from copy import deepcopy
from lxml import etree
import zipfile
-from tempfile import mkdtemp
+from tempfile import mkdtemp, NamedTemporaryFile
from shutil import rmtree
-import sys
-
-from librarian import XMLNamespace, RDFNS, DCNS, WLNS, NCXNS, OPFNS, XHTMLNS, NoDublinCore
-from librarian.dcparser import BookInfo
+from librarian import RDFNS, WLNS, NCXNS, OPFNS, XHTMLNS, OutputFile
+from librarian.cover import ImageCover as WLCover
from librarian import functions, get_resource
return text.replace(u"\ufeff", u"")\
.replace("---", u"\u2014")\
.replace("--", u"\u2013")\
- .replace(",,", u"\u201E")\
+ .replace(",,", u"“")\
.replace('"', u"\u201D")\
.replace("'", u"\u2019")
if node.tag in ('uwaga', 'extra'):
last_node_part = False
for one_part in main_text:
name = one_part.tag
- if name == 'naglowek_czesc':
- yield part_xml
- last_node_part = True
- main_xml_part[:] = [deepcopy(one_part)]
- elif not last_node_part and name in ("naglowek_rozdzial", "naglowek_akt", "srodtytul"):
- yield part_xml
- main_xml_part[:] = [deepcopy(one_part)]
- else:
+ #if name == 'naglowek_czesc':
+ # yield part_xml
+ # last_node_part = True
+ # main_xml_part[:] = [deepcopy(one_part)]
+ #elif not last_node_part and name in ("naglowek_rozdzial", "naglowek_akt", "srodtytul"):
+ # yield part_xml
+ # main_xml_part[:] = [deepcopy(one_part)]
+ #else:
+ if True:
main_xml_part.append(deepcopy(one_part))
last_node_part = False
yield part_xml
""" transforms one chunk, returns a HTML string, a TOC object and a set of used characters """
toc = TOC()
- for element in chunk_xml[0]:
- if element.tag in ("naglowek_czesc", "naglowek_rozdzial", "naglowek_akt", "srodtytul"):
- toc.add(node_name(element), "part%d.html" % chunk_no)
- elif element.tag in ('naglowek_podrozdzial', 'naglowek_scena'):
- subnumber = toc.add(node_name(element), "part%d.html" % chunk_no, level=1, is_part=False)
- element.set('sub', str(subnumber))
+ #for element in chunk_xml[0]:
+ # if element.tag in ("naglowek_czesc", "naglowek_rozdzial", "naglowek_akt", "srodtytul"):
+ # toc.add(node_name(element), "part%d.html" % chunk_no)
+ # elif element.tag in ('naglowek_podrozdzial', 'naglowek_scena'):
+ # subnumber = toc.add(node_name(element), "part%d.html" % chunk_no, level=1, is_part=False)
+ # element.set('sub', str(subnumber))
if empty:
if not _empty_html_static:
_empty_html_static.append(open(get_resource('epub/emptyChunk.html')).read())
chars = set()
output_html = _empty_html_static[0]
else:
- find_annotations(annotations, chunk_xml, chunk_no)
- replace_by_verse(chunk_xml)
- html_tree = xslt(chunk_xml, get_resource('epub/xsltScheme.xsl'))
+ if chunk_no == 1:
+ html_tree = xslt(chunk_xml, get_resource('epub/xsltScheme-FoC.xsl'))
+ else:
+ find_annotations(annotations, chunk_xml, chunk_no)
+ replace_by_verse(chunk_xml)
+ html_tree = xslt(chunk_xml, get_resource('epub/xsltScheme.xsl'))
chars = used_chars(html_tree.getroot())
output_html = etree.tostring(html_tree, method="html", pretty_print=True)
return output_html, toc, chars
-def transform(provider, slug=None, file_path=None, output_file=None, output_dir=None, make_dir=False, verbose=False,
+def transform(wldoc, verbose=False,
style=None, html_toc=False,
sample=None, cover=None, flags=None):
""" produces a EPUB file
- provider: a DocProvider
- slug: slug of file to process, available by provider
- output_file: file-like object or path to output file
- output_dir: path to directory to save output file to; either this or output_file must be present
- make_dir: writes output to <output_dir>/<author>/<slug>.epub instead of <output_dir>/<slug>.epub
sample=n: generate sample e-book (with at least n paragraphs)
