format node. ProtoNode maintains compatibility with it's original implementation as a go-ipld-format only node, which included some mutability, namely the the ability to add/remove links in place

TODO: We should be able to eventually replace this implementation with * go-codec-dagpb for basic DagPB encode/decode to go-ipld-prime * go-unixfsnode ADLs for higher level DAGPB functionality For the time being however, go-unixfsnode is read only and this mutable protonode implementation is needed to support go-unixfs, the only library that implements both read and write for UnixFS v1.

func DecodeProtobuf

func DecodeProtobuf(encoded []byte) (*ProtoNode, error)

DecodeProtobuf decodes raw data and returns a new Node instance.

func NodeWithData

func NodeWithData(d []byte) *ProtoNode

NodeWithData builds a new Protonode with the given data.

func (n *ProtoNode) AddNodeLink(name string, that format.Node) error

AddNodeLink adds a link to another node. The link will be added in sorted order.

If sorting has not already been applied to this node (because it was deserialized from a form that did not have sorted links), the links list will be sorted. If a ProtoNode was deserialized from a badly encoded form that did not already have its links sorted, calling AddNodeLink and then RemoveNodeLink for the same link, will not result in an identically encoded form as the links will have been sorted.

func (n *ProtoNode) AddRawLink(name string, l *format.Link) error

AddRawLink adds a copy of a link to this node. The link will be added in sorted order.

If sorting has not already been applied to this node (because it was deserialized from a form that did not have sorted links), the links list will be sorted. If a ProtoNode was deserialized from a badly encoded form that did not already have its links sorted, calling AddRawLink and then RemoveNodeLink for the same link, will not result in an identically encoded form as the links will have been sorted.

func (*ProtoNode) AsBool added in v0.4.0

func (n *ProtoNode) AsBool() (bool, error)

func (*ProtoNode) AsBytes added in v0.4.0

func (n *ProtoNode) AsBytes() ([]byte, error)

func (*ProtoNode) AsFloat added in v0.4.0

func (n *ProtoNode) AsFloat() (float64, error)

func (*ProtoNode) AsInt added in v0.4.0

func (n *ProtoNode) AsInt() (int64, error)
func (n *ProtoNode) AsLink() (ipld.Link, error)

func (*ProtoNode) AsString added in v0.4.0

func (n *ProtoNode) AsString() (string, error)

func (*ProtoNode) Cid

func (n *ProtoNode) Cid() cid.Cid

Cid returns the node's Cid, calculated according to its prefix and raw data contents.

Note that this method may return a CID representing a zero-length byte slice if there is an error performing the encode. To check whether such an error may have occurred, use node.EncodeProtobuf(false), instead (or prior to calling RawData) and check for its returned error value; the result of EncodeProtobuf is cached so there is minimal overhead when invoking both methods.

func (*ProtoNode) CidBuilder

func (n *ProtoNode) CidBuilder() cid.Builder

CidBuilder returns the CID Builder for this ProtoNode, it is never nil

func (*ProtoNode) Copy

func (n *ProtoNode) Copy() format.Node

Copy returns a copy of the node. The resulting node will have a properly sorted Links list regardless of whether the original came from a badly serialized form that didn't have a sorted list. NOTE: This does not make copies of Node objects in the links.

func (*ProtoNode) Data

func (n *ProtoNode) Data() []byte

Data returns the data stored by this node.

func (*ProtoNode) EncodeProtobuf

func (n *ProtoNode) EncodeProtobuf(force bool) ([]byte, error)

EncodeProtobuf returns the encoded raw data version of a Node instance. It may use a cached encoded version, unless the force flag is given.

func (*ProtoNode) GetLinkedNode

func (n *ProtoNode) GetLinkedNode(ctx context.Context, ds format.DAGService, name string) (format.Node, error)

GetLinkedNode returns a copy of the IPLD Node with the given name.

func (*ProtoNode) GetLinkedProtoNode

func (n *ProtoNode) GetLinkedProtoNode(ctx context.Context, ds format.DAGService, name string) (*ProtoNode, error)

GetLinkedProtoNode returns a copy of the ProtoNode with the given name.

func (n *ProtoNode) GetNodeLink(name string) (*format.Link, error)

