blob42/gosuki
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

debug and print trees, wip firefox

Browse Source
This commit is contained in:
Chakib Ben Ziane 2018-10-25 18:09:03 +02:00
parent 09dea0b655
commit dab486fefd
5 changed files with 119 additions and 25 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
Makefile
View File

@ -3,7 +3,7 @@
TARGET=gomark
all: build docs
all: build
run:
@go run *.go

35
chrome.go
View File

@ -81,7 +81,7 @@ func NewChromeBrowser() IBrowser {
browser.baseDir = ChromeData.BookmarkDir
browser.bkFile = ChromeData.BookmarkFile
browser.Stats = &ParserStats{}
browser.NodeTree = &Node{Name: "root", Parent: nil}
browser.NodeTree = &Node{Name: "root", Parent: nil, Type: "root"}
// Across jobs buffer
browser.InitBuffer()
@ -112,7 +112,7 @@ func (bw *ChromeBrowser) Load() {
func (bw *ChromeBrowser) Run() {
// Rebuild node tree
bw.NodeTree = &Node{Name: "root", Parent: nil}
bw.NodeTree = &Node{Name: "root", Parent: nil, Type: "root"}
// Load bookmark file
bookmarkPath := path.Join(bw.baseDir, bw.bkFile)
@ -144,7 +144,7 @@ func (bw *ChromeBrowser) Run() {
rawNode := new(RawNode)
rawNode.parseItems(node)
log.Debugf("Parsing root folder %s", rawNode.name)
//log.Debugf("Parsing root folder %s", rawNode.name)
currentNode := rawNode.getNode()
@ -164,7 +164,7 @@ func (bw *ChromeBrowser) Run() {
// Finished parsing this root, it is not anymore a parent
_, parentNodes = parentNodes[len(parentNodes)-1], parentNodes[:len(parentNodes)-1]
log.Debugf("Parsed root %s folder", rawNode.name)
//log.Debugf("Parsed root %s folder", rawNode.name)
return nil
}
@ -184,12 +184,12 @@ func (bw *ChromeBrowser) Run() {
rawNode.parseItems(node)
currentNode := rawNode.getNode()
log.Debugf("parsing node %s", currentNode.Name)
//log.Debugf("parsing node %s", currentNode.Name)
// if parents array is not empty
if len(parentNodes) != 0 {
parent := parentNodes[len(parentNodes)-1]
log.Debugf("Adding current node to parent %s", parent.Name)
//log.Debugf("Adding current node to parent %s", parent.Name)
// Add current node to closest parent
currentNode.Parent = parent
@ -201,13 +201,13 @@ func (bw *ChromeBrowser) Run() {
// if node is a folder with children
if rawNode.childrenType == jsonparser.Array && len(rawNode.children) > 2 { // if len(children) > len("[]")
log.Debugf("Started folder %s", rawNode.name)
//log.Debugf("Started folder %s", rawNode.name)
parentNodes = append(parentNodes, currentNode)
// Process recursively all child nodes of this folder node
jsonparser.ArrayEach(node, parseChildren, jsonNodePaths.Children)
log.Debugf("Finished folder %s", rawNode.name)
//log.Debugf("Finished folder %s", rawNode.name)
_, parentNodes = parentNodes[len(parentNodes)-1], parentNodes[:len(parentNodes)-1]
}
@ -243,7 +243,7 @@ func (bw *ChromeBrowser) Run() {
// we check if the hash(name) changed meaning
// the data changed
} else {
log.Debugf("URL Found in index")
//log.Debugf("URL Found in index")
nodeVal = iVal.(*Node)
// hash(name) is different meaning new commands/tags could
@ -266,7 +266,7 @@ func (bw *ChromeBrowser) Run() {
//If parent is folder, add it as tag and add current node as child
//And add this link as child
if currentNode.Parent.Type == jsonNodeTypes.Folder {
log.Debug("Parent is folder, parsing as tag ...")
//log.Debug("Parent is folder, parsing as tag ...")
currentNode.Tags = append(currentNode.Tags, currentNode.Parent.Name)
}
@ -281,11 +281,12 @@ func (bw *ChromeBrowser) Run() {
start := time.Now()
jsonparser.ObjectEach(rootsData, jsonParseRoots)
bw.Stats.lastParseTime = time.Since(start)
log.Debugf("Parsed tree in %s", bw.Stats.lastParseTime)
log.Debugf("<%s> parsed tree in %s", bw.name, bw.Stats.lastParseTime)
// Finished node tree building job
// Debug walk tree
//go WalkNode(bw.nodeTree)
//go WalkNode(bw.NodeTree)
go PrintTree(bw.NodeTree)
// Reset the index to represent the nodetree
bw.RebuildIndex()
@ -297,14 +298,14 @@ func (bw *ChromeBrowser) Run() {
bw.ResetStats()
//Add nodeTree to Cache
log.Debugf("Buffer content")
bw.BufferDB.Print()
//log.Debugf("<%s> buffer content", bw.name)
//bw.BufferDB.Print()
log.Debugf("syncing to buffer")
log.Debugf("<%s> syncing to buffer", bw.name)
syncTreeToBuffer(bw.NodeTree, bw.BufferDB)
log.Debugf("Tree synced to buffer")
log.Debugf("<%s> tree synced to buffer", bw.name)
bw.BufferDB.Print()
//bw.BufferDB.Print()
// cacheDB represents bookmarks across all browsers
// From browsers it should support: add/update

