mirrors/DeDRM_tools
mirrors/DeDRM_tools
1
0
Fork 0
mirror of https://github.com/noDRM/DeDRM_tools.git synced 2024-12-28 12:07:39 +06:00
Code Issues Projects Releases Wiki Activity

Fixes a bug that sometimes caused the plugin to fail

Browse Source
This commit is contained in:
NoDRM 2021-11-19 12:42:29 +01:00
parent a44b50d1d8
commit 2edde54c44
3 changed files with 79 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
.github/workflows/main.yml vendored
View File

@ -10,7 +10,7 @@ jobs:
- name: Checkout
uses: actions/checkout@v2
- name: Package
run: python3 make_release.py 10.0.0
run: python3 make_release.py 10.0.1
- name: Upload
uses: actions/upload-artifact@v2
with:

15
DeDRM_plugin/__init__.py
View File

@ -6,7 +6,7 @@
# Copyright © 2021 NoDRM
__license__ = 'GPL v3'
__version__ = '10.0.0'
__version__ = '10.0.1'
__docformat__ = 'restructuredtext en'
@ -79,6 +79,7 @@ __docformat__ = 'restructuredtext en'
# 7.2.0 - Update for latest KFX changes, and Python 3 Obok fixes.
# 7.2.1 - Whitespace!
# 10.0.0 - First forked version by NoDRM. See CHANGELOG.md for details.
# 10.0.1 - Fixes a bug in the watermark code.
"""
Decrypt DRMed ebooks.
@ -210,6 +211,8 @@ class DeDRM(FileTypePlugin):
# It does stuff like de-obfuscating fonts (by calling checkFonts)
# or removing watermarks.
postProcessStart = time.time()
try:
import calibre_plugins.dedrm.prefs as prefs
dedrmprefs = prefs.DeDRM_Prefs()
@ -224,13 +227,17 @@ class DeDRM(FileTypePlugin):
# Remove Tolino's CDP watermark file
path_to_ebook = watermark.removeCDPwatermark(self, path_to_ebook) or path_to_ebook
# Remove watermarks (currently just Amazon) from the OPF file
# Remove watermarks (Amazon or LemonInk) from the OPF file
path_to_ebook = watermark.removeOPFwatermarks(self, path_to_ebook) or path_to_ebook
# Remove watermarks (currently just Adobe's resource ID) from all HTML and XHTML files
# Remove watermarks (Adobe or LemonInk) from all HTML and XHTML files
path_to_ebook = watermark.removeHTMLwatermarks(self, path_to_ebook) or path_to_ebook
return path_to_ebook
postProcessEnd = time.time()
print("{0} v{1}: Post-processing took {2:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION, postProcessEnd-postProcessStart))
return path_to_ebook
except:
print("Error while checking settings")

