Update README.md

README.md

@@ -6,12 +6,11 @@ Introduction
 
 CBconvert is a [Comic Book](http://en.wikipedia.org/wiki/Comic_Book_Archive_file) convert tool.
 
-
 Features
 --------
 
- - reads rar, zip, 7z, gz, bz2, cbr, cbz, cb7, cbt, pdf and plain directory
- - always saves processed comic in cbz (zip) format
+ - reads RAR, ZIP, 7Z, GZ, BZ2, CBR, CBZ, CB7, CBT, PDF, EPUB, XPS and plain directory
+ - always saves processed comic in CBZ (ZIP) format
  - images can be converted to JPEG, PNG or 4-Bit BMP (16 colors) format
  - choose resize algorithm (NearestNeighbor, Bilinear, Bicubic, MitchellNetravali, Lanczos2/3)
  - export covers from comics
@@ -20,33 +19,8 @@ Features
 Download
 --------
 
- - [Windows static build](https://github.com/gen2brain/cbconvert/releases/download/0.1.0/bukanir-0.1.0.zip)
- - [Linux 64bit build](https://github.com/gen2brain/cbconvert/releases/download/0.1.0/bukanir-0.1.0.tar.gz)
-
-Compile
--------
-
-Install poppler, poppler-glib, cairo, libarchive and imagemagick dev packages:
-
-    apt-get install libpoppler-glib-dev libcairo2-dev libarchive-dev libmagickcore-dev libmagickwand-dev
-
-Install go package:
-
-    go get github.com/gen2brain/cbconvert
-    go install github.com/gen2brain/cbconvert && cbconvert
-
-Dependencies
-------------
-
-	go get github.com/MStoykov/go-libarchive
-	go get github.com/cheggaaa/go-poppler
-	go get github.com/cheggaaa/pb
-	go get github.com/gographics/imagick/imagick
-	go get github.com/hotei/bmp
-	go get github.com/nfnt/resize
-	go get github.com/skarademir/naturalsort
-	go get github.com/ungerik/go-cairo
-    go get gopkg.in/alecthomas/kingpin.v2
+ - [Windows static build](https://github.com/gen2brain/cbconvert/releases/download/0.2.0/cbconvert-0.2.0.zip)
+ - [Linux 64bit build](https://github.com/gen2brain/cbconvert/releases/download/0.2.0/cbconvert-0.2.0.tar.gz)
 
 Using
 -----
@@ -71,7 +45,7 @@ Using
       -o, --outdir="."       output directory
       -m, --size=0           process only files larger then size (in MB)
       -r, --recursive        process subdirectories recursively
-      -v, --quiet            hide console output
+      -Q, --quiet            hide console output
 
     Args:
       <args>  filename or directory
@@ -80,14 +54,55 @@ Using
 Examples
 --------
 
+Rescale images to 1200px for all supported files found in directory with size larger then 60MB:
+
     cbconvert --recursive --width 1200 --size 60 /media/comics/Thorgal/
 
-Rescale images to 1200px for all supported files found in directory with size larger then 60MB.
+Convert all images in archive to 4bit BMP image and save result in ~/comics directory:
 
     cbconvert --bmp --outdir ~/comics /media/comics/Garfield/Garfield_01.cbz
 
-Convert all images in archive to 4bit BMP image and save result in ~/comics directory. [BMP](http://en.wikipedia.org/wiki/BMP_file_format) format is uncompressed, for black&white pages very good choice. Archive size can be smaller 2-3x and file will be readable by comic readers.
+[BMP](http://en.wikipedia.org/wiki/BMP_file_format) format is uncompressed, for black&white pages very good choice. Archive size can be smaller 2-3x and file will be readable by comic readers.
+
+Generate thumbnails by freedesktop specification in ~/.thumbnails/normal directory, Lanczos3 algorithm is used for resizing:
 
     cbconvert --interpolation=5 --outdir ~/.thumbnails/normal --thumbnail /media/comics/GrooTheWanderer/
 
