Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-08 02:34:06 -04:00
Code Issues Packages Projects Releases Wiki Activity

Pull from driver-dev

Browse Source
This commit is contained in:
Kovid Goyal 2009-06-20 19:46:33 -07:00
commit 2d2fadcbd6
7 changed files with 86 additions and 56 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
src/calibre/devices/prs505/books.py
View File

@ -403,9 +403,8 @@ def fix_ids(main, carda, cardb):
for child in db.root_element.childNodes:
if child.nodeType == child.ELEMENT_NODE and child.hasAttribute('id'):
id_map[child.getAttribute('id')] = str(cid)
child.setAttribute('sourceid', '1')
#child.setAttribute("sourceid",
# '0' if getattr(child, 'tagName', '').endswith('playlist') else '1')
child.setAttribute("sourceid",
'0' if getattr(child, 'tagName', '').endswith('playlist') else '1')
child.setAttribute('id', str(cid))
cid += 1

17
src/calibre/devices/prs505/driver.py
View File

@ -180,16 +180,25 @@ class PRS505(CLI, Device):
return zip(paths, sizes, ctimes, cycle([on_card]))
@classmethod
def add_books_to_metadata(cls, locations, metadata, booklists):
def add_books_to_metadata(self, locations, metadata, booklists):
metadata = iter(metadata)
for location in locations:
info = metadata.next()
path = location[0]
blist = 2 if location[3] == 'cardb' else 1 if location[3] == 'carda' else 0
name = path.rpartition(os.sep)[2]
name = (cls.CARD_PATH_PREFIX+'/' if blist else 'database/media/books/') + name
if path.startswith(self._main_prefix):
name = path.replace(self._main_prefix, '')
elif path.startswith(self._card_a_prefix):
name = path.replace(self._card_a_prefix, '')
elif path.startswith(self._card_b_prefix):
name = path.replace(self._card_b_prefix, '')
name = name.replace('\\', '/')
name = name.replace('//', '/')
if name.startswith('/'):
name = name[1:]
booklists[blist].add_book(info, name, *location[1:-1])
fix_ids(*booklists)

14
src/calibre/ebooks/conversion/preprocess.py
View File

@ -1,12 +1,12 @@
#!/usr/bin/env python
# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:ai
from __future__ import with_statement
__license__ = 'GPL v3'
__copyright__ = '2009, Kovid Goyal <kovid@kovidgoyal.net>'
__docformat__ = 'restructuredtext en'
import re, functools
import functools
import re
from calibre import entity_to_unicode
@ -52,7 +52,12 @@ def line_length(raw, percent):
for line in lines:
if len(line) > 0:
lengths.append(len(line))
if not lengths:
return 0
total = sum(lengths)
print total
avg = total / len(lengths)
max_line = avg * 2
@ -178,9 +183,12 @@ class HTMLPreProcessor(object):
elif self.is_book_designer(html):
rules = self.BOOK_DESIGNER
elif self.is_pdftohtml(html):
length = line_length(html, .3)
line_length_rules = []
if length:
line_length_rules = [
# Un wrap using punctuation
(re.compile(r'(?<=.{%i}[a-z,;:-IA])\s*(?P<ital></(i|b|u)>)?\s*(<p.*?>)\s*(?=(<(i|b|u)>)?[\w\d])' % line_length(html, .3), re.UNICODE), wrap_lines),
(re.compile(r'(?<=.{%i}[a-z,;:-IA])\s*(?P<ital></(i|b|u)>)?\s*(<p.*?>)\s*(?=(<(i|b|u)>)?[\w\d])' % length, re.UNICODE), wrap_lines),
]
rules = self.PDFTOHTML + line_length_rules

4
src/calibre/ebooks/fb2/fb2ml.py
View File

@ -82,10 +82,6 @@ class FB2MLizer(object):
return images
def clean_text(self, text):
for entity in set(re.findall('&.+?;', text)):
mo = re.search('(%s)' % entity[1:-1], text)
text = text.replace(entity, entity_to_unicode(mo))
text = text.replace('&', '')
return text

