bump to 9.7.0

* add youtube, discord, humble buttons

* restore margins of top buttons

* fix button height on windows

* don't bold messages

.github/workflows/package-linux.yml

@@ -35,7 +35,7 @@ jobs:
       run: |
         sudo apt-get update 
         sudo apt-get install -y libpng-dev libjpeg-dev p7zip-full p7zip-rar python3-pip squashfs-tools libfuse2 libxcb-cursor0
-        python -m pip install --upgrade pip setuptools wheel certifi pyinstaller --no-binary pyinstaller 
+        python -m pip install --upgrade pip certifi pyinstaller --no-binary pyinstaller 
         python -m pip install -r requirements.txt
     - name: build binary
       run: |
@@ -68,7 +68,7 @@ jobs:
       if: startsWith(github.ref, 'refs/tags/')
       with:
         prerelease: true
-        generate_release_notes: true
+        generate_release_notes: false
         files: |
           LICENSE.txt
           *.AppImage*

.github/workflows/package-macos.yml

@@ -38,7 +38,7 @@ jobs:
           cache: 'pip'
       - name: Install python dependencies
         run: |
-          python -m pip install --upgrade pip setuptools wheel pyinstaller certifi
+          python -m pip install --upgrade pip pyinstaller certifi
           pip install -r requirements.txt
       - name: Install the Apple certificate and provisioning profile
 #        TODO signing
@@ -89,7 +89,7 @@ jobs:
         if: startsWith(github.ref, 'refs/tags/')
         with:
           prerelease: true
-          generate_release_notes: true
+          generate_release_notes: false
           files: |
             dist/*.dmg
       - name: Clean up keychain and provisioning profile

.github/workflows/package-osx-legacy.yml

@@ -40,7 +40,7 @@ jobs:
       - name: Install Python dependencies
         run: |
           python3 --version
-          pip3 install --upgrade pip setuptools wheel pyinstaller certifi
+          pip3 install --upgrade pip pyinstaller certifi
           pip3 install --upgrade -r requirements-osx-legacy.txt
           ./gen_ui_files.sh
       - uses: actions/setup-node@v6
@@ -60,7 +60,7 @@ jobs:
         if: startsWith(github.ref, 'refs/tags/')
         with:
           prerelease: true
-          generate_release_notes: true
+          generate_release_notes: false
           files: |
             LICENSE.txt
             dist/*.dmg

.github/workflows/package-windows.yml

@@ -45,7 +45,7 @@ jobs:
         env:
           PYINSTALLER_COMPILE_BOOTLOADER: 1
         run: |
-          python -m pip install --upgrade pip setuptools wheel
+          python -m pip install --upgrade pip
           pip install -r requirements.txt
           pip install certifi pyinstaller --no-binary pyinstaller
       - name: build binary
@@ -73,6 +73,6 @@ jobs:
         if: startsWith(github.ref, 'refs/tags/')
         with:
           prerelease: true
-          generate_release_notes: true
+          generate_release_notes: false
           files: |
             dist/*.exe

.github/workflows/package-windows7.yml

@@ -37,7 +37,7 @@ jobs:
         env:
           PYINSTALLER_COMPILE_BOOTLOADER: 1
         run: |
-          python -m pip install --upgrade pip setuptools wheel
+          python -m pip install --upgrade pip
           pip install -r requirements-win7.txt
           pip install certifi pyinstaller --no-binary pyinstaller
           .\gen_ui_files.bat
@@ -50,11 +50,22 @@ jobs:
         with:
           name: windows7-build
           path: dist/*.exe
+      - id: optional_step_id
+        uses: signpath/github-action-submit-signing-request@v2.0
+        if: ${{ github.repository == 'ciromattia/kcc' }}
+        with:
+         api-token: '${{ secrets.SIGNPATH_API_TOKEN }}'
+         organization-id: '1dc1bad6-4a8c-4f85-af30-5c5d3d392ea6'
+         project-slug: 'kcc'
+         signing-policy-slug: 'release-signing'
+         github-artifact-id: '${{ steps.upload-unsigned-artifact.outputs.artifact-id }}'
+         wait-for-completion: true
+         output-artifact-directory: 'dist/'
       - name: Release
         uses: softprops/action-gh-release@v2
         if: startsWith(github.ref, 'refs/tags/')
         with:
           prerelease: true
-          generate_release_notes: true
+          generate_release_notes: false
           files: |
             dist/*.exe

Dockerfile

@@ -4,7 +4,7 @@ FROM python:3.13-slim-bullseye AS builder
 
 # Install system dependencies
 RUN set -x && \
-    BUILD_DEPS="build-essential cmake libffi-dev libfreetype6-dev libfontconfig1-dev libpng-dev libjpeg-dev libssl-dev libxft-dev make python3-dev python3-setuptools python3-wheel" && \
+    BUILD_DEPS="build-essential cmake libffi-dev libfreetype6-dev libfontconfig1-dev libpng-dev libjpeg-dev libssl-dev libxft-dev make python3-dev" && \
     RUNTIME_DEPS="bash ca-certificates chrpath locales locales-all libfreetype6 libfontconfig1 p7zip-full python3 python3-pip libgl1" && \
     DEBIAN_FRONTEND=noninteractive apt-get update -y && \
     apt-get install -y --no-install-recommends ${BUILD_DEPS} ${RUNTIME_DEPS}

README.md

@@ -13,6 +13,13 @@ Pages display in fullscreen without margins,
 with proper fixed layout support.
 Supported input formats include JPG/PNG image files in folders, archives, or PDFs.
 Supported output formats include MOBI/AZW3, EPUB, KEPUB, CBZ, and PDF.
+KCC runs on Windows, macOS, and Linux.
+
+Just drop your input files into the KCC window, hit convert, and USB drop the output files onto your device's `documents` folder!
+
+https://github.com/user-attachments/assets/da73d625-e082-482d-91a4-ae4765e96fd7
+
+**WARNING**: Kindle Scribe 2025 support may not be possible. Does not work well currently.
 
 **NEW**: PDF output is now supported for direct conversion to reMarkable devices! 
 When using a reMarkable profile (Rmk1, Rmk2, RmkPP), the format automatically defaults to PDF 
@@ -34,6 +41,7 @@ KCC avoids many common formatting issues (some of which occur [even on the Kindl
 3) Not utilizing the full 1860x2480 resolution of the 10" Kindle Scribe
 4) incorrect page turn direction for manga that's read right to left
 5) unaligned two page spreads in landscape, where pages are shifted over by 1
+6) Removing without blur the rainbow effect on color eink Kaleido 3 due to manga screentones
 
 The GUI looks like this, built in Qt6, with my most commonly used settings:
 
@@ -46,7 +54,9 @@ You can change the default output directory by holding `Shift` while clicking th
 Then just drag and drop the generated output files onto your device's documents folder via USB.
 If you are on macOS and use a 2022+ Kindle, you may need to use Amazon USB File Manager for Mac.
 
-YouTube tutorial (please subscribe): https://www.youtube.com/watch?v=IR2Fhcm9658
+YouTube tutorial (please subscribe): https://www.youtube.com/watch?v=QQ6zJcMF2Iw
+
+Installation tutorial: https://www.youtube.com/watch?v=IR2Fhcm9658
 
 ### A word of warning
 **KCC** _is not_ [Amazon's Kindle Comic Creator](http://www.amazon.com/gp/feature.html?ie=UTF8&docId=1001103761) nor is in any way endorsed by Amazon.
@@ -98,7 +108,7 @@ There are also legacy macOS 10.14+ and Windows 7 experimental versions available
 
 The `c2e` and `c2p` versions are command line tools for power users.
 
-On Mac, follow: https://support.apple.com/guide/mac-help/open-a-mac-app-from-an-unknown-developer-mh40616/mac
+On macOS, if you get a `can't be opened` error, follow: https://support.apple.com/guide/mac-help/open-a-mac-app-from-an-unknown-developer-mh40616/mac
 
 For flatpak, Docker, and AppImage versions, refer to the wiki: https://github.com/ciromattia/kcc/wiki/Installation
 
@@ -238,6 +248,8 @@ MAIN:
 
 PROCESSING:
   -n, --noprocessing    Do not modify image and ignore any profile or processing option
+  --pdfextract          Use legacy PDF image extraction method from KCC 8 and earlier.
+  --pdfwidth            Render vector PDFs based on device width instead of height.
   -u, --upscale         Resize images smaller than device's resolution
   -s, --stretch         Stretch images to device's resolution
   -r SPLITTER, --splitter SPLITTER
@@ -258,8 +270,14 @@ PROCESSING:
                         Crop empty sections. 0: Disabled 1: Horizontally 2: Both [Default=0]
   --blackborders        Disable autodetection and force black borders
   --whiteborders        Disable autodetection and force white borders
+  --smartcovercrop      Attempt to crop main cover from wide image
+  --coverfill           Center-crop only the cover to fill target device screen
   --forcecolor          Don't convert images to grayscale
-  --forcepng            Create PNG files instead JPEG
+  --forcepng            Create PNG files instead JPEG for black and white images
+  --webp                Replace JPG with lossy WEBP and PNG with lossless WEBP
+  --force-png-rgb       Force color images to be saved as PNG
+  --pnglegacy           Use a more compatible 8 bit PNG instead of 4 bit.
+  --noquantize          Don't quantize PNG images to 16 colors
   --mozjpeg             Create JPEG files using mozJpeg
   --jpeg-quality        The JPEG quality, on a scale from 0 (worst) to 95 (best). Default 85 for most devices.
   --maximizestrips      Turn 1x4 strips to 2x2 strips
@@ -280,6 +298,7 @@ OUTPUT SETTINGS:
                         Split output into multiple files. 0: Don't split 1: Automatic mode 2: Consider every subdirectory as separate volume [Default=0]
   --spreadshift         Shift first page to opposite side in landscape for two page spread alignment
   --norotate            Do not rotate double page spreads in spread splitter option.
+  --rotateright         Rotate double page spreads in opposite direction.
   --rotatefirst         Put rotated spread first in spread splitter option.
   --filefusion          Combines all input files into a single file.
   --eraserainbow       Erase rainbow effect on color eink screen by attenuating interfering frequencies

kindlecomicconverter/KCC_gui.py

@@ -42,7 +42,7 @@ from raven import Client
 from tempfile import gettempdir
 
 from .shared import HTMLStripper, sanitizeTrace, walkLevel, subprocess_run
-from .comicarchive import SEVENZIP, available_archive_tools
+from .comicarchive import SEVENZIP, TAR, available_archive_tools
 from . import __version__
 from . import comic2ebook
 from . import metadata
@@ -195,7 +195,7 @@ class VersionThread(QThread):
                         icon = 'bindle'
                     if category == 'kofi':
                         icon = 'kofi'
-                    message = f"<b>{payload.get('name')}</b>"
+                    message = f"{payload.get('name')}"
                     if payload.get('link'):
                         message = '<a href="{}"><b>{}</b></a>'.format(payload.get('link'), payload.get('name'))
                     if payload.get('showDeadline'):
@@ -327,6 +327,14 @@ class WorkerThread(QThread):
             options.maximizestrips = True
         if GUI.disableProcessingBox.isChecked():
             options.noprocessing = True
+        if GUI.pdfExtractBox.isChecked():
+            options.pdfextract = True
+        if GUI.pdfWidthBox.isChecked():
+            options.pdfwidth = True
+        if GUI.smartCoverCropBox.isChecked():
+            options.smartcovercrop = True
+        if GUI.coverFillBox.isChecked():
+            options.coverfill = True
         if GUI.metadataTitleBox.checkState() == Qt.CheckState.PartiallyChecked:
             options.metadatatitle = 1
         elif GUI.metadataTitleBox.checkState() == Qt.CheckState.Checked:
@@ -341,12 +349,22 @@ class WorkerThread(QThread):
             options.filefusion = False
         if GUI.noRotateBox.isChecked():
             options.norotate = True
+        if GUI.rotateRightBox.isChecked():
+            options.rotateright = True
         if GUI.rotateFirstBox.isChecked():
             options.rotatefirst = True
+        if GUI.forcePngRgbBox.isChecked():
+            options.force_png_rgb = True
         if GUI.mozJpegBox.checkState() == Qt.CheckState.PartiallyChecked:
             options.forcepng = True
         elif GUI.mozJpegBox.checkState() == Qt.CheckState.Checked:
             options.mozjpeg = True
+        if GUI.webpBox.isChecked():
+            options.webp = True
+        if GUI.pngLegacyBox.isChecked():
+            options.pnglegacy = True
+        if GUI.noQuantizeBox.isChecked():
+            options.noquantize = True
         if GUI.jpegQualityBox.isChecked():
             options.jpegquality = GUI.jpegQualitySpinBox.value()
         if GUI.currentMode > 2:
@@ -521,6 +539,7 @@ class WorkerThread(QThread):
                             if os.path.exists(item.replace('.epub', '.mobi')):
                                 os.remove(item.replace('.epub', '.mobi'))
                         MW.addMessage.emit('KindleGen failed to create MOBI!', 'error', False)
+                        MW.addMessage.emit(self.kindlegenErrorCode[1], 'error', False)
                         MW.addTrayMessage.emit('KindleGen failed to create MOBI!', 'Critical')
                         if self.kindlegenErrorCode[0] == 1 and self.kindlegenErrorCode[1] != '':
                             MW.showDialog.emit("KindleGen error:\n\n" + self.kindlegenErrorCode[1], 'error')
@@ -677,6 +696,18 @@ class KCCGUI(KCC_ui.Ui_mainWindow):
         # noinspection PyCallByClass
         QDesktopServices.openUrl(QUrl('https://ko-fi.com/eink_dude'))
 
