1
0
mirror of https://github.com/ciromattia/kcc synced 2026-06-11 17:10:34 +00:00

Compare commits

..

9 Commits

Author SHA1 Message Date
Alex Xu 21249854b9 Revert "upgrade 7z to 7zz (#1005)"
This reverts commit 17c0a73f9f.
2025-07-03 12:24:54 -07:00
Alex Xu 0abf620698 Create FUNDING.yml 2025-07-03 11:42:46 -07:00
Alex Xu 69d3bf3278 simplify removeNonImages (#1009) 2025-07-02 17:28:03 -07:00
Alex Xu 793992f408 bump to 8.0.0 2025-07-02 10:18:30 -07:00
Alex Xu f41d5327e0 remove non images early (#1007) 2025-07-02 10:17:54 -07:00
Alex Xu 6f960aa1d0 bump mozjpeg 2025-07-01 08:12:32 -07:00
Alex Xu 17c0a73f9f upgrade 7z to 7zz (#1005) 2025-07-01 08:12:01 -07:00
Adrian 1fa5a5b19b Improved color detection (#1003)
* Improved color detection

* use pure python

---------

Co-authored-by: Alex Xu <alexkurosakimh3@gmail.com>
2025-06-30 17:18:04 -07:00
Alex Xu e8d05c16aa Update README.md 2025-06-29 14:03:43 -07:00
7 changed files with 71 additions and 63 deletions
+15
View File
@@ -0,0 +1,15 @@
# These are supported funding model platforms
github: # Replace with up to 4 GitHub Sponsors-enabled usernames e.g., [user1, user2]
patreon: # Replace with a single Patreon username
open_collective: # Replace with a single Open Collective username
ko_fi: eink_dude
tidelift: # Replace with a single Tidelift platform-name/package-name e.g., npm/babel
community_bridge: # Replace with a single Community Bridge project-name e.g., cloud-foundry
liberapay: # Replace with a single Liberapay username
issuehunt: # Replace with a single IssueHunt username
lfx_crowdfunding: # Replace with a single LFX Crowdfunding project-name e.g., cloud-foundry
polar: # Replace with a single Polar username
buy_me_a_coffee: # Replace with a single Buy Me a Coffee username
thanks_dev: # Replace with a single thanks.dev username
custom: # Replace with up to 4 custom sponsorship URLs e.g., ['link1', 'link2']
+1 -1
View File
@@ -27,7 +27,7 @@ on underpowered ereaders with small storage capacities.
KCC avoids many common formatting issues (some of which occur [even on the Kindle Store](https://github.com/ciromattia/kcc/wiki/Kindle-Store-bad-formatting)), such as: KCC avoids many common formatting issues (some of which occur [even on the Kindle Store](https://github.com/ciromattia/kcc/wiki/Kindle-Store-bad-formatting)), such as:
1) faded black levels causing unneccessarily low contrast, which is hard to see and can cause eyestrain. 1) faded black levels causing unneccessarily low contrast, which is hard to see and can cause eyestrain.
2) unneccessary margins at the bottom of the screen 2) unneccessary margins at the bottom of the screen
3) Not utilizing the full 1860x2480 resolution of the 10" Kindle Scribe (feature in progress) 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 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 5) unaligned two page spreads in landscape, where pages are shifted over by 1
+1 -1
View File
@@ -1,4 +1,4 @@
__version__ = '7.6.0' __version__ = '8.0.0'
__license__ = 'ISC' __license__ = 'ISC'
__copyright__ = '2012-2022, Ciro Mattia Gonano <ciromattia@gmail.com>, Pawel Jastrzebski <pawelj@iosphe.re>, darodi' __copyright__ = '2012-2022, Ciro Mattia Gonano <ciromattia@gmail.com>, Pawel Jastrzebski <pawelj@iosphe.re>, darodi'
__docformat__ = 'restructuredtext en' __docformat__ = 'restructuredtext en'
+21 -19
View File
@@ -647,7 +647,7 @@ def imgFileProcessing(work):
img.autocontrastImage() img.autocontrastImage()
img.resizeImage() img.resizeImage()
img.optimizeForDisplay(opt.reducerainbow) img.optimizeForDisplay(opt.reducerainbow)
if opt.forcecolor and workImg.color: if opt.forcecolor and img.color:
