aboutsummaryrefslogtreecommitdiff
path: root/test/unit/mbyte_spec.lua
diff options
context:
space:
mode:
authorbfredl <bjorn.linse@gmail.com>2023-11-06 14:52:27 +0100
committerbfredl <bjorn.linse@gmail.com>2023-11-17 12:58:57 +0100
commitb522cb1ac3fbdf6e68eed5d0b6e1cbeaf3ac2254 (patch)
tree434ec27e069ba57406ce9f6d194627e95c3d315c /test/unit/mbyte_spec.lua
parent20ec4c776a07492c2e3b995e10b40b1cdb52bc7a (diff)
downloadrneovim-b522cb1ac3fbdf6e68eed5d0b6e1cbeaf3ac2254.tar.gz
rneovim-b522cb1ac3fbdf6e68eed5d0b6e1cbeaf3ac2254.tar.bz2
rneovim-b522cb1ac3fbdf6e68eed5d0b6e1cbeaf3ac2254.zip
refactor(grid): make screen rendering more multibyte than ever before
Problem: buffer text with composing chars are converted from UTF-8 to an array of up to seven UTF-32 values and then converted back to UTF-8 strings. Solution: Convert buffer text directly to UTF-8 based schar_T values. The limit of the text size is now in schar_T bytes, which is currently 31+1 but easily could be raised as it no longer multiplies the size of the entire screen grid when not used, the full size is only required for temporary scratch buffers. Also does some general cleanup to win_line text handling, which was unnecessarily complicated due to multibyte rendering being an "opt-in" feature long ago. Nowadays, a char is just a char, regardless if it consists of one ASCII byte or multiple bytes.
Diffstat (limited to 'test/unit/mbyte_spec.lua')
-rw-r--r--test/unit/mbyte_spec.lua243
1 files changed, 64 insertions, 179 deletions
diff --git a/test/unit/mbyte_spec.lua b/test/unit/mbyte_spec.lua
index fdb1bceab0..cd94624570 100644
--- a/test/unit/mbyte_spec.lua
+++ b/test/unit/mbyte_spec.lua
@@ -4,17 +4,9 @@ local itp = helpers.gen_itp(it)
local ffi = helpers.ffi
local eq = helpers.eq
-local mbyte = helpers.cimport("./src/nvim/mbyte.h")
-local charset = helpers.cimport('./src/nvim/charset.h')
+local lib = helpers.cimport('./src/nvim/mbyte.h', './src/nvim/charset.h', './src/nvim/grid.h')
describe('mbyte', function()
- -- Array for composing characters
- local intp = ffi.typeof('int[?]')
- local function to_intp()
- -- how to get MAX_MCO from globals.h?
- return intp(7, 1)
- end
-
-- Convert from bytes to string
local function to_string(bytes)
local s = {}
@@ -30,14 +22,14 @@ describe('mbyte', function()
itp('utf_ptr2char', function()
-- For strings with length 1 the first byte is returned.
for c = 0, 255 do
- eq(c, mbyte.utf_ptr2char(to_string({c, 0})))
+ eq(c, lib.utf_ptr2char(to_string({c, 0})))
end
-- Some ill formed byte sequences that should not be recognized as UTF-8
-- First byte: 0xc0 or 0xc1
-- Second byte: 0x80 .. 0xbf
- --eq(0x00c0, mbyte.utf_ptr2char(to_string({0xc0, 0x80})))
- --eq(0x00c1, mbyte.utf_ptr2char(to_string({0xc1, 0xbf})))
+ --eq(0x00c0, lib.utf_ptr2char(to_string({0xc0, 0x80})))
+ --eq(0x00c1, lib.utf_ptr2char(to_string({0xc1, 0xbf})))
--
-- Sequences with more than four bytes
end)
@@ -47,240 +39,133 @@ describe('mbyte', function()
local char_p = ffi.typeof('char[?]')