+    def openHumble(self):
+        # noinspection PyCallByClass
+        QDesktopServices.openUrl(QUrl('https://humblebundleinc.sjv.io/3JaR3A'))
+
+    def openYouTube(self):
+        # noinspection PyCallByClass
+        QDesktopServices.openUrl(QUrl('https://www.youtube.com/@eink-dude'))
+    
+    def openDiscord(self):
+        # noinspection PyCallByClass
+        QDesktopServices.openUrl(QUrl('https://discord.gg/um5JRKwmGT'))
+
     def modeChange(self, mode):
         if mode == 1:
             self.currentMode = 1
@@ -811,6 +842,14 @@ class KCCGUI(KCC_ui.Ui_mainWindow):
                     if bad_format in current_format:
                         self.addMessage('Scribe PNG MOBI/EPUB has a lot of problems like blank pages/sections. Use JPG instead.', 'warning')
                         break
+            GUI.pngLegacyBox.setEnabled(True)
+            GUI.noQuantizeBox.setEnabled(True)
+            GUI.forcePngRgbBox.setEnabled(True)
+        else:
+            GUI.pngLegacyBox.setEnabled(False)
+            GUI.noQuantizeBox.setEnabled(False)
+            GUI.forcePngRgbBox.setEnabled(False)           
+
 
     def togglechunkSizeCheckBox(self, value):
         GUI.chunkSizeWidget.setVisible(value)
@@ -878,6 +917,10 @@ class KCCGUI(KCC_ui.Ui_mainWindow):
                 if bad_format in current_format:
                     self.addMessage('Colorsoft MOBI/EPUB can have blank pages. Just go back a few pages, exit, and reenter book.', 'info')
                     break
+        elif profile['Label'] == 'KDX':
+            GUI.mozJpegBox.setCheckState(Qt.CheckState.PartiallyChecked)
+            GUI.borderBox.setCheckState(Qt.CheckState.PartiallyChecked)
+            GUI.pngLegacyBox.setChecked(True)
         if not profile['PVOptions']:
             GUI.qualityBox.setChecked(False)
         if str(GUI.deviceBox.currentText()) == 'Other':
@@ -904,7 +947,7 @@ class KCCGUI(KCC_ui.Ui_mainWindow):
         elif not GUI.webtoonBox.isChecked():
             GUI.chunkSizeCheckBox.setEnabled(True)
         if GUI.formats[str(GUI.formatBox.currentText())]['format'] in ('CBZ', 'PDF') and not GUI.webtoonBox.isChecked():
-            self.addMessage("Partially check W/B Margins if you don't want KCC to extend the image margins.", 'info')   
+            self.addMessage("Partially check W/B Margins if you don't want KCC to extend the image margins.", 'info')
 
     def stripTags(self, html):
         s = HTMLStripper()
@@ -1032,9 +1075,18 @@ class KCCGUI(KCC_ui.Ui_mainWindow):
                                            'colorBox': GUI.colorBox.checkState(),
                                            'eraseRainbowBox': GUI.eraseRainbowBox.checkState(),
                                            'disableProcessingBox': GUI.disableProcessingBox.checkState(),
+                                           'pdfExtractBox': GUI.pdfExtractBox.checkState(),
+                                           'pdfWidthBox': GUI.pdfWidthBox.checkState(),
+                                           'smartCoverCropBox': GUI.smartCoverCropBox.checkState(),
+                                           'coverFillBox': GUI.coverFillBox.checkState(),
                                            'metadataTitleBox': GUI.metadataTitleBox.checkState(),
                                            'mozJpegBox': GUI.mozJpegBox.checkState(),
+                                           'forcePngRgbBox': GUI.forcePngRgbBox.checkState(),
+                                           'webpBox': GUI.webpBox.checkState(),
+                                           'pngLegacyBox': GUI.pngLegacyBox.checkState(),
+                                           'noQuantizeBox': GUI.noQuantizeBox.checkState(),
                                            'jpegQualityBox': GUI.jpegQualityBox.checkState(),
+                                           'jpegQuality': GUI.jpegQualitySpinBox.value(),
                                            'widthBox': GUI.widthBox.value(),
                                            'heightBox': GUI.heightBox.value(),
                                            'deleteBox': GUI.deleteBox.checkState(),
@@ -1042,6 +1094,7 @@ class KCCGUI(KCC_ui.Ui_mainWindow):
                                            'fileFusionBox': GUI.fileFusionBox.checkState(),
                                            'defaultOutputFolderBox': GUI.defaultOutputFolderBox.checkState(),
                                            'noRotateBox': GUI.noRotateBox.checkState(),
+                                           'rotateRightBox': GUI.rotateRightBox.checkState(),
                                            'rotateFirstBox': GUI.rotateFirstBox.checkState(),
                                            'maximizeStrips': GUI.maximizeStrips.checkState(),
                                            'gammaSlider': float(self.gammaValue) * 100,
@@ -1086,7 +1139,8 @@ class KCCGUI(KCC_ui.Ui_mainWindow):
                 if message[-1] == '/':
                     message = message[:-1]
             self.handleMessage(message)
-        GUI.jobList.sortItems()
+        # sorting may conflict with manual file fusion order
+        # GUI.jobList.sortItems()
 
