Commit 2e04a40b authored by Jim Fulton's avatar Jim Fulton

Fixed to allow list elements to introduce examples.

Restructured _str using continue to avoid excessive nesting.
parent b8da5880
......@@ -170,7 +170,7 @@ Special symbology is used to indicate special constructs:
Together with the previous rule this allows easy coding of references or
end notes.
$Id: StructuredText.py,v 1.22 1999/08/02 22:01:28 jim Exp $'''
$Id: StructuredText.py,v 1.23 1999/08/03 20:49:05 jim Exp $'''
# Copyright
#
# Copyright 1996 Digital Creations, L.C., 910 Princess Anne
......@@ -222,6 +222,11 @@ $Id: StructuredText.py,v 1.22 1999/08/02 22:01:28 jim Exp $'''
# (540) 371-6909
#
# $Log: StructuredText.py,v $
# Revision 1.23 1999/08/03 20:49:05 jim
# Fixed to allow list elements to introduce examples.
#
# Restructured _str using continue to avoid excessive nesting.
#
# Revision 1.22 1999/08/02 22:01:28 jim
# Fixed a bunch of bugs introduced by making ts_regex actually thread
# safe.
......@@ -598,44 +603,49 @@ class HTML(StructuredText):
):
r=''
for s in structure:
# print s[0],'\n', len(s[1]), '\n\n'
ts_results = bullet(s[0], (1,))
if ts_results:
p = ts_results[1]
r=self.ul(r,p,self._str(s[1],level))
if s[0][-2:]=='::' and s[1]: ps=self.pre(s[1])
else: ps=self._str(s[1],level)
r=self.ul(r,p,ps)
continue
ts_results = ol(s[0], (3,))
if ts_results:
p = ts_results[1]
if s[0][-2:]=='::' and s[1]: ps=self.pre(s[1])
else: ps=self._str(s[1],level)
r=self.ol(r,p,ps)
continue
ts_results = olp(s[0], (1,))
if ts_results:
p = ts_results[1]
if s[0][-2:]=='::' and s[1]: ps=self.pre(s[1])
else: ps=self._str(s[1],level)
r=self.ol(r,p,ps)
continue
ts_results = dl(s[0], (1,2))
if ts_results:
t,d = ts_results[1]
r=self.dl(r,t,d,self._str(s[1],level))
continue
if example(s[0]) >= 0 and s[1]:
# Introduce an example, using pre tags:
r=self.normal(r,s[0],self.pre(s[1]))
continue
if s[0][-2:]=='::' and s[1]:
# Introduce an example, using pre tags:
r=self.normal(r,s[0][:-1],self.pre(s[1]))
else:
ts_results = ol(s[0], (3,))
if ts_results:
p = ts_results[1]
r=self.ol(r,p,self._str(s[1],level))
if nl(s[0]) < 0 and s[1] and s[0][-1:] != ':':
# Treat as a heading
t=s[0]
r=self.head(r,t,level,
self._str(s[1],level and level+1))
else:
ts_results = olp(s[0], (1,))
if ts_results:
p = ts_results[1]
r=self.ol(r,p,self._str(s[1],level))
else:
ts_results = dl(s[0], (1,2))
if ts_results:
t,d = ts_results[1]
r=self.dl(r,t,d,self._str(s[1],level))
else:
if example(s[0]) >= 0 and s[1]:
# Introduce an example, using pre tags:
r=self.normal(r,s[0],self.pre(s[1]))
else:
if s[0][-2:]=='::' and s[1]:
# Introduce an example, using pre tags:
r=self.normal(r,s[0][:-1],self.pre(s[1]))
else:
if nl(s[0]) < 0 and s[1] and s[0][-1:] != ':':
# Treat as a heading
t=s[0]
r=self.head(r,t,level,
self._str(s[1],level and level+1))
else:
r=self.normal(r,s[0],self._str(s[1],level))
r=self.normal(r,s[0],self._str(s[1],level))
return r
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment