Hopiu/linkchecker
1
0
Fork 0
mirror of https://github.com/Hopiu/linkchecker.git synced 2026-05-13 09:03:11 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

resolve number entity refs in unhtmlify

Browse source 
git-svn-id: https://linkchecker.svn.sourceforge.net/svnroot/linkchecker/trunk/linkchecker@799 e7d03fd6-7b0d-0410-9947-9c21f3af8025
This commit is contained in:
calvin 2003-03-05 01:08:01 +00:00
parent fc82bc9362
commit 11b662858a
1 changed files with 22 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

26
linkcheck/StringUtil.py
View file

@ -20,11 +20,8 @@ import re, sys, htmlentitydefs
markup_re = re.compile("<.*?>", re.DOTALL)
entities = htmlentitydefs.entitydefs.items()
HtmlTable = map(lambda x: (x[1], "&"+x[0]+";"), entities)
UnHtmlTable = map(lambda x: ("&"+x[0]+";", x[1]), entities)
# order matters!
HtmlTable.sort()
UnHtmlTable.sort()
UnHtmlTable.reverse()
# standard xml entities
entities = {
'lt': '<',
@ -118,8 +115,29 @@ def htmlify (s):
return applyTable(HtmlTable, s)
is_charref = re.compile(r'(?i)x?(?P<num>\d+)').match
def resolve_entity (mo):
ent = mo.group(0).lower()
if htmlentitydefs.entitydefs.has_key(ent):
return htmlentitydefs.entitydefs[ent]
mo = is_charref(ent)
if mo:
# convert to number
num = mo.group("num")
if ent.startswith('x'):
radix = 16
else:
radix = 10
num = int(num, radix)
# check char range
if 0<=num<=255:
return chr(num)
return ent
def unhtmlify (s):
return applyTable(UnHtmlTable, s)
return re.sub(r'(?i)&(x?\d+|[a-z]+);', s, resolve_entity)
def xmlify (s):