Skip to content

Commit

Permalink
Implement utf8 library
Browse files Browse the repository at this point in the history
Run pattern-matching and UTF-8 tests from Lua test suite.

Updates #74
  • Loading branch information
zombiezen committed Feb 12, 2025
1 parent 3b0061d commit 938ba2f
Show file tree
Hide file tree
Showing 6 changed files with 856 additions and 58 deletions.
2 changes: 1 addition & 1 deletion internal/lua/auxlib.go
Original file line number Diff line number Diff line change
Expand Up @@ -438,9 +438,9 @@ func OpenLibraries(ctx context.Context, l *State) error {
{TableLibraryName, OpenTable},
{StringLibraryName, OpenString},
{MathLibraryName, NewOpenMath(nil)},
{UTF8LibraryName, OpenUTF8},
// {IOLibraryName, NewIOLibrary().OpenLibrary},
// {OSLibraryName, NewOSLibrary().OpenLibrary},
// {UTF8LibraryName, OpenUTF8},
// {DebugLibraryName, OpenDebug},
// {PackageLibraryName, OpenPackage},
}
Expand Down
11 changes: 10 additions & 1 deletion internal/lua/lua_test.go
Original file line number Diff line number Diff line change
Expand Up @@ -1187,7 +1187,9 @@ func TestRotate(t *testing.T) {
func TestSuite(t *testing.T) {
names := []string{
"math",
"pm",
"strings",
"utf8",
}

for _, name := range names {
Expand All @@ -1207,6 +1209,13 @@ func TestSuite(t *testing.T) {
t.Fatal(err)
}

// Message handler.
l.PushClosure(0, func(ctx context.Context, l *State) (int, error) {
msg, _ := l.ToString(1)
l.PushStringContext(Traceback(l, msg, 1), l.StringContext(1))
return 1, nil
})

sourcePath := filepath.Join("testdata", "testsuite", name+".lua")
sourceData, err := os.ReadFile(sourcePath)
if err != nil {
Expand All @@ -1216,7 +1225,7 @@ func TestSuite(t *testing.T) {
if err != nil {
t.Fatal(err)
}
if err := l.Call(ctx, 0, 0); err != nil {
if err := l.PCall(ctx, 0, 0, -2); err != nil {
t.Fatal(err)
}
})
Expand Down
119 changes: 64 additions & 55 deletions internal/lua/testdata/testsuite/pm.lua
Original file line number Diff line number Diff line change
Expand Up @@ -115,10 +115,11 @@ local function f1 (s, p)
return string.sub(s, t[1], t[#t] - 1)
end

assert(f1('alo alx 123 b\0o b\0o', '(..*) %1') == "b\0o b\0o")
assert(f1('axz123= 4= 4 34', '(.+)=(.*)=%2 %1') == '3= 4= 4 3')
assert(f1('=======', '^(=*)=%1$') == '=======')
assert(not string.match('==========', '^([=]*)=%1$'))
-- XXX: Backreferences not supported.
-- assert(f1('alo alx 123 b\0o b\0o', '(..*) %1') == "b\0o b\0o")
-- assert(f1('axz123= 4= 4 34', '(.+)=(.*)=%2 %1') == '3= 4= 4 3')
-- assert(f1('=======', '^(=*)=%1$') == '=======')
-- assert(not string.match('==========', '^([=]*)=%1$'))

local function range (i, j)
if i <= j then
Expand All @@ -145,8 +146,9 @@ assert(strset('[^%W]') == strset('[%w]'))
assert(strset('[]%%]') == '%]')
assert(strset('[a%-z]') == '-az')
assert(strset('[%^%[%-a%]%-b]') == '-[]^ab')
assert(strset('%Z') == strset('[\1-\255]'))
assert(strset('.') == strset('[\1-\255%z]'))
-- XXX: Undocumented %z not supported.
-- assert(strset('%Z') == strset('[\1-\255]'))
assert(strset('.') == strset('[\0-\255]'))
print('+');

assert(string.match("alo xyzK", "(%w+)K") == "xyz")
Expand Down Expand Up @@ -227,13 +229,14 @@ local r = string.gsub(s, '()(%w+)()', function (a,w,b)
assert(s == r and t[1] == 1 and t[3] == 3 and t[7] == 4 and t[13] == 4)


local function isbalanced (s)
return not string.find(string.gsub(s, "%b()", ""), "[()]")
end
-- XXX: Balances not supported.
-- local function isbalanced (s)
-- return not string.find(string.gsub(s, "%b()", ""), "[()]")
-- end

assert(isbalanced("(9 ((8))(\0) 7) \0\0 a b ()(c)() a"))
assert(not isbalanced("(9 ((8) 7) a b (\0 c) a"))
assert(string.gsub("alo 'oi' alo", "%b''", '"') == 'alo " alo')
-- assert(isbalanced("(9 ((8))(\0) 7) \0\0 a b ()(c)() a"))
-- assert(not isbalanced("(9 ((8) 7) a b (\0 c) a"))
-- assert(string.gsub("alo 'oi' alo", "%b''", '"') == 'alo " alo')


local t = {"apple", "orange", "lime"; n=0}
Expand All @@ -252,8 +255,9 @@ assert(t[1] == "first" and t[2] == "second" and t[3] == undef)
checkerror("invalid replacement value %(a table%)",
string.gsub, "alo", ".", {a = {}})
checkerror("invalid capture index %%2", string.gsub, "alo", ".", "%2")
checkerror("invalid capture index %%0", string.gsub, "alo", "(%0)", "a")
checkerror("invalid capture index %%1", string.gsub, "alo", "(%1)", "a")
-- XXX: Backreferences not supported.
-- checkerror("invalid capture index %%0", string.gsub, "alo", "(%0)", "a")
-- checkerror("invalid capture index %%1", string.gsub, "alo", "(%1)", "a")
checkerror("invalid use of '%%'", string.gsub, "alo", ".", "%x")


