🎨 Add block content statistics template function statBlock https://github.com/siyuan-note/siyuan/issues/13438

This commit is contained in:
Daniel 2024-12-11 23:59:58 +08:00
parent e2017a9fba
commit 11d3516aa7
No known key found for this signature in database
GPG key ID: 86211BA83DF03017
6 changed files with 237 additions and 167 deletions

223
kernel/filesys/stat.go Normal file
View file

@ -0,0 +1,223 @@
// SiYuan - Refactor your thinking
// Copyright (c) 2020-present, b3log.org
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU Affero General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Affero General Public License for more details.
//
// You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <https://www.gnu.org/licenses/>.
package filesys
import (
"bytes"
"github.com/88250/lute"
"github.com/88250/lute/ast"
"github.com/88250/lute/parse"
"github.com/siyuan-note/siyuan/kernel/av"
"github.com/siyuan-note/siyuan/kernel/treenode"
"github.com/siyuan-note/siyuan/kernel/util"
)
func ContentStat(content string) (ret *util.BlockStatResult) {
luteEngine := util.NewLute()
return contentStat(content, luteEngine)
}
func contentStat(content string, luteEngine *lute.Lute) (ret *util.BlockStatResult) {
tree := luteEngine.BlockDOM2Tree(content)
runeCnt, wordCnt, linkCnt, imgCnt, refCnt := tree.Root.Stat()
return &util.BlockStatResult{
RuneCount: runeCnt,
WordCount: wordCnt,
LinkCount: linkCnt,
ImageCount: imgCnt,
RefCount: refCnt,
}
}
func StatBlock(id string) (ret *util.BlockStatResult) {
trees := LoadTrees([]string{id})
if 1 > len(trees) {
return
}
tree := trees[id]
if nil == tree {
return
}
node := treenode.GetNodeInTree(tree, id)
if nil == node {
return
}
if ast.NodeDocument == node.Type {
return statTree(tree)
}
runeCnt, wordCnt, linkCnt, imgCnt, refCnt := node.Stat()
ret = &util.BlockStatResult{
runeCnt,
wordCnt,
linkCnt,
imgCnt,
refCnt,
1,
}
return
}
func StatTree(id string) (ret *util.BlockStatResult) {
trees := LoadTrees([]string{id})
if 1 > len(trees) {
return
}
tree := trees[id]
if nil == tree {
return
}
return statTree(tree)
}
func statTree(tree *parse.Tree) (ret *util.BlockStatResult) {
blockCount := 0
var databaseBlockNodes []*ast.Node
ast.Walk(tree.Root, func(n *ast.Node, entering bool) ast.WalkStatus {
if !entering {
return ast.WalkContinue
}
if n.IsBlock() {
blockCount++
}
if ast.NodeAttributeView != n.Type {
return ast.WalkContinue
}
databaseBlockNodes = append(databaseBlockNodes, n)
return ast.WalkContinue
})
luteEngine := util.NewLute()
var dbRuneCnt, dbWordCnt, dbLinkCnt, dbImgCnt, dbRefCnt int
for _, n := range databaseBlockNodes {
if "" == n.AttributeViewID {
continue
}
attrView, _ := av.ParseAttributeView(n.AttributeViewID)
if nil == attrView {
continue
}
content := bytes.Buffer{}
for _, kValues := range attrView.KeyValues {
for _, v := range kValues.Values {
switch kValues.Key.Type {
case av.KeyTypeURL:
if v.IsEmpty() {
continue
}
dbLinkCnt++
content.WriteString(v.URL.Content)
case av.KeyTypeMAsset:
if v.IsEmpty() {
continue
}
for _, asset := range v.MAsset {
if av.AssetTypeImage == asset.Type {
dbImgCnt++
}
}
case av.KeyTypeBlock:
if v.IsEmpty() {
continue
}
if !v.IsDetached {
dbRefCnt++
}
content.WriteString(v.Block.Content)
case av.KeyTypeText:
if v.IsEmpty() {
continue
}
content.WriteString(v.Text.Content)
case av.KeyTypeNumber:
if v.IsEmpty() {
continue
}
v.Number.FormatNumber()
content.WriteString(v.Number.FormattedContent)
case av.KeyTypeEmail:
if v.IsEmpty() {
continue
}
content.WriteString(v.Email.Content)
case av.KeyTypePhone:
if v.IsEmpty() {
continue
}
content.WriteString(v.Phone.Content)
}
}
}
dbStat := contentStat(content.String(), luteEngine)
dbRuneCnt += dbStat.RuneCount
dbWordCnt += dbStat.WordCount
}
runeCnt, wordCnt, linkCnt, imgCnt, refCnt := tree.Root.Stat()
runeCnt += dbRuneCnt
wordCnt += dbWordCnt
linkCnt += dbLinkCnt
imgCnt += dbImgCnt
refCnt += dbRefCnt
return &util.BlockStatResult{
RuneCount: runeCnt,
WordCount: wordCnt,
LinkCount: linkCnt,
ImageCount: imgCnt,
RefCount: refCnt,
BlockCount: blockCount,
}
}
func BlocksWordCount(ids []string) (ret *util.BlockStatResult) {
ret = &util.BlockStatResult{}
trees := LoadTrees(ids)
for _, id := range ids {
tree := trees[id]
if nil == tree {
continue
}
node := treenode.GetNodeInTree(tree, id)
if nil == node {
continue
}
runeCnt, wordCnt, linkCnt, imgCnt, refCnt := node.Stat()
ret.RuneCount += runeCnt
ret.WordCount += wordCnt
ret.LinkCount += linkCnt
ret.ImageCount += imgCnt
ret.RefCount += refCnt
}
ret.BlockCount = len(ids)
return
}