     def forceShutdown(self):
         self.saveSettings(None)
@@ -1163,7 +1217,7 @@ class KCCGUI(KCC_ui.Ui_mainWindow):
                             'convertButton', 'formatBox']:
                 getattr(GUI, element).setMinimumSize(QSize(0, 0))
             GUI.gridLayout.setContentsMargins(-1, -1, -1, -1)
-            for element in ['gridLayout_2', 'gridLayout_3', 'gridLayout_4', 'horizontalLayout', 'horizontalLayout_2']:
+            for element in ['gridLayout_2', 'gridLayout_3', 'gridLayout_4', 'gridLayout_6', 'horizontalLayout_2']:
                 getattr(GUI, element).setContentsMargins(-1, 0, -1, 0)
             if self.windowSize == '0x0':
                 MW.resize(500, 500)
@@ -1173,6 +1227,7 @@ class KCCGUI(KCC_ui.Ui_mainWindow):
             "EPUB": {'icon': 'EPUB', 'format': 'EPUB'},
             "CBZ": {'icon': 'CBZ', 'format': 'CBZ'},
             "PDF": {'icon': 'EPUB', 'format': 'PDF'},
+            "PDF (200MB limit)": {'icon': 'EPUB', 'format': 'PDF-200MB'},
             "KFX (does not work)": {'icon': 'KFX', 'format': 'KFX'},
             "MOBI + EPUB": {'icon': 'MOBI', 'format': 'MOBI+EPUB'},
             "EPUB (200MB limit)": {'icon': 'EPUB', 'format': 'EPUB-200MB'},
@@ -1195,6 +1250,9 @@ class KCCGUI(KCC_ui.Ui_mainWindow):
             "Kindle 1920x1920": {
                 'PVOptions': True, 'ForceExpert': False, 'DefaultFormat': 0, 'DefaultUpscale': False, 'ForceColor': False, 'Label': 'KS1920',
             },
+            "Kindle 1240x1860": {
+                'PVOptions': True, 'ForceExpert': False, 'DefaultFormat': 0, 'DefaultUpscale': False, 'ForceColor': False, 'Label': 'KS1240',
+            },
             "Kindle Scribe 1/2": {
                 'PVOptions': True, 'ForceExpert': False, 'DefaultFormat': 0, 'DefaultUpscale': False, 'ForceColor': False, 'Label': 'KS',
             },
@@ -1304,6 +1362,7 @@ class KCCGUI(KCC_ui.Ui_mainWindow):
             "Separator",
             "Kindle 1920x1920",
             "Kindle 1860x1920",
+            "Kindle 1240x1860",
             "Kindle 8/10",
             "Kindle Oasis 8",
             "Kindle Paperwhite 7/10",
@@ -1351,7 +1410,7 @@ class KCCGUI(KCC_ui.Ui_mainWindow):
                             '<a href="https://github.com/ciromattia/kcc/wiki/Important-tips">important tips</a>.',
                             'info')
         
-        self.tar = 'tar' in available_archive_tools()
+        self.tar = TAR in available_archive_tools()
         self.sevenzip = SEVENZIP in available_archive_tools()
         if not any([self.tar, self.sevenzip]):
             self.addMessage('<a href="https://github.com/ciromattia/kcc#7-zip">Install 7z (link)</a>'
@@ -1366,6 +1425,9 @@ class KCCGUI(KCC_ui.Ui_mainWindow):
         GUI.editorButton.clicked.connect(self.selectFileMetaEditor)
         GUI.wikiButton.clicked.connect(self.openWiki)
         GUI.kofiButton.clicked.connect(self.openKofi)
+        GUI.humbleButton.clicked.connect(self.openHumble)
+        GUI.youtubeButton.clicked.connect(self.openYouTube)
+        GUI.discordButton.clicked.connect(self.openDiscord)
         GUI.convertButton.clicked.connect(self.convertStart)
         GUI.gammaSlider.valueChanged.connect(self.changeGamma)
         GUI.gammaBox.stateChanged.connect(self.togglegammaBox)
@@ -1433,6 +1495,8 @@ class KCCGUI(KCC_ui.Ui_mainWindow):
                     GUI.croppingPowerSlider.setValue(int(self.options[option]))
                     self.changeCroppingPower(int(self.options[option]))
                     GUI.preserveMarginBox.setValue(self.options.get('preserveMarginBox', 0))
+            elif str(option) == "jpegQuality":
+                GUI.jpegQualitySpinBox.setValue(int(self.options[option]))
             elif str(option) == "chunkSizeBox":
                 GUI.chunkSizeBox.setValue(int(self.options[option]))
             else:

kindlecomicconverter/__init__.py

@@ -1,4 +1,4 @@
-__version__ = '9.4.2'
+__version__ = '9.7.0'
 __license__ = 'ISC'
 __copyright__ = '2012-2022, Ciro Mattia Gonano <ciromattia@gmail.com>, Pawel Jastrzebski <pawelj@iosphe.re>, darodi'
 __docformat__ = 'restructuredtext en'

kindlecomicconverter/comic2ebook.py

@@ -48,6 +48,7 @@ from .comicarchive import SEVENZIP, available_archive_tools
 from . import comic2panel
 from . import image
 from . import comicarchive
+from . import pdfjpgextract
 from . import dualmetafix
 from . import metadata
 from . import kindle
@@ -352,6 +353,8 @@ def buildOPF(dstdir, title, filelist, originalpath, cover=None):
             mt = 'image/png'
         elif '.gif' == filename[1]:
             mt = 'image/gif'
+        elif '.webp' == filename[1]:
+            mt = 'image/webp'
         else:
             mt = 'image/jpeg'
         f.write("<item id=\"img_" + str(uniqueid) + "\" href=\"" + folder + "/" + path[1] + "\" media-type=\"" +
@@ -659,7 +662,7 @@ def imgDirectoryProcessing(path, job_progress=''):
             raise UserWarning("Conversion interrupted.")
         if len(workerOutput) > 0:
             rmtree(os.path.join(path, '..', '..'), True)
-            raise RuntimeError("One of workers crashed. Cause: " + workerOutput[0][0], workerOutput[0][1])
+            raise RuntimeError("One of workers crashed. Maybe restart PC. Cause: " + workerOutput[0][0], workerOutput[0][1])
     else:
         rmtree(os.path.join(path, '..', '..'), True)
         raise UserWarning("C2E: Source directory is empty.")
@@ -688,7 +691,6 @@ def imgFileProcessing(work):
         workImg = image.ComicPageParser((dirpath, afile), opt)
         for i in workImg.payload:
             img = image.ComicPage(opt, *i)
-            is_color = (opt.forcecolor and img.color)
             if opt.cropping == 2 and not opt.webtoon:
                 img.cropPageNumber(opt.croppingp, opt.croppingm)
             if opt.cropping == 1 and not opt.webtoon:
@@ -698,18 +700,24 @@ def imgFileProcessing(work):
 
             img.gammaCorrectImage()
 
+            if not img.colorOutput:
+                img.convertToGrayscale()
+
             img.autocontrastImage()
             img.resizeImage()
-            img.optimizeForDisplay(opt.eraserainbow, is_color)
+            img.optimizeForDisplay(opt.eraserainbow, img.colorOutput)
 