30
firefox.go
View File

@ -2,8 +2,7 @@ package main
import (
"path"
"github.com/OneOfOne/xxhash"
"time"
)
var Firefox = BrowserPaths{
@ -34,7 +33,7 @@ func NewFFBrowser() IBrowser {
browser.baseDir = Firefox.BookmarkDir
browser.bkFile = Firefox.BookmarkFile
browser.Stats = &ParserStats{}
browser.NodeTree = &Node{Name: "root", Parent: nil}
browser.NodeTree = &Node{Name: "root", Parent: nil, Type: "root"}
// Across jobs buffer
browser.InitBuffer()
@ -135,8 +134,8 @@ func getFFBookmarks(bw *FFBrowser) {
urlNode.Parent = tagMap[tagId]
tagMap[tagId].Children = append(tagMap[tagId].Children, urlNode)
// Check if url already in index
iVal, found := bw.URLIndex.Get(urlNode.URL)
// Check if url already in index TODO: should be done in new pass
//iVal, found := bw.URLIndex.Get(urlNode.URL)
/*
* The fields where tags may change are hashed together
@ -145,7 +144,20 @@ func getFFBookmarks(bw *FFBrowser) {
* (tags) for this url then hash their concatenation
*/
nameHash := xxhash.ChecksumString64(urlNode.Name)
//nameHash := xxhash.ChecksumString64(urlNode.Name)
// TODO: No guarantee we finished gathering tags !!
// We should check again against index in a new pass
// This pass needs to finish until we have full count
// of tags for each bookmark
//parents := urlNode.GetParentTags()
//if len(parents) > 4 {
//tags := make([]string, 0)
//for _, v := range parents {
//tags = append(tags, v.Name)
//}
////log.Debugf("<%s> --> [%s]", urlNode.URL, strings.Join(tags, "|"))
//}
bw.Stats.currentUrlCount++
bw.Stats.currentNodeCount++
@ -202,10 +214,16 @@ func (bw *FFBrowser) Run() {
bw._places = placesDB
// Parse bookmarks to a flat tree (for compatibility with tree system)
start := time.Now()
getFFBookmarks(bw)
// Finished parsing
bw.Stats.lastParseTime = time.Since(start)
log.Debugf("<%s> parsed %d bookmarks and %d nodes", bw.name, bw.Stats.currentUrlCount, bw.Stats.currentNodeCount)
log.Debugf("<%s> parsed tree in %s", bw.name, bw.Stats.lastParseTime)
//go PrintTree(bw.NodeTree)
bw.ResetStats()
}

2
parse.go
View File

@ -43,7 +43,7 @@ func ParseTags(node *Node) {
//res := regex.FindAllStringSubmatch(bk.Metadata, -1)
if len(node.Tags) > 0 {
log.Debugf("[Title] found following tags: %s", node.Tags)
log.Debugf("[in title] found following tags: %s", node.Tags)
}
}

75
tree.go
View File

@ -1,5 +1,11 @@
package main
import (
"fmt"
"github.com/xlab/treeprint"
)
type Node struct {
Name string
Type string // folder, tag, url
@ -22,6 +28,75 @@ func (node *Node) GetBookmark() *Bookmark {
}
}
func (node *Node) GetRoot() *Node {
nodePtr := node
for {
if nodePtr.Name == "root" {
break
}
nodePtr = nodePtr.Parent
}
return nodePtr
}
// Returns all parent tags for URL nodes
func (node *Node) GetParentTags() []*Node {
var parents []*Node
var walk func(node *Node)
var nodePtr *Node
root := node.GetRoot()
walk = func(n *Node) {
nodePtr = n
//log.Debugf("type of %s --> %s", nodePtr.Type, nodePtr.Name)
if nodePtr.Type == "url" {
return
}
if len(nodePtr.Children) == 0 {
return
}
for _, v := range nodePtr.Children {
if v.URL == node.URL &&
nodePtr.Type == "tag" {
parents = append(parents, nodePtr)
}
walk(v)
}
}
walk(root)
return parents
}
func PrintTree(root *Node) {
var walk func(node *Node, tree treeprint.Tree)
tree := treeprint.New()
walk = func(node *Node, t treeprint.Tree) {
if len(node.Children) > 0 {
t = t.AddBranch(fmt.Sprintf("%s <%s>", node.Type, node.Name))
for _, child := range node.Children {
walk(child, t)
}
} else {
t.AddNode(fmt.Sprintf("%s <%s>", node.Type, node.URL))
}
}
walk(root, tree)
log.Debug(tree.String())
}
// Debuggin bookmark node tree
// TODO: Better usage of node trees
func WalkNode(node *Node) {