Merge branch 'master' of github.com:ciromattia/kcc

2025-12-15 18:56:28 +00:00 · 2013-05-28 12:18:58 +02:00
parent 7c3a762107 0b90af77da
commit 97b5d8a2ee
3 changed files with 266 additions and 209 deletions
--- a/README.md
+++ b/README.md
@@ -1,4 +1,4 @@
-# KCC
+# KCC

 `KCC` (a.k.a. `KindleComicConverter`) is a Python app to convert comic files or folders to ePub or Panel View MOBI.  
 It was initally developed for Kindle but since v2.2 it outputs valid ePub 2.0 so _**despite its name, KCC is
@@ -16,9 +16,14 @@ _kc2_ in no way is a replacement for **KCC** so you can be quite confident we'll
 You can find the latest released binary at the following links:  
 - OS X: [https://dl.dropbox.com/u/16806101/KindleComicConverter_osx_2.9.zip](https://dl.dropbox.com/u/16806101/KindleComicConverter_osx_2.9.zip)
 - Win64: [https://dl.dropbox.com/u/16806101/KindleComicConverter_win-amd64_2.9.zip](https://dl.dropbox.com/u/16806101/KindleComicConverter_win-amd64_2.9.zip)
- Win32: [http://pawelj.vulturis.eu/Shared/KindleComicConverter_win-x86_2.9.zip](http://pawelj.vulturis.eu/Shared/KindleComicConverter_win-x86_2.9.zip) *(thanks to [AcidWeb](https://github.com/AcidWeb))*
+- Win32: [http://pawelj.vulturis.eu/Shared/KindleComicConverter_win-x86_2.9.zip](http://pawelj.vulturis.eu/Shared/KindleComicConverter_win-x86_2.9.zip)
 - Linux: Just download sourcecode and launch `python kcc.py` *(Provided you have Python and Pillow installed)*

+## AWKCC .NET GUI
+![AWKCC](http://pawelj.vulturis.eu/Shared/CurrentAWKCC.png)
+
+[All-in-one package for Windows users](http://www.mobileread.com/forums/showpost.php?p=2444957&postcount=3)
+
 ## INPUT FORMATS
 `kcc` can understand and convert, at the moment, the following file types:
 - PNG, JPG, GIF, TIFF, BMP
@@ -53,7 +58,7 @@ Options:
  --version             show program's version number and exit
  -h, --help            show this help message and exit
  -p PROFILE, --profile=PROFILE
-                        Device profile (Choose one among K1, K2, K3, K4NT, K4T, KDX, KDXG or KHD) [Default=KHD]
+                        Device profile (Choose one among K1, K2, K3, K4NT, K4T, KDX, KDXG, KHD, KF, KFHD, KFHD8) [Default=KHD]
  -t TITLE, --title=TITLE
                        Comic title [Default=filename]
  -m, --manga-style     Manga style (Right-to-left reading and splitting) [Default=False]
@@ -88,55 +93,92 @@ The app relies and includes the following scripts/binaries:
 - `magic.py` from [python-magic](https://github.com/ahupp/python-magic) library

 ## CHANGELOG
-  - 1.00: Initial version
-  - 1.10: Added support for CBZ/CBR files in comic2ebook.py
-  - 1.11: Added support for CBZ/CBR files in KindleComicConverter
-  - 1.20: Comic optimizations! Split pages not target-oriented (landscape with portrait target or portrait
-   with landscape target), add palette and other image optimizations from Mangle.  
-   WARNING: PIL is required for all image mangling!
-  - 1.30: Fixed an issue in OPF generation for device resolution  
-   Reworked options system (call with -h option to get the inline help)
-  - 1.40: Added some options for controlling image optimization  
-        Further optimization (ImageOps, page numbering cut, autocontrast)
-  - 1.41: Fixed a serious bug on resizing when img ratio was bigger than device one
-  - 1.50: Added subfolder support for multiple chapters.
-  - 2.0: GUI! AppleScript is gone and Tk is used to provide cross-platform GUI support.
-  - 2.1: Added basic error reporting
-  - 2.2: Added (valid!) ePub 2.0 output  
-        Rename .zip files to .cbz to avoid overwriting
-  - 2.3: Fixed win32 ePub generation, folder handling, filenames with spaces and subfolders
-  - 2.4: Use temporary directory as workdir (fixes converting from external volumes and zipfiles renaming)  
-        Fixed "add folders" from GUI.
-  - 2.5: Added --black-borders option to set added borders black when page's ratio is not the device's one (#11).  
-        Fixes epub containing zipped itself (#10)  
-  - 2.6: Added --rotate option to rotate landscape images instead of splitting them (#16, #24)  
-        Added --output option to customize ePub output dir/file (#22)  
-        Add rendition:layout and rendition:orientation ePub meta tags (supported by new kindlegen 2.8)  
-        Fixed natural sorting for files (#18)
-  - 2.7: Lots of GUI improvements (#27, #13)  
-        Added gamma support within --gamma option (defaults to profile-specified gamma) (#26, #27)  
-        Added --nodithering option to prevent dithering optimizations (#27)  
-        Epub margins support (#30)  
-        Fixed no file added if file has no spaces on Windows (#25)  
-        Gracefully exit if unrar missing (#15)  
-        Do not call kindlegen if source epub is bigger than 320MB (#17)  
-        Get filetype from magic number (#14)   
-        PDF conversion works again  
-  - 2.8: updated rarfile library  
-        Panel View support + HQ support (#36) - new option: --nopanelviewhq
-        Split profiles for K4NT and K4T  
-        Rewrite of Landscape Mode support (huge readability improvement for KPW)  
-        Upscale use now BILINEAR method  
-        Added generic CSS file  
-        Optimized archive extraction for zip/rar files (#40)  
-  - 2.9: Added support for generating a plain CBZ (skipping all the EPUB/Mobi generation) (#45)  
-        Prevent output file overwriting the source one: if a duplicate name is detected, append _kcc to the name  
-        Rarfile library updated to 2.6  
-        Added GIF, TIFF and BMP to supported formats (#42)  
-        Filenames slugifications (#28, #31, #9, #8)
-  - 2.10: Kindle Fire support (color ePub/Mobi)
-        Panel View support for horizontal content
+####1.00
+* Initial version

