X-Git-Url: https://git.mdrn.pl/wolnelektury.git/blobdiff_plain/4defd5d35b69d1ad01ca05b2a1e174abc2fcebfb..d3db8ba2750af385acc9f247da53ec3851637cba:/lib/librarian/html.py?ds=inline diff --git a/lib/librarian/html.py b/lib/librarian/html.py index ea71bbb1d..9763428d0 100644 --- a/lib/librarian/html.py +++ b/lib/librarian/html.py @@ -55,7 +55,7 @@ def transform(input_filename, output_filename): result = doc.xslt(style) add_anchors(result.getroot()) add_table_of_contents(result.getroot()) - result.write(output_filename, xml_declaration=True, pretty_print=True, encoding='utf-8') + result.write(output_filename, xml_declaration=False, pretty_print=True, encoding='utf-8') class Fragment(object): @@ -109,7 +109,7 @@ def extract_fragments(input_filename): for event, element in etree.iterparse(input_filename, events=('start', 'end')): # Process begin and end elements - if element.tag == 'span' and element.get('class', '') in ('theme-begin', 'theme-end'): + if element.get('class', '') in ('theme-begin', 'theme-end'): if not event == 'end': continue # Process elements only once, on end event # Open new fragment @@ -117,9 +117,9 @@ def extract_fragments(input_filename): fragment = Fragment(id=element.get('fid'), themes=element.text) # Append parents - if element.getparent().tag != 'body': + if element.getparent().get('id', None) != 'book-text': parents = [element.getparent()] - while parents[-1].getparent().tag != 'body': + while parents[-1].getparent().get('id', None) != 'book-text': parents.append(parents[-1].getparent()) parents.reverse() @@ -171,12 +171,18 @@ def add_anchor(element, number): element.insert(0, anchor_target) +def any_ancestor(element, test): + for ancestor in element.iterancestors(): + if test(ancestor): + return True + return False + + def add_anchors(root): counter = 1 for element in root.iterdescendants(): - if element.getparent().tag in 'div' and 'note' in element.getparent().get('class', ''): - continue - if element.getparent().tag in 'blockquote': + if any_ancestor(element, lambda e: e.get('class') in ('note', 'motto', 'motto_podpis', 'dedication') + or e.tag == 'blockquote'): continue if element.tag == 'p' and 'verse' in element.get('class', ''):