def findComponents(lines, container, components=[]):
"""return a list of (name, props, subcomponents) and remaining lines
# @@TODO: make this a generator
"""
props = []
subs = []
while 1:
try:
n, p, v = parseLine(lines.next(), downcase=False)
except StopIteration:
break
#print "finding...", n, p, v
#print >>sys.stderr, "began", container, n, p, v
if n == 'END':
# @@hmm... found an extra space after END:DAYLIGHT
# in test/20030410querymtg.ics
# where did that come from? allow it, or fix test data?
v = v.rstrip().upper()
if v != container:
raise ValueError, 'expected "%s" but found "%s"' % \
(container, v)
components.append((container, props, subs))
return
elif n == 'BEGIN':
findComponents(lines, v, subs)
else:
props.append((n, p, v))
def findComponents(lines, container, components=[]):
"""return a list of (name, props, subcomponents) and remaining lines
# @@TODO: make this a generator
"""
props = []
subs = []
while 1:
try:
n, p, v = parseLine(lines.next(), downcase=False)
except StopIteration:
break
# print "finding...", n, p, v
# print >>sys.stderr, "began", container, n, p, v
if n == "END":
# @@hmm... found an extra space after END:DAYLIGHT
# in test/20030410querymtg.ics
# where did that come from? allow it, or fix test data?
v = v.rstrip().upper()
if v != container:
raise ValueError, 'expected "%s" but found "%s"' % (container, v)
components.append((container, props, subs))
return
elif n == "BEGIN":
findComponents(lines, v, subs)
else:
props.append((n, p, v))
def _test():
import sys
from pprint import pprint
import doctest, fromIcal
doctest.testmod(fromIcal)
lines = unbreak(open(sys.argv[1]))
n, p, v = parseLine(lines.next())
c = []
findComponents(lines, v, c)
pprint(c)
def _test():
import sys
from pprint import pprint
import doctest, fromIcal
doctest.testmod(fromIcal)
lines = unbreak(open(sys.argv[1]))
n, p, v = parseLine(lines.next())
c = []
findComponents(lines, v, c)
pprint(c)
def interpret(sx, fp, base=None, suppressed=[]):
lines = unbreak(fp)
n, p, v = parseLine(lines.next())
if v != "VCALENDAR":
raise SyntaxError("Expected CALENDAR but found: %s" % (v))
calendars = []
findComponents(lines, v, calendars)
attrs = {}
RDF.bindAttr("rdf", attrs)
iCalendar.bindAttr("", attrs)
if base:
attrs["xml:base"] = base
sx.startElement("rdf:RDF", attrs)
doComponents(sx, calendars, iCalendarDefs, suppressed=suppressed)
sx.endElement("rdf:RDF")
def interpret(sx, fp, base=None, suppressed=[]):
lines = unbreak(fp)
n, p, v = parseLine(lines.next())
if v != 'VCALENDAR':
raise SyntaxError('Expected CALENDAR but found: %s' % (v))
calendars = []
findComponents(lines, v, calendars)
attrs = {}
RDF.bindAttr('rdf', attrs)
iCalendar.bindAttr('', attrs)
if base:
attrs['xml:base'] = base
sx.startElement('rdf:RDF', attrs)
doComponents(sx, calendars, iCalendarDefs, suppressed=suppressed)
sx.endElement('rdf:RDF')
def extract(path):
global nochange
global verbose
total = 0
wr( """# n3 http://www.w3.org/DesignIssues/Notation3.
# From vCard data in %s
# Extracted by $Id: ics2txt.py,v 1.1 2014-07-23 13:23:05 timbl Exp $
@prefix : <#>.
@prefix loc: <#loc_>.
@prefix s: <http://www.w3.org/2000/01/rdf-schema#> .
@prefix log: <http://www.w3.org/2000/10/swap/log#>.
@prefix v: <http://www.w3.org/2006/vcard/ns#>.
