evin/secretary
3
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

Merge branch 'DieterBuysAI-master' into development

Browse source
This commit is contained in:
Christopher Ramírez 2015-04-29 11:55:18 -06:00
commit 029bc9b82d
1 changed files with 22 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

28
secretary.py
View file

@ -26,6 +26,7 @@ from os import path
from mimetypes import guess_type, guess_extension from mimetypes import guess_type, guess_extension
from uuid import uuid4 from uuid import uuid4
from xml.dom.minidom import parseString from xml.dom.minidom import parseString
from xml.parsers.expat import ExpatError
from jinja2 import Environment, Undefined from jinja2 import Environment, Undefined
# Test python versions and normalize calls to basestring, unicode, etc. # Test python versions and normalize calls to basestring, unicode, etc.
@ -306,20 +307,28 @@ class Renderer(object):
parent.removeChild(discard) parent.removeChild(discard)
def _unescape_entities(self, xml_text): @staticmethod
# unescape XML entities gt and lt def _unescape_entities(xml_text):
"""
Strips tags of the form <text:span ...> from inside Jinja elements
and unescapes HTML codes for >, <, & and "
"""
unescape_rules = { unescape_rules = {
r'(?is)({[{|%].*)(&gt;)(.*[%|}]})': r'\1>\3', r'(?is)({[{|%].*)(&gt;)(.*[%|}]})': r'\1>\3',
r'(?is)({[{|%].*)(&lt;)(.*[%|}]})': r'\1<\3', r'(?is)({[{|%].*)(&lt;)(.*[%|}]})': r'\1<\3',
r'(?is)({[{|%].*)(<.?text:s.?>)(.*[%|}]})': r'\1 \3', r'(?is)({[{|%].*)(&amp;)(.*[%|}]})': r'\1&\3',
r'(?is)({[{|%].*)(&quot;)(.*[%|}]})': r'\1"\3'
} }
for p, r in unescape_rules.items(): for regexp, replacement in unescape_rules.items():
xml_text = re.sub(p, r, xml_text) subs_made = True
while subs_made:
xml_text, subs_made = re.subn(regexp, replacement, xml_text)
return xml_text return xml_text
def _encode_escape_chars(self, xml_text): @staticmethod
def _encode_escape_chars(xml_text):
# Replace line feed and/or tabs within text span entities. # Replace line feed and/or tabs within text span entities.
find_pattern = r'(?is)<text:([\S]+?)>([^>]*?([\n|\t])[^<]*?)</text:\1>' find_pattern = r'(?is)<text:([\S]+?)>([^>]*?([\n|\t])[^<]*?)</text:\1>'
for m in re.findall(find_pattern, xml_text): for m in re.findall(find_pattern, xml_text):
@ -443,6 +452,7 @@ class Renderer(object):
self._prepare_template_tags(xml_document) self._prepare_template_tags(xml_document)
template_string = self._unescape_entities(xml_document.toxml()) template_string = self._unescape_entities(xml_document.toxml())
jinja_template = self.environment.from_string(template_string) jinja_template = self.environment.from_string(template_string)
result = jinja_template.render(**kwargs) result = jinja_template.render(**kwargs)
result = self._encode_escape_chars(result) result = self._encode_escape_chars(result)
@ -451,9 +461,15 @@ class Renderer(object):
self.replace_images(final_xml) self.replace_images(final_xml)
return final_xml return final_xml
except ExpatError as e:
near = result.split('\n')[e.lineno -1][e.offset-50:e.offset+50]
raise ExpatError('ExpatError at line %d, column %d\nNear of: "[...]%s[...]"' % \
(e.lineno, e.offset, near))
except: except:
self.log.error('Error rendering template:\n%s', self.log.error('Error rendering template:\n%s',
xml_document.toprettyxml(), exc_info=True) xml_document.toprettyxml(), exc_info=True)
self.log.error('Unescaped template was:\n{}'.format(template_string))
raise raise
finally: finally:
self.log.debug('Rendering xml object finished') self.log.debug('Rendering xml object finished')