Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-06-23 15:30:45 -04:00
Code Issues Packages Projects Releases Wiki Activity

MOBI Input: Dont auto-convert images in PNG/GIF formats to JPEG

Browse Source
This commit is contained in:
Kovid Goyal 2020-03-10 13:37:21 +05:30
parent 0e6e5afe8d
commit 3df15e222a
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 28 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

37
src/calibre/ebooks/mobi/reader/mobi6.py
View File

@ -10,7 +10,7 @@ import shutil, os, re, struct, textwrap, io
from lxml import html, etree from lxml import html, etree
from calibre import (xml_entity_to_unicode, entity_to_unicode) from calibre import xml_entity_to_unicode, entity_to_unicode, guess_type
from calibre.utils.cleantext import clean_ascii_chars, clean_xml_chars from calibre.utils.cleantext import clean_ascii_chars, clean_xml_chars
from calibre.ebooks import DRMError, unit_convert from calibre.ebooks import DRMError, unit_convert
from calibre.ebooks.chardet import strip_encoding_declarations from calibre.ebooks.chardet import strip_encoding_declarations
@ -178,7 +178,7 @@ class MobiReader(object):
self.processed_html = strip_encoding_declarations(self.processed_html) self.processed_html = strip_encoding_declarations(self.processed_html)
self.processed_html = re.sub(r'&(\S+?);', xml_entity_to_unicode, self.processed_html = re.sub(r'&(\S+?);', xml_entity_to_unicode,
self.processed_html) self.processed_html)
self.extract_images(processed_records, output_dir) image_name_map = self.extract_images(processed_records, output_dir)
self.replace_page_breaks() self.replace_page_breaks()
self.cleanup_html() self.cleanup_html()
@ -272,7 +272,7 @@ class MobiReader(object):
head.insert(0, title) head.insert(0, title)
head.text = '\n\t' head.text = '\n\t'
self.upshift_markup(root) self.upshift_markup(root, image_name_map)
guides = root.xpath('//guide') guides = root.xpath('//guide')
guide = guides[0] if guides else None guide = guides[0] if guides else None
metadata_elems = root.xpath('//metadata') metadata_elems = root.xpath('//metadata')
@ -389,8 +389,9 @@ class MobiReader(object):
raw += unit raw += unit
return raw return raw
def upshift_markup(self, root): def upshift_markup(self, root, image_name_map=None):
self.log.debug('Converting style information to CSS...') self.log.debug('Converting style information to CSS...')
image_name_map = image_name_map or {}
size_map = { size_map = {
'xx-small': '0.5', 'xx-small': '0.5',
'x-small': '1', 'x-small': '1',
@ -510,10 +511,11 @@ class MobiReader(object):
recindex = attrib.pop(attr, None) or recindex recindex = attrib.pop(attr, None) or recindex
if recindex is not None: if recindex is not None:
try: try:
recindex = '%05d'%int(recindex) recindex = int(recindex)
except: except Exception:
pass pass
attrib['src'] = 'images/%s.jpg' % recindex else:
attrib['src'] = 'images/' + image_name_map.get(recindex, '%05d.jpg' % recindex)
for attr in ('width', 'height'): for attr in ('width', 'height'):
if attr in attrib: if attr in attrib:
val = attrib[attr] val = attrib[attr]
@ -674,7 +676,7 @@ class MobiReader(object):
for i in getattr(self, 'image_names', []): for i in getattr(self, 'image_names', []):
path = os.path.join(bp, 'images', i) path = os.path.join(bp, 'images', i)
added.add(path) added.add(path)
manifest.append((path, 'image/jpeg')) manifest.append((path, guess_type(path)[0] or 'image/jpeg'))
if cover_copied is not None: if cover_copied is not None:
manifest.append((cover_copied, 'image/jpeg')) manifest.append((cover_copied, 'image/jpeg'))
@ -870,6 +872,7 @@ class MobiReader(object):
os.makedirs(output_dir) os.makedirs(output_dir)
image_index = 0 image_index = 0
self.image_names = [] self.image_names = []
image_name_map = {}
start = getattr(self.book_header, 'first_image_index', -1) start = getattr(self.book_header, 'first_image_index', -1)
if start > self.num_sections or start < 0: if start > self.num_sections or start < 0:
# BAEN PRC files have bad headers # BAEN PRC files have bad headers
@ -882,18 +885,30 @@ class MobiReader(object):
image_index += 1 image_index += 1
if data[:4] in {b'FLIS', b'FCIS', b'SRCS', b'\xe9\x8e\r\n', if data[:4] in {b'FLIS', b'FCIS', b'SRCS', b'\xe9\x8e\r\n',
b'RESC', b'BOUN', b'FDST', b'DATP', b'AUDI', b'VIDE'}: b'RESC', b'BOUN', b'FDST', b'DATP', b'AUDI', b'VIDE'}:
# This record is a known non image type, not need to try to # This record is a known non image type, no need to try to
# load the image # load the image
continue continue
path = os.path.join(output_dir, '%05d.jpg' % image_index)
try: try:
if what(None, data) not in {'jpg', 'jpeg', 'gif', 'png', 'bmp'}: imgfmt = what(None, data)
except Exception:
continue continue
if imgfmt not in {'jpg', 'jpeg', 'gif', 'png', 'bmp'}:
continue
if imgfmt == 'jpeg':
imgfmt = 'jpg'
path = os.path.join(output_dir, '%05d.%s' % (image_index, imgfmt))
image_name_map[image_index] = os.path.basename(path)
if imgfmt in ('gif', 'png'):
with open(path, 'wb') as f:
f.write(data)
else:
try:
save_cover_data_to(data, path, minify_to=(10000, 10000)) save_cover_data_to(data, path, minify_to=(10000, 10000))
except Exception: except Exception:
continue continue
self.image_names.append(os.path.basename(path)) self.image_names.append(os.path.basename(path))
return image_name_map
def test_mbp_regex(): def test_mbp_regex():