@prefix vc: <http://www.w3.org/2006/vcard/class#>.
@prefix abl: <http://www.w3.org/2006/vcard/abl#>.
@prefix user: <#>.
""" % path)
input = open(path, "r")
b = input.read()
input.close()
wr("# Length: " + `len(b)`+ "starts ")
wr(" ".join(["%2x"%ord(ch) for ch in b[:8]]))
wr("\n")
if ord(b[0])==0 and ord(b[1]) == ord('B'): # UTF16 with MSB byte order unmarked
d = "\xfe\xff" # Add byte order mark
buf = (d+b).decode('utf-16')
wr( " #Warning: UTF-16 was not byte order marked.\n")
else:
buf = b.decode('utf-8')
group_line = re.compile(r'^([a-zA-Z0-9]+)\.(.*)')
field_value = re.compile(r'^([A-Za-z0-9_-]*):(.*)')
group = None #
cardData = ""
groupData = {}
groupPred = {}
def readBareLine(buf):
"Return None for EOF or a line, even including a final widowed line"
global bufpo # can't pass by ref
begin = bufpo
if begin == len(buf): return None
bufpo = buf.find('\n', begin)
if bufpo < 0:
bufpo = len(buf)
line = buf[begin:]
else:
line = buf[begin:bufpo]
bufpo += 1 # After \n
while line [-1:] == "\r": line = line[:-1] # Strip triling CRs
return line
def startGroup(g):
# print "# Start group <%s>" % `g`
groupData[g] = [] # Pairs of p and o
groupPred[g] = "loc:%s" % munge(g) # Unless overwritten
def endGroup(g):
#print "# End group <%s> data:%s" % (`g`, groupData[g])
pos = groupData[g]
kludges, data = [], []
for i in range(len(pos)):
p, o = pos[i]
if p in kludgeTags: kludges.append((p,o))
else: data.append((p,o))
if len(data) == 1: # The AddressBook model, one data item + kludges
dp, do = data[0]
for p,o in kludges:
if p == 'v:x-ablabel':
dp = o # Override predicate
if p == 'v:x-abadr':
assert do[-1:] == ']';
do = do[:-1]+' '+ p +' '+ o +';]' # annnotate object
return "%s %s;\n" % (dp, do)
if len(kludges) != 0:
raise ValueError("Unknown Group pattern:"+`pos`)
res = ""
for p,o in pos:
res += " %s %s;" %(p,o)
return "%s [ # %s\n%s];\n" % (groupPred[g], g, res)
def lineFold(str1, str2):
x = str1.rfind('\n')
if x < 0: x = 0
if len(str1) - x + len(str2) > lineLength:
return str1+ "\n\t" + str2
return str1 + str2
def orderedFields(value, map):
cardData = ""
beg = 0
for i in range(len(map)):
end = beg
while 1:
end = value.find(";", end)
if end>0 and value[end-1] == "\\":
end += 1
continue
break
if end < 0:
end = len(value)
st = " ".join(splitBy(value[beg:end], ','))
if st: cardData = lineFold(cardData, ' v:%s %s;' % \
(map[i], stringToN3(st, singleLine=1)))
beg=end+1
if beg > len(value):
break
return cardData
def predicateObject(n, props, value):
"Return a pair of the predicate and object as N3 strings"
modifiers = ""
datatype = None
classes = []
for prop, val in props:
if prop == 'type':
vals = val.lower()
for val in splitBy(vals, ','):
if val == 'internet' and n == 'email':
pass
elif val == 'pref': # Preferred @@ - how represent?
pass
elif val in typeFields.get(n, []):
if relationshipModifiers.get(val, 0):
if modifiers: print "# @@ multiple modifiers in: "+line
modifiers = val + '-' + modifiers
else: classes.append('vc:'+val[0].upper()+val[1:])
else:
raise ValueError("Unhandled type %s in: %s" %(val, line))
elif prop == 'value': # This means datatype
datatype = val
if val == 'date':
pass # Date-times from AB certainly look like w3c not iCal dates
elif val == 'uri':
pass
else:
raise ValueError ('Unimplemented data type:'+val)
elif prop == 'base64' or (prop == 'encoding' and val.lower() == 'b'):
value = value.replace(' ','')
res = ""
while value:
res += value[:lineLength] + "\n"
value = value[lineLength:]
return 'v:'+n, '[ v:base64 """%s"""]\n' % (res) # Special case
else: raise ValueError('Unknown property %s with value %s' & (prop, val))
classSpec = ""
if classes: classSpec = 'a '+(', '.join(classes))
map = fieldProperties.get(n,None)
pred = 'v:%s%s' % (modifiers, n)
if map:
if classSpec: classSpec = '\n\t'+classSpec
if n == 'n': # Special case
assert classSpec == ""
return '', orderedFields(value, map) # Naked fields - see notes
return pred, '[' + orderedFields(value, map) + classSpec + ']'
if n == 'version':
assert value == "3.0", "value found: "+`value`
return "", ""
if n == 'x.ablabel':
return "", "" # used elsewhere
if n == 'categories': # Really should relate these to classes, but this roundtrips
obj = ", ".join(['"'+x+'"' for x in splitBy(value, ',')])
return pred, obj
unesc = splitBy(value, ';')
if len(unesc) != 1: raise ValueError("Unescaped semicolon in value: "+ value)
unesc = unesc[0]
obj = None
if n == 'tel':
if value[0] != '+':
print "# @@ Warning: not international form tel: "+value
obj = '<tel:%s>' % (value.replace(' ','-'))
elif n == 'url':
obj = '<%s>' % (value)
elif n == 'email':
obj = '<mailto:%s>' % (value)
if obj: # Any case so far is a form of URI
if classSpec: wr('%s %s.\n' %(obj, classSpec))
return pred, obj
elif n in singleTextField : # Single text
if classSpec:
raiseValueError("Unexpected class on %s: %s"%(n,`classSpec`))
return pred, stringToN3(unesc, singleLine=0) # @@@ N3 escaping
raise ValueError('Unknown tag:'+n)
global bufpo
bufpo = 0
nextLine = readBareLine(buf)
while 1:
line = nextLine
while 1:
nextLine = readBareLine(buf)
if not nextLine or nextLine[0] != ' ': break
line += nextLine[1:]
if line is None : break # EOF
# wr( "# line: " +line[:100])
m = group_line.match(line)
if m:
g = m.group(1)
line = m.group(2)
if group != g:
if group is not None: cardData += endGroup(group)
if g is not None: startGroup(g)
group = g
n, props, value = icslex.parseLine(line)
# for prop, val in props:
# if prop == 'type':
# val = val.lower()
# if group .startswith("item"): # AB hack
# groupPred[group] = "loc:"+val
if n == 'x-ablabel':
pred = 'v:'+n
if value[:4] == "_$!<" and value[-4:] == ">!$_": # [sic]
obj = "abl:"+munge(value[4:-4]).lower()
else: # User generated
obj = "user:"******""
cardID = "[]"
elif n == "uid":
cardID = "<uid:%s>" % value
elif n == 'end':
wr("%s %s." % (cardID, cardData))
else:
if n == 'n': # ugh special case
map = fieldProperties.get(n,None)
cardData += orderedFields(value, map) +'\n' # Naked fields - see notes
else:
p, o = predicateObject(n, props, value)
if p: cardData+= " %s %s;\n" %(p, o)
wr("\n\n#ends\n")
input.close()
def extract(path):
global nochange
global verbose
total = 0
wr("""# n3 http://www.w3.org/DesignIssues/Notation3.
# From vCard data in %s
# Extracted by $Id: ics2txt.py,v 1.1 2014-07-23 13:23:05 timbl Exp $
@prefix : <#>.
@prefix loc: <#loc_>.
@prefix s: <http://www.w3.org/2000/01/rdf-schema#> .
@prefix log: <http://www.w3.org/2000/10/swap/log#>.
@prefix v: <http://www.w3.org/2006/vcard/ns#>.
@prefix vc: <http://www.w3.org/2006/vcard/class#>.
@prefix abl: <http://www.w3.org/2006/vcard/abl#>.
@prefix user: <#>.
""" % path)
input = open(path, "r")
b = input.read()
input.close()
wr("# Length: " + ` len(b) ` + "starts ")
wr(" ".join(["%2x" % ord(ch) for ch in b[:8]]))
wr("\n")
if ord(b[0]) == 0 and ord(
b[1]) == ord('B'): # UTF16 with MSB byte order unmarked
d = "\xfe\xff" # Add byte order mark
buf = (d + b).decode('utf-16')
wr(" #Warning: UTF-16 was not byte order marked.\n")
else:
buf = b.decode('utf-8')
group_line = re.compile(r'^([a-zA-Z0-9]+)\.(.*)')
field_value = re.compile(r'^([A-Za-z0-9_-]*):(.*)')
group = None #
cardData = ""
groupData = {}
groupPred = {}
def readBareLine(buf):
"Return None for EOF or a line, even including a final widowed line"
global bufpo # can't pass by ref
begin = bufpo
if begin == len(buf): return None
bufpo = buf.find('\n', begin)
if bufpo < 0:
bufpo = len(buf)
line = buf[begin:]
else:
line = buf[begin:bufpo]
bufpo += 1 # After \n
while line[-1:] == "\r":
line = line[:-1] # Strip triling CRs
return line
def startGroup(g):
# print "# Start group <%s>" % `g`
groupData[g] = [] # Pairs of p and o
groupPred[g] = "loc:%s" % munge(g) # Unless overwritten
def endGroup(g):
#print "# End group <%s> data:%s" % (`g`, groupData[g])
pos = groupData[g]
kludges, data = [], []
for i in range(len(pos)):
p, o = pos[i]
if p in kludgeTags: kludges.append((p, o))
else: data.append((p, o))
if len(data) == 1: # The AddressBook model, one data item + kludges
dp, do = data[0]
for p, o in kludges:
if p == 'v:x-ablabel':
dp = o # Override predicate
if p == 'v:x-abadr':
assert do[-1:] == ']'
do = do[:-1] + ' ' + p + ' ' + o + ';]' # annnotate object
return "%s %s;\n" % (dp, do)
if len(kludges) != 0:
raise ValueError("Unknown Group pattern:" + ` pos `)
res = ""
for p, o in pos:
res += " %s %s;" % (p, o)
return "%s [ # %s\n%s];\n" % (groupPred[g], g, res)
def lineFold(str1, str2):
x = str1.rfind('\n')
if x < 0: x = 0
if len(str1) - x + len(str2) > lineLength:
return str1 + "\n\t" + str2
return str1 + str2
def orderedFields(value, map):
cardData = ""
beg = 0
for i in range(len(map)):
end = beg
while 1:
end = value.find(";", end)
if end > 0 and value[end - 1] == "\\":
end += 1
continue
break
if end < 0:
end = len(value)
st = " ".join(splitBy(value[beg:end], ','))
if st: cardData = lineFold(cardData, ' v:%s %s;' % \
(map[i], stringToN3(st, singleLine=1)))
beg = end + 1
if beg > len(value):
break
return cardData
def predicateObject(n, props, value):
"Return a pair of the predicate and object as N3 strings"
modifiers = ""
datatype = None
classes = []
for prop, val in props:
if prop == 'type':
vals = val.lower()
for val in splitBy(vals, ','):
if val == 'internet' and n == 'email':
pass
elif val == 'pref': # Preferred @@ - how represent?
pass
elif val in typeFields.get(n, []):
if relationshipModifiers.get(val, 0):
if modifiers:
print "# @@ multiple modifiers in: " + line
modifiers = val + '-' + modifiers
else:
classes.append('vc:' + val[0].upper() + val[1:])
else:
raise ValueError("Unhandled type %s in: %s" %
(val, line))
elif prop == 'value': # This means datatype
datatype = val
if val == 'date':
pass # Date-times from AB certainly look like w3c not iCal dates
elif val == 'uri':
pass
else:
raise ValueError('Unimplemented data type:' + val)
elif prop == 'base64' or (prop == 'encoding'
and val.lower() == 'b'):
value = value.replace(' ', '')
res = ""
while value:
res += value[:lineLength] + "\n"
value = value[lineLength:]
return 'v:' + n, '[ v:base64 """%s"""]\n' % (res
) # Special case
else:
raise ValueError('Unknown property %s with value %s'
& (prop, val))
classSpec = ""
if classes: classSpec = 'a ' + (', '.join(classes))
map = fieldProperties.get(n, None)
pred = 'v:%s%s' % (modifiers, n)
if map:
if classSpec: classSpec = '\n\t' + classSpec
if n == 'n': # Special case
assert classSpec == ""
return '', orderedFields(value,
map) # Naked fields - see notes
return pred, '[' + orderedFields(value, map) + classSpec + ']'
if n == 'version':
assert value == "3.0", "value found: " + ` value `
return "", ""
if n == 'x.ablabel':
return "", "" # used elsewhere
if n == 'categories': # Really should relate these to classes, but this roundtrips
obj = ", ".join(['"' + x + '"' for x in splitBy(value, ',')])
return pred, obj
unesc = splitBy(value, ';')
if len(unesc) != 1:
raise ValueError("Unescaped semicolon in value: " + value)
unesc = unesc[0]
obj = None
if n == 'tel':
if value[0] != '+':
print "# @@ Warning: not international form tel: " + value
obj = '<tel:%s>' % (value.replace(' ', '-'))
elif n == 'url':
obj = '<%s>' % (value)
elif n == 'email':
obj = '<mailto:%s>' % (value)
if obj: # Any case so far is a form of URI
if classSpec: wr('%s %s.\n' % (obj, classSpec))
return pred, obj
elif n in singleTextField: # Single text
if classSpec:
raiseValueError("Unexpected class on %s: %s" %
(n, ` classSpec `))
return pred, stringToN3(unesc, singleLine=0) # @@@ N3 escaping
raise ValueError('Unknown tag:' + n)
global bufpo
bufpo = 0
nextLine = readBareLine(buf)
while 1:
line = nextLine
while 1:
nextLine = readBareLine(buf)
if not nextLine or nextLine[0] != ' ': break
line += nextLine[1:]
if line is None: break # EOF
# wr( "# line: " +line[:100])
m = group_line.match(line)
if m:
g = m.group(1)
line = m.group(2)
if group != g:
if group is not None: cardData += endGroup(group)
if g is not None: startGroup(g)
group = g
n, props, value = icslex.parseLine(line)
# for prop, val in props:
# if prop == 'type':
# val = val.lower()
# if group .startswith("item"): # AB hack
# groupPred[group] = "loc:"+val
if n == 'x-ablabel':
pred = 'v:' + n
if value[:4] == "_$!<" and value[-4:] == ">!$_": # [sic]
obj = "abl:" + munge(value[4:-4]).lower()
else: # User generated
obj = "user:"******""
cardID = "[]"
elif n == "uid":
cardID = "<uid:%s>" % value
elif n == 'end':
wr("%s %s." % (cardID, cardData))
else:
if n == 'n': # ugh special case
map = fieldProperties.get(n, None)
cardData += orderedFields(
value, map) + '\n' # Naked fields - see notes
else:
p, o = predicateObject(n, props, value)
if p: cardData += " %s %s;\n" % (p, o)
wr("\n\n#ends\n")
input.close()