def transform(wldoc, flags=None, **options):
"""
Transforms input_file in XML to output_file in TXT.
possible flags: raw-text,
"""
# Parse XSLT
style_filename = os.path.join(os.path.dirname(__file__), 'xslt/book2txt.xslt')
style = etree.parse(style_filename)
document = copy.deepcopy(wldoc)
del wldoc
document.swap_endlines()
if flags:
for flag in flags:
document.edoc.getroot().set(flag, 'yes')
result = document.transform(style, **options)
if not flags or 'raw-text' not in flags:
if document.book_info:
parsed_dc = document.book_info
description = parsed_dc.description
url = document.book_info.url
license_description = parsed_dc.license_description
license = parsed_dc.license
if license:
license_description = u"Ten utwór jest udostepniony na licencji %s: \n%s" % (license_description, license)
else:
license_description = u"Ten utwór nie jest chroniony prawem autorskim i znajduje się w domenie publicznej, co oznacza że możesz go swobodnie wykorzystywać, publikować i rozpowszechniać. Jeśli utwór opatrzony jest dodatkowymi materiałami (przypisy, motywy literackie etc.), które podlegają prawu autorskiemu, to te dodatkowe materiały udostępnione są na licencji Creative Commons Uznanie Autorstwa – Na Tych Samych Warunkach 3.0 PL (http://creativecommons.org/licenses/by-sa/3.0/)"
source = parsed_dc.source_name
if source:
source = "\n\nTekst opracowany na podstawie: " + source
else:
source = ''
contributors = ', '.join(person.readable() for person in
sorted(set(p for p in (parsed_dc.technical_editors + parsed_dc.editors) if p)))
if contributors:
contributors = "\n\nOpracowanie redakcyjne i przypisy: %s" % contributors
else:
description = 'Publikacja zrealizowana w ramach projektu Wolne Lektury (http://wolnelektury.pl).'
url = '*' * 10
license = ""
license_description = ""
source = ""
contributors = ""
return OutputFile.from_string((TEMPLATE % {
'description': description,
'url': url,
'license_description': license_description,
'text': unicode(result),
'source': source,
'contributors': contributors,
}).encode('utf-8'))
else:
return OutputFile.from_string(unicode(result).encode('utf-8'))
def transform(wldoc, verbose=False, cover=None, flags=None):
""" produces a FB2 file
cover: a cover.Cover object or True for default
flags: less-advertising, working-copy
"""
document = deepcopy(wldoc)
del wldoc
if flags:
for flag in flags:
document.edoc.getroot().set(flag, 'yes')
document.clean_ed_note()
document.clean_ed_note('abstrakt')
style_filename = os.path.join(os.path.dirname(__file__), 'fb2/fb2.xslt')
style = etree.parse(style_filename)
replace_by_verse(document.edoc)
sectionify(document.edoc)
result = document.transform(style)
return OutputFile.from_bytes(six.text_type(result).encode('utf-8'))
def transform(wldoc, verbose=False,
cover=None, flags=None):
""" produces a FB2 file
cover: a cover.Cover object or True for default
flags: less-advertising, working-copy
"""
document = deepcopy(wldoc)
del wldoc
if flags:
for flag in flags:
document.edoc.getroot().set(flag, 'yes')
document.clean_ed_note()
document.clean_ed_note('abstrakt')
style_filename = os.path.join(os.path.dirname(__file__), 'fb2/fb2.xslt')
style = etree.parse(style_filename)
replace_by_verse(document.edoc)
sectionify(document.edoc)
result = document.transform(style)
return OutputFile.from_string(unicode(result).encode('utf-8'))
def transform(wldoc,
verbose=False,
sample=None,
cover=None,
use_kindlegen=False,
flags=None,
hyphenate=True,
ilustr_path='',
converter_path=None):
""" produces a MOBI file
wldoc: a WLDocument
sample=n: generate sample e-book (with at least n paragraphs)
cover: a cover.Cover factory overriding default
flags: less-advertising,
converter_path: override path to MOBI converter,
either ebook-convert or kindlegen
"""
document = deepcopy(wldoc)
del wldoc
epub = document.as_epub(verbose=verbose,
sample=sample,
html_toc=True,
cover=cover or True,
flags=flags,
hyphenate=hyphenate,
ilustr_path=ilustr_path,
output_type='mobi')
if verbose:
kwargs = {}
else:
devnull = open("/dev/null", 'w')
kwargs = {"stdout": devnull, "stderr": devnull}
output_file = NamedTemporaryFile(prefix='librarian',
suffix='.mobi',
delete=False)
output_file.close()
if use_kindlegen:
output_file_basename = os.path.basename(output_file.name)
subprocess.check_call([
converter_path or 'kindlegen', '-c2',
epub.get_filename(), '-o', output_file_basename
], **kwargs)
else:
subprocess.check_call([
converter_path or 'ebook-convert',
epub.get_filename(), output_file.name, '--no-inline-toc',
'--mobi-file-type=both', '--mobi-ignore-margins'
], **kwargs)
return OutputFile.from_filename(output_file.name)
def transform(wldoc, stylesheet='legacy', options=None, flags=None):
"""Transforms the WL document to XHTML.