-Generate thumbnails by freedesktop specification in ~/.thumbnails/normal directory, Lanczos3 algorithm is used for resizing.
+Compile
+-------
+
+Install imagemagick dev packages:
+
+    apt-get install libmagickcore-dev libmagickwand-dev
+
+Compile latest MuPDF:
+
+    git clone git://git.ghostscript.com/mupdf.git && cd mupdf
+    git submodule update --init --recursive
+    HAVE_X11=no HAVE_GLFW=no HAVE_GLUT=no WANT_CURL=no make && make install
+
+Compile unarr library:
+
+    git clone https://github.com/zeniko/unarr && cd unarr
+    mkdir lzma920 && cd lzma920 && curl -L http://www.7-zip.org/a/lzma920.tar.bz2 | tar -xjvp && cd ..
+    curl -L http://zlib.net/zlib-1.2.8.tar.gz | tar -xzvp
+    curl -L http://www.bzip.org/1.0.6/bzip2-1.0.6.tar.gz | tar -xzvp
+    curl -L https://gist.githubusercontent.com/gen2brain/89fe506863be3fb139e8/raw/8783a7d81e22ad84944d146c5e33beab6dffc641/unarr-makefile.patch | patch -p1
+    CFLAGS="-DHAVE_7Z -DHAVE_ZLIB -DHAVE_BZIP2 -I./lzma920/C -I./zlib-1.2.8 -I./bzip2-1.0.6" make
+    cp build/debug/libunarr.a /usr/lib64/ && cp unarr.h /usr/include
+
+Install dependencies:
+
+    go get github.com/cheggaaa/pb
+	go get github.com/gen2brain/go-fitz
+    go get github.com/gen2brain/go-unarr
+    go get github.com/gographics/imagick/imagick
+    go get github.com/hotei/bmp
+    go get github.com/nfnt/resize
+    go get github.com/skarademir/naturalsort
+    go get gopkg.in/alecthomas/kingpin.v2
+
+Install go package:
+
+    go get github.com/gen2brain/cbconvert
+    go install github.com/gen2brain/cbconvert

cbconvert.go

@@ -39,9 +39,9 @@ import (
 	"sync"
 	"syscall"
 
-	"github.com/MStoykov/go-libarchive"
-	"github.com/cheggaaa/go-poppler"
 	"github.com/cheggaaa/pb"
+	"github.com/gen2brain/go-fitz"
+	"github.com/gen2brain/go-unarr"
 	"github.com/gographics/imagick/imagick"
 	_ "github.com/hotei/bmp"
 	"github.com/nfnt/resize"
@@ -154,17 +154,17 @@ func convertImage(img image.Image, index int, pathName string) {
 	<-throttle
 }
 
-// Converts pdf file to cbz
-func convertPDF(file string) {
+// Converts PDF/EPUB/XPS document to CBZ
+func convertDocument(file string) {
 	workdir, _ = ioutil.TempDir(os.TempDir(), "cbc")
 
-	doc, err := poppler.Open(file)
+	doc, err := fitz.NewDocument(file)
 	if err != nil {
 		fmt.Fprintf(os.Stderr, "Skipping %s, error: %v", file, err.Error())
 		return
 	}
 
-	npages := doc.GetNPages()
+	npages := doc.Pages()
 
 	var bar *pb.ProgressBar
 	if !opts.Quiet {
@@ -179,89 +179,79 @@ func convertPDF(file string) {
 			bar.Increment()
 		}
 
-		page := doc.GetPage(n)
-		images := page.Images()
+		img, err := doc.Image(n)
 
-		if len(images) == 1 {
+		if err == nil && img != nil {
 			throttle <- 1
 			wg.Add(1)
 
-			surface := images[0].GetSurface()
-			go convertImage(surface.GetImage(), page.Index(), "")
-		} else {
-			// FIXME merge images?
+			go convertImage(img, n, "")
 		}
 	}
 	wg.Wait()
 }
 
