X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/08a62c79e1e283f446130c02b92aa7920d7d7949..0efff78112a63c9321b297ff9ff98a8861d3cdf6:/librarian/html.py?ds=sidebyside diff --git a/librarian/html.py b/librarian/html.py index 0eeb76b..6a6e3db 100644 --- a/librarian/html.py +++ b/librarian/html.py @@ -4,6 +4,7 @@ # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information. # import os +import re import cStringIO import copy @@ -22,12 +23,15 @@ STYLESHEETS = { 'partial': 'xslt/wl2html_partial.xslt' } + def get_stylesheet(name): return os.path.join(os.path.dirname(__file__), STYLESHEETS[name]) + def html_has_content(text): return etree.ETXPath('//p|//{%(ns)s}p|//h1|//{%(ns)s}h1' % {'ns': str(XHTMLNS)})(text) + def transform(wldoc, stylesheet='legacy', options=None, flags=None): """Transforms the WL document to XHTML. @@ -49,19 +53,20 @@ def transform(wldoc, stylesheet='legacy', options=None, flags=None): document.edoc.getroot().set(flag, 'yes') document.clean_ed_note() + document.clean_ed_note('abstrakt') if not options: options = {} result = document.transform(style, **options) - del document # no longer needed large object :) + 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')) + return OutputFile.from_string(etree.tostring( + result, method='html', xml_declaration=False, pretty_print=True, encoding='utf-8')) else: return None except KeyError: @@ -69,6 +74,7 @@ def transform(wldoc, stylesheet='legacy', options=None, flags=None): except (XMLSyntaxError, XSLTApplyError), e: raise ParseError(e) + class Fragment(object): def __init__(self, id, themes): super(Fragment, self).__init__() @@ -97,7 +103,8 @@ class Fragment(object): result = [] for event, element in self.closed_events(): if event == 'start': - result.append(u'<%s %s>' % (element.tag, ' '.join('%s="%s"' % (k, v) for k, v in element.attrib.items()))) + result.append(u'<%s %s>' % ( + element.tag, ' '.join('%s="%s"' % (k, v) for k, v in element.attrib.items()))) if element.text: result.append(element.text) elif event == 'end': @@ -127,7 +134,8 @@ def extract_fragments(input_filename): for event, element in etree.iterparse(buf, events=('start', 'end')): # Process begin and end elements if element.get('class', '') in ('theme-begin', 'theme-end'): - if not event == 'end': continue # Process elements only once, on end event + if not event == 'end': + continue # Process elements only once, on end event # Open new fragment if element.get('class', '') == 'theme-begin': @@ -163,7 +171,6 @@ def extract_fragments(input_filename): for fragment_id in open_fragments: open_fragments[fragment_id].append('text', element.tail) - # Process all elements except begin and end else: # Omit annotation tags @@ -208,9 +215,10 @@ def any_ancestor(element, test): def add_anchors(root): counter = 1 for element in root.iterdescendants(): - if any_ancestor(element, lambda e: e.get('class') in ('note', 'motto', 'motto_podpis', 'dedication') - or e.get('id') == 'nota_red' - or e.tag == 'blockquote'): + def f(e): + return e.get('class') in ('note', 'motto', 'motto_podpis', 'dedication') or \ + e.get('id') == 'nota_red' or e.tag == 'blockquote' + if any_ancestor(element, f): continue if element.tag == 'p' and 'verse' in element.get('class', ''): @@ -235,7 +243,8 @@ def add_table_of_contents(root): counter = 1 for element in root.iterdescendants(): if element.tag in ('h2', 'h3'): - if any_ancestor(element, lambda e: e.get('id') in ('footnotes', 'nota_red') or e.get('class') in ('person-list',)): + if any_ancestor(element, + lambda e: e.get('id') in ('footnotes', 'nota_red') or e.get('class') in ('person-list',)): continue element_text = raw_printable_text(element) @@ -258,9 +267,9 @@ def add_table_of_contents(root): if len(subsections): subsection_list = etree.SubElement(section_element, 'ol') - for n, subsection, text, _ in subsections: + for n1, subsection, subtext, _ in subsections: subsection_element = etree.SubElement(subsection_list, 'li') - add_anchor(subsection_element, "s%d" % n, with_target=False, link_text=text) + add_anchor(subsection_element, "s%d" % n1, with_target=False, link_text=subtext) root.insert(0, toc) @@ -269,7 +278,8 @@ def add_table_of_themes(root): try: from sortify import sortify except ImportError: - sortify = lambda x: x + def sortify(x): + return x book_themes = {} for fragment in root.findall('.//a[@class="theme-begin"]'): @@ -292,17 +302,43 @@ def add_table_of_themes(root): root.insert(0, themes_div) - def extract_annotations(html_path): - """For each annotation, yields a tuple: anchor, text, html.""" + """Extracts annotations from HTML for annotations dictionary. + + For each annotation, yields a tuple of: + anchor, footnote type, valid qualifiers, text, html. + + """ + from .fn_qualifiers import FN_QUALIFIERS + parser = etree.HTMLParser(encoding='utf-8') tree = etree.parse(html_path, parser) footnotes = tree.find('//*[@id="footnotes"]') + re_qualifier = re.compile(ur'[^\u2014]+\s+\(([^\)]+)\)\s+\u2014') if footnotes is not None: for footnote in footnotes.findall('div'): - anchor = footnote.find('a[@name]').get('name') + fn_type = footnote.get('class').split('-')[1] + anchor = footnote.find('a[@class="annotation"]').get('href')[1:] del footnote[:2] - text_str = etree.tostring(footnote, method='text', encoding='utf-8').strip() - html_str = etree.tostring(footnote, method='html', encoding='utf-8') - yield anchor, text_str, html_str + footnote.text = None + if len(footnote) and footnote[-1].tail == '\n': + footnote[-1].tail = None + text_str = etree.tostring(footnote, method='text', encoding=unicode).strip() + html_str = etree.tostring(footnote, method='html', encoding=unicode).strip() + + match = re_qualifier.match(text_str) + if match: + qualifier_str = match.group(1) + qualifiers = [] + for candidate in re.split('[;,]', qualifier_str): + candidate = candidate.strip() + if candidate in FN_QUALIFIERS: + qualifiers.append(candidate) + elif candidate.startswith('z '): + subcandidate = candidate.split()[1] + if subcandidate in FN_QUALIFIERS: + qualifiers.append(subcandidate) + else: + qualifiers = [] + yield anchor, fn_type, qualifiers, text_str, html_str