X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/48acd2e1b4ad0a64661091dff97122b2a53e4685..abbf124db7e7957dd12da2b050faa0fd020dc10d:/librarian/packagers.py?ds=inline diff --git a/librarian/packagers.py b/librarian/packagers.py index 36a7b60..f57a983 100644 --- a/librarian/packagers.py +++ b/librarian/packagers.py @@ -4,42 +4,43 @@ # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information. # import os -from copy import deepcopy -from lxml import etree -from librarian import pdf, epub, DirDocProvider, ParseError, cover +from librarian import pdf, epub, mobi, DirDocProvider, ParseError from librarian.parser import WLDocument +from util import makedirs + class Packager(object): cover = None flags = None @classmethod - def prepare_file(cls, main_input, output_dir, verbose=False): + def transform(cls, *args, **kwargs): + return cls.converter.transform(*args, **kwargs) + + @classmethod + def prepare_file(cls, main_input, output_dir, verbose=False, overwrite=False): path, fname = os.path.realpath(main_input).rsplit('/', 1) provider = DirDocProvider(path) slug, ext = os.path.splitext(fname) if output_dir != '': - try: - os.makedirs(output_dir) - except: - pass + makedirs(output_dir) outfile = os.path.join(output_dir, slug + '.' + cls.ext) + if os.path.exists(outfile) and not overwrite: + return doc = WLDocument.from_file(main_input, provider=provider) - output_file = cls.converter.transform(doc, - cover=cls.cover, flags=cls.flags) + output_file = cls.transform(doc, cover=cls.cover, flags=cls.flags) doc.save_output_file(output_file, output_path=outfile) - @classmethod - def prepare(cls, input_filenames, output_dir='', verbose=False): + def prepare(cls, input_filenames, output_dir='', verbose=False, overwrite=False): try: for main_input in input_filenames: if verbose: print main_input - cls.prepare_file(main_input, output_dir, verbose) + cls.prepare_file(main_input, output_dir, verbose, overwrite) except ParseError, e: print '%(file)s:%(name)s:%(message)s' % { 'file': main_input, @@ -52,98 +53,16 @@ class EpubPackager(Packager): converter = epub ext = 'epub' -class PdfPackager(Packager): - converter = pdf - ext = 'pdf' - -class GandalfEpubPackager(EpubPackager): - cover = cover.GandalfCover +class MobiPackager(Packager): + converter = mobi + ext = 'mobi' -class GandalfPdfPackager(PdfPackager): - cover = cover.GandalfCover -class BookotekaEpubPackager(EpubPackager): - cover = cover.BookotekaCover - -class PrestigioEpubPackager(EpubPackager): - cover = cover.PrestigioCover - flags = ('less-advertising',) - -class PrestigioPdfPackager(PdfPackager): - cover = cover.PrestigioCover - flags = ('less-advertising',) - - -class VirtualoEpubPackager(Packager): - @staticmethod - def utf_trunc(text, limit): - """ truncates text to at most `limit' bytes in utf-8 """ - if text is None: - return text - if len(text.encode('utf-8')) > limit: - newlimit = limit - 3 - while len(text.encode('utf-8')) > newlimit: - text = text[:(newlimit - len(text.encode('utf-8'))) / 4] - text += '...' - return text +class PdfPackager(Packager): + converter = pdf + ext = 'pdf' @classmethod - def prepare(cls, input_filenames, output_dir='', verbose=False): - xml = etree.fromstring(""" - """) - product = etree.fromstring(""" - - - - - - - Jan - Kowalski - - - 0.0 - PL - """) - - try: - for main_input in input_filenames: - if verbose: - print main_input - path, fname = os.path.realpath(main_input).rsplit('/', 1) - provider = DirDocProvider(path) - slug, ext = os.path.splitext(fname) - - outfile_dir = os.path.join(output_dir, slug) - os.makedirs(os.path.join(output_dir, slug)) - - doc = WLDocument.from_file(main_input, provider=provider) - info = doc.book_info - - product_elem = deepcopy(product) - product_elem[0].text = cls.utf_trunc(slug, 100) - product_elem[1].text = cls.utf_trunc(info.title, 255) - product_elem[2].text = cls.utf_trunc(info.description, 255) - product_elem[3].text = cls.utf_trunc(info.source_name, 3000) - product_elem[4][0][0].text = cls.utf_trunc(u' '.join(info.author.first_names), 100) - product_elem[4][0][1].text = cls.utf_trunc(info.author.last_name, 100) - xml.append(product_elem) - - cover.VirtualoCover(info).save(os.path.join(outfile_dir, slug+'.jpg')) - outfile = os.path.join(outfile_dir, '1.epub') - outfile_sample = os.path.join(outfile_dir, '1.sample.epub') - doc.save_output_file(epub.transform(doc), - output_path=outfile) - doc.save_output_file(epub.transform(doc, sample=25), - output_path=outfile_sample) - except ParseError, e: - print '%(file)s:%(name)s:%(message)s' % { - 'file': main_input, - 'name': e.__class__.__name__, - 'message': e.message - } - - xml_file = open(os.path.join(output_dir, 'import_products.xml'), 'w') - xml_file.write(etree.tostring(xml, pretty_print=True, encoding=unicode).encode('utf-8')) - xml_file.close() + def transform(cls, *args, **kwargs): + return cls.converter.transform(*args, morefloats='new', **kwargs)