summaryrefslogtreecommitdiffstats
path: root/Tools/unicode/mkstringprep.py
diff options
context:
space:
mode:
Diffstat (limited to 'Tools/unicode/mkstringprep.py')
-rw-r--r--Tools/unicode/mkstringprep.py137
1 files changed, 67 insertions, 70 deletions
diff --git a/Tools/unicode/mkstringprep.py b/Tools/unicode/mkstringprep.py
index 4271883..99fcae7 100644
--- a/Tools/unicode/mkstringprep.py
+++ b/Tools/unicode/mkstringprep.py
@@ -1,14 +1,16 @@
-import re
-from unicodedata import ucd_3_2_0 as unicodedata
+import re, unicodedata, sys
+
+if sys.maxunicode == 65535:
+ raise RuntimeError("need UCS-4 Python")
def gen_category(cats):
for i in range(0, 0x110000):
- if unicodedata.category(chr(i)) in cats:
+ if unicodedata.category(unichr(i)) in cats:
yield(i)
def gen_bidirectional(cats):
for i in range(0, 0x110000):
- if unicodedata.bidirectional(chr(i)) in cats:
+ if unicodedata.bidirectional(unichr(i)) in cats:
yield(i)
def compact_set(l):
@@ -35,20 +37,16 @@ def compact_set(l):
tuple.append((prev,prev+span+1))
else:
single.append(prev)
- if not single and len(tuple) == 1:
- tuple = "range(%d,%d)" % tuple[0]
- else:
- tuple = " + ".join("list(range(%d,%d))" % t for t in tuple)
+ tuple = " + ".join(["range(%d,%d)" % t for t in tuple])
if not single:
return "set(%s)" % tuple
if not tuple:
- return "set(%r)" % (single,)
- return "set(%r + %s)" % (single, tuple)
+ return "set(%s)" % repr(single)
+ return "set(%s + %s)" % (repr(single),tuple)
############## Read the tables in the RFC #######################
-with open("rfc3454.txt") as f:
- data = f.readlines()
+data = open("rfc3454.txt").readlines()
tables = []
curname = None
@@ -57,7 +55,8 @@ for l in data:
if not l:
continue
# Skip RFC page breaks
- if l.startswith(("Hoffman & Blanchet", "RFC 3454")):
+ if l.startswith("Hoffman & Blanchet") or\
+ l.startswith("RFC 3454"):
continue
# Find start/end lines
m = re.match("----- (Start|End) Table ([A-Z](.[0-9])+) -----", l)
@@ -72,8 +71,6 @@ for l in data:
else:
if not curname:
raise RuntimeError("End without start", l)
- if curname != m.group(2):
- raise RuntimeError("Unexpected end", l)
curname = None
continue
if not curname:
@@ -109,17 +106,17 @@ for l in data:
########### Generate compact Python versions of the tables #############
-print("""# This file is generated by mkstringprep.py. DO NOT EDIT.
+print """# This file is generated by mkstringprep.py. DO NOT EDIT.
\"\"\"Library that exposes various tables found in the StringPrep RFC 3454.
There are two kinds of tables: sets, for which a member test is provided,
and mappings, for which a mapping function is provided.
\"\"\"
-from unicodedata import ucd_3_2_0 as unicodedata
-""")
+import unicodedata
+"""
-print("assert unicodedata.unidata_version == %r" % (unicodedata.unidata_version,))
+print "assert unicodedata.unidata_version == %s" % repr(unicodedata.unidata_version)
# A.1 is the table of unassigned characters
# XXX Plane 15 PUA is listed as unassigned in Python.
@@ -137,24 +134,24 @@ Cn -= set(range(0xFFFF, 0x110000, 0x10000))
# assert table == Cn
-print("""
+print """
def in_table_a1(code):
if unicodedata.category(code) != 'Cn': return False
c = ord(code)
if 0xFDD0 <= c < 0xFDF0: return False
return (c & 0xFFFF) not in (0xFFFE, 0xFFFF)
-""")
+"""
# B.1 cannot easily be derived
name, table = tables[0]
del tables[0]
assert name == "B.1"
table = sorted(table.keys())
-print("""
+print """
b1_set = """ + compact_set(table) + """
def in_table_b1(code):
return ord(code) in b1_set
-""")
+"""
