mirror of https://github.com/mahmoud/boltons.git
thought better of the html2text behavior and reverted
This commit is contained in:
parent
2a5cc74a1e
commit
93ae692393
|
@ -544,8 +544,7 @@ class HTMLTextExtractor(HTMLParser):
|
|||
try:
|
||||
codepoint = htmlentitydefs.name2codepoint[name]
|
||||
except KeyError:
|
||||
# likely not a real entity (possibly an unescaped part of a URL)
|
||||
self.result.append(u'&' + name)
|
||||
self.result.append(u'&' + name + u';')
|
||||
else:
|
||||
self.result.append(unichr(codepoint))
|
||||
|
||||
|
|
Loading…
Reference in New Issue