73
DeDRM_plugin/epubwatermark.py
View File

@ -29,6 +29,12 @@ def removeHTMLwatermarks(object, path_to_ebook):
modded_names = []
modded_contents = []
count_adept = 0
count_lemonink_invisible = 0
count_lemonink_visible = 0
lemonink_trackingID = None
for file in namelist:
if not (file.endswith('.html') or file.endswith('.xhtml') or file.endswith('.xml')):
continue
@ -40,8 +46,33 @@ def removeHTMLwatermarks(object, path_to_ebook):
# Remove Adobe ADEPT watermarks
# Match optional newline at the beginning, then a "meta" tag with name = "Adept.expected.resource" or "Adept.resource"
# and either a "value" or a "content" element with an Adobe UUID
pre_remove = str_new
str_new = re.sub(r'((\r\n|\r|\n)\s*)?\<meta\s+name=\"(Adept\.resource|Adept\.expected\.resource)\"\s+(content|value)=\"urn:uuid:[0-9a-fA-F\-]+\"\s*\/>', '', str_new)
str_new = re.sub(r'((\r\n|\r|\n)\s*)?\<meta\s+(content|value)=\"urn:uuid:[0-9a-fA-F\-]+\"\s+name=\"(Adept\.resource|Adept\.expected\.resource)\"\s*\/>', '', str_new)
if (str_new != pre_remove):
count_adept += 1
# Remove eLibri / LemonInk watermark
# Run this in a loop, as it is possible a file has been watermarked twice ...
while True:
pre_remove = str_new
unique_id = re.search(r'<body[^>]+class="[^"]*(t0x[0-9a-fA-F]{25})[^"]*"[^>]*>', str_new)
if (unique_id):
lemonink_trackingID = unique_id.groups()[0]
count_lemonink_invisible += 1
str_new = re.sub(lemonink_trackingID, '', str_new)
pre_remove = str_new
pm = r'(<body[^>]+class="[^"]*"[^>]*>)'
pm += r'\<div style\=\'padding\:0\;border\:0\;text\-indent\:0\;line\-height\:normal\;margin\:0 1cm 0.5cm 1cm\;[^\']*text\-decoration\:none\;[^\']*background\:none\;[^\']*\'\>(.*?)</div>'
pm += r'\<div style\=\'padding\:0\;border\:0\;text\-indent\:0\;line\-height\:normal\;margin\:0 1cm 0.5cm 1cm\;[^\']*text\-decoration\:none\;[^\']*background\:none\;[^\']*\'\>(.*?)</div>'
str_new = re.sub(pm, r'\1', str_new)
if (str_new != pre_remove):
count_lemonink_visible += 1
else:
break
except:
traceback.print_exc()
continue
@ -51,6 +82,7 @@ def removeHTMLwatermarks(object, path_to_ebook):
modded_names.append(file)
modded_contents.append(str_new)
if len(modded_names) == 0:
# No file modified, return original
@ -58,7 +90,7 @@ def removeHTMLwatermarks(object, path_to_ebook):
if len(modded_names) != len(modded_contents):
# Something went terribly wrong, return original
print("Watermark: Error during ADEPT watermark removal")
print("Watermark: Error during watermark removal")
return path_to_ebook
# Re-package with modified files:
@ -105,12 +137,20 @@ def removeHTMLwatermarks(object, path_to_ebook):
traceback.print_exc()
return path_to_ebook
if (count_adept > 0):
print("Watermark: Successfully stripped {0} ADEPT watermark(s) from ebook.".format(count_adept))
if (count_lemonink_invisible > 0 or count_lemonink_visible > 0):
print("Watermark: Successfully stripped {0} visible and {1} invisible LemonInk watermark(s) (\"{2}\") from ebook."
.format(count_lemonink_visible, count_lemonink_invisible, lemonink_trackingID))
return output
except:
traceback.print_exc()
return path_to_ebook
print("Watermark: Successfully stripped {0} ADEPT watermark(s) from ebook.".format(len(modded_names)))
return output
# Finds the main OPF file, then uses RegEx to remove watermarks
@ -141,10 +181,27 @@ def removeOPFwatermarks(object, path_to_ebook):
container_str = inf.read(opf_path).decode("utf-8")
container_str_new = container_str
had_amazon = False
had_elibri = False
# Remove Amazon hex watermarks
# Match optional newline at the beginning, then spaces, then a "meta" tag with name = "Watermark" or "Watermark_(hex)" and a "content" element.
container_str_new = re.sub(r'((\r\n|\r|\n)\s*)?\<meta\s+name=\"Watermark(_\(hex\))?\"\s+content=\"[0-9a-fA-F]+\"\s*\/>', '', container_str_new)
container_str_new = re.sub(r'((\r\n|\r|\n)\s*)?\<meta\s+content=\"[0-9a-fA-F]+\"\s+name=\"Watermark(_\(hex\))?\"\s*\/>', '', container_str_new)
# This regex also matches DuMont watermarks with meta name="watermark", with the case-insensitive match on the "w" in watermark.
pre_remove = container_str_new
container_str_new = re.sub(r'((\r\n|\r|\n)\s*)?\<meta\s+name=\"[Ww]atermark(_\(hex\))?\"\s+content=\"[0-9a-fA-F]+\"\s*\/>', '', container_str_new)
container_str_new = re.sub(r'((\r\n|\r|\n)\s*)?\<meta\s+content=\"[0-9a-fA-F]+\"\s+name=\"[Ww]atermark(_\(hex\))?\"\s*\/>', '', container_str_new)
if pre_remove != container_str_new:
had_amazon = True
# Remove elibri / lemonink watermark
# Lemonink replaces all "id" fields in the opf with "idX_Y", with X being the watermark and Y being a number for that particular ID.
# This regex replaces all "idX_Y" IDs with "id_Y", removing the watermark IDs.
pre_remove = container_str_new
container_str_new = re.sub(r'((\r\n|\r|\n)\s*)?\<\!\-\-\s*Wygenerowane przez elibri dla zamówienia numer [0-9a-fA-F]+\s*\-\-\>', '', container_str_new)
container_str_new = re.sub(r'\=\"id[0-9]+_([0-9]+)\"', r'="id_\1"', container_str_new)
if pre_remove != container_str_new:
had_elibri = True
except:
traceback.print_exc()
return path_to_ebook
@ -191,7 +248,11 @@ def removeOPFwatermarks(object, path_to_ebook):
traceback.print_exc()
return path_to_ebook
print("Watermark: Successfully stripped Amazon watermark from OPF file.")
if had_elibri:
print("Watermark: Successfully stripped eLibri watermark from OPF file.")
if had_amazon:
print("Watermark: Successfully stripped Amazon watermark from OPF file.")
return output