Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

Run preprocess rules when viewing PDF files

Browse Source
This commit is contained in:
Kovid Goyal 2009-05-24 21:58:20 -07:00
parent 352b5d24ed
commit aa25d2a814
2 changed files with 13 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
src/calibre/ebooks/oeb/iterator.py
View File

@ -130,6 +130,9 @@ class EbookIterator(object):
plumber.opts.dont_package = True plumber.opts.dont_package = True
if hasattr(plumber.opts, 'no_process'): if hasattr(plumber.opts, 'no_process'):
plumber.opts.no_process = True plumber.opts.no_process = True
if hasattr(plumber.input_plugin, '_preprocess_html_for_viewer'):
plumber.input_plugin._preprocess_html_for_viewer = True
self.pathtoopf = plumber.input_plugin(open(plumber.input, 'rb'), self.pathtoopf = plumber.input_plugin(open(plumber.input, 'rb'),
plumber.opts, plumber.input_fmt, self.log, plumber.opts, plumber.input_fmt, self.log,
{}, self.base) {}, self.base)

14
src/calibre/ebooks/pdf/input.py
View File

@ -12,19 +12,25 @@ from calibre.ebooks.pdf.pdftohtml import pdftohtml
from calibre.ebooks.metadata.opf2 import OPFCreator from calibre.ebooks.metadata.opf2 import OPFCreator
class PDFInput(InputFormatPlugin): class PDFInput(InputFormatPlugin):
name = 'PDF Input' name = 'PDF Input'
author = 'John Schember' author = 'John Schember'
description = 'Convert PDF files to HTML' description = 'Convert PDF files to HTML'
file_types = set(['pdf']) file_types = set(['pdf'])
_preprocess_html_for_viewer = False
def convert(self, stream, options, file_ext, log, def convert(self, stream, options, file_ext, log,
accelerators): accelerators):
html = pdftohtml(stream.name) html = pdftohtml(stream.name)
if self._preprocess_html_for_viewer:
from calibre.ebooks.conversion.preprocess import HTMLPreProcessor
prepro = HTMLPreProcessor(lambda x:x, False)
html = prepro(html)
with open('index.html', 'wb') as index: with open('index.html', 'wb') as index:
index.write(html) index.write(html)
from calibre.ebooks.metadata.meta import get_metadata from calibre.ebooks.metadata.meta import get_metadata
mi = get_metadata(stream, 'pdf') mi = get_metadata(stream, 'pdf')
opf = OPFCreator(os.getcwd(), mi) opf = OPFCreator(os.getcwd(), mi)
@ -32,5 +38,5 @@ class PDFInput(InputFormatPlugin):
opf.create_spine(['index.html']) opf.create_spine(['index.html'])
with open('metadata.opf', 'wb') as opffile: with open('metadata.opf', 'wb') as opffile:
opf.render(opffile) opf.render(opffile)
return os.path.join(os.getcwd(), 'metadata.opf') return os.path.join(os.getcwd(), 'metadata.opf')