bump 7.2.3

Commit f952634971 moved image corruption detection out from the ComicPage constructor and into a standalone detectCorruption() function. This led to a performance regression because now corruption detection happens in a single thread when it used to be distributed across worker threads, and because a source image is now loaded twice in memory: once during corruption detection and once when actually going to process the image.

Image file corruption detection is now back inside the ComicPage constructor and the extra load() has been removed because the convert() call will automatically invoke load() and most likely throw the same exceptions.

.github/workflows/package-windows-with-docker.yml

@@ -25,6 +25,11 @@ jobs:
 #      - name: build binary
 #        run: |
 #          pyi-makespec -F -i icons\\comic2ebook.ico -n KCC_test -w --noupx kcc.py
+      - name: Package Application
+        uses: JackMcKew/pyinstaller-action-windows@main
+        with:
+          path: .
+          spec: ./kcc.spec
       - name: Package Application
         uses: JackMcKew/pyinstaller-action-windows@main
         with:
@@ -38,6 +43,7 @@ jobs:
       - name: rename binaries
         run: |   
           version_built=$(cat kindlecomicconverter/__init__.py | grep version | awk '{print $3}' | sed "s/[^.0-9b]//g")
+          mv dist/windows/kcc.exe dist/windows/KCC_${version_built}.exe 
           mv dist/windows/kcc-c2e.exe dist/windows/KCC_c2e_${version_built}.exe 
           mv dist/windows/kcc-c2p.exe dist/windows/KCC_c2p_${version_built}.exe
       - name: upload build

kcc.spec

@@ -0,0 +1,39 @@
+# -*- mode: python ; coding: utf-8 -*-
+
+
+block_cipher = None
+
+
+a = Analysis(['kcc.py'],
+             pathex=['.'],
+             binaries=[],
+             datas=[],
+             hiddenimports=['_cffi_backend'],
+             hookspath=[],
+             runtime_hooks=[],
+             excludes=[],
+             win_no_prefer_redirects=False,
+             win_private_assemblies=False,
+             cipher=block_cipher,
+             noarchive=False)
+pyz = PYZ(a.pure, a.zipped_data,
+             cipher=block_cipher)
+
+exe = EXE(pyz,
+          a.scripts,
+          a.binaries,
+          a.zipfiles,
+          a.datas,  
+          [],
+          name='kcc',
+          debug=False,
+          bootloader_ignore_signals=False,
+          strip=False,
+          upx=False,
+          upx_exclude=[],
+          runtime_tmpdir=None,
+          console=False,
+          disable_windowed_traceback=False,
+          target_arch=None,
+          codesign_identity=None,
+          entitlements_file=None , icon='icons\\comic2ebook.ico')

kindlecomicconverter/__init__.py

@@ -1,4 +1,4 @@
-__version__ = '7.2.0'
+__version__ = '7.2.3'
 __license__ = 'ISC'
 __copyright__ = '2012-2022, Ciro Mattia Gonano <ciromattia@gmail.com>, Pawel Jastrzebski <pawelj@iosphe.re>, darodi'
 __docformat__ = 'restructuredtext en'

kindlecomicconverter/comic2ebook.py

@@ -645,6 +645,7 @@ def getWorkFolder(afile):
         if afile.lower().endswith('.pdf'):
             pdf = pdfjpgextract.PdfJpgExtract(afile)
             path, njpg = pdf.extract()
+            workdir = path
             sanitizePermissions(path)
             if njpg == 0:
                 rmtree(path, True)
@@ -656,16 +657,14 @@ def getWorkFolder(afile):
                 path = cbx.extract(workdir)
                 sanitizePermissions(path)
                 tdir = os.listdir(workdir)
-                is_nested_single_dir = False
                 if len(tdir) == 2 and 'ComicInfo.xml' in tdir:
                     tdir.remove('ComicInfo.xml')
-                    is_nested_single_dir = os.path.isdir(os.path.join(workdir, tdir[0]))
-                    if is_nested_single_dir:
+                    if os.path.isdir(os.path.join(workdir, tdir[0])):
                         os.replace(
                             os.path.join(workdir, 'ComicInfo.xml'),
                             os.path.join(workdir, tdir[0], 'ComicInfo.xml')
                         )
-                if len(tdir) == 1 and is_nested_single_dir:
+                if len(tdir) == 1 and os.path.isdir(os.path.join(workdir, tdir[0])):
                     path = os.path.join(workdir, tdir[0])           
             except OSError as e:
                 rmtree(workdir, True)
@@ -673,8 +672,7 @@ def getWorkFolder(afile):
     else:
         raise UserWarning("Failed to open source file/directory.")
     newpath = mkdtemp('', 'KCC-', os.path.dirname(afile))
-    copytree(path, os.path.join(newpath, 'OEBPS', 'Images'))
-    rmtree(workdir, True)
+    os.renames(path, os.path.join(newpath, 'OEBPS', 'Images'))
     return newpath
 
 
@@ -828,12 +826,11 @@ def sanitizePermissions(filetree):
             os.chmod(os.path.join(root, name), S_IWRITE | S_IREAD | S_IEXEC)
 
 
-def batch_directory(path):
+def chunk_directory(path):
     level = -1
     for root, _, files in os.walk(os.path.join(path, 'OEBPS', 'Images')):
         for f in files:
-            if f.endswith('.jpg') or f.endswith('.jpeg') or f.endswith('.png') or f.endswith('.gif') or \
-                    f.endswith('.webp'):
+            if getImageFileName(f):
                 newLevel = os.path.join(root, f).replace(os.path.join(path, 'OEBPS', 'Images'), '').count(os.sep)
                 if level != -1 and level != newLevel:
                     level = 0
@@ -842,16 +839,16 @@ def batch_directory(path):
                     level = newLevel
     if level > 0:
         parent = pathlib.Path(path).parent
-        batcher = batch_process(os.path.join(path, 'OEBPS', 'Images'), level, parent)
+        chunker = chunk_process(os.path.join(path, 'OEBPS', 'Images'), level, parent)
         path = [path]
-        for tome in batcher:
+        for tome in chunker:
             path.append(tome)
         return path
     else:
         raise UserWarning('Unsupported directory structure.')
 
 
-def batch_process(path, mode, parent):
+def chunk_process(path, mode, parent):
     output = []
     currentSize = 0
     currentTarget = path
@@ -890,8 +887,7 @@ def batch_process(path, mode, parent):
                     firstTome = False
     return output
 
-
-def detectCorruption(tmppath, orgpath):
+def detectSuboptimalProcessing(tmppath, orgpath):
     imageNumber = 0
     imageSmaller = 0
     alreadyProcessed = False
@@ -907,9 +903,6 @@ def detectCorruption(tmppath, orgpath):
                     raise RuntimeError('Image file %s is corrupted.' % pathOrg)
                 try:
                     img = Image.open(path)
-                    img.verify()
-                    img = Image.open(path)
-                    img.load()
                     imageNumber += 1
                     if options.profileData[1][0] > img.size[0] and options.profileData[1][1] > img.size[1]:
                         imageSmaller += 1
@@ -1181,7 +1174,7 @@ def makeBook(source, qtgui=None):
     path = getWorkFolder(source)
     print("Checking images...")
     getComicInfo(os.path.join(path, "OEBPS", "Images"), source)
-    detectCorruption(os.path.join(path, "OEBPS", "Images"), source)
+    detectSuboptimalProcessing(os.path.join(path, "OEBPS", "Images"), source)
     if options.webtoon:
         y = image.ProfileData.Profiles[options.profile][1][1]
         comic2panel.main(['-y ' + str(y), '-i', '-m', path], qtgui)
@@ -1196,7 +1189,7 @@ def makeBook(source, qtgui=None):
         GUI.progressBarTick.emit('1')
     chapterNames = sanitizeTree(os.path.join(path, 'OEBPS', 'Images'))
     if options.batchsplit > 0:
-        tomes = batch_directory(path)
+        tomes = chunk_directory(path)
     else:
         tomes = [path]
     filepath = []

kindlecomicconverter/comicarchive.py

@@ -76,6 +76,8 @@ class ComicArchive:
                 ['unar', self.filepath, '-f', '-o', targetdir]
             )
 
+        extraction_commands.reverse()
+
         if distro.id() == 'fedora' or distro.like() == 'fedora':
             extraction_commands.append(
                 ['unrar', 'x', '-y', '-x__MACOSX', '-x.DS_Store', '-xthumbs.db', '-xThumbs.db', self.filepath, targetdir]

kindlecomicconverter/image.py

@@ -141,7 +141,13 @@ class ComicPageParser:
         self.source = source
         self.size = self.opt.profileData[1]
         self.payload = []
-        self.image = Image.open(os.path.join(source[0], source[1])).convert('RGB')
+
+        # Detect corruption in source image, let caller catch any exceptions triggered.
+        srcImgPath = os.path.join(source[0], source[1])
+        self.image = Image.open(srcImgPath)
+        self.image.verify()
+        self.image = Image.open(srcImgPath).convert('RGB')
+
         self.color = self.colorCheck()
         self.fill = self.fillCheck()
         # backwards compatibility for Pillow >9.1.0

kindlecomicconverter/shared.py

@@ -49,7 +49,7 @@ class HTMLStripper(HTMLParser):
 def getImageFileName(imgfile):
     name, ext = os.path.splitext(imgfile)
     ext = ext.lower()
-    if (name.startswith('.') and len(name) == 1) or ext not in ['.png', '.jpg', '.jpeg', '.gif', '.webp']:
+    if (name.startswith('.') and len(name) == 1) or ext not in ['.png', '.jpg', '.jpeg', '.gif', '.webp', '.jp2', '.j2k', '.jpx']:
         return None
     return [name, ext]