pass pass
elif opt.forcepng: elif opt.forcepng:
img.quantizeImage() img.quantizeImage()
@@ -690,16 +690,6 @@ def getWorkFolder(afile):
cbx = comicarchive.ComicArchive(afile) cbx = comicarchive.ComicArchive(afile)
path = cbx.extract(workdir) path = cbx.extract(workdir)
sanitizePermissions(path) sanitizePermissions(path)
tdir = os.listdir(workdir)
if len(tdir) == 2 and 'ComicInfo.xml' in tdir:
tdir.remove('ComicInfo.xml')
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 os.path.isdir(os.path.join(workdir, tdir[0])):
path = os.path.join(workdir, tdir[0])
except OSError as e: except OSError as e:
rmtree(workdir, True) rmtree(workdir, True)
raise UserWarning(e) raise UserWarning(e)
@@ -822,6 +812,22 @@ def getPanelViewSize(deviceres, size):
return str(int(x)), str(int(y)) return str(int(x)), str(int(y))
def removeNonImages(filetree):
# clean dot from original file
dot_clean(filetree)
for root, dirs, files in os.walk(filetree):
for name in files:
_, ext = getImageFileName(name)
if ext not in ('.png', '.jpg', '.jpeg', '.gif', '.webp', '.jp2'):
if os.path.exists(os.path.join(root, name)):
os.remove(os.path.join(root, name))
# remove empty nested folders
for root, dirs, files in os.walk(filetree, False):
if not files and not dirs:
os.rmdir(root)
def sanitizeTree(filetree): def sanitizeTree(filetree):
chapterNames = {} chapterNames = {}
page = 1 page = 1
@@ -830,13 +836,13 @@ def sanitizeTree(filetree):
dirs.sort(key=OS_SORT_KEY) dirs.sort(key=OS_SORT_KEY)
files.sort(key=OS_SORT_KEY) files.sort(key=OS_SORT_KEY)
for name in files: for name in files:
splitname = os.path.splitext(name) _, ext = getImageFileName(name)
# 9999 page limit # 9999 page limit
slugified = f'kcc-{page:04}' slugified = f'kcc-{page:04}'
page += 1 page += 1
newKey = os.path.join(root, slugified + splitname[1]) newKey = os.path.join(root, slugified + ext)
key = os.path.join(root, name) key = os.path.join(root, name)
if key != newKey: if key != newKey:
os.replace(key, newKey) os.replace(key, newKey)
@@ -870,8 +876,7 @@ def sanitizePermissions(filetree):
os.chmod(os.path.join(root, name), S_IWRITE | S_IREAD) os.chmod(os.path.join(root, name), S_IWRITE | S_IREAD)
for name in dirs: for name in dirs:
os.chmod(os.path.join(root, name), S_IWRITE | S_IREAD | S_IEXEC) os.chmod(os.path.join(root, name), S_IWRITE | S_IREAD | S_IEXEC)
# clean dot from original file
dot_clean(filetree)
def dot_clean(filetree): def dot_clean(filetree):
for root, _, files in os.walk(filetree, topdown=False): for root, _, files in os.walk(filetree, topdown=False):
@@ -990,10 +995,6 @@ def detectSuboptimalProcessing(tmppath, orgpath):
os.remove(os.path.join(root, name)) os.remove(os.path.join(root, name))
except OSError as e: except OSError as e:
raise RuntimeError(f"{name}: {e}") raise RuntimeError(f"{name}: {e}")
# remove empty nested folders
for root, dirs, files in os.walk(tmppath, False):
if not files and not dirs:
os.rmdir(root)
if alreadyProcessed: if alreadyProcessed:
print("WARNING: Source files are probably created by KCC. The second conversion will decrease quality.") print("WARNING: Source files are probably created by KCC. The second conversion will decrease quality.")
if GUI: if GUI:
@@ -1309,6 +1310,7 @@ def makeBook(source, qtgui=None):
path = getWorkFolder(source) path = getWorkFolder(source)
print("Checking images...") print("Checking images...")
getComicInfo(os.path.join(path, "OEBPS", "Images"), source) getComicInfo(os.path.join(path, "OEBPS", "Images"), source)
removeNonImages(os.path.join(path, "OEBPS", "Images"))
detectSuboptimalProcessing(os.path.join(path, "OEBPS", "Images"), source) detectSuboptimalProcessing(os.path.join(path, "OEBPS", "Images"), source)
chapterNames, cover_path = sanitizeTree(os.path.join(path, 'OEBPS', 'Images')) chapterNames, cover_path = sanitizeTree(os.path.join(path, 'OEBPS', 'Images'))
cover = image.Cover(cover_path, options) cover = image.Cover(cover_path, options)
+32 -35
View File
@@ -20,7 +20,9 @@
# along with this program. If not, see <http://www.gnu.org/licenses/>. # along with this program. If not, see <http://www.gnu.org/licenses/>.
import io import io
import os import os
import numpy as np
from pathlib import Path from pathlib import Path
from functools import cached_property
import mozjpeg_lossless_optimization import mozjpeg_lossless_optimization
from PIL import Image, ImageOps, ImageStat, ImageChops, ImageFilter, ImageDraw from PIL import Image, ImageOps, ImageStat, ImageChops, ImageFilter, ImageDraw
from .page_number_crop_alg import get_bbox_crop_margin_page_number, get_bbox_crop_margin from .page_number_crop_alg import get_bbox_crop_margin_page_number, get_bbox_crop_margin
@@ -146,11 +148,9 @@ class ComicPageParser:
# Detect corruption in source image, let caller catch any exceptions triggered. # Detect corruption in source image, let caller catch any exceptions triggered.
srcImgPath = os.path.join(source[0], source[1]) srcImgPath = os.path.join(source[0], source[1])
Image.open(srcImgPath).verify()
self.image = Image.open(srcImgPath) self.image = Image.open(srcImgPath)
self.image.verify()
self.image = Image.open(srcImgPath).convert('RGB')
self.color = self.colorCheck()
self.fill = self.fillCheck() self.fill = self.fillCheck()
# backwards compatibility for Pillow >9.1.0 # backwards compatibility for Pillow >9.1.0
if not hasattr(Image, 'Resampling'): if not hasattr(Image, 'Resampling'):
@@ -181,13 +181,13 @@ class ComicPageParser:
new_image = Image.new("RGB", (int(width / 2), int(height*2))) new_image = Image.new("RGB", (int(width / 2), int(height*2)))
new_image.paste(pageone, (0, 0)) new_image.paste(pageone, (0, 0))
new_image.paste(pagetwo, (0, height)) new_image.paste(pagetwo, (0, height))
self.payload.append(['N', self.source, new_image, self.color, self.fill]) self.payload.append(['N', self.source, new_image, self.fill])
elif (width > height) != (dstwidth > dstheight) and width <= dstheight and height <= dstwidth \ elif (width > height) != (dstwidth > dstheight) and width <= dstheight and height <= dstwidth \
and not self.opt.webtoon and self.opt.splitter == 1: and not self.opt.webtoon and self.opt.splitter == 1:
spread = self.image spread = self.image
if not self.opt.norotate: if not self.opt.norotate:
spread = spread.rotate(90, Image.Resampling.BICUBIC, True) spread = spread.rotate(90, Image.Resampling.BICUBIC, True)
self.payload.append(['R', self.source, spread, self.color, self.fill]) self.payload.append(['R', self.source, spread, self.fill])
elif (width > height) != (dstwidth > dstheight) and not self.opt.webtoon: elif (width > height) != (dstwidth > dstheight) and not self.opt.webtoon:
if self.opt.splitter != 1: if self.opt.splitter != 1:
if width > height: if width > height:
@@ -202,38 +202,15 @@ class ComicPageParser:
else: else:
pageone = self.image.crop(leftbox) pageone = self.image.crop(leftbox)
pagetwo = self.image.crop(rightbox) pagetwo = self.image.crop(rightbox)
self.payload.append(['S1', self.source, pageone, self.color, self.fill]) self.payload.append(['S1', self.source, pageone, self.fill])
self.payload.append(['S2', self.source, pagetwo, self.color, self.fill]) self.payload.append(['S2', self.source, pagetwo, self.fill])
if self.opt.splitter > 0: if self.opt.splitter > 0:
spread = self.image spread = self.image
