def articulate_re(text, articulate, articulators, depth=0):
"Articulate text using a simple regular expression lexer"
bottom = len(articulators)
# move down the stack until a pattern is matched ...
while True:
texts = [t for t in articulators[depth].split(text) if t]
depth += 1
if len(texts) > 1:
break
if depth == bottom:
# no match found, bottom of the stack reached.
return text
field = set()
if depth == bottom:
# bottom of the stack reached, simply split
name = pns_model.pns_name(netstring.encode(texts), field)
else:
# not yet at the bottom of the stack, recurse ...
name = pns_model.pns_name(
netstring.encode((articulate_re(t, articulate, articulators, depth) for t in texts)), field
)
# validate the articulated name(s) as a Public Names
if len(field) > 1:
articulate((len(field), field, name, text))
return name
def articulate_chunk(text, articulate, articulators, CHUNK, depth=0):
"Articulate in chunks, usefull for larger text articulations"
bottom = len(articulators)
# move down the stack until a pattern is matched ...
while True:
texts = [t for t in articulators[depth].split(text) if t]
depth += 1
if len(texts) > 1:
break
if depth == bottom:
# no match found, bottom of the stack reached, this
# is supposed to chunk, not articulate, do not name
# because it is inarticulated junk ...
return ""
if depth == bottom:
# bottom of the stack reached, simply split do not articulate
# any further, there is nothing to chunk and it is a flat
# articulation ...
return pns_model.pns_name(netstring.encode((t for t in texts if len(t) <= CHUNK)), set())
# not yet at the bottom of the stack, recurse
for t in texts:
if len(t) > CHUNK:
articulate_chunk(t, articulate, articulators, CHUNK, depth)
else:
articulate_re(t, articulate, articulators, depth)
return None
def xml_utf8_name (element, dom):
# articulate a context's name one from its children's name(s)
field = set ()
return pns_model.pns_name (netstring.encode ((
child.pns_name
for child in element.xml_children
if child.pns_name
)), field)
from allegra import netstring, pns_model
context = '7:Rushing,3:Sam,'
print context == pns_model.pns_name (context, set ())
context = '3:Sam,7:Rushing,'
print context == pns_model.pns_name (context, set ())
context = '4:Kill,4:Kill,8:Pussycat,'
print context == pns_model.pns_name (context, set ())
field = set ()
context = netstring.netstrings ((
'An', 'example', 'of', (
'Simple', 'Articulated', 'Text'
), 'and', 'a', 'bit', 'of', 'dispersion'
))
print netstring.netoutline (context)
name = pns_model.pns_name (context, field)
print netstring.netoutline (name)
print 'field: %r' % field
print 'horizon: %d' % len (field)
def xml_utf8_name(element, dom):
# articulate a context's name one from its children's name(s)
field = set()
return pns_model.pns_name(
netstring.encode((child.pns_name for child in element.xml_children
if child.pns_name)), field)
def pns_articulate_route (encoded):
names = netstring.netlist (encoded)
name = pns_model.pns_name (netstring.encode (names[1:]), set ())
return (names[0], name)