If output_filename is None, returns an XML,
otherwise returns True if file has been written,False if it hasn't.
File won't be written if it has no content.
"""
# Parse XSLT
try:
style_filename = get_stylesheet(stylesheet)
style = etree.parse(style_filename)
document = copy.deepcopy(wldoc)
del wldoc
document.swap_endlines()
if flags:
for flag in flags:
document.edoc.getroot().set(flag, 'yes')
document.clean_ed_note()
document.clean_ed_note('abstrakt')
if not options:
options = {}
options.setdefault('gallery', "''")
result = document.transform(style, **options)
del document # no longer needed large object :)
if html_has_content(result):
add_anchors(result.getroot())
add_table_of_themes(result.getroot())
add_table_of_contents(result.getroot())
return OutputFile.from_bytes(
etree.tostring(result,
method='html',
xml_declaration=False,
pretty_print=True,
encoding='utf-8'))
else:
return None
except KeyError:
raise ValueError("'%s' is not a valid stylesheet.")
except (XMLSyntaxError, XSLTApplyError) as e:
raise ParseError(e)
def transform(wldoc, verbose=False,
sample=None, cover=None, flags=None):
""" produces a MOBI file
wldoc: a WLDocument
sample=n: generate sample e-book (with at least n paragraphs)
cover: a cover.Cover factory overriding default
flags: less-advertising,
"""
document = deepcopy(wldoc)
del wldoc
book_info = document.book_info
# provide a cover by default
if not cover:
cover = DefaultEbookCover
cover_file = NamedTemporaryFile(suffix='.png', delete=False)
bound_cover = cover(book_info)
bound_cover.save(cover_file)
if bound_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)
if not flags:
flags = []
flags = list(flags) + ['with-full-fonts']
epub = document.as_epub(verbose=verbose, sample=sample, html_toc=True,
flags=flags, style=get_resource('epub/style.css'))
if verbose:
kwargs = {}
else:
devnull = open("/dev/null", 'w')
kwargs = {"stdout": devnull, "stderr": devnull}
output_file = NamedTemporaryFile(prefix='librarian', suffix='.mobi', delete=False)
output_file.close()
subprocess.check_call(['ebook-convert', epub.get_filename(), output_file.name,
'--no-inline-toc', '--mobi-file-type=both', '--subset-embedded-fonts', '--mobi-ignore-margins', '--cover=%s' % cover_file.name], **kwargs)
os.unlink(cover_file.name)
return OutputFile.from_filename(output_file.name)
def transform(wldoc, stylesheet='legacy', options=None, flags=None):
"""Transforms the WL document to XHTML.
If output_filename is None, returns an XML,
otherwise returns True if file has been written,False if it hasn't.
File won't be written if it has no content.
"""
# Parse XSLT
try:
style_filename = get_stylesheet(stylesheet)
style = etree.parse(style_filename)
document = copy.deepcopy(wldoc)
del wldoc
document.swap_endlines()
if flags:
for flag in flags:
document.edoc.getroot().set(flag, 'yes')
document.clean_ed_note()
document.clean_ed_note('abstrakt')
if not options:
options = {}
options.setdefault('gallery', "''")
result = document.transform(style, **options)
del document # no longer needed large object :)
if html_has_content(result):
add_anchors(result.getroot())
add_table_of_themes(result.getroot())
add_table_of_contents(result.getroot())
return OutputFile.from_string(etree.tostring(
result, method='html', xml_declaration=False, pretty_print=True, encoding='utf-8'))
else:
return None
except KeyError:
raise ValueError("'%s' is not a valid stylesheet.")