- cover: a cover.Cover object
- flags: less-advertising, without-fonts
+ cover: a cover.Cover object or True for default
+ flags: less-advertising, without-fonts, working-copy
"""
- def transform_file(input_xml, chunk_counter=1, first=True, sample=None):
+ def transform_file(wldoc, chunk_counter=1, first=True, sample=None):
""" processes one input file and proceeds to its children """
- replace_characters(input_xml.getroot())
-
- children = [child.text for child in input_xml.findall('.//'+DCNS('relation.hasPart'))]
+ replace_characters(wldoc.edoc.getroot())
# every input file will have a TOC entry,
# pointing to starting chunk
- toc = TOC(node_name(input_xml.find('.//'+DCNS('title'))), "part%d.html" % chunk_counter)
+
+ # hack for FoC:
+ if wldoc.book_info.author is not None:
+ toc_title = "%s, %s" % (wldoc.book_info.author.readable(), wldoc.book_info.title)
+ note = wldoc.edoc.find('//dzielo_nadrzedne')
+ if note is not None:
+ toc_title += " (%s)" % note.text
+ else:
+ toc_title = wldoc.book_info.title
+ toc = TOC(toc_title, "part%d.html" % chunk_counter)
chars = set()
if first:
# write book title page
- html_tree = xslt(input_xml, get_resource('epub/xsltTitle.xsl'))
+ html_tree = xslt(wldoc.edoc, get_resource('epub/xsltTitle.xsl'))
chars = used_chars(html_tree.getroot())
zip.writestr('OPS/title.html',
etree.tostring(html_tree, method="html", pretty_print=True))
# add a title page TOC entry
- toc.add(u"Strona tytułowa", "title.html")
- elif children:
+ toc.add(u"Title page", "title.html")
+ toc.add(u"Dear readers!", "part1.html")
+ elif wldoc.book_info.parts:
# write title page for every parent
if sample is not None and sample <= 0:
chars = set()
html_string = open(get_resource('epub/emptyChunk.html')).read()
else:
- html_tree = xslt(input_xml, get_resource('epub/xsltChunkTitle.xsl'))
+ html_tree = xslt(wldoc.edoc, get_resource('epub/xsltChunkTitle.xsl'))
chars = used_chars(html_tree.getroot())
html_string = etree.tostring(html_tree, method="html", pretty_print=True)
zip.writestr('OPS/part%d.html' % chunk_counter, html_string)
add_to_spine(spine, chunk_counter)
chunk_counter += 1
- if len(input_xml.getroot()) > 1:
+ if len(wldoc.edoc.getroot()) > 1:
# rdf before style master
- main_text = input_xml.getroot()[1]
+ main_text = wldoc.edoc.getroot()[1]
else:
# rdf in style master
- main_text = input_xml.getroot()[0]
+ main_text = wldoc.edoc.getroot()[0]
if main_text.tag == RDFNS('RDF'):
main_text = None
add_to_spine(spine, chunk_counter)
chunk_counter += 1
- if children:
- for child in children:
- child_xml = etree.parse(provider.by_uri(child))
- child_toc, chunk_counter, chunk_chars, sample = transform_file(child_xml, chunk_counter, first=False, sample=sample)
- toc.append(child_toc)
- chars = chars.union(chunk_chars)
+ for child in wldoc.parts():
+ child_toc, chunk_counter, chunk_chars, sample = transform_file(
+ child, chunk_counter, first=False, sample=sample)
+ toc.append(child_toc)
+ chars = chars.union(chunk_chars)
return toc, chunk_counter, chars, sample
- # read metadata from the first file
- if file_path:
- if slug:
- raise ValueError('slug or file_path should be specified, not both')
- f = open(file_path, 'r')
- input_xml = etree.parse(f)
- f.close()
- else:
- if not slug:
- raise ValueError('either slug or file_path should be specified')
- input_xml = etree.parse(provider[slug])
+
+ document = deepcopy(wldoc)
+ del wldoc
if flags:
for flag in flags:
- input_xml.getroot().set(flag, 'yes')
-
- metadata = input_xml.find('.//'+RDFNS('Description'))
- if metadata is None:
- raise NoDublinCore('Document has no DublinCore - which is required.')