28
src/calibre/ebooks/pdf/input.py
View File

@ -6,7 +6,7 @@ __docformat__ = 'restructuredtext en'
import os
from calibre.customize.conversion import InputFormatPlugin
from calibre.customize.conversion import InputFormatPlugin, OptionRecommendation
from calibre.ebooks.pdf.pdftohtml import pdftohtml
from calibre.ebooks.metadata.opf2 import OPFCreator
@ -17,17 +17,33 @@ class PDFInput(InputFormatPlugin):
description = 'Convert PDF files to HTML'
file_types = set(['pdf'])
options = set([
OptionRecommendation(name='no_images', recommended_value=False,
help=_('Do not extract images from the document')),
])
def convert(self, stream, options, file_ext, log,
accelerators):
html = pdftohtml(stream.name)
with open('index.html', 'wb') as index:
index.write(html)
# The main html file will be named index.html
pdftohtml(os.getcwd(), stream.name, options.no_images)
from calibre.ebooks.metadata.meta import get_metadata
mi = get_metadata(stream, 'pdf')
opf = OPFCreator(os.getcwd(), mi)
opf.create_manifest([('index.html', None)])
manifest = [('index.html', None)]
images = os.listdir(os.getcwd())
images.remove('index.html')
for i in images:
# Remove the - from the file name because it causes problems.
# The referenec to the image with the - will be changed to not
# include it later in the conversion process.
new_i = i.replace('-', '')
os.rename(i, new_i)
manifest.append((new_i, None))
opf.create_manifest(manifest)
opf.create_spine(['index.html'])
with open('metadata.opf', 'wb') as opffile:
opf.render(opffile)

26
src/calibre/ebooks/pdf/pdftohtml.py
View File

@ -14,7 +14,6 @@ from functools import partial
from calibre.ebooks import ConversionError, DRMError
from calibre import isosx, iswindows, islinux
from calibre import CurrentDir
from calibre.ptempfile import TemporaryDirectory
PDFTOHTML = 'pdftohtml'
popen = subprocess.Popen
@ -26,10 +25,11 @@ if iswindows and hasattr(sys, 'frozen'):
if islinux and getattr(sys, 'frozen_path', False):
PDFTOHTML = os.path.join(getattr(sys, 'frozen_path'), 'pdftohtml')
def pdftohtml(pdf_path):
def pdftohtml(output_dir, pdf_path, no_images):
'''
Convert the pdf into html using the pdftohtml app.
@return: The HTML as a unicode string.
This will write the html as index.html into output_dir.
It will also wirte all extracted images to the output_dir
'''
if isinstance(pdf_path, unicode):
@ -37,14 +37,14 @@ def pdftohtml(pdf_path):
if not os.access(pdf_path, os.R_OK):
raise ConversionError('Cannot read from ' + pdf_path)
with TemporaryDirectory('_pdftohtml') as tdir:
index = os.path.join(tdir, 'index.html')
with CurrentDir(output_dir):
index = os.path.join(os.getcwd(), 'index.html')
# This is neccessary as pdftohtml doesn't always (linux) respect absolute paths
pdf_path = os.path.abspath(pdf_path)
cmd = (PDFTOHTML, '-enc', 'UTF-8', '-noframes', '-p', '-nomerge', '-i', '-q', pdf_path, os.path.basename(index))
cwd = os.getcwd()
cmd = [PDFTOHTML, '-enc', 'UTF-8', '-noframes', '-p', '-nomerge', '-nodrm', '-q', pdf_path, os.path.basename(index)]
if no_images:
cmd.append('-i')
with CurrentDir(tdir):
try:
p = popen(cmd, stderr=subprocess.PIPE)
except OSError, err:
@ -69,9 +69,9 @@ def pdftohtml(pdf_path):
if not os.path.exists(index) or os.stat(index).st_size < 100:
raise DRMError()
with open(index, 'rb') as i:
with open(index, 'rb+wb') as i:
raw = i.read()
if not '<br' in raw[:4000]:
raise ConversionError(os.path.basename(pdf_path) + _(' is an image based PDF. Only conversion of text based PDFs is supported.'))
return '<!-- created by calibre\'s pdftohtml -->\n' + raw
raw = '<!-- created by calibre\'s pdftohtml -->\n' + raw
i.seek(0)
i.truncate()
i.write(raw)

2
src/calibre/gui2/widgets.py
View File

@ -227,6 +227,7 @@ class LocationModel(QAbstractListModel):
self.free[1] = fs[1] if fs[1] is not None and cpa is not None else -1
self.free[2] = fs[2] if fs[2] is not None and cpb is not None else -1
self.reset()
self.emit(SIGNAL('devicesChanged()'))
def location_changed(self, row):
self.highlight_row = row
@ -253,6 +254,7 @@ class LocationView(QListView):
self.connect(self, SIGNAL('entered(QModelIndex)'), self.item_entered)
self.connect(self, SIGNAL('viewportEntered()'), self.viewport_entered)
self.connect(self.eject_button, SIGNAL('clicked()'), lambda: self.emit(SIGNAL('umount_device()')))
self.connect(self.model(), SIGNAL('devicesChanged()'), self.eject_button.hide)
def count_changed(self, new_count):
self.model().count = new_count