# B.2 and B.3 is case folding.
# It takes CaseFolding.txt into account, which is
@@ -176,25 +173,25 @@ assert name == "B.3"
b3_exceptions = {}
for k,v in table_b2.items():
- if list(map(ord, chr(k).lower())) != v:
- b3_exceptions[k] = "".join(map(chr,v))
+ if map(ord, unichr(k).lower()) != v:
+ b3_exceptions[k] = u"".join(map(unichr,v))
b3 = sorted(b3_exceptions.items())
-print("""
-b3_exceptions = {""")
-for i, kv in enumerate(b3):
- print("0x%x:%a," % kv, end=' ')
+print """
+b3_exceptions = {"""
+for i,(k,v) in enumerate(b3):
+ print "0x%x:%s," % (k, repr(v)),
if i % 4 == 3:
- print()
-print("}")
+ print
+print "}"
-print("""
+print """
def map_table_b3(code):
r = b3_exceptions.get(ord(code))
if r is not None: return r
return code.lower()
-""")
+"""
def map_table_b3(code):
r = b3_exceptions.get(ord(code))
@@ -208,7 +205,7 @@ def map_table_b3(code):
def map_table_b2(a):
al = map_table_b3(a)
b = unicodedata.normalize("NFKC", al)
- bl = "".join([map_table_b3(ch) for ch in b])
+ bl = u"".join([map_table_b3(ch) for ch in b])
c = unicodedata.normalize("NFKC", bl)
if b != c:
return c
@@ -217,23 +214,23 @@ def map_table_b2(a):
specials = {}
for k,v in table_b2.items():
- if list(map(ord, map_table_b2(chr(k)))) != v:
+ if map(ord, map_table_b2(unichr(k))) != v:
specials[k] = v
# B.3 should not add any additional special cases
assert specials == {}
-print("""
+print """
def map_table_b2(a):
al = map_table_b3(a)
b = unicodedata.normalize("NFKC", al)
- bl = "".join([map_table_b3(ch) for ch in b])
+ bl = u"".join([map_table_b3(ch) for ch in b])
c = unicodedata.normalize("NFKC", bl)
if b != c:
return c
else:
return al
-""")
+"""
# C.1.1 is a table with a single character
name, table = tables[0]
@@ -241,10 +238,10 @@ del tables[0]
assert name == "C.1.1"
assert table == {0x20:0x20}
-print("""
+print """
def in_table_c11(code):
- return code == " "
-""")
+ return code == u" "
+"""
# C.1.2 is the rest of all space characters
name, table = tables[0]
@@ -252,16 +249,16 @@ del tables[0]
assert name == "C.1.2"
# table = set(table.keys())
-# Zs = set(gen_category(["Zs"])) - {0x20}
+# Zs = set(gen_category(["Zs"])) - set([0x20])
# assert Zs == table
-print("""
+print """
def in_table_c12(code):
- return unicodedata.category(code) == "Zs" and code != " "
+ return unicodedata.category(code) == "Zs" and code != u" "
def in_table_c11_c12(code):
return unicodedata.category(code) == "Zs"
-""")
+"""
# C.2.1 ASCII control characters
name, table_c21 = tables[0]
@@ -273,10 +270,10 @@ Cc_ascii = Cc & set(range(128))
table_c21 = set(table_c21.keys())
assert Cc_ascii == table_c21
-print("""
+print """
def in_table_c21(code):
return ord(code) < 128 and unicodedata.category(code) == "Cc"
-""")
+"""
# C.2.2 Non-ASCII control characters. It also includes
# a number of characters in category Cf.
@@ -291,7 +288,7 @@ assert len(Cc_nonascii - table_c22) == 0
specials = list(table_c22 - Cc_nonascii)
specials.sort()
-print("""c22_specials = """ + compact_set(specials) + """
+print """c22_specials = """ + compact_set(specials) + """
def in_table_c22(code):
c = ord(code)
if c < 128: return False
@@ -301,7 +298,7 @@ def in_table_c22(code):
def in_table_c21_c22(code):
return unicodedata.category(code) == "Cc" or \\
ord(code) in c22_specials
-""")
+"""
# C.3 Private use
name, table = tables[0]
@@ -311,10 +308,10 @@ assert name == "C.3"
Co = set(gen_category(["Co"]))
assert set(table.keys()) == Co
-print("""
+print """
def in_table_c3(code):
return unicodedata.category(code) == "Co"
-""")
+"""
# C.4 Non-character code points, xFFFE, xFFFF
# plus process internal codes
@@ -322,19 +319,19 @@ name, table = tables[0]
del tables[0]
assert name == "C.4"
-nonchar = set(range(0xFDD0,0xFDF0))
-nonchar.update(range(0xFFFE,0x110000,0x10000))
-nonchar.update(range(0xFFFF,0x110000,0x10000))
+nonchar = set(range(0xFDD0,0xFDF0) +
+ range(0xFFFE,0x110000,0x10000) +
+ range(0xFFFF,0x110000,0x10000))
table = set(table.keys())
assert table == nonchar
-print("""
+print """
def in_table_c4(code):
c = ord(code)
if c < 0xFDD0: return False
if c < 0xFDF0: return True
return (ord(code) & 0xFFFF) in (0xFFFE, 0xFFFF)
-""")
+"""
# C.5 Surrogate codes
name, table = tables[0]
@@ -344,10 +341,10 @@ assert name == "C.5"
Cs = set(gen_category(["Cs"]))
assert set(table.keys()) == Cs
-print("""
+print """
def in_table_c5(code):
return unicodedata.category(code) == "Cs"
-""")
+"""
# C.6 Inappropriate for plain text
name, table = tables[0]
@@ -356,11 +353,11 @@ assert name == "C.6"
table = sorted(table.keys())
-print("""
+print """
c6_set = """ + compact_set(table) + """
def in_table_c6(code):
return ord(code) in c6_set
-""")
+"""
# C.7 Inappropriate for canonical representation
name, table = tables[0]
@@ -369,11 +366,11 @@ assert name == "C.7"
table = sorted(table.keys())
-print("""
+print """
c7_set = """ + compact_set(table) + """
def in_table_c7(code):
return ord(code) in c7_set
-""")
+"""
# C.8 Change display properties or are deprecated
name, table = tables[0]
@@ -382,11 +379,11 @@ assert name == "C.8"
table = sorted(table.keys())
-print("""
+print """
c8_set = """ + compact_set(table) + """
def in_table_c8(code):
return ord(code) in c8_set
-""")
+"""
# C.9 Tagging characters
name, table = tables[0]
@@ -395,11 +392,11 @@ assert name == "C.9"
table = sorted(table.keys())
-print("""
+print """
c9_set = """ + compact_set(table) + """
def in_table_c9(code):
return ord(code) in c9_set
-""")
+"""
# D.1 Characters with bidirectional property "R" or "AL"
name, table = tables[0]
@@ -409,10 +406,10 @@ assert name == "D.1"
RandAL = set(gen_bidirectional(["R","AL"]))
assert set(table.keys()) == RandAL
-print("""
+print """
def in_table_d1(code):
return unicodedata.bidirectional(code) in ("R","AL")
-""")
+"""
# D.2 Characters with bidirectional property "L"
name, table = tables[0]
@@ -422,7 +419,7 @@ assert name == "D.2"
L = set(gen_bidirectional(["L"]))
assert set(table.keys()) == L
-print("""
+print """
def in_table_d2(code):
return unicodedata.bidirectional(code) == "L"
-""")
+"""
generated by cgit v0.12 at 2025-09-15 00:07:50 (GMT)