mirror of
https://github.com/unidoc/unipdf.git
synced 2025-04-27 13:48:51 +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
58 lines
1.8 KiB
Go
58 lines
1.8 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 extractor
|
|
|
|
import (
|
|
"github.com/unidoc/unipdf/v3/model"
|
|
)
|
|
|
|
// Extractor stores and offers functionality for extracting content from PDF pages.
|
|
type Extractor struct {
|
|
// stream contents and resources for page
|
|
contents string
|
|
resources *model.PdfPageResources
|
|
|
|
// fontCache is a simple LRU cache that is used to prevent redundant constructions of PdfFont's from
|
|
// PDF objects. NOTE: This is not a conventional glyph cache. It only caches PdfFont's.
|
|
fontCache map[string]fontEntry
|
|
|
|
// text results from running extractXYText on forms within the page.
|
|
// TODO(peterwilliams): Cache this map accross all pages in a PDF to speed up processig.
|
|
formResults map[string]textResult
|
|
|
|
// accessCount is used to set fontEntry.access to an incrementing number.
|
|
accessCount int64
|
|
|
|
// textCount is an incrementing number used to identify XYTest objects.
|
|
textCount int64
|
|
}
|
|
|
|
// New returns an Extractor instance for extracting content from the input PDF page.
|
|
func New(page *model.PdfPage) (*Extractor, error) {
|
|
contents, err := page.GetAllContentStreams()
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
|
|
// Uncomment these lines to see the contents of the page. For debugging.
|
|
// fmt.Println("========================= +++ =========================")
|
|
// fmt.Printf("%s\n", contents)
|
|
// fmt.Println("========================= ::: =========================")
|
|
|
|
return NewFromContents(contents, page.Resources)
|
|
}
|
|
|
|
// NewFromContents creates a new extractor from contents and page resources.
|
|
func NewFromContents(contents string, resources *model.PdfPageResources) (*Extractor, error) {
|
|
e := &Extractor{
|
|
contents: contents,
|
|
resources: resources,
|
|
fontCache: map[string]fontEntry{},
|
|
formResults: map[string]textResult{},
|
|
}
|
|
return e, nil
|
|
}
|