1 Star 0 Fork 0

coodder / unipdf

加入 Gitee
与超过 1200万 开发者一起发现、参与优秀开源项目,私有仓库也完全免费 :)
免费加入
克隆/下载
combine_duplicate_streams.go 1.36 KB
一键复制 编辑 原始数据 按行查看 历史
jhonm 提交于 2023-08-07 15:31 . init
package optimize
import (
"crypto/md5"
"gitee.com/coodder/unipdf/core"
)
// CombineDuplicateStreams combines duplicated streams by its data hash.
// It implements interface model.Optimizer.
type CombineDuplicateStreams struct {
}
// Optimize optimizes PDF objects to decrease PDF size.
func (dup *CombineDuplicateStreams) Optimize(objects []core.PdfObject) (optimizedObjects []core.PdfObject, err error) {
replaceTable := make(map[core.PdfObject]core.PdfObject)
toDelete := make(map[core.PdfObject]struct{})
streamsByHash := make(map[string][]*core.PdfObjectStream)
for _, obj := range objects {
if stream, isStreamObj := obj.(*core.PdfObjectStream); isStreamObj {
hasher := md5.New()
hasher.Write([]byte(stream.Stream))
hash := string(hasher.Sum(nil))
streamsByHash[hash] = append(streamsByHash[hash], stream)
}
}
for _, streams := range streamsByHash {
if len(streams) < 2 {
continue
}
firstStream := streams[0]
for i := 1; i < len(streams); i++ {
stream := streams[i]
replaceTable[stream] = firstStream
toDelete[stream] = struct{}{}
}
}
optimizedObjects = make([]core.PdfObject, 0, len(objects)-len(toDelete))
for _, obj := range objects {
if _, found := toDelete[obj]; found {
continue
}
optimizedObjects = append(optimizedObjects, obj)
}
replaceObjectsInPlace(optimizedObjects, replaceTable)
return optimizedObjects, nil
}
Go
1
https://gitee.com/coodder/unipdf.git
git@gitee.com:coodder/unipdf.git
coodder
unipdf
unipdf
v1.2.0

搜索帮助

53164aa7 5694891 3bd8fe86 5694891