except (XMLSyntaxError, XSLTApplyError), e:
raise ParseError(e)
def transform(wldoc, verbose=False, cover=None, flags=None):
""" produces a FB2 file
cover: a cover.Cover object or True for default
flags: less-advertising, working-copy
"""
document = deepcopy(wldoc)
del wldoc
if flags:
for flag in flags:
document.edoc.getroot().set(flag, "yes")
style_filename = os.path.join(os.path.dirname(__file__), "fb2/fb2.xslt")
style = etree.parse(style_filename)
result = document.transform(style)
return OutputFile.from_string(unicode(result).encode("utf-8"))
def transform(wldoc, verbose=False, sample=None, cover=None, use_kindlegen=False, flags=None):
""" produces a MOBI file
wldoc: a WLDocument
sample=n: generate sample e-book (with at least n paragraphs)
cover: a cover.Cover factory overriding default
flags: less-advertising,
"""
document = deepcopy(wldoc)
del wldoc
epub = document.as_epub(verbose=verbose, sample=sample, html_toc=True, cover=cover or True, flags=flags)
if verbose:
kwargs = {}
else:
devnull = open("/dev/null", "w")
kwargs = {"stdout": devnull, "stderr": devnull}
output_file = NamedTemporaryFile(prefix="librarian", suffix=".mobi", delete=False)
output_file.close()
if use_kindlegen:
output_file_basename = os.path.basename(output_file.name)
subprocess.check_call(["kindlegen", "-c2", epub.get_filename(), "-o", output_file_basename], **kwargs)
else:
subprocess.check_call(
[
"ebook-convert",
epub.get_filename(),
output_file.name,
"--no-inline-toc",
"--mobi-file-type=both",
"--mobi-ignore-margins",
],
**kwargs
)
return OutputFile.from_filename(output_file.name)
def transform(wldoc, verbose=False, style=None, html_toc=False,
sample=None, cover=None, flags=None, hyphenate=False, ilustr_path='', output_type='epub'):
""" produces a EPUB file
sample=n: generate sample e-book (with at least n paragraphs)
cover: a cover.Cover factory or True for default
flags: less-advertising, without-fonts, working-copy
"""
def transform_file(wldoc, chunk_counter=1, first=True, sample=None):
""" processes one input file and proceeds to its children """
replace_characters(wldoc.edoc.getroot())
hyphenator = set_hyph_language(wldoc.edoc.getroot()) if hyphenate else None
hyphenate_and_fix_conjunctions(wldoc.edoc.getroot(), hyphenator)
# every input file will have a TOC entry,
# pointing to starting chunk
toc = TOC(wldoc.book_info.title, "part%d.html" % chunk_counter)
chars = set()
if first:
# write book title page
html_tree = xslt(wldoc.edoc, get_resource('epub/xsltTitle.xsl'), outputtype=output_type)
chars = used_chars(html_tree.getroot())
zip.writestr(
'OPS/title.html',
etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"' +
' "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
)
)
# add a title page TOC entry
toc.add(u"Strona tytułowa", "title.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(wldoc.edoc, get_resource('epub/xsltChunkTitle.xsl'))
chars = used_chars(html_tree.getroot())
html_string = etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"' +
' "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
)
zip.writestr('OPS/part%d.html' % chunk_counter, html_string)
add_to_manifest(manifest, chunk_counter)
add_to_spine(spine, chunk_counter)
chunk_counter += 1
if len(wldoc.edoc.getroot()) > 1:
# rdf before style master
main_text = wldoc.edoc.getroot()[1]
else:
# rdf in style master
main_text = wldoc.edoc.getroot()[0]
if main_text.tag == RDFNS('RDF'):
main_text = None
if main_text is not None:
for chunk_xml in chop(main_text):
empty = False
if sample is not None:
if sample <= 0:
empty = True
else:
sample -= len(chunk_xml.xpath('//strofa|//akap|//akap_cd|//akap_dialog'))
chunk_html, chunk_toc, chunk_chars = transform_chunk(chunk_xml, chunk_counter, annotations, empty)
toc.extend(chunk_toc)
chars = chars.union(chunk_chars)
zip.writestr('OPS/part%d.html' % chunk_counter, chunk_html)
add_to_manifest(manifest, chunk_counter)
add_to_spine(spine, chunk_counter)
chunk_counter += 1
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
document = deepcopy(wldoc)
del wldoc
if flags:
for flag in flags:
document.edoc.getroot().set(flag, 'yes')
document.clean_ed_note()
document.clean_ed_note('abstrakt')
# add editors info
editors = document.editors()
if editors:
document.edoc.getroot().set('editors', u', '.join(sorted(
editor.readable() for editor in editors)))
if document.book_info.funders:
document.edoc.getroot().set('funders', u', '.join(
document.book_info.funders))
if document.book_info.thanks:
document.edoc.getroot().set('thanks', document.book_info.thanks)
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)
functions.reg_mathml_epub(zip)
if os.path.isdir(ilustr_path):
for i, filename in enumerate(os.listdir(ilustr_path)):
file_path = os.path.join(ilustr_path, filename)
zip.write(file_path, os.path.join('OPS', filename))
image_id = 'image%s' % i
manifest.append(etree.fromstring(
'<item id="%s" href="%s" media-type="%s" />' % (image_id, filename, guess_type(file_path)[0])))
# write static elements
mime = zipfile.ZipInfo()
mime.filename = 'mimetype'
mime.compress_type = zipfile.ZIP_STORED
mime.extra = ''
zip.writestr(mime, 'application/epub+zip')
zip.writestr(
'META-INF/container.xml',
'<?xml version="1.0" ?>'
'<container version="1.0" '
'xmlns="urn:oasis:names:tc:opendocument:xmlns:container">'
'<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'))
if not style:
style = get_resource('epub/style.css')
zip.write(style, os.path.join('OPS', 'style.css'))
if cover:
if cover is True:
cover = DefaultEbookCover
cover_file = StringIO()
bound_cover = cover(document.book_info)
bound_cover.save(cover_file)
cover_name = 'cover.%s' % bound_cover.ext()
zip.writestr(os.path.join('OPS', cover_name), cover_file.getvalue())
del cover_file
cover_tree = etree.parse(get_resource('epub/cover.html'))
cover_tree.find('//' + XHTMLNS('img')).set('src', cover_name)
zip.writestr('OPS/cover.html', etree.tostring(
cover_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" ' +
'"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
))
if bound_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)
manifest.append(etree.fromstring(
'<item id="cover" href="cover.html" media-type="application/xhtml+xml" />'))
manifest.append(etree.fromstring(
'<item id="cover-image" href="%s" media-type="%s" />' % (cover_name, bound_cover.mime_type())))
spine.insert(0, etree.fromstring('<itemref idref="cover"/>'))
opf.getroot()[0].append(etree.fromstring('<meta name="cover" content="cover-image"/>'))
guide.append(etree.fromstring('<reference href="cover.html" type="cover" title="Okładka"/>'))
annotations = etree.Element('annotations')
toc_file = etree.fromstring(
'<?xml version="1.0" encoding="utf-8"?><!DOCTYPE ncx PUBLIC '
'"-//NISO//DTD ncx 2005-1//EN" '
'"http://www.daisy.org/z3986/2005/ncx-2005-1.dtd">'
'<ncx xmlns="http://www.daisy.org/z3986/2005/ncx/" xml:lang="pl" '
'version="2005-1"><head></head><docTitle></docTitle><navMap>'
'</navMap></ncx>'
)
nav_map = toc_file[-1]
if html_toc:
manifest.append(etree.fromstring(
'<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"/>'))
toc, chunk_counter, chars, sample = transform_file(document, sample=sample)
if len(toc.children) < 2:
toc.add(u"Początek utworu", "part1.html")
# Last modifications in container files and EPUB creation
if len(annotations) > 0:
toc.add("Przypisy", "annotations.html")
manifest.append(etree.fromstring(
'<item id="annotations" href="annotations.html" media-type="application/xhtml+xml" />'))
spine.append(etree.fromstring(
'<itemref idref="annotations" />'))
replace_by_verse(annotations)
html_tree = xslt(annotations, get_resource('epub/xsltAnnotations.xsl'))
chars = chars.union(used_chars(html_tree.getroot()))
zip.writestr('OPS/annotations.html', etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" ' +
'"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
))
toc.add("Wesprzyj Wolne Lektury", "support.html")
manifest.append(etree.fromstring(
'<item id="support" href="support.html" media-type="application/xhtml+xml" />'))
spine.append(etree.fromstring(
'<itemref idref="support" />'))
html_string = open(get_resource('epub/support.html')).read()
chars.update(used_chars(etree.fromstring(html_string)))
zip.writestr('OPS/support.html', html_string)
toc.add("Strona redakcyjna", "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(document.edoc, get_resource('epub/xsltLast.xsl'), outputtype=output_type)
chars.update(used_chars(html_tree.getroot()))
zip.writestr('OPS/last.html', etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" ' +
'"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
))
if not flags or 'without-fonts' not in flags:
# strip fonts
tmpdir = mkdtemp('-librarian-epub')
try:
cwd = os.getcwd()
except OSError:
cwd = None
os.chdir(os.path.join(os.path.dirname(os.path.realpath(__file__)), 'font-optimizer'))
for fname in 'DejaVuSerif.ttf', 'DejaVuSerif-Bold.ttf', 'DejaVuSerif-Italic.ttf', 'DejaVuSerif-BoldItalic.ttf':
optimizer_call = ['perl', 'subset.pl', '--chars',
''.join(chars).encode('utf-8'),
get_resource('fonts/' + fname),
os.path.join(tmpdir, fname)]
if verbose:
print "Running font-optimizer"
subprocess.check_call(optimizer_call)
else:
dev_null = open(os.devnull, 'w')
subprocess.check_call(optimizer_call, stdout=dev_null, stderr=dev_null)
zip.write(os.path.join(tmpdir, fname), os.path.join('OPS', fname))
manifest.append(etree.fromstring(
'<item id="%s" href="%s" media-type="application/x-font-truetype" />' % (fname, fname)))
