Revert "huge speed optimization on HDD by removing md5 (#845)"

This reverts commit 01625904d1.

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

@@ -46,25 +46,11 @@ jobs:
           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_${version_built}.exe
-
-      - name: upload-unsigned-artifact
-        id: upload-unsigned-artifact
+      - name: upload build
         uses: actions/upload-artifact@v4
         with:
           name: windows-build
           path: dist/windows/*.exe
-      
-      - id: optional_step_id
-        uses: signpath/github-action-submit-signing-request@v1.1
-        with:
-          api-token: '${{ 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/windows/signed/'
-      
       - name: Release
         uses: softprops/action-gh-release@v2
         if: startsWith(github.ref, 'refs/tags/')
@@ -74,4 +60,4 @@ jobs:
           files: |
             CHANGELOG.md
             LICENSE.txt
-            dist/windows/signed/*.exe
+            dist/windows/*.exe

README.md

@@ -33,9 +33,6 @@ If you find **KCC** valuable you can consider donating to the authors:
 - Alex Xu (active 2023-Present)
   - [![Donate PayPal](https://img.shields.io/badge/Donate-PayPal-green.svg)](https://www.paypal.com/donate/?business=QFJVE7A6LCP6U&no_recurring=0&item_name=Kindle+Comic+Converter&currency_code=USD)
 
-## Sponsors
-
-- Free code signing on Windows provided by [SignPath.io](https://about.signpath.io/), certificate by [SignPath Foundation](https://signpath.org/)
 
 ## DOWNLOADS
 

kindlecomicconverter/__init__.py

@@ -1,4 +1,4 @@
-__version__ = '7.3.1'
+__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

@@ -40,7 +40,7 @@ from subprocess import STDOUT, PIPE
 from psutil import virtual_memory, disk_usage
 from html import escape as hescape
 
-from .shared import available_archive_tools, getImageFileName, walkSort, walkLevel, sanitizeTrace, subprocess_run
+from .shared import available_archive_tools, md5Checksum, getImageFileName, walkSort, walkLevel, sanitizeTrace, subprocess_run
 from . import comic2panel
 from . import image
 from . import comicarchive
@@ -77,14 +77,15 @@ def main(argv=None):
     return 0
 
 
-def buildHTML(path, imgfile):
+def buildHTML(path, imgfile, imgfilepath):
+    imgfilepath = md5Checksum(imgfilepath)
     filename = getImageFileName(imgfile)
     deviceres = options.profileData[1]
-    if not options.noprocessing and "Rotated" in imgfile:
+    if not options.noprocessing and "Rotated" in options.imgMetadata[imgfilepath]:
         rotatedPage = True
     else:
         rotatedPage = False
-    if not options.noprocessing and "BlackBackground" in imgfile:
+    if not options.noprocessing and "BlackBackground" in options.imgMetadata[imgfilepath]:
         additionalStyle = 'background-color:#000000;'
     else:
         additionalStyle = ''
@@ -424,6 +425,7 @@ def buildOPF(dstdir, title, filelist, cover=None):
                   "</container>"])
     f.close()
 
+
 def buildEPUB(path, chapternames, tomenumber, ischunked):
     filelist = []
     chapterlist = []
@@ -504,7 +506,6 @@ 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)
@@ -514,12 +515,10 @@ 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'), afile))
+                chapterlist.append((dirpath.replace('Images', 'Text'), filelist[-1][1]))
                 chapter = True
-            filelist.append(buildHTML(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 = []
@@ -540,7 +539,7 @@ def buildEPUB(path, chapternames, tomenumber, ischunked):
             global_diff = 0
 
             for x in range(0, pageid + cur_diff + 1):
-                if '-KCC-B' in filelist[x][1]:
+                if '-kcc-b' in filelist[x][1]:
                     pageid += diff_delta
                     global_diff += diff_delta
 
@@ -556,6 +555,8 @@ def imgDirectoryProcessing(path):
     global workerPool, workerOutput
     workerPool = Pool(maxtasksperchild=100)
     workerOutput = []
+    options.imgMetadata = {}
+    options.imgOld = []
     work = []
     pagenumber = 0
     for dirpath, _, filenames in os.walk(path):
@@ -565,19 +566,19 @@ 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.")
         if len(workerOutput) > 0:
             rmtree(os.path.join(path, '..', '..'), True)
             raise RuntimeError("One of workers crashed. Cause: " + workerOutput[0][0], workerOutput[0][1])
+        for file in options.imgOld:
+            if os.path.isfile(file):
+                os.remove(file)
     else:
         rmtree(os.path.join(path, '..', '..'), True)
         raise UserWarning("Source directory is empty.")
@@ -587,7 +588,11 @@ def imgFileProcessingTick(output):
     if isinstance(output, tuple):
         workerOutput.append(output)
         workerPool.terminate()
-
+    else:
+        for page in output:
+            if page is not None:
+                options.imgMetadata[page[0]] = page[1]
+                options.imgOld.append(page[2])
     if GUI:
         GUI.progressBarTick.emit('tick')
         if not GUI.conversionAlive:
@@ -731,7 +736,9 @@ def getComicInfo(path, originalpath):
         except Exception:
             os.remove(xmlPath)
             return
-        if defaultTitle:
+        if xml.data['Title']:
+            options.title = hescape(xml.data['Title'])
+        elif defaultTitle:
             if xml.data['Series']:
                 options.title = hescape(xml.data['Series'])
             if xml.data['Volume']:
@@ -790,9 +797,13 @@ def sanitizeTree(filetree):
         for name in files:
             splitname = os.path.splitext(name)
 
-            # 9999 page limit
+            # file needs kcc at front AND back to avoid renaming issues
             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()
+                    break
 
             newKey = os.path.join(root, slugified + splitname[1])
             key = os.path.join(root, name)
@@ -1178,7 +1189,6 @@ def makeBook(source, qtgui=None):
     print("Checking images...")
     getComicInfo(os.path.join(path, "OEBPS", "Images"), source)
     detectSuboptimalProcessing(os.path.join(path, "OEBPS", "Images"), source)
-    chapterNames = sanitizeTree(os.path.join(path, 'OEBPS', 'Images'))
     if options.webtoon:
         y = image.ProfileData.Profiles[options.profile][1][1]
         comic2panel.main(['-y ' + str(y), '-i', '-m', path], qtgui)
@@ -1191,6 +1201,7 @@ def makeBook(source, qtgui=None):
         imgDirectoryProcessing(os.path.join(path, "OEBPS", "Images"))
     if GUI:
         GUI.progressBarTick.emit('1')
+    chapterNames = sanitizeTree(os.path.join(path, 'OEBPS', 'Images'))
     if options.batchsplit > 0:
         tomes = chunk_directory(path)
     else:

kindlecomicconverter/image.py

@@ -20,9 +20,9 @@
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 import io
 import os
-from pathlib import Path
 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
 
@@ -299,12 +299,13 @@ class ComicPage:
 
     def saveToDir(self):
         try:
+            flags = []
             if not self.opt.forcecolor and not self.opt.forcepng:
                 self.image = self.image.convert('L')
             if self.rotated:
-                self.targetPath += '-Rotated'
+                flags.append('Rotated')
             if self.fill != 'white':
-                self.targetPath += '-BlackBackground'
+                flags.append('BlackBackground')
             if self.opt.forcepng:
                 self.image.info["transparency"] = None
                 self.targetPath += '.png'
@@ -320,9 +321,7 @@ class ComicPage:
                             output_jpeg_file.write(output_jpeg_bytes)
                 else:
                     self.image.save(self.targetPath, 'JPEG', optimize=1, quality=85)
-            if os.path.isfile(self.orgPath):
-                os.remove(self.orgPath)
-            return Path(self.targetPath).name
+            return [md5Checksum(self.targetPath), flags, self.orgPath]
         except IOError as err:
             raise RuntimeError('Cannot save image. ' + str(err))
 

kindlecomicconverter/shared.py

@@ -75,6 +75,16 @@ 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))\