bump 7.3.0

* Eliminate unnecessary use of MD5 checksum

md5checksum() computes the actual checksum of a specified file, which is appropriately expensive, but the code seemed to be using the checksum result as a key into the imgMetadata dictionary to avoid handling image files being renamed during processing steps. This seems like a very expensive way to handle the rename so instead, I now update the imgMetadata keys with the new filename in the one place that the rename happens, and avoid MD5 checksums entirely.

* merge conflicts

* Add missing handling for image path renames due to nested chapter folder name

* merge conflicts

* merge conflicts

* add perf_counters

* imgFileProcessing perf_counter

* use startswith and removeprefix

---------

Co-authored-by: utopiafallen <utopiafallen@gmail.com>

README.md

@@ -68,9 +68,11 @@ If you have issues detecting it, get stuck on the MOBI conversion step, or use L
 
 ### 7-Zip
 
-This is only required for certain files and advanced features. 
+This is optional but will make conversions much faster.
 
-KCC will ask you to install if needed. 
+This is required for certain files and advanced features. 
+
+KCC will ask you to install if needed.
 
 Refer to the wiki to install: https://github.com/ciromattia/kcc/wiki/Installation#7-zip
 

kindlecomicconverter/KCC_gui.py

@@ -37,7 +37,7 @@ from packaging.version import Version
 from raven import Client
 from tempfile import gettempdir
 
-from .shared import HTMLStripper, sanitizeTrace, walkLevel, subprocess_run
+from .shared import HTMLStripper, available_archive_tools, sanitizeTrace, walkLevel, subprocess_run
 from . import __version__
 from . import comic2ebook
 from . import metadata
@@ -317,13 +317,8 @@ class WorkerThread(QThread):
                 GUI.progress.content = ''
                 self.errors = True
                 _, _, traceback = sys.exc_info()
-                if len(err.args) == 1:
+                MW.showDialog.emit("Error during conversion %s:\n\n%s\n\nTraceback:\n%s"
-                    MW.showDialog.emit("Error during conversion %s:\n\n%s\n\nTraceback:\n%s"
+                                   % (jobargv[-1], str(err), sanitizeTrace(traceback)), 'error')
-                                       % (jobargv[-1], str(err), sanitizeTrace(traceback)), 'error')
-                else:
-                    MW.showDialog.emit("Error during conversion %s:\n\n%s\n\nTraceback:\n%s"
-                                       % (jobargv[-1], str(err.args[0]), err.args[1]), 'error')
-                    GUI.sentry.extra_context({'realTraceback': err.args[1]})
                 if ' is corrupted.' not in str(err):
                     GUI.sentry.captureException()
                 MW.addMessage.emit('Error during conversion! Please consult '
@@ -1071,19 +1066,12 @@ class KCCGUI(KCC_ui.Ui_mainWindow):
             self.addMessage('Since you are a new user of <b>KCC</b> please see few '
                             '<a href="https://github.com/ciromattia/kcc/wiki/Important-tips">important tips</a>.',
                             'info')
-        try:
+        
-            subprocess_run(['tar'], stdout=PIPE, stderr=STDOUT)
+        self.tar = 'tar' in available_archive_tools()
-            self.tar = True
+        self.sevenzip = '7z' in available_archive_tools()
-        except FileNotFoundError:
+        if not any([self.tar, self.sevenzip]):
-            self.tar = False
+            self.addMessage('<a href="https://github.com/ciromattia/kcc#7-zip">Install 7z (link)</a>'
-        try:
+                            ' to enable CBZ/CBR/ZIP/etc processing.', 'warning')
-            subprocess_run(['7z'], stdout=PIPE, stderr=STDOUT)
-            self.sevenzip = True
-        except FileNotFoundError:
-            self.sevenzip = False
-            if not self.tar:
-                self.addMessage('<a href="https://github.com/ciromattia/kcc#7-zip">Install 7z (link)</a>'
-                                ' to enable CBZ/CBR/ZIP/etc processing.', 'warning')
         self.detectKindleGen(True)
 
