mirror of
https://github.com/unidoc/unipdf.git
synced 2025-04-26 13:48:55 +08:00
11f692bc3a
* Track runes in IdentityEncoder (for subsetting), track decoded runes * Working with the identity encoder in font_composite.go * Add GetFilterArray to multi encoder. Add comments. * Add NewFromContents constructor to extractor only requiring contents and resources * golint fixes * Optimizer compress streams - improved detection of raw streams * Optimize - CleanContentStream optimizer that removes redundant operands * WIP Optimize - clean fonts Will support both font file reduction and subsetting. (WIP) * Optimize - image processing - try combined DCT and Flate * Update options.go * Update optimizer.go * Create utils.go for optimize with common methods needed for optimization * Optimizer - add font subsetting method Covers XObject Forms, annotaitons etc. Uses extractor package to extract text marks covering what fonts and glyphs are used. Package truetype used for subsetting. * Add some comments * Fix cmap parsing rune conversion * Error checking for extractor. Add some comments. * Update Jenkinsfile * Update modules
53 lines
1.5 KiB
Go
53 lines
1.5 KiB
Go
/*
|
|
* This file is subject to the terms and conditions defined in
|
|
* file 'LICENSE.md', which is part of this source code package.
|
|
*/
|
|
|
|
package optimize
|
|
|
|
import (
|
|
"github.com/unidoc/unipdf/v3/core"
|
|
)
|
|
|
|
// CompressStreams compresses uncompressed streams.
|
|
// It implements interface model.Optimizer.
|
|
type CompressStreams struct {
|
|
}
|
|
|
|
// Optimize optimizes PDF objects to decrease PDF size.
|
|
func (c *CompressStreams) Optimize(objects []core.PdfObject) (optimizedObjects []core.PdfObject, err error) {
|
|
optimizedObjects = make([]core.PdfObject, len(objects))
|
|
copy(optimizedObjects, objects)
|
|
for _, obj := range objects {
|
|
stream, isStreamObj := core.GetStream(obj)
|
|
if !isStreamObj {
|
|
continue
|
|
}
|
|
// Skip objects that are already encoded.
|
|
// TODO: Try filter combinations, and ignoring inefficient filters.
|
|
if obj := stream.Get("Filter"); obj != nil {
|
|
if _, skip := core.GetName(obj); skip {
|
|
continue
|
|
}
|
|
if arr, ok := core.GetArray(obj); ok && arr.Len() > 0 {
|
|
continue
|
|
}
|
|
}
|
|
|
|
encoder := core.NewFlateEncoder() // Most mainstream compressor and probably most robust.
|
|
var data []byte
|
|
data, err = encoder.EncodeBytes(stream.Stream)
|
|
if err != nil {
|
|
return optimizedObjects, err
|
|
}
|
|
dict := encoder.MakeStreamDict()
|
|
// compare compressed and uncompressed sizes
|
|
if len(data)+len(dict.WriteString()) < len(stream.Stream) {
|
|
stream.Stream = data
|
|
stream.PdfObjectDictionary.Merge(dict)
|
|
stream.PdfObjectDictionary.Set("Length", core.MakeInteger(int64(len(stream.Stream))))
|
|
}
|
|
}
|
|
return optimizedObjects, nil
|
|
}
|