def escape_latex_all(o):
if isinstance(o, list):
return map(escape_latex_all, o)
elif isinstance(o, dict):
return dict(map(lambda x: (x[0], escape_latex_all(x[1])), o.items()))
else:
# TODO: HACK HACK HACK. Use Cheetah custom types.
return tex.escape_latex(str(o)).replace('LaTeX', r'\LaTeX').replace('\\textbackslash{}', '\\').replace('\\{', '{').replace('\\}', '}')
def escape_tex(s):
s = HYPERLINK_REGEX.sub(r'\href{\2}{\1}', s)
s = tex.escape_latex(s)
s = s.replace('LaTeX', r'\LaTeX')
# A bit of a hack... have to reverse some of tex.escape_latex escapes
s = s.replace('\\textbackslash{}', '\\')
s = s.replace('\\{', '{').replace('\\}', '}')
return s
def escape_tex(s):
s = HYPERLINK_REGEX.sub(r'\href{\2}{\1}', s)
s = tex.escape_latex(s)
s = s.replace('LaTeX', r'\LaTeX')
# A bit of a hack... have to reverse some of tex.escape_latex escapes
s = s.replace('\\textbackslash{}', '\\')
s = s.replace('\\{', '{').replace('\\}', '}')
return s
def escape_latex_all(o):
if isinstance(o, list):
return map(escape_latex_all, o)
elif isinstance(o, dict):
return dict(map(lambda x: (x[0], escape_latex_all(x[1])), o.items()))
else:
# TODO: HACK HACK HACK. Use Cheetah custom types.
return tex.escape_latex(str(o)).replace('LaTeX', r'\LaTeX').replace(
'\\textbackslash{}', '\\').replace('\\{', '{').replace('\\}', '}')
def run(self):
"""
Actual fetcher. Could this be threaded?
"""
url = view_text_json % self.url
logger.debug("Fetching URL: " + url)
raw_content = urllib2.urlopen(url)
json_content = json.loads(raw_content.read())
test_content = json_content['content']
test_content = unicodedata.normalize('NFKD',
test_content).encode('ascii', 'ignore')
# Here, we begin constructing the syntax tree
parser = MyHTMLParser()
parser.feed(test_content)
string = """
\\documentclass[12pt]{article}
\\begin{document}
{\\Large %s}
""" % clean(json_content['title'])
for line in parser.lines:
line = escape_latex(line)
string += line
string += """
\\end{document}
"""
with open('temp.latex', 'wb+') as f:
f.write(string)
pdf = latex2pdf(string)
with open('temp.pdf', 'wb+') as f:
f.write(pdf)