GetNodeLink returns a copy of the link with the given name.

func (*ProtoNode) GetPBNode added in v0.2.4

func (n *ProtoNode) GetPBNode() *pb.PBNode

GetPBNode converts *ProtoNode into it's protocol buffer variant. If you plan on mutating the data of the original node, it is recommended that you call ProtoNode.Copy() before calling ProtoNode.GetPBNode()

func (*ProtoNode) IsAbsent added in v0.4.0

func (n *ProtoNode) IsAbsent() bool

Absent nodes are returned when traversing a struct field that is defined by a schema but unset in the data. (Absent nodes are not possible otherwise; you'll only see them from `schema.TypedNode`.) The absent flag is necessary so iterating over structs can unambiguously make the distinction between values that are present-and-null versus values that are absent.

Absent nodes respond to `Kind()` as `ipld.Kind_Null`, for lack of any better descriptive value; you should therefore always check IsAbsent rather than just a switch on kind when it may be important to handle absent values distinctly.

func (*ProtoNode) IsNull added in v0.4.0

func (n *ProtoNode) IsNull() bool

func (*ProtoNode) Kind added in v0.4.0

func (n *ProtoNode) Kind() ipld.Kind

Kind returns a value from the Kind enum describing what the essential serializable kind of this node is (map, list, integer, etc). Most other handling of a node requires first switching upon the kind.

func (*ProtoNode) Length added in v0.4.0

func (n *ProtoNode) Length() int64

Length returns the length of a list, or the number of entries in a map, or -1 if the node is not of list nor map kind.

func (n *ProtoNode) Links() []*format.Link

Links returns a copy of the node's links.

func (*ProtoNode) ListIterator added in v0.4.0

func (n *ProtoNode) ListIterator() ipld.ListIterator

ListIterator returns an iterator which yields key-value pairs traversing the node. If the node kind is anything other than a list, nil will be returned.

The iterator will yield every entry in the list; that is, it can be expected that itr.Next will be called node.Length times before itr.Done becomes true.

func (*ProtoNode) Loggable

func (n *ProtoNode) Loggable() map[string]interface{}

Loggable implements the ipfs/go-log.Loggable interface.

func (*ProtoNode) LookupByIndex added in v0.4.0

func (n *ProtoNode) LookupByIndex(idx int64) (ipld.Node, error)

LookupByIndex is the equivalent of LookupByString but for indexing into a list. As with LookupByString, the returned Node may be any of the Kind: a primitive (string, int64, etc), a map, a list, or a link.

If the Kind of this Node is not Kind_List, a nil node and an error will be returned.

If idx is out of range, a nil node and an error will be returned.

func (*ProtoNode) LookupByNode added in v0.4.0

func (n *ProtoNode) LookupByNode(key ipld.Node) (ipld.Node, error)

LookupByNode is the equivalent of LookupByString, but takes a reified Node as a parameter instead of a plain string. This mechanism is useful if working with typed maps (if the key types have constraints, and you already have a reified `schema.TypedNode` value, using that value can save parsing and validation costs); and may simply be convenient if you already have a Node value in hand.

(When writing generic functions over Node, a good rule of thumb is: when handling a map, check for `schema.TypedNode`, and in this case prefer the LookupByNode(Node) method; otherwise, favor LookupByString; typically implementations will have their fastest paths thusly.)

func (*ProtoNode) LookupBySegment added in v0.4.0

func (n *ProtoNode) LookupBySegment(seg ipld.PathSegment) (ipld.Node, error)

LookupBySegment is will act as either LookupByString or LookupByIndex, whichever is contextually appropriate.

Using LookupBySegment may imply an "atoi" conversion if used on a list node, or an "itoa" conversion if used on a map node. If an "itoa" conversion takes place, it may error, and this method may return that error.

func (*ProtoNode) LookupByString added in v0.4.0

func (n *ProtoNode) LookupByString(key string) (ipld.Node, error)

LookupByString looks up a child object in this node and returns it. The returned Node may be any of the Kind: a primitive (string, int64, etc), a map, a list, or a link.

If the Kind of this Node is not Kind_Map, a nil node and an error will be returned.