         APP.messageFromOtherInstance.connect(self.handleMessage)

kindlecomicconverter/__init__.py

@@ -1,4 +1,4 @@
-__version__ = '7.2.2'
+__version__ = '7.3.0'
 __license__ = 'ISC'
 __copyright__ = '2012-2022, Ciro Mattia Gonano <ciromattia@gmail.com>, Pawel Jastrzebski <pawelj@iosphe.re>, darodi'
 __docformat__ = 'restructuredtext en'

kindlecomicconverter/comic2ebook.py

@@ -23,7 +23,7 @@ import pathlib
 import re
 import sys
 from argparse import ArgumentParser
-from time import strftime, gmtime
+from time import perf_counter, strftime, gmtime
 from copy import copy
 from glob import glob, escape
 from re import sub
@@ -33,14 +33,14 @@ from tempfile import mkdtemp, gettempdir, TemporaryFile
 from shutil import move, copytree, rmtree, copyfile
 from multiprocessing import Pool
 from uuid import uuid4
-from natsort import os_sorted
+from natsort import os_sort_keygen
 from slugify import slugify as slugify_ext
 from PIL import Image, ImageFile
 from subprocess import STDOUT, PIPE
 from psutil import virtual_memory, disk_usage
 from html import escape as hescape
 
-from .shared import md5Checksum, getImageFileName, walkSort, walkLevel, sanitizeTrace, subprocess_run
+from .shared import available_archive_tools, getImageFileName, walkSort, walkLevel, sanitizeTrace, subprocess_run
 from . import comic2panel
 from . import image
 from . import comicarchive
@@ -51,7 +51,7 @@ from . import kindle
 from . import __version__
 
 ImageFile.LOAD_TRUNCATED_IMAGES = True
-
+OS_SORT_KEY = os_sort_keygen()
 
 def main(argv=None):
     global options
@@ -78,7 +78,6 @@ def main(argv=None):
 
 
 def buildHTML(path, imgfile, imgfilepath):
-    imgfilepath = md5Checksum(imgfilepath)
     filename = getImageFileName(imgfile)
     deviceres = options.profileData[1]
     if not options.noprocessing and "Rotated" in options.imgMetadata[imgfilepath]:
@@ -425,7 +424,6 @@ def buildOPF(dstdir, title, filelist, cover=None):
                   "</container>"])
     f.close()
 
-
 def buildEPUB(path, chapternames, tomenumber, ischunked):
     filelist = []
     chapterlist = []
@@ -506,6 +504,7 @@ def buildEPUB(path, chapternames, tomenumber, ischunked):
                       "display: none;\n",
                       "}\n"])
     f.close()
+    build_html_start = perf_counter()
     for dirpath, dirnames, filenames in os.walk(os.path.join(path, 'OEBPS', 'Images')):
         chapter = False
         dirnames, filenames = walkSort(dirnames, filenames)
@@ -515,10 +514,12 @@ def buildEPUB(path, chapternames, tomenumber, ischunked):
                                      'cover' + getImageFileName(afile)[1])
                 options.covers.append((image.Cover(os.path.join(dirpath, afile), cover, options,
                                                    tomenumber), options.uuid))
-            filelist.append(buildHTML(dirpath, afile, os.path.join(dirpath, afile)))
             if not chapter:
-                chapterlist.append((dirpath.replace('Images', 'Text'), filelist[-1][1]))
+                chapterlist.append((dirpath.replace('Images', 'Text'), afile))
                 chapter = True
+            filelist.append(buildHTML(dirpath, afile, os.path.join(dirpath, afile)))
+    build_html_end = perf_counter()
+    print(f"buildHTML: {build_html_end - build_html_start} seconds")
     # Overwrite chapternames if tree is flat and ComicInfo.xml has bookmarks
     if not chapternames and options.chapters and not ischunked:
         chapterlist = []
@@ -566,10 +567,13 @@ def imgDirectoryProcessing(path):
     if GUI:
         GUI.progressBarTick.emit(str(pagenumber))
     if len(work) > 0:
+        img_processing_start = perf_counter()
         for i in work:
             workerPool.apply_async(func=imgFileProcessing, args=(i,), callback=imgFileProcessingTick)
         workerPool.close()
         workerPool.join()
+        img_processing_end = perf_counter()
+        print(f"imgFileProcessing: {img_processing_end - img_processing_start} seconds")
         if GUI and not GUI.conversionAlive:
             rmtree(os.path.join(path, '..', '..'), True)
             raise UserWarning("Conversion interrupted.")
@@ -657,16 +661,14 @@ 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 os.path.isdir(os.path.join(workdir, tdir[0])):
-                    if is_nested_single_dir:
                         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)
@@ -674,8 +676,7 @@ 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'))
+    os.renames(path, os.path.join(newpath, 'OEBPS', 'Images'))
-    rmtree(workdir, True)
     return newpath
 