+####1.10
+* Added support for CBZ/CBR files in comic2ebook.py
+
+####1.11
+* Added support for CBZ/CBR files in KindleComicConverter
+
+####1.20
+* Comic optimizations! Split pages not target-oriented (landscape with portrait target or portrait with landscape target), add palette and other image optimizations from Mangle. WARNING: PIL is required for all image mangling!
+
+####1.30
+* Fixed an issue in OPF generation for device resolution  
+* Reworked options system (call with -h option to get the inline help)
+
+####1.40
+* Added some options for controlling image optimization  
+* Further optimization (ImageOps, page numbering cut, autocontrast)
+
+####1.41
+* Fixed a serious bug on resizing when img ratio was bigger than device one
+
+####1.50
+* Added subfolder support for multiple chapters.
+
+####2.0
+* GUI! AppleScript is gone and Tk is used to provide cross-platform GUI support.
+
+####2.1
+* Added basic error reporting
+
+#### 2.2:
+* Added (valid!) ePub 2.0 output  
+* Rename .zip files to .cbz to avoid overwriting
+
+####2.3
+* Fixed win32 ePub generation, folder handling, filenames with spaces and subfolders
+
+####2.4
+* Use temporary directory as workdir (fixes converting from external volumes and zipfiles renaming)  
+* Fixed "add folders" from GUI.
+
+####2.5
+* Added --black-borders option to set added borders black when page's ratio is not the device's one (#11).  
+* Fixes epub containing zipped itself (#10)  
+
+####2.6
+* Added --rotate option to rotate landscape images instead of splitting them (#16, #24)  
+* Added --output option to customize ePub output dir/file (#22)  
+* Add rendition:layout and rendition:orientation ePub meta tags (supported by new kindlegen 2.8)  
+* Fixed natural sorting for files (#18)
+
+####2.7
+* Lots of GUI improvements (#27, #13)  
+* Added gamma support within --gamma option (defaults to profile-specified gamma) (#26, #27)  
+* Added --nodithering option to prevent dithering optimizations (#27)  
+* Epub margins support (#30)  
+* Fixed no file added if file has no spaces on Windows (#25)  
+* Gracefully exit if unrar missing (#15)  
+* Do not call kindlegen if source epub is bigger than 320MB (#17)  
+* Get filetype from magic number (#14)   
+* PDF conversion works again  
+
+####2.8
+* Updated rarfile library  
+* Panel View support + HQ support (#36) - new option: --nopanelviewhq
+* Split profiles for K4NT and K4T  
+* Rewrite of Landscape Mode support (huge readability improvement for KPW)  
+* Upscale use now BILINEAR method  
+* Added generic CSS file  
+* Optimized archive extraction for zip/rar files (#40)  
+
+####2.9
+* Added support for generating a plain CBZ (skipping all the EPUB/Mobi generation) (#45)  
+* Prevent output file overwriting the source one: if a duplicate name is detected, append _kcc to the name  
+* Rarfile library updated to 2.6  
+* Added GIF, TIFF and BMP to supported formats (#42)  
+* Filenames slugifications (#28, #31, #9, #8)
+
+####2.10:
+* Multiprocessing support
+* Kindle Fire support (color ePub/Mobi)
+* Panel View support for horizontal content
+* Fixed panel order for horizontal pages when --rotate is enabled
+* Disabled cropping and page number cutting for blank pages

 ## COPYRIGHT

--- a/kcc/comic2ebook.py
+++ b/kcc/comic2ebook.py
@@ -32,6 +32,7 @@ from shutil import copytree
 from shutil import rmtree
 from shutil import make_archive
 from optparse import OptionParser
+from multiprocessing import Pool
 import image
 import cbxarchive
 import pdfjpgextract
@@ -40,6 +41,11 @@ import pdfjpgextract
 def buildHTML(path, imgfile):
    filename = getImageFileName(imgfile)
    if filename is not None:
+        # All files marked with this sufix need horizontal Panel View.
+        if "_rotated" in str(filename):
+            rotate = True
+        else:
+            rotate = False
        htmlpath = ''
        postfix = ''
        backref = 1
@@ -70,7 +76,7 @@ def buildHTML(path, imgfile):
                      imgfile, "\" class=\"singlePage\"/></div>\n"
                      ])
        if options.panelview:
-            if options.panelviewhorizontal:
+            if rotate:
                if options.righttoleft:
                    f.writelines(["<div id=\"BoxTL\"><a class=\"app-amzn-magnify\" data-app-amzn-magnify=",
                                  "'{\"targetId\":\"BoxTL-Panel-Parent\", \"ordinal\":1}'></a></div>\n",
@@ -250,7 +256,7 @@ def buildOPF(profile, dstdir, title, filelist, cover=None):
    f.write("</manifest>\n<spine toc=\"ncx\">\n")
    splitCountUsed = 1
    for entry in reflist:
-        if entry.endswith("-1"):
+        if entry.endswith("-000001") or entry.endswith("-1"):
            # noinspection PyRedundantParentheses
            if ((options.righttoleft and facing == 'left') or (not options.righttoleft and facing == 'right')) and\
                    options.landscapemode:
@@ -260,7 +266,7 @@ def buildOPF(profile, dstdir, title, filelist, cover=None):
                f.write("<itemref idref=\"page_" + entry + "\" properties=\"page-spread-" + facing1 + "\"/>\n")
            else:
                f.write("<itemref idref=\"page_" + entry + "\"/>\n")
-        elif entry.endswith("-2"):
+        elif entry.endswith("-000002") or entry.endswith("-2"):
            if options.landscapemode:
                f.write("<itemref idref=\"page_" + entry + "\" properties=\"page-spread-" + facing2 + "\"/>\n")
            else:
@@ -309,16 +315,7 @@ def getImageFileName(imgfile):
    return filename


-def isInFilelist(filename, filelist):
-    filename = os.path.splitext(filename)
-    seen = False
-    for item in filelist:
-        if filename[0] == item[0]:
-            seen = True
-    return seen
-
-
-def applyImgOptimization(img, isSplit=False, toRight=False):
+def applyImgOptimization(img, isSplit, toRight, options):
    img.cropWhiteSpace(10.0)
    if options.cutpagenumbers:
        img.cutPageNumber()
@@ -331,51 +328,62 @@ def applyImgOptimization(img, isSplit=False, toRight=False):

 def dirImgProcess(path):
    global options, splitCount
-    if options.righttoleft:
-        facing = "right"
-    else:
-        facing = "left"
-
+    work = []
+    pagenumber = 0
+    pagenumbermodifier = 0
+    pool = Pool()
    for (dirpath, dirnames, filenames) in os.walk(path):
        for afile in filenames:
            if getImageFileName(afile) is not None:
-                if options.verbose:
-                    print "Optimizing " + afile + " for " + options.profile
-                else:
-                    print ".",
-                img = image.ComicPage(os.path.join(dirpath, afile), options.profile)
-                if options.nosplitrotate:
-                    split = None
-                else:
-                    split = img.splitPage(dirpath, options.righttoleft, options.rotate)
-                if split is not None:
-                    if options.verbose:
-                        print "Splitted " + afile
-                    if options.righttoleft:
-                        toRight1 = False
-                        toRight2 = True
-                        if facing == "left":
-                            splitCount += 1
-                        facing = "right"
-                    else:
-                        toRight1 = True
-                        toRight2 = False
-                        if facing == "right":
-                            splitCount += 1
-                        facing = "left"
-                    img0 = image.ComicPage(split[0], options.profile)
-                    applyImgOptimization(img0, True, toRight1)
-                    img0.saveToDir(dirpath, options.forcepng, options.forcecolor)
-                    img1 = image.ComicPage(split[1], options.profile)
-                    applyImgOptimization(img1, True, toRight2)
-                    img1.saveToDir(dirpath, options.forcepng, options.forcecolor)
-                else:
-                    if facing == "right":
-                        facing = "left"
-                    else:
-                        facing = "right"
-                    applyImgOptimization(img)
-                    img.saveToDir(dirpath, options.forcepng, options.forcecolor)
+                pagenumber += 1
+                work.append([afile, dirpath, pagenumber, options])
+    splitpages = pool.map(fileImgProcess, work)
+    pool.close()
+    pool.join()
+    splitpages = filter(None, splitpages)
+    splitpages.sort()
+    for page in splitpages:
+        if (page + pagenumbermodifier) % 2 == 0:
+            splitCount += 1
+            pagenumbermodifier += 1
+        pagenumbermodifier += 1
+
+
+def fileImgProcess(work):
+    afile = work[0]
+    dirpath = work[1]
+    pagenumber = work[2]
+    options = work[3]
+    output = None
+    if options.verbose:
+        print "Optimizing " + afile + " for " + options.profile
+    else:
+        print ".",
+    img = image.ComicPage(os.path.join(dirpath, afile), options.profile)
+    if options.nosplitrotate:
+        split = None
+    else:
+        split = img.splitPage(dirpath, options.righttoleft, options.rotate)
+    if split is not None and split is not "R":
+        if options.verbose:
+            print "Splitted " + afile
+        if options.righttoleft:
+            toRight1 = False
+            toRight2 = True
+        else:
+            toRight1 = True
+            toRight2 = False
+        output = pagenumber
+        img0 = image.ComicPage(split[0], options.profile)
+        applyImgOptimization(img0, True, toRight1, options)
+        img0.saveToDir(dirpath, options.forcepng, options.forcecolor)
+        img1 = image.ComicPage(split[1], options.profile)
+        applyImgOptimization(img1, True, toRight2, options)
+        img1.saveToDir(dirpath, options.forcepng, options.forcecolor)
+    else:
+        applyImgOptimization(img, False, False, options)
+        img.saveToDir(dirpath, options.forcepng, options.forcecolor, split)
+    return output