-// Converts archive to cbz
+// Converts archive to CBZ
 func convertArchive(file string) {
 	workdir, _ = ioutil.TempDir(os.TempDir(), "cbc")
 
-	f, err := os.Open(file)
-	if err != nil {
-		fmt.Fprintf(os.Stderr, "Error Open: %v\n", err.Error())
-		return
-	}
-	defer f.Close()
+	ncontents := len(listArchive(file))
 
-	reader, err := archive.NewReader(f)
+	archive, err := unarr.NewArchive(file)
 	if err != nil {
 		fmt.Fprintf(os.Stderr, "Error NewReader: %v\n", err.Error())
 	}
-	defer reader.Free()
-	defer reader.Close()
+	defer archive.Close()
 
 	var bar *pb.ProgressBar
 	if !opts.Quiet {
-		s, _ := f.Stat()
-		bar = pb.New(int(s.Size()))
-		bar.SetUnits(pb.U_BYTES)
+		bar = pb.New(ncontents)
 		bar.ShowTimeLeft = false
 		bar.Prefix(fmt.Sprintf("Converting %d of %d: ", current, nfiles))
 		bar.Start()
 	}
 
 	for {
-		entry, err := reader.Next()
+		err := archive.Entry()
 		if err != nil {
-			if err == archive.ErrArchiveEOF {
+			if err == io.EOF {
 				break
 			} else {
-				fmt.Fprintf(os.Stderr, "Error Next: %v\n", err.Error())
+				fmt.Fprintf(os.Stderr, "Error Entry: %v\n", err.Error())
 				continue
 			}
 		}
 
-		stat := entry.Stat()
-		if stat.Mode()&os.ModeType != 0 || stat.IsDir() {
-			continue
-		}
-
 		if !opts.Quiet {
-			size := reader.Size()
-			bar.Set(size)
+			bar.Increment()
 		}
 
-		pathname := entry.PathName()
+		size := archive.Size()
+		pathname := archive.Name()
+
+		buf := make([]byte, size)
+		for size > 0 {
+			n, err := archive.Read(buf)
+			if err != nil && err != io.EOF {
+				break
+			}
+			size -= n
+		}
+
+		if size > 0 {
+			fmt.Printf("Error Read\n")
+			continue
+		}
 
 		if isImage(pathname) {
-			buf := new(bytes.Buffer)
-			_, err := buf.ReadFrom(reader)
-			if err != nil {
-				fmt.Fprintf(os.Stderr, "Error ReadFrom: %v\n", err.Error())
-				continue
-			}
-
-			img, err := decodeImage(bytes.NewReader(buf.Bytes()), pathname)
+			img, err := decodeImage(bytes.NewReader(buf), pathname)
 			if err != nil {
 				fmt.Fprintf(os.Stderr, "Error Decode: %v\n", err.Error())
 				continue
 			}
 
 			if opts.NoRGB && !isGrayScale(img) {
-				copyFile(bytes.NewReader(buf.Bytes()), filepath.Join(workdir, filepath.Base(pathname)))
+				copyFile(bytes.NewReader(buf), filepath.Join(workdir, filepath.Base(pathname)))
 				continue
 			}
 
@@ -271,13 +261,13 @@ func convertArchive(file string) {
 				go convertImage(img, 0, pathname)
 			}
 		} else {
-			copyFile(reader, filepath.Join(workdir, filepath.Base(pathname)))
+			copyFile(bytes.NewReader(buf), filepath.Join(workdir, filepath.Base(pathname)))
 		}
 	}
 	wg.Wait()
 }
 
-// Converts directory to cbz
+// Converts directory to CBZ
 func convertDirectory(path string) {
 	workdir, _ = ioutil.TempDir(os.TempDir(), "cbc")
 
@@ -324,7 +314,7 @@ func convertDirectory(path string) {
 	wg.Wait()
 }
 
-// Saves workdir to cbz archive
+// Saves workdir to CBZ archive
 func saveArchive(file string) {
 	defer os.RemoveAll(workdir)
 
@@ -368,64 +358,72 @@ func saveArchive(file string) {
 	z.Close()
 }
 
-// Unpacks archive to directory
-func unpackArchive(file string, dir string) {
-	f, err := os.Open(file)
-	if err != nil {
-		fmt.Fprintf(os.Stderr, "Error Open: %v\n", err.Error())
-		return
-	}
-	defer f.Close()
-
-	reader, err := archive.NewReader(f)
+// Lists contents of archive
+func listArchive(file string) []string {
+	var contents []string
+	archive, err := unarr.NewArchive(file)
 	if err != nil {
 		fmt.Fprintf(os.Stderr, "Error NewReader: %v\n", err.Error())
-		return
 	}
-	defer reader.Free()
-	defer reader.Close()
+	defer archive.Close()
 
 	for {
-		entry, err := reader.Next()
+		err := archive.Entry()
 		if err != nil {
-			if err == archive.ErrArchiveEOF {
+			if err == io.EOF {
 				break
 			} else {
+				fmt.Fprintf(os.Stderr, "Error Entry: %v\n", err.Error())
 				continue
 			}
 		}
 
-		if entry.Stat().Mode()&os.ModeType == 0 {
-			err = copyFile(reader, filepath.Join(dir, entry.PathName()))
-		}
-
-		if err != nil {
-			fmt.Fprintf(os.Stderr, "Error: %v\n", err.Error())
-		}
+		pathname := archive.Name()
+		contents = append(contents, pathname)
 	}
+
+	return contents
 }
 
 // Extracts cover from archive
 func coverArchive(file string) (image.Image, error) {
-	tmpdir, _ := ioutil.TempDir(os.TempDir(), "cbc")
-	defer os.RemoveAll(tmpdir)
+	var images []string
 
-	unpackArchive(file, tmpdir)
-
-	images := getImages(tmpdir)
-	if len(images) == 0 {
-		return nil, errors.New("No images")
+	contents := listArchive(file)
+	for _, c := range contents {
+		if isImage(c) {
+			images = append(images, c)
+		}
 	}
 
 	cover := getCover(images)
 
-	p, err := os.Open(cover)
+	archive, err := unarr.NewArchive(file)
 	if err != nil {
 		return nil, err
 	}
-	defer p.Close()
+	defer archive.Close()
 
-	img, err := decodeImage(p, cover)
+	err = archive.EntryFor(cover)
+	if err != nil {
+		return nil, err
+	}
+
+	size := archive.Size()
+	buf := make([]byte, size)
+	for size > 0 {
+		n, err := archive.Read(buf)
+		if err != nil && err != io.EOF {
+			break
+		}
+		size -= n
+	}
+
+	if size > 0 {
+		return nil, errors.New("Error Read")
+	}
+
+	img, err := decodeImage(bytes.NewReader(buf), cover)
 	if err != nil {
 		return nil, err
 	}
@@ -433,28 +431,23 @@ func coverArchive(file string) (image.Image, error) {
 	return img, nil
 }
 
-// Extracts cover from pdf
-func coverPDF(file string) (image.Image, error) {
-	doc, err := poppler.Open(file)
+// Extracts cover from document
+func coverDocument(file string) (image.Image, error) {
+	doc, err := fitz.NewDocument(file)
 	if err != nil {
 		return nil, err
 	}
 
-	page := doc.GetPage(0)
-	images := page.Images()
-
-	if len(images) == 1 {
-		surface := images[0].GetSurface()
-		img := surface.GetImage()
-
-		if img == nil {
-			return nil, errors.New("Image is nil")
-		}
-
-		return img, nil
+	img, err := doc.Image(0)
+	if err != nil {
+		return nil, err
 	}
 
-	return nil, nil
+	if img == nil {
+		return nil, errors.New("Image is nil")
+	}
+
+	return img, nil
 }
 
 // Extracts cover from directory
@@ -487,7 +480,7 @@ func getFiles() []string {
 
 	walkFiles := func(fp string, f os.FileInfo, err error) error {
 		if !f.IsDir() {
-			if isComic(fp) {
+			if isArchive(fp) || isDocument(fp) {
 				if isSize(f.Size()) {
 					files = append(files, fp)
 				}
@@ -505,7 +498,7 @@ func getFiles() []string {
 		}
 
 		if !stat.IsDir() {
-			if isComic(path) {
+			if isArchive(path) || isDocument(path) {
 				if isSize(stat.Size()) {
 					files = append(files, path)
 				}
@@ -516,7 +509,7 @@ func getFiles() []string {
 			} else {
 				fs, _ := ioutil.ReadDir(path)
 				for _, f := range fs {
-					if isComic(f.Name()) {
+					if isArchive(f.Name()) || isArchive(f.Name()) {
 						if isSize(f.Size()) {
 							files = append(files, f.Name())
 						}
@@ -534,7 +527,7 @@ func getFiles() []string {
 	return files
 }
 
-// Returns sorted list of found image files for given directory
+// Returns list of found image files for given directory
 func getImages(path string) []string {
 	var images []string
 
@@ -578,9 +571,20 @@ func getCover(images []string) string {
 }
 
 // Checks if file is comic
-func isComic(f string) bool {
-	var types = []string{".rar", ".zip", ".7z", ".gz", ".bz2",
-		".cbr", ".cbz", ".cb7", ".cbt", ".pdf"}
+func isArchive(f string) bool {
+	var types = []string{".rar", ".zip", ".7z", ".gz",
+		".bz2", ".cbr", ".cbz", ".cb7", ".cbt"}
+	for _, t := range types {
+		if strings.ToLower(filepath.Ext(f)) == t {
+			return true
+		}
+	}
+	return false
+}
+
+// Checks if file is document
+func isDocument(f string) bool {
+	var types = []string{".pdf", ".epub", ".xps"}
 	for _, t := range types {
 		if strings.ToLower(filepath.Ext(f)) == t {
 			return true
@@ -663,8 +667,8 @@ func extractCover(file string, info os.FileInfo) {
 	var cover image.Image
 	if info.IsDir() {
 		cover, err = coverDirectory(file)
-	} else if strings.ToLower(filepath.Ext(file)) == ".pdf" {
-		cover, err = coverPDF(file)
+	} else if isDocument(file) {
+		cover, err = coverDocument(file)
 	} else {
 		cover, err = coverArchive(file)
 	}
@@ -696,8 +700,8 @@ func extractThumbnail(file string, info os.FileInfo) {
 	var cover image.Image
 	if info.IsDir() {
 		cover, err = coverDirectory(file)
-	} else if strings.ToLower(filepath.Ext(file)) == ".pdf" {
-		cover, err = coverPDF(file)
+	} else if isDocument(file) {
+		cover, err = coverDocument(file)
 	} else {
 		cover, err = coverArchive(file)
 	}
@@ -747,8 +751,8 @@ func convertComic(file string, info os.FileInfo) {
 	if info.IsDir() {
 		convertDirectory(file)
 		saveArchive(file)
-	} else if strings.ToLower(filepath.Ext(file)) == ".pdf" {
-		convertPDF(file)
+	} else if isDocument(file) {
+		convertDocument(file)
 		saveArchive(file)
 	} else {
 		convertArchive(file)
@@ -759,7 +763,7 @@ func convertComic(file string, info os.FileInfo) {
 // Parses command line flags
 func parseFlags() {
 	opts = options{}
-	kingpin.Version("CBconvert 0.1.0")
+	kingpin.Version("CBconvert 0.3.0")
 	kingpin.CommandLine.Help = "Comic Book convert tool."
 	kingpin.Flag("png", "encode images to PNG instead of JPEG").Short('p').BoolVar(&opts.ToPNG)
 	kingpin.Flag("bmp", "encode images to 4-Bit BMP instead of JPEG").Short('b').BoolVar(&opts.ToBMP)
@@ -775,7 +779,7 @@ func parseFlags() {
 	kingpin.Flag("outdir", "output directory").Default(".").Short('o').StringVar(&opts.Outdir)
 	kingpin.Flag("size", "process only files larger then size (in MB)").Short('m').Default(strconv.Itoa(0)).Int64Var(&opts.Size)
 	kingpin.Flag("recursive", "process subdirectories recursively").Short('r').BoolVar(&opts.Recursive)
-	kingpin.Flag("quiet", "hide console output").Short('v').BoolVar(&opts.Quiet)
+	kingpin.Flag("quiet", "hide console output").Short('Q').BoolVar(&opts.Quiet)
 	kingpin.Arg("args", "filename or directory").Required().ExistingFilesOrDirsVar(&arguments)
 	kingpin.Parse()
 }

make.bash

@@ -5,14 +5,17 @@ mkdir -p build
 CGO_ENABLED=1 GOOS=linux GOARCH=amd64 go build -o build/cbconvert
 strip build/cbconvert
 
+#CGO_LDFLAGS="-lm -lz -ldl -lltdl -lfreetype -static-libgcc" CGO_ENABLED=1 GOOS=linux GOARCH=amd64 go build -v -x -o build/cbconvert-static --ldflags '-extldflags "-static"'
+#strip build/cbconvert-static
+
 CGO_LDFLAGS="-L/usr/i686-pc-mingw32/usr/lib" \
-CGO_CFLAGS="-I/usr/i686-pc-mingw32/usr/include -I/usr/i686-pc-mingw32/usr/include/poppler/glib -I/usr/i686-pc-mingw32/usr/include/glib-2.0 -I/usr/i686-pc-mingw32/usr/include/cairo" \
-CGO_CXXFLAGS="-I/usr/i686-pc-mingw32/usr/include -I/usr/i686-pc-mingw32/usr/include/poppler/glib -I/usr/i686-pc-mingw32/usr/include/glib-2.0 -I/usr/i686-pc-mingw32/usr/include/cairo" \
-CGO_CPPFLAGS="-I/usr/i686-pc-mingw32/usr/include -I/usr/i686-pc-mingw32/usr/include/poppler/glib -I/usr/i686-pc-mingw32/usr/include/glib-2.0 -I/usr/i686-pc-mingw32/usr/include/cairo" \
+CGO_CFLAGS="-I/usr/i686-pc-mingw32/usr/include -Wno-poison-system-directories" \
+CGO_CXXFLAGS="-I/usr/i686-pc-mingw32/usr/include -Wno-poison-system-directories" \
+CGO_CPPFLAGS="-I/usr/i686-pc-mingw32/usr/include -Wno-poison-system-directories" \
 PKG_CONFIG=/usr/bin/i686-pc-mingw32-pkg-config \
 PKG_CONFIG_PATH=/usr/i686-pc-mingw32/usr/lib/pkgconfig \
 PKG_CONFIG_LIBDIR=/usr/i686-pc-mingw32/usr/lib/pkgconfig \
 CC="i686-pc-mingw32-gcc" CXX="i686-pc-mingw32-g++" \
 CC_FOR_TARGET=i686-pc-mingw32-gcc CXX_FOR_TARGET=i686-pc-mingw32-g++ \
-CGO_ENABLED=1 GOOS=windows GOARCH=386 go build -x -work -o build/cbconvert.exe -ldflags "-linkmode external -extldflags -static"
+CGO_ENABLED=1 GOOS=windows GOARCH=386 go build -o build/cbconvert.exe -ldflags "-linkmode external -extldflags -static"
 i686-pc-mingw32-strip build/cbconvert.exe