- book_info = BookInfo.from_element(input_xml)
- metadata = etree.ElementTree(metadata)
-
- # if output to dir, create the file
- if output_dir is not None:
- if make_dir:
- author = unicode(book_info.author)
- output_dir = os.path.join(output_dir, author)
- try:
- os.makedirs(output_dir)
- except OSError:
- pass
- if slug:
- output_file = open(os.path.join(output_dir, '%s.epub' % slug), 'w')
- else:
- output_file = open(os.path.join(output_dir, os.path.splitext(os.path.basename(file_path))[0] + '.epub'), 'w')
+ document.edoc.getroot().set(flag, 'yes')
+ opf = xslt(document.book_info.to_etree(), get_resource('epub/xsltContent.xsl'))
+ manifest = opf.find('.//' + OPFNS('manifest'))
+ guide = opf.find('.//' + OPFNS('guide'))
+ spine = opf.find('.//' + OPFNS('spine'))
+
+ output_file = NamedTemporaryFile(prefix='librarian', suffix='.epub', delete=False)
zip = zipfile.ZipFile(output_file, 'w', zipfile.ZIP_DEFLATED)
# write static elements
'<rootfiles><rootfile full-path="OPS/content.opf" ' \
'media-type="application/oebps-package+xml" />' \
'</rootfiles></container>')
- zip.write(get_resource('res/wl-logo-small.png'), os.path.join('OPS', 'logo_wolnelektury.png'))
- zip.write(get_resource('res/jedenprocent.png'), os.path.join('OPS', 'jedenprocent.png'))
+ #zip.write(get_resource('res/wl-logo-small.png'), os.path.join('OPS', 'logo_wolnelektury.png'))
+ #zip.write(get_resource('res/jedenprocent.png'), os.path.join('OPS', 'jedenprocent.png'))
+ zip.write('logo.png', os.path.join('OPS', 'logo.png'))
if not style:
style = get_resource('epub/style.css')
zip.write(style, os.path.join('OPS', 'style.css'))
- opf = xslt(metadata, get_resource('epub/xsltContent.xsl'))
- manifest = opf.find('.//' + OPFNS('manifest'))
- guide = opf.find('.//' + OPFNS('guide'))
- spine = opf.find('.//' + OPFNS('spine'))
-
if cover:
+ if cover is True:
+ cover = WLCover
+ if cover.uses_dc_cover:
+ if document.book_info.cover_by:
+ document.edoc.getroot().set('data-cover-by', document.book_info.cover_by)
+ if document.book_info.cover_source:
+ document.edoc.getroot().set('data-cover-source', document.book_info.cover_source)
+
cover_file = StringIO()
- c = cover(book_info.author.readable(), book_info.title)
+ c = cover(document.book_info)
+ import Image
+ c.im = Image.open('cover.jpg')
+ c.ext = lambda: 'jpg'
c.save(cover_file)
c_name = 'cover.%s' % c.ext()
zip.writestr(os.path.join('OPS', c_name), cover_file.getvalue())
'<item id="html_toc" href="toc.html" media-type="application/xhtml+xml" />'))
spine.append(etree.fromstring(
'<itemref idref="html_toc" />'))
- guide.append(etree.fromstring('<reference href="toc.html" type="toc" title="Spis treści"/>'))
+ guide.append(etree.fromstring('<reference href="toc.html" type="toc" title="Table of Contents"/>'))
- toc, chunk_counter, chars, sample = transform_file(input_xml, sample=sample)
+ toc, chunk_counter, chars, sample = transform_file(document, sample=sample)
if len(toc.children) < 2:
toc.add(u"Początek utworu", "part1.html")
zip.writestr('OPS/annotations.html', etree.tostring(
html_tree, method="html", pretty_print=True))
- toc.add("Strona redakcyjna", "last.html")
+ toc.add("Editorial page", "last.html")
manifest.append(etree.fromstring(
'<item id="last" href="last.html" media-type="application/xhtml+xml" />'))
spine.append(etree.fromstring(
'<itemref idref="last" />'))
- html_tree = xslt(input_xml, get_resource('epub/xsltLast.xsl'))
+ html_tree = xslt(document.edoc, get_resource('epub/xsltLast.xsl'))
chars.update(used_chars(html_tree.getroot()))
zip.writestr('OPS/last.html', etree.tostring(
html_tree, method="html", pretty_print=True))
os.chdir(cwd)
zip.writestr('OPS/content.opf', etree.tostring(opf, pretty_print=True))
- contents = []
- title = node_name(etree.ETXPath('.//'+DCNS('title'))(input_xml)[0])
+ title = document.book_info.title
attributes = "dtb:uid", "dtb:depth", "dtb:totalPageCount", "dtb:maxPageNumber"
for st in attributes:
meta = toc_file.makeelement(NCXNS('meta'))
# write TOC
if html_toc:
- toc.add(u"Spis treści", "toc.html", index=1)
+ toc.add(u"Table of Contents", "toc.html", index=1)
zip.writestr('OPS/toc.html', toc.html().encode('utf-8'))
toc.write_to_xml(nav_map)
zip.writestr('OPS/toc.ncx', etree.tostring(toc_file, pretty_print=True))
zip.close()
+
+ return OutputFile.from_filename(output_file.name)
fnp / librarian.git / blobdiff