rmtree(tmpdir)
if cwd is not None:
os.chdir(cwd)
zip.writestr('OPS/content.opf', etree.tostring(opf, pretty_print=True,
xml_declaration=True, encoding="utf-8"))
title = document.book_info.title
attributes = "dtb:uid", "dtb:depth", "dtb:totalPageCount", "dtb:maxPageNumber"
for st in attributes:
meta = toc_file.makeelement(NCXNS('meta'))
meta.set('name', st)
meta.set('content', '0')
toc_file[0].append(meta)
toc_file[0][0].set('content', str(document.book_info.url))
toc_file[0][1].set('content', str(toc.depth()))
set_inner_xml(toc_file[1], ''.join(('<text>', title, '</text>')))
# write TOC
if html_toc:
toc.add(u"Spis treści", "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,
xml_declaration=True, encoding="utf-8"))
zip.close()
return OutputFile.from_filename(output_file.name)
def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
cover=None, flags=None, customizations=None):
""" produces a PDF file with XeLaTeX
wldoc: a WLDocument
verbose: prints all output from LaTeX
save_tex: path to save the intermediary LaTeX file to
morefloats (old/new/none): force specific morefloats
cover: a cover.Cover factory or True for default
flags: less-advertising,
customizations: user requested customizations regarding various formatting parameters (passed to wl LaTeX class)
"""
# Parse XSLT
try:
book_info = wldoc.book_info
document = load_including_children(wldoc)
root = document.edoc.getroot()
if cover:
if cover is True:
cover = DefaultEbookCover
bound_cover = cover(book_info, width=1200)
root.set('data-cover-width', str(bound_cover.width))
root.set('data-cover-height', str(bound_cover.height))
if bound_cover.uses_dc_cover:
if book_info.cover_by:
root.set('data-cover-by', book_info.cover_by)
if book_info.cover_source:
root.set('data-cover-source',
book_info.cover_source)
if flags:
for flag in flags:
root.set('flag-' + flag, 'yes')
# check for LaTeX packages
if morefloats:
root.set('morefloats', morefloats.lower())
elif package_available('morefloats', 'maxfloats=19'):
root.set('morefloats', 'new')
# add customizations
if customizations is not None:
root.set('customizations', u','.join(customizations))
# add editors info
editors = document.editors()
if editors:
root.set('editors', u', '.join(sorted(
editor.readable() for editor in editors)))
if document.book_info.funders:
root.set('funders', u', '.join(document.book_info.funders))
if document.book_info.thanks:
root.set('thanks', document.book_info.thanks)
# hack the tree
move_motifs_inside(document.edoc)
hack_motifs(document.edoc)
parse_creator(document.edoc)
substitute_hyphens(document.edoc)
fix_hanging(document.edoc)
fix_tables(document.edoc)
# wl -> TeXML
style_filename = get_stylesheet("wl2tex")
style = etree.parse(style_filename)
functions.reg_mathml_latex()
# TeXML -> LaTeX
temp = mkdtemp('-wl2pdf')
for sponsor in book_info.sponsors:
ins = etree.Element("data-sponsor", name=sponsor)
logo = sponsor_logo(sponsor)
if logo:
fname = 'sponsor-%s' % os.path.basename(logo)
shutil.copy(logo, os.path.join(temp, fname))
ins.set('src', fname)
root.insert(0, ins)
if book_info.sponsor_note:
root.set("sponsor-note", book_info.sponsor_note)
texml = document.transform(style)
if cover:
with open(os.path.join(temp, 'cover.png'), 'w') as f:
bound_cover.save(f, quality=80)
del document # no longer needed large object :)
tex_path = os.path.join(temp, 'doc.tex')
fout = open(tex_path, 'w')
process(StringIO(texml), fout, 'utf-8')
fout.close()
del texml
if save_tex:
shutil.copy(tex_path, save_tex)
# LaTeX -> PDF
shutil.copy(get_resource('pdf/wl.cls'), temp)
shutil.copy(get_resource('res/wl-logo.png'), temp)
try:
cwd = os.getcwd()
except OSError:
cwd = None
os.chdir(temp)
if verbose:
p = call(['xelatex', tex_path])
else:
p = call(['xelatex', '-interaction=batchmode', tex_path], stdout=PIPE, stderr=PIPE)
if p:
raise ParseError("Error parsing .tex file")
if cwd is not None:
os.chdir(cwd)
output_file = NamedTemporaryFile(prefix='librarian', suffix='.pdf', delete=False)
pdf_path = os.path.join(temp, 'doc.pdf')
shutil.move(pdf_path, output_file.name)
shutil.rmtree(temp)
return OutputFile.from_filename(output_file.name)
except (XMLSyntaxError, XSLTApplyError), e:
raise ParseError(e)
def output_file(self, *args, **kwargs):
imgstr = BytesIO()
self.save(imgstr, *args, **kwargs)
return OutputFile.from_bytes(imgstr.getvalue())
def output_file(self, *args, **kwargs):
imgstr = StringIO()
self.save(imgstr, *args, **kwargs)
return OutputFile.from_string(imgstr.getvalue())
def output_file(self, *args, **kwargs):
imgstr = BytesIO()
self.save(imgstr, *args, **kwargs)
return OutputFile.from_bytes(imgstr.getvalue())
def transform(wldoc, flags=None, **options):
"""
Transforms input_file in XML to output_file in TXT.
