def article_html_generator(file_xml_path: str, dest_path: str) -> None:
logger.debug("file: %s", file_xml_path)
parsed_xml = XML(file_xml_path, no_network=False)
html_generator = HTMLGenerator.parse(
parsed_xml,
valid_only=False,
css="https://new.scielo.br/static/css/scielo-article.css",
print_css="https://new.scielo.br/static/css/scielo-bundle-print.css",
js="https://new.scielo.br/static/js/scielo-article-min.js",
)
for lang, trans_result in html_generator:
fpath, fname = os.path.split(file_xml_path)
fname, fext = fname.rsplit(".", 1)
out_fname = ".".join([fname, lang, "html"])
new_file_html_path = os.path.join(dest_path, out_fname)
files.write_file(
new_file_html_path,
etree.tostring(
trans_result,
doctype=u"<!DOCTYPE html>",
pretty_print=True,
encoding="utf-8",
method="html",
).decode("utf-8"),
)
def html_languages(self):
"""
Extract all the HTML languages in XML content.
"""
try:
generator = HTMLGenerator.parse(self._content, valid_only=False)
except ValueError as e:
logger.error('Error getting htmlgenerator: {}.'.format(e.message))
else:
return [
{'type': 'html', 'lang': lang}
for lang, __ in generator
]
def render_html_from_xml(article, lang):
result = fetch_data(normalize_ssm_url(article.xml))
xml = etree.parse(BytesIO(result))
generator = HTMLGenerator.parse(xml, valid_only=False)
# Criamos um objeto do tip soup
soup = BeautifulSoup(
etree.tostring(generator.generate(lang),
encoding="UTF-8",
method="html"), 'html.parser')
# Fatiamos o HTML pelo div com class: articleTxt
return soup.find('div', {'id': 'standalonearticle'}), generator.languages
def register_htmls(self):
"""
Register HTML contents from XML for all the text languages.
"""
try:
generator = HTMLGenerator.parse(
self._content,
valid_only=False,
css=config.OPAC_PROC_ARTICLE_CSS_URL,
print_css=config.OPAC_PROC_ARTICLE_PRINT_CSS_URL,
js=config.OPAC_PROC_ARTICLE_JS_URL)
except ValueError as e:
logger.error('Error getting htmlgenerator: {}.'.format(e.message))
return None
registered_htmls = []
for lang, trans_result in generator:
html_as_bytes = None
try:
html = etree.tostring(trans_result,
pretty_print=True,
encoding='utf-8',
method='html',
doctype="<!DOCTYPE html>")
html_as_bytes = BytesIO(html)
except Exception as e:
logger.error(
'Error converting etree {} to string. '.format(lang))
else:
metadata = self.get_metadata()
metadata.update({
'bucket_name': self.bucket_name,
'type': 'html',
'version': 'xml'
})
__, html_url = self._register_ssm_asset(
html_as_bytes, self._get_file_name('html', lang), 'html',
metadata)
registered_htmls.append({
'type': 'html',
'lang': lang,
'url': html_url
})
return registered_htmls
def render_html_converted_view(request):
file_xml_path = os.path.join(config.get("CONVERSION_PATH"),
request.matchdict["file_xml"])
parsed_xml = XML(file_xml_path, no_network=False)
html_generator = HTMLGenerator.parse(
parsed_xml,
valid_only=False,
css="/static/css/scielo-article.css",
print_css="/static/css/scielo-bundle-print.css",
js="/static/js/scielo-article-min.js",
)
html = html_generator.generate(request.matchdict["language"])
return Response(
etree.tostring(
html,
doctype=u"<!DOCTYPE html>",
pretty_print=True,
encoding="utf-8",
method="html",
).decode("utf-8"))