-            if is_color:
+            if img.colorOutput:
                 pass
             elif opt.forcepng:
-                img.convertToGrayscale()
-                if opt.format != 'PDF':
+                if not opt.noquantize:
                     img.quantizeImage()
-            else:
-                img.convertToGrayscale()
+                if opt.format == 'PDF':
+                    img.convertToGrayscale()
+                elif opt.profile == 'KDX' and opt.format == 'CBZ':
+                    img.convertToGrayscale()
+                elif opt.pnglegacy:
+                    img.convertToGrayscale()
             output.append(img.saveToDir())
         return output
     except Exception:
@@ -739,7 +747,9 @@ def render_page(vector):
     cpu = vector[1]  # number of CPUs
     filename = vector[2]  # document filename
     output_dir = vector[3]
-    target_height = vector[4]
+    target_width = vector[4]
+    target_height = vector[5]
+    pdf_width = vector[6]
     with pymupdf.open(filename) as doc:  # open the document
         num_pages = doc.page_count  # get number of pages
 
@@ -750,7 +760,10 @@ def render_page(vector):
 
         for i in range(seg_from, seg_to):  # work through our page segment
             page = doc[i]
-            zoom = target_height / page.rect.height
+            if not pdf_width or page.rect.width > page.rect.height:
+                zoom = target_height / page.rect.height
+            else:
+                zoom = target_width / page.rect.width
             mat = pymupdf.Matrix(zoom, zoom)
             # TODO: decide colorspace earlier so later color check is cheaper.
             # This is actually pretty hard when you have to deal with color vector text
@@ -800,9 +813,7 @@ def extract_page(vector):
             if len(image_list) > 1:
                 raise UserWarning("mupdf_pdf_extract_page_image() function can be used only with single image pages.")
             if not image_list:
-                width, height = int(page.rect.width), int(page.rect.height)
-                blank_page = Image.new("RGB", (width, height), "white")
-                blank_page.save(output_path)
+                continue
             else:
                 xref = image_list[0][0]
                 d = doc.extract_image(xref)
@@ -817,7 +828,7 @@ def extract_page(vector):
 
 
 
-def mupdf_pdf_process_pages_parallel(filename, output_dir, target_height):
+def mupdf_pdf_process_pages_parallel(filename, output_dir, target_width, target_height):
     render = False
     with pymupdf.open(filename) as doc:
         for page in doc:
@@ -837,7 +848,7 @@ def mupdf_pdf_process_pages_parallel(filename, output_dir, target_height):
     cpu = cpu_count()
 
     # make vectors of arguments for the processes
-    vectors = [(i, cpu, filename, output_dir, target_height) for i in range(cpu)]
+    vectors = [(i, cpu, filename, output_dir, target_width, target_height, options.pdfwidth) for i in range(cpu)]
     print("Starting %i processes for '%s'." % (cpu, filename))
 
 
@@ -854,6 +865,7 @@ def mupdf_pdf_process_pages_parallel(filename, output_dir, target_height):
 def getWorkFolder(afile, workdir=None):
     if not workdir:
         workdir = mkdtemp('', 'KCC-')
+        # workdir = mkdtemp('', 'KCC-', os.path.dirname(afile))
         fullPath = os.path.join(workdir, 'OEBPS', 'Images')
     else:
         fullPath = workdir
@@ -875,13 +887,21 @@ def getWorkFolder(afile, workdir=None):
                 os.makedirs(fullPath)
             path = workdir
             sanitizePermissions(path)
-            target_height = options.profileData[1][1]
+            if options.pdfextract:
+                pdf = pdfjpgextract.PdfJpgExtract(afile, fullPath)
+                njpg = pdf.extract()
+                if njpg == 0:
+                    raise UserWarning("Failed to extract images from PDF file.")
+                return workdir
+            target_width, target_height = options.profileData[1]
             if options.cropping == 1:
-                target_height = target_height + target_height*0.20 #Account for possible margin at the top and bottom
+                target_height *= 1.2 #Account for possible margin at the top and bottom
+                target_width *= 1.2
             elif options.cropping == 2:
-                target_height = target_height + target_height*0.25 #Account for possible margin at the top and bottom with page number
+                target_height *= 1.25 #Account for possible margin at the top and bottom with page number
+                target_width *= 1.25
             try:
-                mupdf_pdf_process_pages_parallel(afile, fullPath, target_height)
+                mupdf_pdf_process_pages_parallel(afile, fullPath, target_width, target_height)
             except Exception as e:
                 rmtree(path, True)
                 raise UserWarning(f"Failed to extract images from PDF file. {e}")
@@ -957,6 +977,13 @@ def getOutputFilename(srcpath, wantedname, ext, tomenumber):
         while os.path.isfile(basename + '_kcc' + str(counter) + ext):
             counter += 1
         filename = basename + '_kcc' + str(counter) + ext
+    elif options.format == 'MOBI' and ext == '.epub':
+        counter = 0
+        basename = os.path.splitext(filename)[0]
+        if os.path.isfile(basename + '.mobi'):
+            while os.path.isfile(basename + '_kcc' + str(counter) + '.mobi'):
+                counter += 1
+            filename = basename + '_kcc' + str(counter) + ext      
     return filename
 
 
@@ -1066,7 +1093,7 @@ def removeNonImages(filetree):
         raise UserWarning('No images detected, nested archives are not supported.')
 
 
-def sanitizeTree(filetree):
+def sanitizeTree(filetree, prefix='kcc'):
     chapterNames = {}
     page = 1
     cover_path = None
@@ -1076,7 +1103,7 @@ def sanitizeTree(filetree):
             _, ext = getImageFileName(name)
 
