X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/74dd02ea5d106d125877b16cd8b5712c5361714f..3cae7a60cf91b6e2b84628c5f3b4b2df77eda5d6:/librarian/epub.py diff --git a/librarian/epub.py b/librarian/epub.py index c802ec7..333b56f 100644 --- a/librarian/epub.py +++ b/librarian/epub.py @@ -29,6 +29,10 @@ functions.reg_person_name() functions.reg_lang_code_3to2() +def squeeze_whitespace(s): + return re.sub(r'\s+', ' ', s) + + def set_hyph_language(source_tree): def get_short_lng_code(text): result = '' @@ -186,7 +190,7 @@ class Stanza(object): tail = self.stanza.tail self.stanza.clear() self.stanza.tail = tail - self.stanza.extend(self.verses) + self.stanza.extend(verse for verse in self.verses if verse.text or len(verse) > 0) def open_normal_verse(self): self.open_verse = self.stanza.makeelement("wers_normalny") @@ -437,15 +441,13 @@ def transform(wldoc, verbose=False, style=None, html_toc=False, # 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='' - ) + html_string = etree.tostring( + html_tree, pretty_print=True, xml_declaration=True, + encoding="utf-8", + doctype='' ) + zip.writestr('OPS/title.html', squeeze_whitespace(html_string)) # add a title page TOC entry toc.add(u"Strona tytułowa", "title.html") elif wldoc.book_info.parts: @@ -462,7 +464,7 @@ def transform(wldoc, verbose=False, style=None, html_toc=False, doctype='' ) - zip.writestr('OPS/part%d.html' % chunk_counter, html_string) + zip.writestr('OPS/part%d.html' % chunk_counter, squeeze_whitespace(html_string)) add_to_manifest(manifest, chunk_counter) add_to_spine(spine, chunk_counter) chunk_counter += 1 @@ -488,7 +490,7 @@ def transform(wldoc, verbose=False, style=None, html_toc=False, toc.extend(chunk_toc) chars = chars.union(chunk_chars) - zip.writestr('OPS/part%d.html' % chunk_counter, chunk_html) + zip.writestr('OPS/part%d.html' % chunk_counter, squeeze_whitespace(chunk_html)) add_to_manifest(manifest, chunk_counter) add_to_spine(spine, chunk_counter) chunk_counter += 1 @@ -645,7 +647,7 @@ def transform(wldoc, verbose=False, style=None, html_toc=False, '')) html_string = open(get_resource('epub/support.html')).read() chars.update(used_chars(etree.fromstring(html_string))) - zip.writestr('OPS/support.html', html_string) + zip.writestr('OPS/support.html', squeeze_whitespace(html_string)) toc.add("Strona redakcyjna", "last.html") manifest.append(etree.fromstring( @@ -654,12 +656,12 @@ def transform(wldoc, verbose=False, style=None, html_toc=False, '')) 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( + zip.writestr('OPS/last.html', squeeze_whitespace(etree.tostring( html_tree, pretty_print=True, xml_declaration=True, encoding="utf-8", doctype='' - )) + ))) if not flags or 'without-fonts' not in flags: # strip fonts