possible flags: raw-text,
"""
# Parse XSLT
style_filename = os.path.join(os.path.dirname(__file__), 'xslt/book2txt.xslt')
style = etree.parse(style_filename)
document = copy.deepcopy(wldoc)
del wldoc
document.swap_endlines()
if flags:
for flag in flags:
document.edoc.getroot().set(flag, 'yes')
if 'wrapping' in options:
options['wrapping'] = str(options['wrapping'])
result = document.transform(style, **options)
if not flags or 'raw-text' not in flags:
if document.book_info:
parsed_dc = document.book_info
description = parsed_dc.description
url = document.book_info.url
license_description = parsed_dc.license_description
license = parsed_dc.license
if license:
license_description = u"Ten utwór jest udostepniony na licencji %s: \n%s" % (
license_description, license)
else:
license_description = u"Ten utwór nie jest objęty majątkowym prawem autorskim i znajduje się " \
u"w domenie publicznej, co oznacza że możesz go swobodnie wykorzystywać, " \
u"publikować i rozpowszechniać. Jeśli utwór opatrzony jest dodatkowymi " \
u"materiałami (przypisy, motywy literackie etc.), które podlegają prawu " \
u"autorskiemu, to te dodatkowe materiały udostępnione są na licencji " \
u"Creative Commons Uznanie Autorstwa – Na Tych Samych Warunkach 3.0 PL " \
u"(http://creativecommons.org/licenses/by-sa/3.0/)"
source = parsed_dc.source_name
if source:
source = "\n\nTekst opracowany na podstawie: " + source
else:
source = ''
contributors = ', '.join(person.readable() for person in
sorted(set(p for p in (parsed_dc.technical_editors + parsed_dc.editors) if p)))
if contributors:
contributors = "\n\nOpracowanie redakcyjne i przypisy: %s." % contributors
funders = ', '.join(parsed_dc.funders)
if funders:
funders = u"\n\nPublikację wsparli i wsparły: %s." % funders
publisher = '\n\nWydawca: ' + ', '.join(parsed_dc.publisher)
isbn = getattr(parsed_dc, 'isbn_txt', None)
if isbn:
isbn = '\n\n' + isbn
else:
isbn = ''
else:
description = 'Publikacja zrealizowana w ramach projektu Wolne Lektury (http://wolnelektury.pl).'
url = '*' * 10
license_description = ""
source = ""
contributors = ""
funders = ""
publisher = ""
isbn = ""
result = (TEMPLATE % {
'description': description,
'url': url,
'license_description': license_description,
'text': unicode(result),
'source': source,
'contributors': contributors,
'funders': funders,
'publisher': publisher,
'isbn': isbn,
}).encode('utf-8')
else:
result = unicode(result).encode('utf-8')
return OutputFile.from_string("\r\n".join(result.splitlines()) + "\r\n")
def transform(wldoc, verbose=False,
style=None, html_toc=False,
sample=None, cover=None, flags=None):
""" produces a EPUB file
sample=n: generate sample e-book (with at least n paragraphs)
cover: a cover.Cover factory or True for default
flags: less-advertising, without-fonts, working-copy
"""
def transform_file(wldoc, chunk_counter=1, first=True, sample=None):
""" processes one input file and proceeds to its children """
replace_characters(wldoc.edoc.getroot())
hyphenator = set_hyph_language(wldoc.edoc.getroot())
hyphenate_and_fix_conjunctions(wldoc.edoc.getroot(), hyphenator)
# every input file will have a TOC entry,
# pointing to starting chunk
toc = TOC(wldoc.book_info.title, "part%d.html" % chunk_counter)
chars = set()
if first:
# write book title page
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, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"' +
' "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
)
)
# add a title page TOC entry
toc.add(u"Strona tytułowa", "title.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(wldoc.edoc, get_resource('epub/xsltChunkTitle.xsl'))
chars = used_chars(html_tree.getroot())
html_string = etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"' +
' "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
)
zip.writestr('OPS/part%d.html' % chunk_counter, html_string)
add_to_manifest(manifest, chunk_counter)
add_to_spine(spine, chunk_counter)
chunk_counter += 1
if len(wldoc.edoc.getroot()) > 1:
# rdf before style master
main_text = wldoc.edoc.getroot()[1]
else:
# rdf in style master
main_text = wldoc.edoc.getroot()[0]
if main_text.tag == RDFNS('RDF'):
main_text = None
if main_text is not None:
for chunk_xml in chop(main_text):
empty = False
if sample is not None:
if sample <= 0:
empty = True
else:
sample -= len(chunk_xml.xpath('//strofa|//akap|//akap_cd|//akap_dialog'))
chunk_html, chunk_toc, chunk_chars = transform_chunk(chunk_xml, chunk_counter, annotations, empty)
toc.extend(chunk_toc)
chars = chars.union(chunk_chars)
zip.writestr('OPS/part%d.html' % chunk_counter, chunk_html)
add_to_manifest(manifest, chunk_counter)
add_to_spine(spine, chunk_counter)
chunk_counter += 1
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
document = deepcopy(wldoc)
del wldoc
if flags:
for flag in flags:
