1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
|
-- Test utf routines
context("UTF8 check functions", function()
local ffi = require("ffi")
ffi.cdef [[
unsigned int rspamd_str_lc_utf8 (char *str, unsigned int size);
unsigned int rspamd_str_lc (char *str, unsigned int size);
void rspamd_fast_utf8_library_init (unsigned flags);
void ottery_rand_bytes(void *buf, size_t n);
double rspamd_get_ticks(int allow);
size_t rspamd_fast_utf8_validate (const unsigned char *data, size_t len);
size_t rspamd_fast_utf8_validate_ref (const unsigned char *data, size_t len);
char * rspamd_str_make_utf_valid (const char *src, size_t slen, size_t *dstlen, void *);
]]
local cases = {
{ "АбЫрвАлг", "абырвалг" },
{ "АAБBвc", "аaбbвc" },
--{"STRASSE", "straße"}, XXX: NYI
{ "KEÇİ", "keçi" },
}
for i, c in ipairs(cases) do
test("UTF lowercase " .. tostring(i), function()
local buf = ffi.new("char[?]", #c[1] + 1)
ffi.copy(buf, c[1])
local nlen = ffi.C.rspamd_str_lc_utf8(buf, #c[1])
local s = ffi.string(buf, nlen)
assert_equal(s, c[2])
end)
end
cases = {
{ "AbCdEf", "abcdef" },
{ "A", "a" },
{ "AaAa", "aaaa" },
{ "AaAaAaAa", "aaaaaaaa" }
}
for i, c in ipairs(cases) do
test("ASCII lowercase " .. tostring(i), function()
local buf = ffi.new("char[?]", #c[1] + 1)
ffi.copy(buf, c[1])
ffi.C.rspamd_str_lc(buf, #c[1])
local s = ffi.string(buf)
assert_equal(s, c[2])
end)
end
cases = {
{ 'тест', 'тест' },
{ '\200\213\202', '���' },
{ 'тест\200\213\202test', 'тест���test' },
{ '\200\213\202test', '���test' },
{ '\200\213\202test\200\213\202', '���test���' },
{ 'тест\200\213\202test\200\213\202', 'тест���test���' },
{ 'тест\200\213\202test\200\213\202тест', 'тест���test���тест' },
}
local NULL = ffi.new 'void*'
for i, c in ipairs(cases) do
test("Unicode make valid " .. tostring(i), function()
local buf = ffi.new("char[?]", #c[1] + 1)
ffi.copy(buf, c[1])
local s = ffi.string(ffi.C.rspamd_str_make_utf_valid(buf, #c[1], NULL, NULL))
local function to_hex(s)
return (s:gsub('.', function(c)
return string.format('%02X', string.byte(c))
end))
end
print(to_hex(s))
print(to_hex(c[2]))
assert_equal(s, c[2])
end)
end
-- Enable sse and avx2
ffi.C.rspamd_fast_utf8_library_init(3)
local valid_cases = {
"a",
"\xc3\xb1",
"\xe2\x82\xa1",
"\xf0\x90\x8c\xbc",
"안녕하세요, 세상"
}
for i, c in ipairs(valid_cases) do
test("Unicode validate success: " .. tostring(i), function()
local buf = ffi.new("char[?]", #c + 1)
ffi.copy(buf, c)
local ret = ffi.C.rspamd_fast_utf8_validate(buf, #c)
assert_equal(ret, 0)
end)
end
local invalid_cases = {
"\xc3\x28",
"\xa0\xa1",
"\xe2\x28\xa1",
"\xe2\x82\x28",
"\xf0\x28\x8c\xbc",
"\xf0\x90\x28\xbc",
"\xf0\x28\x8c\x28",
"\xc0\x9f",
"\xf5\xff\xff\xff",
"\xed\xa0\x81",
"\xf8\x90\x80\x80\x80",
"123456789012345\xed",
"123456789012345\xf1",
"123456789012345\xc2",
"\xC2\x7F"
}
for i, c in ipairs(invalid_cases) do
test("Unicode validate fail: " .. tostring(i), function()
local buf = ffi.new("char[?]", #c + 1)
ffi.copy(buf, c)
local ret = ffi.C.rspamd_fast_utf8_validate(buf, #c)
assert_not_equal(ret, 0)
end)
end
if os.getenv("RSPAMD_LUA_EXPENSIVE_TESTS") then
local speed_iters = 10000
local function test_size(buflen, is_valid, impl)
local logger = require "rspamd_logger"
local test_str
if is_valid then
test_str = table.concat(valid_cases)
else
test_str = table.concat(valid_cases) .. table.concat(invalid_cases)
end
local buf = ffi.new("char[?]", buflen)
if #test_str < buflen then
local t = {}
local len = #test_str
while len < buflen do
t[#t + 1] = test_str
len = len + #test_str
end
test_str = table.concat(t)
end
ffi.copy(buf, test_str:sub(1, buflen))
local tm = 0
for _ = 1, speed_iters do
if impl == 'ref' then
local t1 = ffi.C.rspamd_get_ticks(1)
ffi.C.rspamd_fast_utf8_validate_ref(buf, buflen)
local t2 = ffi.C.rspamd_get_ticks(1)
tm = tm + (t2 - t1)
elseif impl == 'opt' then
local t1 = ffi.C.rspamd_get_ticks(1)
ffi.C.rspamd_fast_utf8_validate(buf, buflen)
local t2 = ffi.C.rspamd_get_ticks(1)
tm = tm + (t2 - t1)
end
end
logger.messagex("%s utf8 %s check (valid = %s): %s ticks per iter, %s ticks per byte",
impl, buflen, is_valid,
tm / speed_iters, tm / speed_iters / buflen)
return 0
end
for _, sz in ipairs({ 78, 512, 65535 }) do
test(string.format("Utf8 test %s %d buffer, %s", 'ref', sz, 'valid'), function()
local res = test_size(sz, true, 'ref')
assert_equal(res, 0)
end)
test(string.format("Utf8 test %s %d buffer, %s", 'ref', sz, 'invalid'), function()
local res = test_size(sz, false, 'ref')
assert_equal(res, 0)
end)
test(string.format("Utf8 test %s %d buffer, %s", 'opt', sz, 'valid'), function()
local res = test_size(sz, true, 'opt')
assert_equal(res, 0)
end)
test(string.format("Utf8 test %s %d buffer, %s", 'opt', sz, 'invalid'), function()
local res = test_size(sz, false, 'opt')
assert_equal(res, 0)
end)
end
end
end)
|