             # 9999 page limit
-            unique_name = f'kcc-{page:04}'
+            unique_name = f'{prefix}-{page:04}'
             page += 1
 
             newKey = os.path.join(root, unique_name + ext)
@@ -1269,12 +1296,15 @@ def makeZIP(zipfilename, basedir, job_progress='', isepub=False):
     zipfilename = os.path.abspath(zipfilename) + '.zip'
     if SEVENZIP in available_archive_tools():
         if isepub:
-            mimetypeFile = open(os.path.join(basedir, 'mimetype'), 'w')
+            mimetypeFile = open(os.path.join(basedir, '!mimetype'), 'w')
             mimetypeFile.write('application/epub+zip')
             mimetypeFile.close()
-        subprocess_run([SEVENZIP, 'a', '-tzip', zipfilename, "*"], capture_output=True, check=True, cwd=basedir)
+        subprocess_run([SEVENZIP, 'a', '-mx0', '-tzip', zipfilename, "*"], capture_output=True, check=True, cwd=basedir)
+        # crazy hack to ensure mimetype is first when using 7zip
+        if isepub:
+            subprocess_run([SEVENZIP, 'rn', zipfilename, '!mimetype', 'mimetype'], capture_output=True, check=True, cwd=basedir)
     else:
-        zipOutput = ZipFile(zipfilename, 'w', ZIP_DEFLATED)
+        zipOutput = ZipFile(zipfilename, 'w', ZIP_STORED)
         if isepub:
             zipOutput.writestr('mimetype', 'application/epub+zip', ZIP_STORED)
         for dirpath, _, filenames in os.walk(basedir):
@@ -1337,11 +1367,21 @@ def makeParser():
                                 help="Shift first page to opposite side in landscape for spread alignment")
     output_options.add_argument("--norotate", action="store_true", dest="norotate", default=False,
                                 help="Do not rotate double page spreads in spread splitter option.")
+    output_options.add_argument("--rotateright", action="store_true", dest="rotateright", default=False,
+                                help="Rotate double page spreads in opposite direction.")
     output_options.add_argument("--rotatefirst", action="store_true", dest="rotatefirst", default=False,
                                 help="Put rotated 2 page spread first in spread splitter option.")
 
     processing_options.add_argument("-n", "--noprocessing", action="store_true", dest="noprocessing", default=False,
                                     help="Do not modify image and ignore any profile or processing option")
+    processing_options.add_argument("--pdfextract", action="store_true", dest="pdfextract", default=False,
+                                    help="Use the legacy PDF image extraction method from KCC 8 and earlier")
+    processing_options.add_argument("--pdfwidth", action="store_true", dest="pdfwidth", default=False,
+                                    help="Render vector PDFs to device width instead of height.")
+    processing_options.add_argument("--smartcovercrop", action="store_true", dest="smartcovercrop", default=False,
+                                    help="Attempt to crop main cover from wide image")
+    processing_options.add_argument("--coverfill", action="store_true", dest="coverfill", default=False,
+                                    help="Crop cover to fill screen")
     processing_options.add_argument("-u", "--upscale", action="store_true", dest="upscale", default=False,
                                     help="Resize images smaller than device's resolution")
     processing_options.add_argument("-s", "--stretch", action="store_true", dest="stretch", default=False,
@@ -1377,7 +1417,15 @@ def makeParser():
     output_options.add_argument("--eraserainbow", action="store_true", dest="eraserainbow", default=False,
                                 help="Erase rainbow effect on color eink screen by attenuating interfering frequencies")
     processing_options.add_argument("--forcepng", action="store_true", dest="forcepng", default=False,
-                                    help="Create PNG files instead JPEG")
+                                    help="Create PNG files instead JPEG for black and white images")
+    processing_options.add_argument("--force-png-rgb", action="store_true", dest="force_png_rgb", default=False,
+                                    help="Force color images to be saved as PNG")
+    processing_options.add_argument("--webp", action="store_true", dest="webp", default=False,
+                                    help="Replace JPG with lossy WEBP and PNG with lossless WEBP")
+    processing_options.add_argument("--pnglegacy", action="store_true", dest="pnglegacy", default=False,
+                                    help="Use a more compatible 8 bit png instead of 4 bit")
+    processing_options.add_argument("--noquantize", action="store_true", dest="noquantize", default=False,
+                                    help="Don't quantize to 16 color PNG")
     processing_options.add_argument("--mozjpeg", action="store_true", dest="mozjpeg", default=False,
                                     help="Create JPEG files using mozJpeg")
     processing_options.add_argument("--jpeg-quality", type=int, dest="jpegquality", 
@@ -1404,6 +1452,11 @@ def checkOptions(options):
     options.isKobo = False
     options.bordersColor = None
     options.keep_epub = False
+    if options.format == 'PDF-200MB':
+        options.targetsize = 195
+        options.format = 'PDF'
+        if options.batchsplit != 2:
+            options.batchsplit = 1
     if options.format == 'EPUB-200MB':
         options.targetsize = 195
         options.format = 'EPUB'
@@ -1415,6 +1468,8 @@ def checkOptions(options):
         options.format = 'MOBI'
         if options.batchsplit != 2:
             options.batchsplit = 1
+    if not options.targetsize and options.profile.startswith('Rmk'):
+        options.targetsize = 95
     if options.format == 'MOBI+EPUB':
         options.keep_epub = True
         options.format = 'MOBI'
@@ -1460,9 +1515,6 @@ def checkOptions(options):
     if 'Ko' in options.profile:
         options.panelview = False
         options.hq = False
-    # CBZ files on Kindle DX/DXG support higher resolution
-    if options.profile == 'KDX' and options.format == 'CBZ':
-        options.customheight = 1200
     # KFX output create EPUB that might be can be by jhowell KFX Output Calibre plugin
     if options.format == 'KFX':
         options.format = 'EPUB'
@@ -1488,6 +1540,15 @@ def checkOptions(options):
             options.jpegquality = 85
     options.kindle_azw3 = options.iskindle and ('MOBI' in options.format or 'EPUB' in options.format)
     options.kindle_scribe_azw3 = options.profile.startswith('KS') and options.kindle_azw3
+
+    options.webp_output = options.format != 'PDF' and not options.kindle_azw3 and options.webp
+
+    # CBZ files on Kindle DX/DXG support higher resolution
+    if options.profile == 'KDX' and options.format == 'CBZ':
+        options.profileData = list(image.ProfileData.Profiles[options.profile])
+        options.profileData[1] = list(options.profileData[1])
+        options.profileData[1][1] = 1200
+
     if options.kindle_scribe_azw3:
         options.profileData = list(image.ProfileData.Profiles[options.profile])
         options.profileData[1] = list(options.profileData[1])
@@ -1539,17 +1600,26 @@ def makeFusion(sources: List[str]):
         fusion_path = first_path.parent.joinpath(first_path.name + ' [fused]')
     print("Running Fusion")
 
