def get_xhtml_from_xml(self, xml_string):
# IV convert the xml tags and attribute to HTML-TEI
# remove comments
content = re.sub(ur'(?musi)<!--.*?-->', ur'', xml_string)
#
self.c = 0
self.conversion_cache = {}
def replace_tag(match):
if match.group(0) in self.conversion_cache:
return self.conversion_cache[match.group(0)]
self.c += 1
if self.c > 10e6:
exit()
ret = match.group(0)
tag = match.group(2)
# don't convert <p>
if tag in ['p', 'span']:
return ret
# any closing tag is /span
if '/' in match.group(1):
return '</span>'
if tag == 'pb':
print self.c
# tag -
ret = ur'<span data-dpt="%s"' % tag
# attribute - assumes " for attribute values
attrs = (re.sub(ur'(?ui)(\w+)(\s*=\s*")', ur'data-dpt-\1\2', match.group(3))).strip()
if attrs:
ret += ' ' + attrs
ret += match.group(4)
# print '', ret
self.conversion_cache[match.group(0)] = ret
return ret
from digipal.utils import re_sub_fct
content = re_sub_fct(content, ur'(?musi)(<\s*/?\s*)(\w+)([^>]*?)(/?\s*>)', replace_tag)
return content
def operation_foliate(self, options, content):
'''
<span data-dpt="margin">fol. 1. b</span>[...]
=>
<p><span data-dpt="location" data-dpt-loctype="locus">1v</span><p>
'''
self._next_locus = u'1r'
def replace(match):
ret = match.group(0)
locus = match.group(1)
parts = re.match(ur'(?musi)^\s?(\d+)\.?\s*(b?)\.?$', locus)
if not parts:
print 'WARNING: no match [%s]' % repr(locus)
else:
lo = parts.group(1)
lon = lo
if parts.group(2) == 'b':
lon = u'%sr' % (int(lo) + 1,)
lo += 'v'
else:
lon = lo + 'v'
lo += 'r'
print '%s ("%s")' % (lo, locus)
if lo != self._next_locus:
print 'WARNING: locus out of sequence, expected %s, got %s' % (self._next_locus, lo)
self._next_locus = lon
ret = u'</p><p><span data-dpt="location" data-dpt-loctype="locus">%s</span></p><p>' % lo
return ret
content = re_sub_fct(content, ur'(?musi)<span data-dpt="margin">\s*fol.([^<]*)</span>', replace)
return content
def operation_pb2locus(self, options, content):
start_page = 1
if options:
start_page = int(options[0])
self.rep_option = start_page
def replace(match):
# !!! ASSUME pb is not in <p> or anything else
number = re.sub(ur'^.*"([^"]+)".*$', ur'\1', match.group(1))
if len(number) == len(match.group(1)):
number = self.rep_option
ret = u'<p><span data-dpt="location" data-dpt-loctype="locus">%s</span></p>' % number
self.rep_option = get_int(number, default=self.rep_option) + 1
return ret
content = re_sub_fct(content, ur'<span\s+data-dpt\s*=\s*"pb"([^>]*)>', replace)
return content