If the key does not exist, a nil node and an error will be returned.

func (*ProtoNode) MapIterator added in v0.4.0

func (n *ProtoNode) MapIterator() ipld.MapIterator

Note that when using codegenerated types, there may be a fifth variant of lookup method on maps: `Get($GeneratedTypeKey) $GeneratedTypeValue`! MapIterator returns an iterator which yields key-value pairs traversing the node. If the node kind is anything other than a map, nil will be returned.

The iterator will yield every entry in the map; that is, it can be expected that itr.Next will be called node.Length times before itr.Done becomes true.

func (*ProtoNode) Marshal

func (n *ProtoNode) Marshal() ([]byte, error)

Marshal encodes a *Node instance into a new byte slice. The conversion uses an intermediate PBNode.

func (*ProtoNode) MarshalJSON

func (n *ProtoNode) MarshalJSON() ([]byte, error)

MarshalJSON returns a JSON representation of the node.

func (*ProtoNode) Multihash

func (n *ProtoNode) Multihash() mh.Multihash

Multihash hashes the encoded data of this node.

Note that this method may return a multihash representing a zero-length byte slice if there is an error performing the encode. To check whether such an error may have occurred, use node.EncodeProtobuf(false), instead (or prior to calling RawData) and check for its returned error value; the result of EncodeProtobuf is cached so there is minimal overhead when invoking both methods.

func (*ProtoNode) Prototype added in v0.4.0

func (n *ProtoNode) Prototype() ipld.NodePrototype

Prototype returns a NodePrototype which can describe some properties of this node's implementation, and also be used to get a NodeBuilder, which can be use to create new nodes with the same implementation as this one.

For typed nodes, the NodePrototype will also implement schema.Type.

For Advanced Data Layouts, the NodePrototype will encapsulate any additional parameters and configuration of the ADL, and will also (usually) implement NodePrototypeSupportingAmend.

Calling this method should not cause an allocation.

func (*ProtoNode) RawData

func (n *ProtoNode) RawData() []byte

RawData returns the encoded byte form of this node.

Note that this method may return an empty byte slice if there is an error performing the encode. To check whether such an error may have occurred, use node.EncodeProtobuf(false), instead (or prior to calling RawData) and check for its returned error value; the result of EncodeProtobuf is cached so there is minimal overhead when invoking both methods.

func (n *ProtoNode) RemoveNodeLink(name string) error

RemoveNodeLink removes a link on this node by the given name. If there are no links with this name, ErrLinkNotFound will be returned. If there are more than one link with this name, they will all be removed.

func (*ProtoNode) Resolve

func (n *ProtoNode) Resolve(path []string) (interface{}, []string, error)

Resolve is an alias for ResolveLink.

func (n *ProtoNode) ResolveLink(path []string) (*format.Link, []string, error)

ResolveLink consumes the first element of the path and obtains the link corresponding to it from the node. It returns the link and the path without the consumed element.

func (*ProtoNode) SetCidBuilder

func (n *ProtoNode) SetCidBuilder(builder cid.Builder) error

SetCidBuilder sets the CID builder if it is non nil, if nil then it is reset to the default value. An error will be returned if the builder is not usable.

func (*ProtoNode) SetData

func (n *ProtoNode) SetData(d []byte)

SetData stores data in this nodes.

func (n *ProtoNode) SetLinks(links []*format.Link) error

SetLinks replaces the node links with a copy of the provided links. Sorting will be applied to the list.

func (*ProtoNode) Size

func (n *ProtoNode) Size() (uint64, error)

Size returns the total size of the data addressed by node, including the total sizes of references.

func (*ProtoNode) Stat

func (n *ProtoNode) Stat() (*format.NodeStat, error)

Stat returns statistics on the node.

func (*ProtoNode) String

func (n *ProtoNode) String() string

String prints the node's Cid.