document.edoc.getroot().set(flag, 'yes')
# add editors info
editors = document.editors()
if editors:
document.edoc.getroot().set('editors', u', '.join(sorted(
editor.readable() for editor in editors)))
if document.book_info.funders:
document.edoc.getroot().set('funders', u', '.join(
document.book_info.funders))
if document.book_info.thanks:
document.edoc.getroot().set('thanks', document.book_info.thanks)
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
mime = zipfile.ZipInfo()
mime.filename = 'mimetype'
mime.compress_type = zipfile.ZIP_STORED
mime.extra = ''
zip.writestr(mime, 'application/epub+zip')
zip.writestr(
'META-INF/container.xml',
'<?xml version="1.0" ?>'
'<container version="1.0" '
'xmlns="urn:oasis:names:tc:opendocument:xmlns:container">'
'<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'))
if not style:
style = get_resource('epub/style.css')
zip.write(style, os.path.join('OPS', 'style.css'))
if cover:
if cover is True:
cover = DefaultEbookCover
cover_file = StringIO()
bound_cover = cover(document.book_info)
bound_cover.save(cover_file)
cover_name = 'cover.%s' % bound_cover.ext()
zip.writestr(os.path.join('OPS', cover_name), cover_file.getvalue())
del cover_file
cover_tree = etree.parse(get_resource('epub/cover.html'))
cover_tree.find('//' + XHTMLNS('img')).set('src', cover_name)
zip.writestr('OPS/cover.html', etree.tostring(
cover_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" ' +
'"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
))
if bound_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)
manifest.append(etree.fromstring(
'<item id="cover" href="cover.html" media-type="application/xhtml+xml" />'))
manifest.append(etree.fromstring(
'<item id="cover-image" href="%s" media-type="%s" />' % (cover_name, bound_cover.mime_type())))
spine.insert(0, etree.fromstring('<itemref idref="cover"/>'))
opf.getroot()[0].append(etree.fromstring('<meta name="cover" content="cover-image"/>'))
guide.append(etree.fromstring('<reference href="cover.html" type="cover" title="Okładka"/>'))
annotations = etree.Element('annotations')
toc_file = etree.fromstring(
'<?xml version="1.0" encoding="utf-8"?><!DOCTYPE ncx PUBLIC '
'"-//NISO//DTD ncx 2005-1//EN" '
'"http://www.daisy.org/z3986/2005/ncx-2005-1.dtd">'
'<ncx xmlns="http://www.daisy.org/z3986/2005/ncx/" xml:lang="pl" '
'version="2005-1"><head></head><docTitle></docTitle><navMap>'
'</navMap></ncx>'
)
nav_map = toc_file[-1]
if html_toc:
manifest.append(etree.fromstring(
'<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"/>'))
toc, chunk_counter, chars, sample = transform_file(document, sample=sample)
if len(toc.children) < 2:
toc.add(u"Początek utworu", "part1.html")
# Last modifications in container files and EPUB creation
if len(annotations) > 0:
toc.add("Przypisy", "annotations.html")
manifest.append(etree.fromstring(
'<item id="annotations" href="annotations.html" media-type="application/xhtml+xml" />'))
spine.append(etree.fromstring(
'<itemref idref="annotations" />'))
replace_by_verse(annotations)
html_tree = xslt(annotations, get_resource('epub/xsltAnnotations.xsl'))
chars = chars.union(used_chars(html_tree.getroot()))
zip.writestr('OPS/annotations.html', etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" ' +
'"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
))
toc.add("Wesprzyj Wolne Lektury", "support.html")
manifest.append(etree.fromstring(
'<item id="support" href="support.html" media-type="application/xhtml+xml" />'))
spine.append(etree.fromstring(
'<itemref idref="support" />'))
html_string = open(get_resource('epub/support.html')).read()
chars.update(used_chars(etree.fromstring(html_string)))
zip.writestr('OPS/support.html', html_string)
toc.add("Strona redakcyjna", "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(document.edoc, get_resource('epub/xsltLast.xsl'))
chars.update(used_chars(html_tree.getroot()))
zip.writestr('OPS/last.html', etree.tostring(
html_tree, pretty_print=True, xml_declaration=True,
encoding="utf-8",
doctype='<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" ' +
'"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
))
if not flags or not 'without-fonts' in flags:
# strip fonts
tmpdir = mkdtemp('-librarian-epub')
try:
cwd = os.getcwd()
except OSError:
cwd = None
os.chdir(os.path.join(os.path.dirname(os.path.realpath(__file__)), 'font-optimizer'))
for fname in 'DejaVuSerif.ttf', 'DejaVuSerif-Bold.ttf', 'DejaVuSerif-Italic.ttf', 'DejaVuSerif-BoldItalic.ttf':
optimizer_call = ['perl', 'subset.pl', '--chars',
''.join(chars).encode('utf-8'),
get_resource('fonts/' + fname),
os.path.join(tmpdir, fname)]
if verbose:
print "Running font-optimizer"
subprocess.check_call(optimizer_call)
else:
subprocess.check_call(optimizer_call, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