View file

@ -0,0 +1,84 @@
// SiYuan - Refactor your thinking
// Copyright (c) 2020-present, b3log.org
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU Affero General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Affero General Public License for more details.
//
// You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <https://www.gnu.org/licenses/>.
package filesys
import (
"math"
"text/template"
"time"
"github.com/88250/go-humanize"
"github.com/Masterminds/sprig/v3"
"github.com/araddon/dateparse"
"github.com/siyuan-note/logging"
"github.com/siyuan-note/siyuan/kernel/treenode"
"github.com/siyuan-note/siyuan/kernel/util"
"github.com/spf13/cast"
)
func BuiltInTemplateFuncs() (ret template.FuncMap) {
ret = sprig.TxtFuncMap()
// 因为安全原因移除一些函数 https://github.com/siyuan-note/siyuan/issues/13426
delete(ret, "env")
delete(ret, "expandenv")
delete(ret, "getHostByName")
ret["Weekday"] = util.Weekday
ret["WeekdayCN"] = util.WeekdayCN
ret["WeekdayCN2"] = util.WeekdayCN2
ret["ISOWeek"] = util.ISOWeek
ret["pow"] = pow
ret["powf"] = powf
ret["log"] = log
ret["logf"] = logf
ret["parseTime"] = parseTime
ret["FormatFloat"] = FormatFloat
ret["getHPathByID"] = getHPathByID
ret["statBlock"] = StatBlock
return
}
func pow(a, b interface{}) int64 { return int64(math.Pow(cast.ToFloat64(a), cast.ToFloat64(b))) }
func powf(a, b interface{}) float64 { return math.Pow(cast.ToFloat64(a), cast.ToFloat64(b)) }
func log(a, b interface{}) int64 {
return int64(math.Log(cast.ToFloat64(a)) / math.Log(cast.ToFloat64(b)))
}
func logf(a, b interface{}) float64 { return math.Log(cast.ToFloat64(a)) / math.Log(cast.ToFloat64(b)) }
func parseTime(dateStr string) time.Time {
now := time.Now()
retTime, err := dateparse.ParseIn(dateStr, now.Location())
if err != nil {
logging.LogWarnf("parse date [%s] failed [%s], return current time instead", dateStr, err)
return now
}
return retTime
}
func FormatFloat(format string, n float64) string {
return humanize.FormatFloat(format, n)
}
func getHPathByID(id string) (ret string) {
bt := treenode.GetBlockTree(id)
if nil == bt {
return
}
ret = bt.HPath
return
}