def unescape(result: MarkedUpText) -> None:
"""
a "private" method to replace HTML codes like > with corresponding symbols in
the resulting plain text
:param result: MarkedUpText containing resulting plain text
"""
new_text = ''
transformations = [
] # type: List[Tuple[Tuple[int, int], Tuple[int, int]]]
last_stop = 0
for match in _charref.finditer(result.text):
replacement = _replace_charref(match)
src_s, src_e = (match.start(), match.end())
end_e = src_s + len(replacement)
if end_e != src_e:
transformations.append(((src_s, src_e), (src_s, end_e)))
new_text += result.text[last_stop:src_s]
new_text += replacement
last_stop = src_e
new_text += result.text[last_stop:len(result.text)]
result.text = new_text
if transformations:
result.apply_transformations(transformations)
def convert_to_styled(self, data):
"""
This scans incoming notes for possible html. It converts a select few
tags into StyledText and removes the rest of the tags. Notes of this
type occur in data from FTM and ancestry.com. Result is a much
cleaner note.
@param data: a string of text possibly containg html
@type data: str
"""
prev = 0
chunkpos = 0
chunks = []
italics = []
bolds = []
unders = []
links = []
reds = []
bldpos = -1
# data = html.unescape(data) # clean up escaped html "<" etc.
for mo in re.finditer(html._charref, data._string):
out = html._replace_charref(mo)
in_start = mo.start()
in_end = mo.end()
data._string = (data._string[:in_start] + out +
data._string[(in_start + len(out)):])
if prev != in_start + len(out):
chunks.append(data[prev:(in_start + len(out))])
chunkpos += (in_start - prev + len(out))
prev = in_end
chunks.append(data[prev:])
data = StyledText().join(chunks)
prev = 0
chunkpos = 0
chunks = []
for mo in re.finditer(self.tok_regex,
data._string,
flags=(re.DOTALL | re.I)):
kind = mo.lastgroup
st_txt = mo.group(kind)
in_start = mo.start()
in_end = mo.end()
if kind == 'SKIP' or kind == 'TABLE':
if prev != in_start:
chunks.append(data[prev:in_start])
chunkpos += (in_start - prev)
elif kind == 'PARAEND':
chunks.append(data[prev:in_start] + '\n')
chunkpos += (in_start - prev + 1)
elif kind == 'ITALIC':
chunks.append(data[prev:in_start] +
data[(in_start + 3):in_end])
newpos = chunkpos - prev + in_end - 3
italics.append((chunkpos + in_start - prev, newpos))
chunkpos = newpos
elif kind == 'BOLD':
chunks.append(data[prev:in_start] +
data[(in_start + 3):in_end])
newpos = chunkpos - prev + in_end - 3
bolds.append((chunkpos + in_start - prev, newpos))
chunkpos = newpos
elif kind == 'UNDER':
chunks.append(data[prev:in_start] +
data[(in_start + 3):in_end])
newpos = chunkpos - prev + in_end - 3
unders.append((chunkpos + in_start - prev, newpos))
chunkpos = newpos
elif kind == 'HTTP': # HTTP found
st_txt = mo.group('HTTP')
oldpos = chunkpos + in_start - prev
chunks.append(data[prev:in_start] + st_txt)
chunkpos += (in_start - prev + len(st_txt))
st_txt = st_txt.rstrip(' .:)')
newpos = oldpos + len(st_txt)
links.append((st_txt, oldpos, newpos))
elif kind == 'HREF': # HREF found
st_txt = mo.group('HREFT')
lk_txt = mo.group('HREFL')
# fix up relative links emmitted by ancestry.com
if (lk_txt.startswith("/search/dbextra")
or lk_txt.startswith("/handler/domain")):
lk_txt = "http://search.ancestry.com" + lk_txt
oldpos = chunkpos + in_start - prev
# if tag (minus any trailing '.') is substring of link
if st_txt[0:-1] in lk_txt:
st_txt = lk_txt # just use the link
else: # use link and tag
st_txt = " " + lk_txt + " (" + st_txt + ")"
newpos = oldpos + len(st_txt)
chunks.append(data[prev:in_start] + st_txt)
chunkpos += (in_start - prev + len(st_txt))
links.append((lk_txt, oldpos, newpos))
elif kind == 'TBLCELL' or kind == 'TBLHDRC': # Table cell break
chunks.append(data[prev:in_start] + ': ')
chunkpos += (in_start - prev + 3)
elif kind == 'TBLHDRB': # header start
if prev != in_start:
chunks.append(data[prev:in_start])
chunkpos += (in_start - prev)
bldpos = chunkpos
elif kind == 'TBLHDRE': # Header end
if bldpos == -1:
if prev != in_start:
chunks.append(data[prev:in_end])
newpos = chunkpos - prev + in_end
reds.append((chunkpos + in_start - prev, newpos))
chunkpos = newpos
print('Invalid table header, no start tag found')
else:
if prev != in_start:
chunks.append(data[prev:in_start])
chunkpos += (in_start - prev)
bolds.append((bldpos, chunkpos))
bldpos = -1
elif kind == 'UNKNWN':
chunks.append(data[prev:in_end])
newpos = chunkpos - prev + in_end
reds.append((chunkpos + in_start - prev, newpos))
chunkpos = newpos
print('Unexpected or unimplemented HTML tag', st_txt)
else:
print("shouldn't get here")
prev = in_end
chunks.append(data[prev:])
result = StyledText().join(chunks)
tags = []
for link in links:
tags.append(
StyledTextTag(StyledTextTagType.LINK, link[0],
[(link[1], link[2])]))
if italics:
tags.append(StyledTextTag(StyledTextTagType.ITALIC, False,
italics))
if bolds:
tags.append(StyledTextTag(StyledTextTagType.BOLD, False, bolds))
if unders:
tags.append(
StyledTextTag(StyledTextTagType.UNDERLINE, False, unders))
if reds:
tags.append(
StyledTextTag(StyledTextTagType.HIGHLIGHT, '#FFFF00', reds))
return StyledText(result._string, tag_merge(result._tags, tags))