def view_selection(self, req, resp, url):
"""
View the highlighted selector (from `action_view`)
"""
from deliverance.selector import Selector
doc = document_fromstring(resp.body)
el = Element('base')
el.set('href', posixpath.dirname(url) + '/')
doc.head.insert(0, el)
selector = Selector.parse(req.GET['selector'])
dummy_type, elements, dummy_attributes = selector(doc)
if not elements:
template = self._not_found_template
else:
template = self._found_template
all_elements = []
els_in_head = False
for index, el in enumerate(elements):
el_in_head = self._el_in_head(el)
if el_in_head:
els_in_head = True
anchor = 'deliverance-selection'
if index:
anchor += '-%s' % index
if el.get('id'):
anchor = el.get('id')
## FIXME: is a <a name> better?
if not el_in_head:
el.set('id', anchor)
else:
anchor = None
## FIXME: add :target CSS rule
## FIXME: or better, some Javascript
all_elements.append((anchor, el))
if not el_in_head:
style = el.get('style', '')
if style:
style += '; '
style += '/* deliverance */ border: 2px dotted #f00'
el.set('style', style)
else:
el.set('DELIVERANCE-MATCH', '1')
def highlight(html_code):
"""Highlights the given code (for use in the template)"""
if isinstance(html_code, _Element):
html_code = tostring(html_code)
return html(
pygments_highlight(html_code, HtmlLexer(),
HtmlFormatter(noclasses=True)))
def format_tag(tag):
"""Highlights the lxml HTML tag"""
return highlight(tostring(tag).split('>')[0] + '>')
def wrap_html(html, width=100):
if isinstance(html, _Element):
html = tostring(html)
lines = html.splitlines()
new_lines = []
def wrap_html_line(line):
if len(line) <= width:
return [line]
match_trail = re.search(r'^[^<]*</.*?>', line, re.S)
if match_trail:
result = [match_trail.group(0)]
result.extend(wrap_html_line(line[match_trail.end():]))
return result
match1 = re.search(r'^[^<]*<[^>]*>', line, re.S)
match2 = re.search(r'<[^>]*>[^<>]*$', line, re.S)
if not match1 or not match2:
return [line]
result = [match1.group(0)]
result.extend(wrap_html_line(
line[match1.end():match2.start()]))
result.append(match2.group(0))
return result
for line in lines:
new_lines.extend(wrap_html_line(line))
return '\n'.join(new_lines)
def mark_deliv_match(highlighted_text):
result = re.sub(
r'(?:<[^/][^>]*>)*<.*?DELIVERANCE-MATCH=.*?>(?:</[^>]*>)*',
lambda match: r'<b style="background-color: #ff8">%s</b>' %
match.group(0), unicode(highlighted_text), re.S)
return html(result)
text = template.substitute(base_url=url,
els_in_head=els_in_head,
doc=doc,
elements=all_elements,
selector=selector,
format_tag=format_tag,
highlight=highlight,
wrap_html=wrap_html,
mark_deliv_match=mark_deliv_match)
message = fromstring(
self._message_template.substitute(message=text, url=url))
if doc.body.text:
message.tail = doc.body.text
doc.body.text = ''
doc.body.insert(0, message)
text = tostring(doc)
return Response(text)
def view_selection(self, req, resp, url):
"""
View the highlighted selector (from `action_view`)
"""
from deliverance.selector import Selector
doc = document_fromstring(resp.body)
el = Element('base')
el.set('href', posixpath.dirname(url) + '/')
doc.head.insert(0, el)
selector = Selector.parse(req.GET['selector'])
dummy_type, elements, dummy_attributes = selector(doc)
if not elements:
template = self._not_found_template
else:
template = self._found_template
all_elements = []
els_in_head = False
for index, el in enumerate(elements):
el_in_head = self._el_in_head(el)
if el_in_head:
els_in_head = True
anchor = 'deliverance-selection'
if index:
anchor += '-%s' % index
if el.get('id'):
anchor = el.get('id')
## FIXME: is a <a name> better?
if not el_in_head:
el.set('id', anchor)
else:
anchor = None
## FIXME: add :target CSS rule
## FIXME: or better, some Javascript
all_elements.append((anchor, el))
if not el_in_head:
style = el.get('style', '')
if style:
style += '; '
style += '/* deliverance */ border: 2px dotted #f00'
el.set('style', style)
else:
el.set('DELIVERANCE-MATCH', '1')
def highlight(html_code):
"""Highlights the given code (for use in the template)"""
if isinstance(html_code, _Element):
html_code = tostring(html_code)
return html(pygments_highlight(html_code, HtmlLexer(),
HtmlFormatter(noclasses=True)))
def format_tag(tag):
"""Highlights the lxml HTML tag"""
return highlight(tostring(tag).split('>')[0]+'>')
def wrap_html(html, width=100):
if isinstance(html, _Element):
html = tostring(html)
lines = html.splitlines()
new_lines = []
def wrap_html_line(line):
if len(line) <= width:
return [line]
match_trail = re.search(r'^[^<]*</.*?>', line, re.S)
if match_trail:
result = [match_trail.group(0)]
result.extend(wrap_html_line(line[match_trail.end():]))
return result
match1 = re.search(r'^[^<]*<[^>]*>', line, re.S)
match2 = re.search(r'<[^>]*>[^<>]*$', line, re.S)
if not match1 or not match2:
return [line]
result = [match1.group(0)]
result.extend(wrap_html_line(line[match1.end():match2.start()]))
result.append(match2.group(0))
return result
for line in lines:
new_lines.extend(wrap_html_line(line))
return '\n'.join(new_lines)
def mark_deliv_match(highlighted_text):
result = re.sub(r'(?:<[^/][^>]*>)*<.*?DELIVERANCE-MATCH=.*?>(?:</[^>]*>)*', lambda match: r'<b style="background-color: #ff8">%s</b>' % match.group(0), unicode(highlighted_text), re.S)
return html(result)
text = template.substitute(
base_url=url,
els_in_head=els_in_head, doc=doc,
elements=all_elements, selector=selector,
format_tag=format_tag, highlight=highlight,
wrap_html=wrap_html, mark_deliv_match=mark_deliv_match)
message = fromstring(
self._message_template.substitute(message=text, url=url))
if doc.body.text:
message.tail = doc.body.text
doc.body.text = ''
doc.body.insert(0, message)
text = tostring(doc)
return Response(text)
def split_and_output(input_root,
template_file_name,
input_file_name,
output_folder=''):
output_tree = html.parse(template_file_name)
# output_root = output_tree.getroot()
# put element lists in dict with file_lable as the key
file_lables_element_lists = {'op': [], 'new_fb': [], 'an': []}
# select all the paragraphs etc within the top levle divs
paragraph_elements = input_root.xpath('//body/div/*')
list_to_add_to = file_lables_element_lists['op']
# look through all the paragraph elemets and find out if any are Annoncements etc
for paragraph_element in paragraph_elements:
# if (paragraph_element.get('class') == 'DocumentTitle'
# and 'PART 2' in paragraph_element.text_content().upper()):
# # start new list for future business
# list_to_add_to = file_lables_element_lists['new_fb']
list_to_add_to.append(paragraph_element)
# build up output trees
for file_lable, element_list in file_lables_element_lists.items():
if len(element_list) != 0:
# copy the template tree and add elements needed for this section
temp_output_tree = deepcopy(output_tree)
# temp_output_tree = html.parse(template_file_name)
temp_output_root = temp_output_tree.getroot()
# change the title
if file_lable == 'op':
title_text = 'Order Paper for ' + DATES.sitting_date_medium
h1_text = 'Order Paper for ' + DATES.sitting_date_long
# elif file_lable == 'new_fb':
# title_text = 'Future Business as of ' + DATES.sitting_date_medium
# h1_text = 'Future Business as of ' + DATES.sitting_date_long
temp_output_root.xpath('//h1[@id="mainTitle"]')[0].text = h1_text
temp_output_root.xpath('//head/title')[0].text = title_text
# get the position (in the template) where we will inject html (from the input)
code_injection_point = temp_output_root.xpath(
'//div[@id="content-goes-here"]')[0]
for element in element_list:
# remove Future Business heading from start of part 2
if 'paraChamberSummaryHeading' in element.classes:
if element.text_content().lower() == 'future business':
continue
# remove the docuemnt headings from the html i.e. part 1 head
if 'DocumentTitle' in element.classes:
text_content = element.text_content().lower()
h2 = Element('h2')
h2.set('class', 'OP-heading-outdent')
if 'part 1' in text_content:
h2.text = 'Part 1: Business Today'
code_injection_point.append(h2)
elif 'part 2' in text_content:
h2.text = 'Part 2: Future Business'
code_injection_point.append(h2)
else:
code_injection_point.append(element)
# Add IDs and perminant ancors to the html
# Added at the request of IDMS
# need to get all the heading elements
xpath = '//h1|//h2|//h3|//h4|//h5|//h6|//*[@class="paraBusinessItemHeading"]' \
'|//*[@class="paraBusinessItemHeading-bulleted"]|//*[@class="FbaLocation"]'
headings = temp_output_root.xpath(xpath)
for i, heading in enumerate(headings):
# generate id text
id_text = f'{DATES.sitting_date_compact}-{i}'
if heading.get('id', default=None):
heading.set('name', heading.get('id'))
heading.set('id', id_text)
# parmalink_span = SubElement(heading, 'span')
# parmalink_span.set('class', 'perma-link')
# anchor = SubElement(parmalink_span, 'a')
anchor = SubElement(heading, 'a')
permalink_for = 'Permalink for ' + heading.text_content()
anchor.set('href', '#' + id_text)
anchor.set('aria-label', 'Anchor')
anchor.set('title', permalink_for)
anchor.set('data-anchor-icon', '§')
anchor.set('class', 'anchor-link')
# create the tables of contents
# This will be overridden by tocbot.
# We still want a ToC even if JavaScript is dissabled...
# find where to put the Toc
nav_xpath_results = temp_output_root.xpath('//nav[@id="toc"][1]')
# look for all the h2's
# // Where to grab the headings to build the table of contents.
# contentSelector: '.js-toc-content'
h2s = temp_output_root.xpath(
'//*[contains(@class, "js-toc-content")]//h2')
if len(nav_xpath_results):
toc_injection_point = nav_xpath_results[0]
ol = SubElement(toc_injection_point, 'ol')
ol.set('class', 'toc-list')
for h2 in h2s:
li = SubElement(ol, 'li')
li.set('class', 'toc-list-item')
a = SubElement(li, 'a')
a.set('href', '#' + h2.get('id', ''))
a.set('class', 'toc-link')
a.text = h2.text_content()
else:
print('no element')
# itterate through tree and remove CR from tail and text
for element in temp_output_root.iter():
if element.tail:
element.tail = element.tail.replace('\r', '')
if element.text:
element.text = element.text.replace('\r', '')
# write out the output html files
# outputfile_name = os.path.join(os.path.dirname(input_file_name),
# file_lable + DATES.sitting_date_compact[2:] + fileextension)
outputfile_name = f'{file_lable}{DATES.sitting_date_compact[2:]}{ fileextension}'
if output_folder:
outputfile_path = Path(output_folder).joinpath(outputfile_name)
# print(outputfile_path)
else:
outputfile_path = Path(input_file_name).parent.joinpath(
outputfile_name)
# created element tree so we can use write method
# temp_output_tree = ElementTree(temp_output_root)
temp_output_tree.write(str(outputfile_path),
doctype=DOCTYPE,
encoding='UTF-8',
method="html",
xml_declaration=False)
print(f'{file_lable} file is at:\t{outputfile_path}')
