Logo Search packages:      
Sourcecode: calibre version File versions  Download package


import os.path
# -*- coding: utf-8 -*-

__license__   = 'GPL v3'
__copyright__ = '2009, John Schember <john@nachtimwald.com>'
__docformat__ = 'restructuredtext en'

import glob
import os
import shutil

from calibre.customize.conversion import InputFormatPlugin
from calibre.ptempfile import TemporaryDirectory
from calibre.utils.zipfile import ZipFile
from calibre.ebooks.pml.pmlconverter import PML_HTMLizer
from calibre.ebooks.metadata.toc import TOC
from calibre.ebooks.metadata.opf2 import OPFCreator

class PMLInput(InputFormatPlugin):

    name        = 'PML Input'
    author      = 'John Schember'
    description = 'Convert PML to OEB'
    # pmlz is a zip file containing pml files and png images.
    file_types  = set(['pml', 'pmlz'])

    def process_pml(self, pml_path, html_path, close_all=False):
        pclose = False
        hclose = False

        if not hasattr(pml_path, 'read'):
            pml_stream = open(pml_path, 'rb')
            pclose = True
            pml_stream = pml_path

        if not hasattr(html_path, 'write'):
            html_stream = open(html_path, 'wb')
            hclose = True
            html_stream = html_path

        ienc = pml_stream.encoding if pml_stream.encoding else 'cp1252'
        if self.options.input_encoding:
            ienc = self.options.input_encoding

        self.log.debug('Converting PML to HTML...')
        hizer = PML_HTMLizer()
        html = hizer.parse_pml(pml_stream.read().decode(ienc), html_path)
        html_stream.write('<html><head><title /></head><body>%s</body></html>' % html.encode('utf-8', 'replace'))

        if pclose:
        if hclose:

        return hizer.get_toc()

    def get_images(self, stream, tdir, top_level=False):
        images = []
        imgs = []

        if top_level:
            imgs = glob.glob(os.path.join(tdir, '*.png'))
        # Images not in top level try bookname_img directory because
        # that's where Dropbook likes to see them.
        if not imgs:
            if hasattr(stream, 'name'):
                imgs = glob.glob(os.path.join(tdir, os.path.splitext(os.path.basename(stream.name))[0] + '_img', '*.png'))
        # No images in Dropbook location try generic images directory
        if not imgs:
            imgs = glob.glob(os.path.join(os.path.join(tdir, 'images'), '*.png'))
        if imgs:
            os.makedirs(os.path.join(os.getcwd(), 'images'))
        for img in imgs:
            pimg_name = os.path.basename(img)
            pimg_path = os.path.join(os.getcwd(), 'images', pimg_name)

            images.append('images/' + pimg_name)

            shutil.copy(img, pimg_path)

        return images

    def convert(self, stream, options, file_ext, log,
        self.options = options
        self.log = log
        pages, images = [], []
        toc = TOC()

        if file_ext == 'pmlz':
            log.debug('De-compressing content to temporary directory...')
            with TemporaryDirectory('_unpmlz') as tdir:
                zf = ZipFile(stream)

                pmls = glob.glob(os.path.join(tdir, '*.pml'))
                for pml in pmls:
                    html_name = os.path.splitext(os.path.basename(pml))[0]+'.html'
                    html_path = os.path.join(os.getcwd(), html_name)

                    log.debug('Processing PML item %s...' % pml)
                    ttoc = self.process_pml(pml, html_path)
                    toc += ttoc
                images = self.get_images(stream, tdir, True)
            toc = self.process_pml(stream, 'index.html')

            if hasattr(stream, 'name'):
                images = self.get_images(stream, os.path.abspath(os.path.dirname(stream.name)))

        # We want pages to be orded alphabetically.

        manifest_items = []
        for item in pages+images:
            manifest_items.append((item, None))

        from calibre.ebooks.metadata.meta import get_metadata
        log.debug('Reading metadata from input file...')
        mi = get_metadata(stream, 'pml')
        if 'images/cover.png' in images:
            mi.cover = 'images/cover.png'
        opf = OPFCreator(os.getcwd(), mi)
        log.debug('Generating manifest...')
        with open('metadata.opf', 'wb') as opffile:
            with open('toc.ncx', 'wb') as tocfile:
                opf.render(opffile, tocfile, 'toc.ncx')

        return os.path.join(os.getcwd(), 'metadata.opf')

Generated by  Doxygen 1.6.0   Back to index