-    for source in sources:
+    # Check if prefix is needed when user-specified ordering differs from OS natural sorting
+    path_names = [Path(s).stem if Path(s).is_file() else Path(s).name for s in sources]
+    needs_prefix = os_sorted(path_names) != path_names
+
+    for index, source in enumerate(sources, start=1):
         print(f"Processing {source}...")
         checkPre(source)
         print("Checking images...")
         source_path = Path(source)
+        # Add the fusion_0001_ prefix to maintain user-specified order if needed
+        prefix = ''
+        if needs_prefix:
+            prefix = f'fusion_{index:04d}_'
         if source_path.is_file():
-            targetpath = fusion_path.joinpath(source_path.stem)
+            targetpath = fusion_path.joinpath(f'{prefix}{source_path.stem}')
         else:
-            targetpath = fusion_path.joinpath(source_path.name)
+            targetpath = fusion_path.joinpath(f'{prefix}{source_path.name}')
+
         getWorkFolder(source, str(targetpath))
-        sanitizeTree(targetpath)
+        sanitizeTree(targetpath, prefix='fusion')
         # TODO: remove flattenTree when subchapters are supported
         flattenTree(targetpath)   
 
@@ -1576,6 +1646,9 @@ def makeBook(source, qtgui=None, job_progress=''):
     removeNonImages(os.path.join(path, "OEBPS", "Images"))
     detectSuboptimalProcessing(os.path.join(path, "OEBPS", "Images"), source)
     chapterNames, cover_path = sanitizeTree(os.path.join(path, 'OEBPS', 'Images'))
+    if options.filefusion:
+        # Strip the fusion_0001_ sort prefix from makeFusion if present
+        chapterNames = {k: sub(r'^fusion_\d{4}_', '', v) for k, v in chapterNames.items()}
     cover = None
     if not options.webtoon:
         cover = image.Cover(cover_path, options)
@@ -1767,4 +1840,3 @@ def makeMOBI(work, qtgui=None):
     makeMOBIWorkerPool.close()
     makeMOBIWorkerPool.join()
     return makeMOBIWorkerOutput
-

kindlecomicconverter/comicarchive.py

@@ -30,6 +30,7 @@ from .shared import IMAGE_TYPES, subprocess_run
 
 EXTRACTION_ERROR = 'Failed to extract archive. Try extracting file outside of KCC.'
 SEVENZIP = '7zz' if platform.system() == 'Darwin' else '7z'
+TAR = 'bsdtar' if platform.system() == 'Linux' else 'tar'
 
 
 class ComicArchive:
@@ -73,7 +74,7 @@ class ComicArchive:
         missing = []
 
         extraction_commands = [
-            ['tar', '--exclude', '__MACOSX', '--exclude', '.DS_Store', '--exclude', 'thumbs.db', '--exclude', 'Thumbs.db', '-xf', self.basename, '-C', targetdir],
+            [TAR, '--exclude', '__MACOSX', '--exclude', '.DS_Store', '--exclude', 'thumbs.db', '--exclude', 'Thumbs.db', '-xf', self.basename, '-C', targetdir],
             [SEVENZIP, 'x', '-y', '-xr!__MACOSX', '-xr!.DS_Store', '-xr!thumbs.db', '-xr!Thumbs.db', '-o' + targetdir, self.basename],
         ]
 
@@ -125,7 +126,7 @@ class ComicArchive:
 def available_archive_tools():
     available = []
 
-    for tool in ['tar', SEVENZIP, 'unar', 'unrar']:
+    for tool in [TAR, SEVENZIP, 'unar', 'unrar']:
         try:
             subprocess_run([tool], stdout=PIPE, stderr=STDOUT)
             available.append(tool)

kindlecomicconverter/image.py

@@ -103,6 +103,7 @@ class ProfileData:
         'KPW5': ("Kindle Paperwhite 5/Signature Edition", (1236, 1648), Palette16, 1.0),
         'KS1860': ("Kindle 1860", (1860, 1920), Palette16, 1.0),
         'KS1920': ("Kindle 1920", (1920, 1920), Palette16, 1.0),
+        'KS1240': ("Kindle 1240", (1240, 1860), Palette16, 1.0),
         'KS': ("Kindle Scribe 1/2", (1860, 2480), Palette16, 1.0),
         'KCS': ("Kindle Colorsoft", (1264, 1680), Palette16, 1.0),
         'KS3': ("Kindle Scribe 3", (1986, 2648), Palette16, 1.0),
@@ -157,7 +158,7 @@ class ComicPageParser:
 
         # Detect corruption in source image, let caller catch any exceptions triggered.
         srcImgPath = os.path.join(source[0], source[1])
-        Image.open(srcImgPath).verify()
+        # Image.open(srcImgPath).verify()
         with Image.open(srcImgPath) as im:
             self.image = im.copy()
 
@@ -196,7 +197,10 @@ class ComicPageParser:
                 and not self.opt.webtoon and self.opt.splitter == 1:
             spread = self.image
             if not self.opt.norotate:
-                spread = spread.rotate(90, Image.Resampling.BICUBIC, True)
+                if not self.opt.rotateright:
+                    spread = spread.rotate(90, Image.Resampling.BICUBIC, True)
+                else:
+                    spread = spread.rotate(-90, Image.Resampling.BICUBIC, True)
             self.payload.append(['R', self.source, spread, self.fill])
         elif (width > height) != (dstwidth > dstheight) and not self.opt.webtoon:
             if self.opt.splitter != 1:
@@ -217,7 +221,10 @@ class ComicPageParser:
             if self.opt.splitter > 0:
                 spread = self.image
                 if not self.opt.norotate:
-                    spread = spread.rotate(90, Image.Resampling.BICUBIC, True)
+                    if not self.opt.rotateright:
+                        spread = spread.rotate(90, Image.Resampling.BICUBIC, True)
+                    else:
+                        spread = spread.rotate(-90, Image.Resampling.BICUBIC, True)
                 self.payload.append(['R', self.source, spread, self.fill])
         else:
             self.payload.append(['N', self.source, self.image, self.fill])
@@ -270,6 +277,8 @@ class ComicPage:
         self.original_color_mode = image.mode
         # TODO: color check earlier
         self.image = image.convert("RGB")
+        self.color = self.colorCheck()
+        self.colorOutput = self.color and self.opt.forcecolor
         self.fill = fill
         self.rotated = False
         self.orgPath = os.path.join(path[0], path[1])
@@ -288,8 +297,7 @@ class ComicPage:
         if not hasattr(Image, 'Resampling'):
             Image.Resampling = Image
 