zip.write(os.path.join(tmpdir, fname), os.path.join('OPS', fname))
manifest.append(etree.fromstring(
'<item id="%s" href="%s" media-type="application/x-font-truetype" />' % (fname, fname)))
rmtree(tmpdir)
if cwd is not None:
os.chdir(cwd)
zip.writestr('OPS/content.opf', etree.tostring(opf, pretty_print=True,
xml_declaration=True, encoding="utf-8"))
title = document.book_info.title
attributes = "dtb:uid", "dtb:depth", "dtb:totalPageCount", "dtb:maxPageNumber"
for st in attributes:
meta = toc_file.makeelement(NCXNS('meta'))
meta.set('name', st)
meta.set('content', '0')
toc_file[0].append(meta)
toc_file[0][0].set('content', str(document.book_info.url))
toc_file[0][1].set('content', str(toc.depth()))
set_inner_xml(toc_file[1], ''.join(('<text>', title, '</text>')))
# write TOC
if html_toc:
toc.add(u"Spis treści", "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,
xml_declaration=True, encoding="utf-8"))
zip.close()
return OutputFile.from_filename(output_file.name)
def output_file(self, *args, **kwargs):
imgstr = StringIO()
self.save(imgstr, *args, **kwargs)
return OutputFile.from_string(imgstr.getvalue())
def transform(wldoc,
verbose=False,
save_tex=None,
morefloats=None,
cover=None,
flags=None,
customizations=None,
ilustr_path='',
latex_dir=False):
""" produces a PDF file with XeLaTeX
wldoc: a WLDocument
verbose: prints all output from LaTeX
save_tex: path to save the intermediary LaTeX file to
morefloats (old/new/none): force specific morefloats
cover: a cover.Cover factory or True for default
flags: less-advertising,
customizations: user requested customizations regarding various formatting parameters (passed to wl LaTeX class)
"""
# Parse XSLT
try:
book_info = wldoc.book_info
document = load_including_children(wldoc)
root = document.edoc.getroot()
if cover:
if cover is True:
cover = make_cover
bound_cover = cover(book_info, width=1200)
root.set('data-cover-width', str(bound_cover.width))
root.set('data-cover-height', str(bound_cover.height))
if bound_cover.uses_dc_cover:
if book_info.cover_by:
root.set('data-cover-by', book_info.cover_by)
if book_info.cover_source:
root.set('data-cover-source', book_info.cover_source)
if flags:
for flag in flags:
root.set('flag-' + flag, 'yes')
# check for LaTeX packages
if morefloats:
root.set('morefloats', morefloats.lower())
elif package_available('morefloats', 'maxfloats=19'):
root.set('morefloats', 'new')
# add customizations
if customizations is not None:
root.set('customizations', u','.join(customizations))
# add editors info
editors = document.editors()
if editors:
root.set(
'editors',
u', '.join(sorted(editor.readable() for editor in editors)))
if document.book_info.funders:
root.set('funders', u', '.join(document.book_info.funders))
if document.book_info.thanks:
root.set('thanks', document.book_info.thanks)
# hack the tree
move_motifs_inside(document.edoc)
hack_motifs(document.edoc)
parse_creator(document.edoc)
substitute_hyphens(document.edoc)
fix_hanging(document.edoc)
fix_tables(document.edoc)
mark_subauthors(document.edoc)
# wl -> TeXML
style_filename = get_stylesheet("wl2tex")
style = etree.parse(style_filename)
functions.reg_mathml_latex()
# TeXML -> LaTeX
temp = mkdtemp('-wl2pdf')
for ilustr in document.edoc.findall("//ilustr"):
shutil.copy(os.path.join(ilustr_path, ilustr.get("src")), temp)
for sponsor in book_info.sponsors:
ins = etree.Element("data-sponsor", name=sponsor)
logo = sponsor_logo(sponsor)
if logo:
fname = 'sponsor-%s' % os.path.basename(logo)
shutil.copy(logo, os.path.join(temp, fname))
ins.set('src', fname)
root.insert(0, ins)
if book_info.sponsor_note:
root.set("sponsor-note", book_info.sponsor_note)
texml = document.transform(style)
if cover:
with open(os.path.join(temp, 'cover.png'), 'w') as f:
bound_cover.save(f, quality=80)
del document # no longer needed large object :)
tex_path = os.path.join(temp, 'doc.tex')
fout = open(tex_path, 'wb')
process(six.BytesIO(texml), fout, 'utf-8')
fout.close()
del texml
if save_tex:
shutil.copy(tex_path, save_tex)
# LaTeX -> PDF
shutil.copy(get_resource('pdf/wl.cls'), temp)
shutil.copy(get_resource('res/wl-logo.png'), temp)
if latex_dir:
return temp
try:
cwd = os.getcwd()
except OSError:
cwd = None
os.chdir(temp)
# some things work better when compiled twice
# (table of contents, [line numbers - disabled])
for run in range(2):
if verbose:
p = call(['xelatex', tex_path])
else:
p = call(['xelatex', '-interaction=batchmode', tex_path],
stdout=PIPE,
stderr=PIPE)
if p:
raise ParseError("Error parsing .tex file")
if cwd is not None:
os.chdir(cwd)
output_file = NamedTemporaryFile(prefix='librarian',
suffix='.pdf',
delete=False)
pdf_path = os.path.join(temp, 'doc.pdf')
shutil.move(pdf_path, output_file.name)
shutil.rmtree(temp)
return OutputFile.from_filename(output_file.name)
except (XMLSyntaxError, XSLTApplyError) as e:
raise ParseError(e)