def parse(self, md_file):
""" Parse a md file into a post object
"""
# compile the templates when we parse the first post. This is needed
# to ensure that plugins get a chance to modify the templates before
# we compile them.
if not self.jinja2:
self.jinja2 = jinja2.Environment(loader=jinja2.DictLoader(
self.templates))
parsed_post = utils.dict_to_objdict()
# parsing frontmatter and getting the md
parsed_post.meta, parsed_post.md = frontmatter.parse(md_file)
# parsing markdown and extractring info
# NOTE: this must called before every parsing
self.renderer.init(self.jinja2, self.code_formatter, self.site,
parsed_post.meta)
parsed_post.html = self.md_parser.parse(parsed_post.md)
parsed_post.text = html2text(parsed_post.html) # used by NLP
parsed_post.meta.statistics = self.renderer.get_stats()
parsed_post.meta.toc = self.renderer.get_json_toc()
parsed_post.elements = self.renderer.get_info()
return parsed_post
def test_broken_html():
snip = '''
</td>test</span>
'''
assert html2text(snip) == 'test'
def test_empty():
assert html2text('') == ''
def test_dangling_tags():
snip = '''
<table><td>
<span>test</span>
'''
assert html2text(snip) == 'test'
def test_upercase_code():
snip = '''
<code>alert(1)</CoDe>
<span>test</span>
'''
assert html2text(snip) == 'test'
def test_none():
assert html2text(None) == ''
def test_pre():
snip = '''
<pre>alert(1)</pre>
<span>test</span>
'''
assert html2text(snip) == 'test'
def test_basic():
assert html2text('<span>test</span>') == 'test'