-    @cached_property
-    def color(self):
+    def colorCheck(self):
         if self.original_color_mode in ("L", "1"):
             return False
         if self.opt.webtoon:
@@ -398,17 +406,23 @@ class ComicPage:
             raise RuntimeError('Cannot save image. ' + str(err))
 
     def save_with_codec(self, image, targetPath):
-        if self.opt.forcepng:
+        if self.opt.forcepng and (not self.colorOutput or self.opt.force_png_rgb):
             image.info.pop('transparency', None)
-            if self.opt.iskindle and ('MOBI' in self.opt.format or 'EPUB' in self.opt.format):
+            if self.opt.webp_output:
+                targetPath += '.webp'
+                image.save(targetPath, 'WEBP', lossless=True, quality=self.opt.jpegquality)
+            elif self.opt.kindle_azw3:
                 targetPath += '.gif'
                 image.save(targetPath, 'GIF', optimize=1, interlace=False)
             else:
                 targetPath += '.png'
                 image.save(targetPath, 'PNG', optimize=1)
         else:
-            targetPath += '.jpg'
-            if self.opt.mozjpeg:
+            if self.opt.webp_output:
+                targetPath += '.webp'
+                image.save(targetPath, 'WEBP', quality=self.opt.jpegquality)
+            elif self.opt.mozjpeg:
+                targetPath += '.jpg'
                 with io.BytesIO() as output:
                     image.save(output, format="JPEG", optimize=1, quality=self.opt.jpegquality)
                     input_jpeg_bytes = output.getvalue()
@@ -416,6 +430,7 @@ class ComicPage:
                     with open(targetPath, "wb") as output_jpeg_file:
                         output_jpeg_file.write(output_jpeg_bytes)
             else:
+                targetPath += '.jpg'
                 image.save(targetPath, 'JPEG', optimize=1, quality=self.opt.jpegquality)
         return targetPath
 
@@ -500,7 +515,9 @@ class ComicPage:
         elif method == Image.Resampling.BICUBIC and not self.opt.upscale:
             pass
         else: # if image bigger than device resolution or smaller with upscaling
-            if abs(ratio_image - ratio_device) < AUTO_CROP_THRESHOLD:
+            if self.opt.profile == 'KDX' and abs(ratio_image - ratio_device) < AUTO_CROP_THRESHOLD * 3:
+                self.image = ImageOps.fit(self.image, self.size, method=method)
+            elif abs(ratio_image - ratio_device) < AUTO_CROP_THRESHOLD:
                 self.image = ImageOps.fit(self.image, self.size, method=method)
             elif (self.opt.format in ('CBZ', 'PDF') or self.opt.kfx) and not self.opt.white_borders:
                 self.image = ImageOps.pad(self.image, self.size, method=method, color=self.fill)
@@ -508,7 +525,7 @@ class ComicPage:
                 self.image = ImageOps.contain(self.image, self.size, method=method)
 
     def resize_method(self):
-        if self.image.size[0] < self.size[0] and self.image.size[1] < self.size[1]:
+        if self.image.size[0] <= self.size[0] and self.image.size[1] <= self.size[1]:
             return Image.Resampling.BICUBIC
         else:
             return Image.Resampling.LANCZOS
@@ -562,13 +579,18 @@ class Cover:
         self.image = ImageOps.autocontrast(self.image, preserve_tone=True)
         if not self.options.forcecolor:
             self.image = self.image.convert('L')
-        self.crop_main_cover()
+        if self.options.smartcovercrop:
+            self.crop_main_cover()
 
         size = list(self.options.profileData[1])
         if self.options.kindle_scribe_azw3:
             size[0] = min(size[0], 1920)
             size[1] = min(size[1], 1920)
-        self.image.thumbnail(tuple(size), Image.Resampling.LANCZOS)
+        if self.options.coverfill and not self.options.kindle_scribe_azw3:
+            # TODO: Kindle Scribe case
+            self.image = ImageOps.fit(self.image, tuple(size), Image.Resampling.LANCZOS, centering=(0.5, 0.5))
+        else:
+            self.image.thumbnail(tuple(size), Image.Resampling.LANCZOS)
 
     def crop_main_cover(self):
         w, h = self.image.size

kindlecomicconverter/page_number_crop_alg.py

@@ -160,6 +160,8 @@ def ignore_pixels_near_edge(bw_img):
     for box in edge_bbox:
         edge = bw_img.crop(box)
         h = edge.histogram()
+        if not edge.height or not edge.width:
+            continue
         imperfections = h[255] / (edge.height * edge.width)
         if imperfections > 0 and imperfections < .02:
             bw_img.paste(im=0, box=box)

kindlecomicconverter/pdfjpgextract.py

@@ -0,0 +1,75 @@
+# -*- coding: utf-8 -*-
+#
+# Copyright (c) 2012-2014 Ciro Mattia Gonano <ciromattia@gmail.com>
+# Copyright (c) 2013-2019 Pawel Jastrzebski <pawelj@iosphe.re>
+#
+# Based upon the code snippet by Ned Batchelder
+# (http://nedbatchelder.com/blog/200712/extracting_jpgs_from_pdfs.html)
+#
+# Permission to use, copy, modify, and/or distribute this software for
+# any purpose with or without fee is hereby granted, provided that the
+# above copyright notice and this permission notice appear in all
+# copies.
+#
+# THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL
+# WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED
+# WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE
+# AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL
+# DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA
+# OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
+# TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
+# PERFORMANCE OF THIS SOFTWARE.
+#
+
+import os
+
+# skip stray images a few pixels in size in some PDFs
+# typical images are many thousands in length
+# https://github.com/ciromattia/kcc/pull/546
+STRAY_IMAGE_LENGTH_THRESHOLD = 300
+
+
+class PdfJpgExtract:
+    def __init__(self, fname, fullPath):
+        self.fname = fname
+        self.path = fullPath
+
+    def getPath(self):
+        return self.path
+
+    def extract(self):
+        pdf = open(self.fname, "rb").read()
+        startmark = b"\xff\xd8"
+        startfix = 0
+        endmark = b"\xff\xd9"
+        endfix = 2
+        i = 0
+        njpg = 0
+        while True:
+            istream = pdf.find(b"stream", i)
+            if istream < 0:
+                break
+            istart = pdf.find(startmark, istream, istream + 20)
+            if istart < 0:
+                i = istream + 20
+                continue
+            iend = pdf.find(b"endstream", istart)
+            if iend < 0:
+                raise Exception("Didn't find end of stream!")
+            iend = pdf.find(endmark, iend - 20)
+            if iend < 0:
+                raise Exception("Didn't find end of JPG!")
+            istart += startfix
+            iend += endfix
+            i = iend
+
+            if iend - istart < STRAY_IMAGE_LENGTH_THRESHOLD:
+                continue
+
+            jpg = pdf[istart:iend]
+            jpgfile = open(os.path.join(self.path, "jpg%d.jpg" % njpg), "wb")
+            jpgfile.write(jpg)
+            jpgfile.close()
+            njpg += 1
+
+        return njpg