 def genEpubStruct(path):
@@ -387,8 +395,8 @@ def genEpubStruct(path):
    sanitizeTree(os.path.join(path, 'OEBPS', 'Images'))
    os.mkdir(os.path.join(path, 'OEBPS', 'Text'))
    f = open(os.path.join(path, 'OEBPS', 'Text', 'style.css'), 'w')
-    #DON'T COMPRESS CSS. KINDLE WILL FAIL TO PARSE IT.
-    #Generic Panel View support + Margins fix for Non-Kindle devices.
+    # DON'T COMPRESS CSS. KINDLE WILL FAIL TO PARSE IT.
+    # Generic Panel View support + Margins fix for Non-Kindle devices.
    f.writelines(["@page {\n",
                  "margin-bottom: 0;\n",
                  "margin-top: 0\n",
@@ -511,13 +519,6 @@ def genEpubStruct(path):
        for afile in filenames:
            filename = getImageFileName(afile)
            if filename is not None:
-                if "credit" in afile.lower():
-                    os.rename(os.path.join(dirpath, afile), os.path.join(dirpath, 'ZZZ999_' + afile))
-                    afile = 'ZZZ999_' + afile
-                if "+" in afile.lower() or "#" in afile.lower():
-                    newfilename = afile.replace('+', '_').replace('#', '_')
-                    os.rename(os.path.join(dirpath, afile), os.path.join(dirpath, newfilename))
-                    afile = newfilename
                filelist.append(buildHTML(dirpath, afile))
                if not chapter:
                    chapterlist.append((dirpath.replace('Images', 'Text'), filelist[-1][1]))
@@ -598,11 +599,11 @@ def sanitizeTree(filetree):

 def Copyright():
    print ('comic2ebook v%(__version__)s. '
-           'Written 2012 by Ciro Mattia Gonano.' % globals())
+           'Written 2013 by Ciro Mattia Gonano and Pawel Jastrzebski.' % globals())


 def Usage():
-    print "Generates HTML, NCX and OPF for a Comic ebook from a bunch of images."
+    print "Generates EPUB/CBZ comic ebook from a bunch of images."
    parser.print_help()


@@ -621,8 +622,6 @@ def main(argv=None):
                      help="Outputs a CBZ archive and does not generate EPUB")
    parser.add_option("--nopanelviewhq", action="store_true", dest="nopanelviewhq", default=False,
                      help="Disable high quality Panel View [Default=False]")
-    parser.add_option("--panelviewhorizontal", action="store_true", dest="panelviewhorizontal", default=False,
-                      help="Enable horizontal Panel View [Default=False]")
    parser.add_option("--noprocessing", action="store_false", dest="imgproc", default=True,
                      help="Do not apply image preprocessing (Page splitting and optimizations) [Default=True]")
    parser.add_option("--forcepng", action="store_true", dest="forcepng", default=False,
@@ -696,25 +695,33 @@ def getOutputFilename(srcpath, wantedname, ext):

 def checkOptions():
    global options
+    # Landscape mode is only supported by Kindle Touch and Paperwhite.
    if options.profile == 'K4T' or options.profile == 'KHD':
        options.landscapemode = True
    else:
        options.landscapemode = False
+    # Older Kindle don't support Virtual Panel View. We providing them configuration that will fake that feature.
    if options.profile == 'K3' or options.profile == 'K4NT':
-        #Real Panel View
+        # Real Panel View
        options.panelview = True
    else:
-        #Virtual Panel View
+        # Virtual Panel View
        options.panelview = False
+    # Older Kindle don't need higher resolution files due lack of Panel View.
+    # Kindle Fire family have very high resolution. Bigger images are not needed.
    if options.profile == 'K1' or options.profile == 'K2' or options.profile == 'KDX' or options.profile == 'KDXG'\
            or options.profile == 'KF' or options.profile == 'KFHD' or options.profile == 'KFHD8':
        options.nopanelviewhq = True
+    # Disabling grayscale conversion for Kindle Fire family.
+    # Forcing JPEG output. For now code can't provide color PNG files.
    if options.profile == 'KF' or options.profile == 'KFHD' or options.profile == 'KFHD8':
        options.forcecolor = True
        options.forcepng = False
    else:
        options.forcecolor = False
-    if options.panelviewhorizontal:
+    # Mixing vertical and horizontal pages require real Panel View.
+    # Landscape mode don't work correcly without Virtual Panel View.
+    if options.rotate:
        options.panelview = True
        options.landscapemode = False

@@ -723,6 +730,7 @@ def getEpubPath():
    global epub_path
    return epub_path

+
 if __name__ == "__main__":
    Copyright()
    main(sys.argv[1:])
--- a/kcc/image.py
+++ b/kcc/image.py
@@ -20,7 +20,7 @@ __copyright__ = '2012-2013, Ciro Mattia Gonano <ciromattia@gmail.com>'
 __docformat__ = 'restructuredtext en'

 import os
-from PIL import Image, ImageOps, ImageStat
+from PIL import Image, ImageOps, ImageStat, ImageChops


 class ImageFlags:
@@ -119,16 +119,21 @@ class ComicPage:
            raise RuntimeError('Cannot read image file %s' % source)
        self.image = self.image.convert('RGB')

-    def saveToDir(self, targetdir, forcepng, color=False):
+    def saveToDir(self, targetdir, forcepng, color, sufix=None):
        filename = os.path.basename(self.origFileName)
        try:
            if not color:
                self.image = self.image.convert('L')    # convert to grayscale
+            # Sufix is used to recognise which files need horizontal Panel View.
+            if sufix == "R":
+                sufix = "_rotated"
+            else:
+                sufix = ""
            os.remove(os.path.join(targetdir, filename))
            if forcepng:
-                self.image.save(os.path.join(targetdir, os.path.splitext(filename)[0] + ".png"), "PNG")
+                self.image.save(os.path.join(targetdir, os.path.splitext(filename)[0] + sufix + ".png"), "PNG")
            else:
-                self.image.save(os.path.join(targetdir, os.path.splitext(filename)[0] + ".jpg"), "JPEG")
+                self.image.save(os.path.join(targetdir, os.path.splitext(filename)[0] + sufix + ".jpg"), "JPEG")
        except IOError as e:
            raise RuntimeError('Cannot write image in directory %s: %s' % (targetdir, e))

@@ -202,7 +207,7 @@ class ComicPage:
        if (width > height) != (dstwidth > dstheight):
            if rotate:
                self.image = self.image.rotate(90)
-                return None
+                return "R"
            else:
                if width > height:
                    # source is landscape, so split by the width
@@ -232,95 +237,97 @@ class ComicPage:
            return None

    def cutPageNumber(self):
-        widthImg, heightImg = self.image.size
-        delta = 2
-        diff = delta
-        fixedThreshold = 5
-        if ImageStat.Stat(self.image).var[0] < 2 * fixedThreshold:
-            return self.image
-        while ImageStat.Stat(self.image.crop((0, heightImg - diff, widthImg, heightImg))).var[0] < fixedThreshold\
-                and diff < heightImg:
-            diff += delta
-        diff -= delta
-        pageNumberCut1 = diff
-        if diff < delta:
+        if ImageChops.invert(self.image).getbbox() is not None:
+            widthImg, heightImg = self.image.size
+            delta = 2
            diff = delta
-        oldStat = ImageStat.Stat(self.image.crop((0, heightImg - diff, widthImg, heightImg))).var[0]
-        diff += delta
-        while ImageStat.Stat(self.image.crop((0, heightImg - diff, widthImg, heightImg))).var[0] - oldStat > 0\
-                and diff < heightImg / 4:
+            fixedThreshold = 5
+            if ImageStat.Stat(self.image).var[0] < 2 * fixedThreshold:
+                return self.image
+            while ImageStat.Stat(self.image.crop((0, heightImg - diff, widthImg, heightImg))).var[0] < fixedThreshold\
+                    and diff < heightImg:
+                diff += delta
+            diff -= delta
+            pageNumberCut1 = diff
+            if diff < delta:
+                diff = delta
            oldStat = ImageStat.Stat(self.image.crop((0, heightImg - diff, widthImg, heightImg))).var[0]
            diff += delta
-        diff -= delta
-        pageNumberCut2 = diff
-        diff += delta
-        oldStat = ImageStat.Stat(self.image.crop((0, heightImg - diff, widthImg, heightImg - pageNumberCut2))).var[0]
-        while ImageStat.Stat(self.image.crop((0, heightImg - diff, widthImg, heightImg - pageNumberCut2))).var[0]\
-                < fixedThreshold + oldStat and diff < heightImg / 4:
+            while ImageStat.Stat(self.image.crop((0, heightImg - diff, widthImg, heightImg))).var[0] - oldStat > 0\
+                    and diff < heightImg / 4:
+                oldStat = ImageStat.Stat(self.image.crop((0, heightImg - diff, widthImg, heightImg))).var[0]
+                diff += delta
+            diff -= delta
+            pageNumberCut2 = diff
            diff += delta
-        diff -= delta
-        pageNumberCut3 = diff
-        delta = 5
-        diff = delta
-        while ImageStat.Stat(self.image.crop((0, heightImg - pageNumberCut2, diff, heightImg))).var[0] < fixedThreshold\
-                and diff < widthImg:
-            diff += delta
-        diff -= delta
-        pageNumberX1 = diff
-        diff = delta
-        while ImageStat.Stat(self.image.crop((widthImg - diff, heightImg - pageNumberCut2,
-                                              widthImg, heightImg))).var[0] < fixedThreshold and diff < widthImg:
-            diff += delta
-        diff -= delta
-        pageNumberX2 = widthImg - diff
-        if pageNumberCut3 - pageNumberCut1 > 2 * delta\
-                and float(pageNumberX2 - pageNumberX1) / float(pageNumberCut2 - pageNumberCut1) <= 9.0\
-                and ImageStat.Stat(self.image.crop((0, heightImg - pageNumberCut3, widthImg, heightImg))).var[0]\
-                / ImageStat.Stat(self.image).var[0] < 0.1\
-                and pageNumberCut3 < heightImg / 4 - delta:
-            diff = pageNumberCut3
-        else:
-            diff = pageNumberCut1
-        self.image = self.image.crop((0, 0, widthImg, heightImg - diff))
+            oldStat = ImageStat.Stat(self.image.crop((0, heightImg - diff, widthImg, heightImg - pageNumberCut2))).var[0]
+            while ImageStat.Stat(self.image.crop((0, heightImg - diff, widthImg, heightImg - pageNumberCut2))).var[0]\
+                    < fixedThreshold + oldStat and diff < heightImg / 4:
+                diff += delta
+            diff -= delta
+            pageNumberCut3 = diff
+            delta = 5
+            diff = delta
+            while ImageStat.Stat(self.image.crop((0, heightImg - pageNumberCut2, diff, heightImg))).var[0] < fixedThreshold\
+                    and diff < widthImg:
+                diff += delta
+            diff -= delta
+            pageNumberX1 = diff
+            diff = delta
+            while ImageStat.Stat(self.image.crop((widthImg - diff, heightImg - pageNumberCut2,
+                                                  widthImg, heightImg))).var[0] < fixedThreshold and diff < widthImg:
+                diff += delta
+            diff -= delta
+            pageNumberX2 = widthImg - diff
+            if pageNumberCut3 - pageNumberCut1 > 2 * delta\
+                    and float(pageNumberX2 - pageNumberX1) / float(pageNumberCut2 - pageNumberCut1) <= 9.0\
+                    and ImageStat.Stat(self.image.crop((0, heightImg - pageNumberCut3, widthImg, heightImg))).var[0]\
+                    / ImageStat.Stat(self.image).var[0] < 0.1\
+                    and pageNumberCut3 < heightImg / 4 - delta:
+                diff = pageNumberCut3
+            else:
+                diff = pageNumberCut1
+            self.image = self.image.crop((0, 0, widthImg, heightImg - diff))
        return self.image

    def cropWhiteSpace(self, threshold):
-        widthImg, heightImg = self.image.size
-        delta = 10
-        diff = delta
-        # top
-        while ImageStat.Stat(self.image.crop((0, 0, widthImg, diff))).var[0] < threshold and diff < heightImg:
-            diff += delta
-        diff -= delta
-        #    print "Top crop: %s"%diff
-        self.image = self.image.crop((0, diff, widthImg, heightImg))
-        widthImg, heightImg = self.image.size
-        diff = delta
-        # left
-        while ImageStat.Stat(self.image.crop((0, 0, diff, heightImg))).var[0] < threshold and diff < widthImg:
-            diff += delta
-        diff -= delta
-        #    print "Left crop: %s"%diff
-        self.image = self.image.crop((diff, 0, widthImg, heightImg))
-        widthImg, heightImg = self.image.size
-        diff = delta
-        # down
-        while ImageStat.Stat(self.image.crop((0, heightImg - diff, widthImg, heightImg))).var[0] < threshold\
-                and diff < heightImg:
-            diff += delta
-        diff -= delta
-        #    print "Down crop: %s"%diff
-        self.image = self.image.crop((0, 0, widthImg, heightImg - diff))
-        widthImg, heightImg = self.image.size
-        diff = delta
-        # right
-        while ImageStat.Stat(self.image.crop((widthImg - diff, 0, widthImg, heightImg))).var[0] < threshold\
-                and diff < widthImg:
-            diff += delta
-        diff -= delta
-        #    print "Right crop: %s"%diff
-        self.image = self.image.crop((0, 0, widthImg - diff, heightImg))
-        #    print "New size: %sx%s"%(self.image.size[0],self.image.size[1])
+        if ImageChops.invert(self.image).getbbox() is not None:
+            widthImg, heightImg = self.image.size
+            delta = 10
+            diff = delta
+            # top
+            while ImageStat.Stat(self.image.crop((0, 0, widthImg, diff))).var[0] < threshold and diff < heightImg:
+                diff += delta
+            diff -= delta
+            #    print "Top crop: %s"%diff
+            self.image = self.image.crop((0, diff, widthImg, heightImg))
+            widthImg, heightImg = self.image.size
+            diff = delta
+            # left
+            while ImageStat.Stat(self.image.crop((0, 0, diff, heightImg))).var[0] < threshold and diff < widthImg:
+                diff += delta
+            diff -= delta
+            #    print "Left crop: %s"%diff
+            self.image = self.image.crop((diff, 0, widthImg, heightImg))
+            widthImg, heightImg = self.image.size
+            diff = delta
+            # down
+            while ImageStat.Stat(self.image.crop((0, heightImg - diff, widthImg, heightImg))).var[0] < threshold\
+                    and diff < heightImg:
+                diff += delta
+            diff -= delta
+            #    print "Down crop: %s"%diff
+            self.image = self.image.crop((0, 0, widthImg, heightImg - diff))
+            widthImg, heightImg = self.image.size
+            diff = delta
+            # right
+            while ImageStat.Stat(self.image.crop((widthImg - diff, 0, widthImg, heightImg))).var[0] < threshold\
+                    and diff < widthImg:
+                diff += delta
+            diff -= delta
+            #    print "Right crop: %s"%diff
+            self.image = self.image.crop((0, 0, widthImg - diff, heightImg))
+            #    print "New size: %sx%s"%(self.image.size[0],self.image.size[1])
        return self.image

    # def addProgressbar(self, file_number, files_totalnumber, size, howoften):