Note that this method may return a CID representing a zero-length byte slice if there is an error performing the encode. To check whether such an error may have occurred, use node.EncodeProtobuf(false), instead (or prior to calling RawData) and check for its returned error value; the result of EncodeProtobuf is cached so there is minimal overhead when invoking both methods.

func (*ProtoNode) Tree

func (n *ProtoNode) Tree(p string, depth int) []string

Tree returns the link names of the ProtoNode. ProtoNodes are only ever one path deep, so anything different than an empty string for p results in nothing. The depth parameter is ignored.

func (*ProtoNode) UnmarshalJSON

func (n *ProtoNode) UnmarshalJSON(b []byte) error

UnmarshalJSON reads the node fields from a JSON-encoded byte slice.

func (n *ProtoNode) UpdateNodeLink(name string, that *ProtoNode) (*ProtoNode, error)

UpdateNodeLink return a copy of the node with the link name set to point to that. The link will be added in sorted order. If a link of the same name existed, it is removed.

If sorting has not already been applied to this node (because it was deserialized from a form that did not have sorted links), the links list will be sorted in the returned copy.

type RawNode

type RawNode struct {
	blocks.Block

	// Always a node/basic Bytes.
	// We can't reference a specific type, as it's not exposed there.
	// If we find that the interface indirection really matters,
	// then we could possibly use dagpb.Bytes.
	ipld.Node
}

RawNode represents a node which only contains data.

func NewRawNode

func NewRawNode(data []byte) *RawNode

NewRawNode creates a RawNode using the default sha2-256 hash function.

func NewRawNodeWPrefix

func NewRawNodeWPrefix(data []byte, builder cid.Builder) (*RawNode, error)

NewRawNodeWPrefix creates a RawNode using the provided cid builder

func (*RawNode) Copy

func (rn *RawNode) Copy() format.Node

Copy performs a deep copy of this node and returns it as an format.Node

func (rn *RawNode) Links() []*format.Link

Links returns nil.

func (*RawNode) MarshalJSON added in v0.0.2

func (rn *RawNode) MarshalJSON() ([]byte, error)

MarshalJSON is required for our "ipfs dag" commands.

func (*RawNode) Resolve

func (rn *RawNode) Resolve(path []string) (interface{}, []string, error)

Resolve returns an error.

func (rn *RawNode) ResolveLink(path []string) (*format.Link, []string, error)

ResolveLink returns an error.

func (*RawNode) Size

func (rn *RawNode) Size() (uint64, error)

Size returns the size of this node

func (*RawNode) Stat

func (rn *RawNode) Stat() (*format.NodeStat, error)

Stat returns some Stats about this node.

func (*RawNode) Tree

func (rn *RawNode) Tree(p string, depth int) []string

Tree returns nil.

type SessionMaker

type SessionMaker interface {
	Session(context.Context) ipld.NodeGetter
}

SessionMaker is an object that can generate a new fetching session.

type WalkOption added in v0.2.1

type WalkOption func(*walkOptions)

WalkOption is a setter for walkOptions

func Concurrency added in v0.2.1

func Concurrency(worker int) WalkOption

Concurrency is a WalkOption indicating that node fetching should be done in parallel, with a specific concurrency factor. NOTE: When using that option, the walk order is *not* guarantee. NOTE: It *does not* make multiple concurrent calls to the passed `visit` function.

func Concurrent added in v0.2.1

func Concurrent() WalkOption

Concurrent is a WalkOption indicating that node fetching should be done in parallel, with the default concurrency factor. NOTE: When using that option, the walk order is *not* guarantee. NOTE: It *does not* make multiple concurrent calls to the passed `visit` function.

func IgnoreErrors added in v0.2.1

func IgnoreErrors() WalkOption

IgnoreErrors is a WalkOption indicating that the walk should attempt to continue even when an error occur.

func IgnoreMissing added in v0.2.1

func IgnoreMissing() WalkOption

IgnoreMissing is a WalkOption indicating that the walk should continue when a node is missing.

func OnError added in v0.2.1

func OnError(handler func(c cid.Cid, err error) error) WalkOption

OnError is a WalkOption adding a custom error handler. If this handler return a nil error, the walk will continue.

func OnMissing added in v0.2.1

func OnMissing(callback func(c cid.Cid)) WalkOption

OnMissing is a WalkOption adding a callback that will be triggered on a missing node.

func SkipRoot added in v0.2.2

func SkipRoot() WalkOption

SkipRoot is a WalkOption indicating that the root node should skipped

Directories

Path Synopsis
Package traverse provides merkledag traversal functions
Package traverse provides merkledag traversal functions

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL