def xml_utf8_sat(element, dom):
if element.xml_children:
element.pns_sat_articulated = []
return pns_sat.articulate_re(
' '.join(xml_utf8.xml_cdatas(element)),
element.pns_sat_articulated.append, element.pns_sat_language
or dom.pns_sat_language)
elif element.xml_first:
element.pns_sat_articulated = []
return pns_sat.articulate_re(
element.xml_first, element.pns_sat_articulated.append,
element.pns_sat_language or dom.pns_sat_language)
def xml_utf8_sat (element, dom):
if element.xml_children:
element.pns_sat_articulated = []
return pns_sat.articulate_re (
' '.join (xml_utf8.xml_cdatas (element)),
element.pns_sat_articulated.append,
element.pns_sat_language or dom.pns_sat_language
)
elif element.xml_first:
element.pns_sat_articulated = []
return pns_sat.articulate_re (
element.xml_first,
element.pns_sat_articulated.append,
element.pns_sat_language or dom.pns_sat_language
)
def articulate_headers (
headers, context, statement, lang,
names=('subject', ), predicates=('from', 'to', 'date', )
):
articulated = []
subject = pns_model.pns_name (netstring.encode ((
pns_sat.articulate_re (
headers.get (n), articulated.append, lang
) for n in names
)))
for predicate in predicates:
object = headers.get (predicate)
if object == None:
continue
statement ((subject, predicate, object), )
# Note about this implementation
#
# This module implements a minimal MIME collector and UTF-8 transcoder
# for PNS/SAT articulator, something practical to articulate plain text
# found in USENET and e-mail messages.
#
# The purpose of pns_mime is to support the Enron mail and comp.lang.python
# test cases, indexing all mail made available by the famous case and
# do the same for as much as comp.lang.python is available on USENET.
#
# The CLI is a pipe that collects a stream of MIME messages from STDIN and
# pipes out articulated PNS statements on STDOUT:
#
# pns_mime < mime 1> pns
#
# or if provided the proper arguments:
#
# pns_mime 127.0.0.1 3534 < mime
#
# to a PNS/TCP metabase.
# <mime>
# <id/>
# <subject/>
# <from/>
# <to/>
# <content-type/>
# </mime>
#
def articulate_headers(headers,
context,
statement,
lang,
names=('subject', ),
predicates=(
'from',
'to',
'date',
)):
articulated = []
subject = pns_model.pns_name(
netstring.encode((pns_sat.articulate_re(headers.get(n),
articulated.append, lang)
for n in names)))
for predicate in predicates:
object = headers.get(predicate)
if object == None:
continue
statement((subject, predicate, object), )
# -*- coding: CP1252 -*-
from allegra import netstring, loginfo, pns_sat
def log_articulated (articulated):
for horizon, field, name, text in articulated:
loginfo.log (text)
loginfo.log (netstring.netoutline (name))
articulated = []
text = (
'A just machine to make big decisions\n'
'Programmed by fellows with compassion and vision'
)
name = pns_sat.articulate_re (
text, articulated.append, pns_sat.language ('en')
)
log_articulated (articulated)
articulated = []
text = (
"Ce qui se conçoit bien s'énonce clairement\n"
"Et les mots pour le dire arrivent aisément"
)
name = pns_sat.articulate_re (
text, articulated.append, pns_sat.language ('fr')
)
log_articulated (articulated)
articulated = []
text = (