 
@@ -793,12 +794,16 @@ def getPanelViewSize(deviceres, size):
 
 def sanitizeTree(filetree):
     chapterNames = {}
-    for root, dirs, files in os.walk(filetree, False):
+    page = 1
-        for i, name in enumerate(os_sorted(files)):
+    for root, dirs, files in os.walk(filetree):
+        dirs.sort(key=OS_SORT_KEY)
+        files.sort(key=OS_SORT_KEY)
+        for name in files:
             splitname = os.path.splitext(name)
 
             # file needs kcc at front AND back to avoid renaming issues
-            slugified = f'kcc-{i:04}'
+            slugified = f'kcc-{page:04}'
+            page += 1
             for suffix in '-KCC', '-KCC-A', '-KCC-B', '-KCC-C':
                 if splitname[0].endswith(suffix):
                     slugified += suffix.lower()
@@ -808,7 +813,8 @@ def sanitizeTree(filetree):
             key = os.path.join(root, name)
             if key != newKey:
                 os.replace(key, newKey)
-        for name in dirs:
+                options.imgMetadata[newKey] = options.imgMetadata.pop(key)
+        for i, name in enumerate(dirs):
             tmpName = name
             slugified = slugify(name)
             while os.path.exists(os.path.join(root, slugified)) and name.upper() != slugified.upper():
@@ -818,6 +824,11 @@ def sanitizeTree(filetree):
             key = os.path.join(root, name)
             if key != newKey:
                 os.replace(key, newKey)
+                dirs[i] = newKey
+                existingImgPathKeys = list(options.imgMetadata.keys())
+                for imgPath in existingImgPathKeys:
+                    if imgPath.startswith(key):
+                        options.imgMetadata[newKey + imgPath.removeprefix(key)] = options.imgMetadata.pop(imgPath)
     return chapterNames
 
 
@@ -890,8 +901,7 @@ def chunk_process(path, mode, parent):
                     firstTome = False
     return output
 
-
+def detectSuboptimalProcessing(tmppath, orgpath):
-def detectCorruption(tmppath, orgpath):
     imageNumber = 0
     imageSmaller = 0
     alreadyProcessed = False
@@ -907,9 +917,6 @@ 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
@@ -953,17 +960,27 @@ def slugify(value):
 
 
 def makeZIP(zipfilename, basedir, isepub=False):
+    start = perf_counter()
     zipfilename = os.path.abspath(zipfilename) + '.zip'
-    zipOutput = ZipFile(zipfilename, 'w', ZIP_DEFLATED)
+    if '7z' in available_archive_tools():
-    if isepub:
+        if isepub:
-        zipOutput.writestr('mimetype', 'application/epub+zip', ZIP_STORED)
+            mimetypeFile = open(os.path.join(basedir, 'mimetype'), 'w')
-    for dirpath, _, filenames in os.walk(basedir):
+            mimetypeFile.write('application/epub+zip')
-        for name in filenames:
+            mimetypeFile.close()
-            path = os.path.normpath(os.path.join(dirpath, name))
+        subprocess_run(['7z', 'a', '-tzip', zipfilename, os.path.join(basedir, "*")], capture_output=True, check=True)
-            aPath = os.path.normpath(os.path.join(dirpath.replace(basedir, ''), name))
+    else:
-            if os.path.isfile(path):
+        zipOutput = ZipFile(zipfilename, 'w', ZIP_DEFLATED)
-                zipOutput.write(path, aPath)
+        if isepub:
-    zipOutput.close()
+            zipOutput.writestr('mimetype', 'application/epub+zip', ZIP_STORED)
+        for dirpath, _, filenames in os.walk(basedir):
+            for name in filenames:
+                path = os.path.normpath(os.path.join(dirpath, name))
+                aPath = os.path.normpath(os.path.join(dirpath.replace(basedir, ''), name))
+                if os.path.isfile(path):
+                    zipOutput.write(path, aPath)
+        zipOutput.close()
+    end = perf_counter()
+    print(f"makeZIP time: {end - start} seconds")
     return zipfilename
 
 
@@ -1138,9 +1155,7 @@ def checkTools(source):
     source = source.upper()
     if source.endswith('.CB7') or source.endswith('.7Z') or source.endswith('.RAR') or source.endswith('.CBR') or \
             source.endswith('.ZIP') or source.endswith('.CBZ'):
-        try:
+        if '7z' not in available_archive_tools():
-            subprocess_run(['7z'], stdout=PIPE, stderr=STDOUT)
-        except FileNotFoundError:
             print('ERROR: 7z is missing!')
             sys.exit(1)
     if options.format == 'MOBI':
@@ -1170,6 +1185,7 @@ def checkPre(source):
 
 
 def makeBook(source, qtgui=None):
+    start = perf_counter()
     global GUI
     GUI = qtgui
     if GUI:
@@ -1181,7 +1197,7 @@ 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)
@@ -1272,6 +1288,8 @@ def makeBook(source, qtgui=None):
         elif os.path.isdir(source):
             rmtree(source)
 
+    end = perf_counter()
+    print(f"makeBook: {end - start} seconds")
     return 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

@@ -22,7 +22,6 @@ import io
 import os
 import mozjpeg_lossless_optimization
 from PIL import Image, ImageOps, ImageStat, ImageChops, ImageFilter
-from .shared import md5Checksum
 from .page_number_crop_alg import get_bbox_crop_margin_page_number, get_bbox_crop_margin
 from .inter_panel_crop_alg import crop_empty_inter_panel
 
@@ -141,7 +140,13 @@ 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
@@ -315,7 +320,7 @@ class ComicPage:
                             output_jpeg_file.write(output_jpeg_bytes)
                 else:
                     self.image.save(self.targetPath, 'JPEG', optimize=1, quality=85)
-            return [md5Checksum(self.targetPath), flags, self.orgPath]
+            return [self.targetPath, flags, self.orgPath]
         except IOError as err:
             raise RuntimeError('Cannot save image. ' + str(err))
 

kindlecomicconverter/shared.py

@@ -18,6 +18,7 @@
 # PERFORMANCE OF THIS SOFTWARE.
 #
 
+from functools import lru_cache
 import os
 from hashlib import md5
 from html.parser import HTMLParser
@@ -74,16 +75,6 @@ def walkLevel(some_dir, level=1):
             del dirs[:]
 
 
-def md5Checksum(fpath):
-    with open(fpath, 'rb') as fh:
-        m = md5()
-        while True:
-            data = fh.read(8192)
-            if not data:
-                break
-            m.update(data)
-        return m.hexdigest()
-
 
 def sanitizeTrace(traceback):
     return ''.join(format_tb(traceback))\
@@ -137,6 +128,19 @@ def dependencyCheck(level):
         print('ERROR: ' + ', '.join(missing) + ' is not installed!')
         sys.exit(1)
 
+@lru_cache
+def available_archive_tools():
+    available = []
+
+    for tool in ['tar', '7z', 'unar', 'unrar']:
+        try:
+            subprocess_run([tool], stdout=subprocess.PIPE, stderr=subprocess.STDOUT)
+            available.append(tool)
+        except FileNotFoundError:
+            pass
+    
+    return available
+
 def subprocess_run(command, **kwargs):
     if (os.name == 'nt'):
         kwargs.setdefault('creationflags', subprocess.CREATE_NO_WINDOW)