Expand Down Expand Up @@ -301,11 +305,12 @@ for w in string.gmatch("first second word", "%w+") do
end
assert(t[1] == "first" and t[2] == "second" and t[3] == "word")

t = {3, 6, 9}
for i in string.gmatch ("xuxx uu ppar r", "()(.)%2") do
assert(i == table.remove(t, 1))
end
assert(#t == 0)
-- XXX: Backreferences not supported.
-- t = {3, 6, 9}
-- for i in string.gmatch ("xuxx uu ppar r", "()(.)%2") do
-- assert(i == table.remove(t, 1))
-- end
-- assert(#t == 0)

t = {}
for i,j in string.gmatch("13 14 10 = 11, 15= 16, 22=23", "(%d+)%s*=%s*(%d+)") do
Expand Down Expand Up @@ -355,12 +360,14 @@ assert(string.gsub("function", "%f[\1-\255]%w", ".") == ".unction")
assert(string.gsub("function", "%f[^\1-\255]", ".") == "function.")

assert(string.find("a", "%f[a]") == 1)
assert(string.find("a", "%f[^%z]") == 1)
-- XXX: Undocumented %z not supported.
assert(string.find("a", "%f[^\0]") == 1)
assert(string.find("a", "%f[^%l]") == 2)
assert(string.find("aba", "%f[a%z]") == 3)
assert(string.find("aba", "%f[%z]") == 4)
assert(not string.find("aba", "%f[%l%z]"))
assert(not string.find("aba", "%f[^%l%z]"))
-- XXX: Undocumented %z not supported.
assert(string.find("aba", "%f[a\0]") == 3)
assert(string.find("aba", "%f[\0]") == 4)
assert(not string.find("aba", "%f[%l\0]"))
assert(not string.find("aba", "%f[^%l\0]"))

local i, e = string.find(" alo aalo allo", "%f[%S].-%f[%s].-%f[%S]")
assert(i == 2 and e == 5)
Expand Down Expand Up @@ -388,8 +395,8 @@ malform("[]")
malform("[^]")
malform("[a%]")
malform("[a%")
malform("%b")
malform("%ba")
malform("%b", "balances not supported")
malform("%ba", "balances not supported")
malform("%")
malform("%f", "missing")

Expand All @@ -398,7 +405,8 @@ assert(string.match("ab\0\1\2c", "[\0-\2]+") == "\0\1\2")
assert(string.match("ab\0\1\2c", "[\0-\0]+") == "\0")
assert(string.find("b$a", "$\0?") == 2)
assert(string.find("abc\0efg", "%\0") == 4)
assert(string.match("abc\0efg\0\1e\1g", "%b\0\1") == "\0efg\0\1e\1")
-- XXX: Balances not supported.
-- assert(string.match("abc\0efg\0\1e\1g", "%b\0\1") == "\0efg\0\1e\1")
assert(string.match("abc\0\0\0", "%\0+") == "\0\0\0")
assert(string.match("abc\0\0\0", "%\0%\0?") == "\0\0")

Expand All @@ -407,34 +415,35 @@ assert(string.find("abc\0\0","\0.") == 4)
assert(string.find("abcx\0\0abc\0abc","x\0\0abc\0a.") == 4)


do -- test reuse of original string in gsub
local s = string.rep("a", 100)
local r = string.gsub(s, "b", "c") -- no match
assert(string.format("%p", s) == string.format("%p", r))

r = string.gsub(s, ".", {x = "y"}) -- no substitutions
assert(string.format("%p", s) == string.format("%p", r))

local count = 0
r = string.gsub(s, ".", function (x)
assert(x == "a")
count = count + 1
return nil -- no substitution
end)
r = string.gsub(r, ".", {b = 'x'}) -- "a" is not a key; no subst.
assert(count == 100)
assert(string.format("%p", s) == string.format("%p", r))

count = 0
r = string.gsub(s, ".", function (x)
assert(x == "a")
count = count + 1
return x -- substitution...
end)
assert(count == 100)
-- no reuse in this case
assert(r == s and string.format("%p", s) ~= string.format("%p", r))
end
-- XXX: Strings do not have a pointer address in our implementation.
-- do -- test reuse of original string in gsub
-- local s = string.rep("a", 100)
-- local r = string.gsub(s, "b", "c") -- no match
-- assert(string.format("%p", s) == string.format("%p", r))

-- r = string.gsub(s, ".", {x = "y"}) -- no substitutions
-- assert(string.format("%p", s) == string.format("%p", r))

-- local count = 0
-- r = string.gsub(s, ".", function (x)
-- assert(x == "a")
-- count = count + 1
-- return nil -- no substitution
-- end)
-- r = string.gsub(r, ".", {b = 'x'}) -- "a" is not a key; no subst.
-- assert(count == 100)
-- assert(string.format("%p", s) == string.format("%p", r))

-- count = 0
-- r = string.gsub(s, ".", function (x)
-- assert(x == "a")
-- count = count + 1
-- return x -- substitution...
-- end)
-- assert(count == 100)
-- -- no reuse in this case
-- assert(r == s and string.format("%p", s) ~= string.format("%p", r))
-- end

print('OK')

3 changes: 2 additions & 1 deletion internal/lua/testdata/testsuite/utf8.lua
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,8 @@

print "testing UTF-8 library"

local utf8 = require'utf8'
-- XXX: utf8 library available as a global instead of being required.
local utf8 = utf8


local function checkerror (msg, f, ...)
Expand Down
Loading

0 comments on commit 938ba2f

Please sign in to comment.