Kindle APNX: Add calculation using pagebreak tag.

This commit is contained in:
John Schember 2014-05-22 09:26:11 -04:00
parent e7387aab1a
commit c15e049684
2 changed files with 50 additions and 14 deletions

View File

@ -8,6 +8,7 @@ __docformat__ = 'restructuredtext en'
Generates and writes an APNX page mapping file. Generates and writes an APNX page mapping file.
''' '''
import re
import struct import struct
from calibre.ebooks.mobi.reader.mobi6 import MobiReader from calibre.ebooks.mobi.reader.mobi6 import MobiReader
@ -22,7 +23,7 @@ class APNXBuilder(object):
Create an APNX file using a pseudo page mapping. Create an APNX file using a pseudo page mapping.
''' '''
def write_apnx(self, mobi_file_path, apnx_path, accurate=True, page_count=0): def write_apnx(self, mobi_file_path, apnx_path, method=None, page_count=0):
''' '''
If you want a fixed number of pages (such as from a custom column) then If you want a fixed number of pages (such as from a custom column) then
pass in a value to page_count, otherwise a count will be estimated pass in a value to page_count, otherwise a count will be estimated
@ -60,15 +61,17 @@ class APNXBuilder(object):
if page_count: if page_count:
pages = self.get_pages_exact(mobi_file_path, page_count) pages = self.get_pages_exact(mobi_file_path, page_count)
else: else:
if accurate: try:
try: if method='accurate':
pages = self.get_pages_accurate(mobi_file_path) pages = self.get_pages_accurate(mobi_file_path)
except: elif method='pagebreak':
# Fall back to the fast parser if we can't pages = self.get_pages_pagebreak_tag(mobi_file_path)
# use the accurate one. Typically this is else:
# due to the file having DRM. raise('no valid accurate method chosen use fast')
pages = self.get_pages_fast(mobi_file_path) except:
else: # Fall back to the fast parser if we can't
# use the accurate one. Typically this is
# due to the file having DRM.
pages = self.get_pages_fast(mobi_file_path) pages = self.get_pages_fast(mobi_file_path)
if not pages: if not pages:
@ -261,3 +264,23 @@ class APNXBuilder(object):
pages.append(lines[i]) pages.append(lines[i])
return pages return pages
def get_pages_pagebreak_tag(self, mobi_file_path):
'''
Determine pages based on the presense of
<mbp:pagebreak>.
'''
pages = []
# Get the MOBI html.
mr = MobiReader(mobi_file_path, default_log)
if mr.book_header.encryption_type != 0:
# DRMed book
return self.get_pages_fast(mobi_file_path)
mr.extract_text()
html = mr.mobi_html.lower()
for m in re.finditer('<\s*(mbp:)?pagebreak[^>]*>'):
pages.append(m.end())
return pages

View File

@ -311,6 +311,14 @@ class KINDLE2(KINDLE):
'generator will produce pages that correspond better to a printed book. ' 'generator will produce pages that correspond better to a printed book. '
'However, this method is slower and will slow down sending files ' 'However, this method is slower and will slow down sending files '
'to the Kindle.'), 'to the Kindle.'),
_('Accurate calculation method') +
':::' +
_('There are multiple methods to accuratly calculate the page numbers. "accurate" which '
'is an estimation based on the number of chapters, paragraphs, and visible lines in the book. '
'This method is designed to simulate an average paperback book where there are 32 lines per '
'page and a maximum of 70 characters per line. \n\n'
'The "pagebreak" method uses the presense of <mbp:pagebreak> tags within the book to '
'determine pages.'),
_('Custom column name to retrieve page counts from') + _('Custom column name to retrieve page counts from') +
':::' + ':::' +
_('If you have a custom column in your library that you use to ' _('If you have a custom column in your library that you use to '
@ -322,13 +330,15 @@ class KINDLE2(KINDLE):
EXTRA_CUSTOMIZATION_DEFAULT = [ EXTRA_CUSTOMIZATION_DEFAULT = [
True, True,
False, False,
'accurate',
'', '',
] ]
OPT_APNX = 0 OPT_APNX = 0
OPT_APNX_ACCURATE = 1 OPT_APNX_ACCURATE = 1
OPT_APNX_CUST_COL = 2 OPT_APNX_ACCURATE_METHOD = 2
OPT_APNX_CUST_COL = 3
# x330 on the PaperWhite # x330 on the PaperWhite
THUMBNAIL_HEIGHT = 330 THUMBNAIL_HEIGHT = 330
# x262 on the Touch. Doesn't choke on x330, though. # x262 on the Touch. Doesn't choke on x330, though.
def formats_to_scan_for(self): def formats_to_scan_for(self):
@ -441,8 +451,11 @@ class KINDLE2(KINDLE):
apnx_path = '%s.apnx' % os.path.join(path, filename) apnx_path = '%s.apnx' % os.path.join(path, filename)
apnx_builder = APNXBuilder() apnx_builder = APNXBuilder()
try: try:
method = None
if opts.extra_customization[self.OPT_APNX_ACCURATE]:
method = opts.extra_customization[self.OPT_APNX_ACCURATE_METHOD]
apnx_builder.write_apnx(filepath, apnx_path, apnx_builder.write_apnx(filepath, apnx_path,
accurate=opts.extra_customization[self.OPT_APNX_ACCURATE], method=method,
page_count=custom_page_count) page_count=custom_page_count)
except: except:
print 'Failed to generate APNX' print 'Failed to generate APNX'