if not self.opt.norotate: if not self.opt.norotate:
spread = spread.rotate(90, Image.Resampling.BICUBIC, True) spread = spread.rotate(90, Image.Resampling.BICUBIC, True)
self.payload.append(['R', self.source, spread, self.payload.append(['R', self.source, spread, self.fill])
self.color, self.fill])
else: else:
self.payload.append(['N', self.source, self.image, self.color, self.fill]) self.payload.append(['N', self.source, self.image, self.fill])
def colorCheck(self):
if self.opt.webtoon:
return True
else:
img = self.image.copy()
bands = img.getbands()
if bands == ('R', 'G', 'B') or bands == ('R', 'G', 'B', 'A'):
thumb = img.resize((40, 40))
SSE, bias = 0, [0, 0, 0]
bias = ImageStat.Stat(thumb).mean[:3]
bias = [b - sum(bias) / 3 for b in bias]
for pixel in thumb.getdata():
mu = sum(pixel) / 3
SSE += sum((pixel[i] - mu - bias[i]) * (pixel[i] - mu - bias[i]) for i in [0, 1, 2])
MSE = float(SSE) / (40 * 40)
if MSE > 22:
return True
else:
return False
else:
return False
def fillCheck(self): def fillCheck(self):
if self.opt.bordersColor: if self.opt.bordersColor:
@@ -275,14 +252,14 @@ class ComicPageParser:
class ComicPage: class ComicPage:
def __init__(self, options, mode, path, image, color, fill): def __init__(self, options, mode, path, image, fill):
self.opt = options self.opt = options
_, self.size, self.palette, self.gamma = self.opt.profileData _, self.size, self.palette, self.gamma = self.opt.profileData
if self.opt.hq: if self.opt.hq:
self.size = (int(self.size[0] * 1.5), int(self.size[1] * 1.5)) self.size = (int(self.size[0] * 1.5), int(self.size[1] * 1.5))
self.kindle_scribe_azw3 = (options.profile == 'KS') and (options.format in ('MOBI', 'EPUB')) self.kindle_scribe_azw3 = (options.profile == 'KS') and (options.format in ('MOBI', 'EPUB'))
self.image = image self.original_color_mode = image.mode
self.color = color self.image = image.convert("RGB")
self.fill = fill self.fill = fill
self.rotated = False self.rotated = False
self.orgPath = os.path.join(path[0], path[1]) self.orgPath = os.path.join(path[0], path[1])
@@ -301,6 +278,26 @@ class ComicPage:
if not hasattr(Image, 'Resampling'): if not hasattr(Image, 'Resampling'):
Image.Resampling = Image Image.Resampling = Image
@cached_property
def color(self):
if self.original_color_mode in ("L", "1"):
return False
img = self.image.convert("YCbCr")
_, cb, cr = img.split()
cb_hist = cb.histogram()
cr_hist = cr.histogram()
cb_nonzero = [i for i, e in enumerate(cb_hist) if e]
cr_nonzero = [i for i, e in enumerate(cr_hist) if e]
cb_spread = cb_nonzero[-1] - cb_nonzero[0] if len(cb_nonzero) else 0
cr_spread = cr_nonzero[-1] - cr_nonzero[0] if len(cr_nonzero) else 0
SPREAD_THRESHOLD=20
if cb_spread < SPREAD_THRESHOLD and cr_spread < SPREAD_THRESHOLD:
return False
else:
return True
def saveToDir(self): def saveToDir(self):
try: try:
flags = [] flags = []
-6
View File
@@ -49,12 +49,6 @@ class HTMLStripper(HTMLParser):
def getImageFileName(imgfile): def getImageFileName(imgfile):
name, ext = os.path.splitext(imgfile) name, ext = os.path.splitext(imgfile)
ext = ext.lower() ext = ext.lower()
if (name.startswith('.') and len(name) == 1):
return None
if name.startswith('._'):
return None
if ext not in ['.png', '.jpg', '.jpeg', '.gif', '.webp', '.jp2', '.j2k', '.jpx']:
return None
return [name, ext] return [name, ext]
+1 -1
View File
@@ -5,7 +5,7 @@ requests>=2.31.0
python-slugify>=1.2.1 python-slugify>=1.2.1
raven>=6.0.0 raven>=6.0.0
packaging>=23.2 packaging>=23.2
mozjpeg-lossless-optimization==1.2.0 mozjpeg-lossless-optimization>=1.2.0
natsort>=8.4.0 natsort>=8.4.0
distro>=1.8.0 distro>=1.8.0
numpy>=1.22.4 numpy>=1.22.4