from tempfile import mkdtemp, NamedTemporaryFile
from shutil import rmtree
-from librarian import RDFNS, WLNS, NCXNS, OPFNS, XHTMLNS, OutputFile
+from librarian import RDFNS, WLNS, NCXNS, OPFNS, XHTMLNS, DCNS, OutputFile
from librarian.cover import DefaultEbookCover
from librarian import functions, get_resource
+from librarian.hyphenator import Hyphenator
+
functions.reg_person_name()
functions.reg_lang_code_3to2()
-
+def set_hyph_language(source_tree):
+ def get_short_lng_code(text):
+ result = ''
+ text = ''.join(text)
+ with open(get_resource('res/ISO-639-2_8859-1.txt'), 'rb') as f:
+ for line in f:
+ list = line.strip().split('|')
+ if list[0] == text:
+ result=list[2]
+ if result == '':
+ return text
+ else:
+ return result
+ bibl_lng = etree.XPath('//dc:language//text()', namespaces = {'dc':str(DCNS)})(source_tree)
+ short_lng = get_short_lng_code(bibl_lng[0])
+ try:
+ return Hyphenator(get_resource('res/hyph-dictionaries/hyph_' + short_lng + '.dic'))
+ except:
+ pass
+
+def hyphenate_and_fix_conjunctions(source_tree, hyph):
+ if hyph is not None:
+ texts = etree.XPath('/utwor/*[2]//text()')(source_tree)
+ for t in texts:
+ parent = t.getparent()
+ newt = ''
+ wlist = re.compile(r'\w+|[^\w]', re.UNICODE).findall(t)
+ for w in wlist:
+ newt += hyph.inserted(w, u'\u00AD')
+ newt = re.sub(r'(?<=\s\w)\s+', u'\u00A0', newt)
+ if t.is_text:
+ parent.text = newt
+ elif t.is_tail:
+ parent.tail = newt
+
def inner_xml(node):
""" returns node's text and children as a string
def replace_chars(text):
if text is None:
return None
- #text = re.sub(r"(?<=\s\w)\s+", u"\u00a0", text) #fix for hanging single letter conjunctions – for future use.
return text.replace(u"\ufeff", u"")\
.replace("---", u"\u2014")\
.replace("--", u"\u2013")\
nav_label = nav_map.makeelement(NCXNS('navLabel'))
text = nav_map.makeelement(NCXNS('text'))
- text.text = child.name
+ if child.name is not None:
+ text.text = re.sub(r'\n', ' ', child.name)
+ else:
+ text.text = child.name
nav_label.append(text)
nav_point.append(nav_label)
yield part_xml
last_node_part = True
main_xml_part[:] = [deepcopy(one_part)]
- elif not last_node_part and name in ("naglowek_scena"):
+ elif not last_node_part and name == "naglowek_scena":
yield part_xml
main_xml_part[:] = [deepcopy(one_part)]
else:
toc = TOC()
for element in chunk_xml[0]:
- if element.tag in ("naglowek_czesc"):
+ if element.tag == "naglowek_czesc":
toc.add(node_name(element), "part%d.html#book-text" % chunk_no)
elif element.tag in ("naglowek_rozdzial", "naglowek_akt", "srodtytul"):
toc.add(node_name(element), "part%d.html" % 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)
+ output_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">')
return output_html, toc, chars
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, with-full-fonts
+ 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)
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))
+ 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:
else:
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)
+ 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)
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, method="html", pretty_print=True))
+ 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:
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, method="html", pretty_print=True))
+ 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(
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))
+ 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
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':
- if not flags or not 'with-full-fonts' in flags:
- 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))
+ 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:
- zip.write(get_resource('fonts/' + fname), os.path.join('OPS', fname))
+ 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'))
+ 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:
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.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)
fnp / librarian.git / blobdiff