for c = n * 0x1000, n * 0x1000 + 0xFFF do
local p = char_p(4, 0)
- mbyte.utf_char2bytes(c, p)
- eq(c, mbyte.utf_ptr2char(p))
- eq(charset.vim_iswordc(c), charset.vim_iswordp(p))
+ lib.utf_char2bytes(c, p)
+ eq(c, lib.utf_ptr2char(p))
+ eq(lib.vim_iswordc(c), lib.vim_iswordp(p))
end
end)
end
- describe('utfc_ptr2char_len', function()
+ describe('utfc_ptr2schar_len', function()
+ local function test_seq(seq)
+ local firstc = ffi.new("int[1]")
+ local buf = ffi.new("char[32]")
+ lib.schar_get(buf, lib.utfc_ptr2schar_len(to_string(seq), #seq, firstc))
+ return {ffi.string(buf), firstc[0]}
+ end
+
+ local function byte(val)
+ return {string.char(val), val}
+ end
itp('1-byte sequences', function()
- local pcc = to_intp()
- for c = 0, 255 do
- eq(c, mbyte.utfc_ptr2char_len(to_string({c}), pcc, 1))
- eq(0, pcc[0])
+ eq({'', 0}, test_seq{0})
+ for c = 1, 127 do
+ eq(byte(c), test_seq{c})
+ end
+ for c = 128, 255 do
+ eq({'', c}, test_seq{c})
end
end)
itp('2-byte sequences', function()
- local pcc = to_intp()
-- No combining characters
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string({0x7f, 0x7f}), pcc, 2))
- eq(0, pcc[0])
+ eq(byte(0x7f), test_seq{0x7f, 0x7f})
-- No combining characters
- pcc = to_intp()
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string({0x7f, 0x80}), pcc, 2))
- eq(0, pcc[0])
+ eq(byte(0x7f), test_seq{0x7f, 0x80})
-- No UTF-8 sequence
- pcc = to_intp()
- eq(0x00c2, mbyte.utfc_ptr2char_len(to_string({0xc2, 0x7f}), pcc, 2))
- eq(0, pcc[0])
+ eq({'', 0xc2}, test_seq{0xc2, 0x7f})
-- One UTF-8 character
- pcc = to_intp()
- eq(0x0080, mbyte.utfc_ptr2char_len(to_string({0xc2, 0x80}), pcc, 2))
- eq(0, pcc[0])
+ eq({'\xc2\x80', 0x80}, test_seq{0xc2, 0x80})
-- No UTF-8 sequence
- pcc = to_intp()
- eq(0x00c2, mbyte.utfc_ptr2char_len(to_string({0xc2, 0xc0}), pcc, 2))
- eq(0, pcc[0])
+ eq({'', 0xc2}, test_seq{0xc2, 0xc0})
end)
itp('3-byte sequences', function()
- local pcc = to_intp()
-
-- No second UTF-8 character
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string({0x7f, 0x80, 0x80}), pcc, 3))
- eq(0, pcc[0])
+ eq(byte(0x7f), test_seq{0x7f, 0x80, 0x80})
-- No combining character
- pcc = to_intp()
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string({0x7f, 0xc2, 0x80}), pcc, 3))
- eq(0, pcc[0])
+ eq(byte(0x7f), test_seq{0x7f, 0xc2, 0x80})
-- Combining character is U+0300
- pcc = to_intp()
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string({0x7f, 0xcc, 0x80}), pcc, 3))
- eq(0x0300, pcc[0])
- eq(0x0000, pcc[1])
+ eq({"\x7f\xcc\x80", 0x7f}, test_seq{0x7f, 0xcc, 0x80})
-- No UTF-8 sequence
- pcc = to_intp()
- eq(0x00c2, mbyte.utfc_ptr2char_len(to_string({0xc2, 0x7f, 0xcc}), pcc, 3))
- eq(0, pcc[0])
+ eq({'', 0xc2}, test_seq{0xc2, 0x7f, 0xcc})
-- Incomplete combining character
- pcc = to_intp()
- eq(0x0080, mbyte.utfc_ptr2char_len(to_string({0xc2, 0x80, 0xcc}), pcc, 3))
- eq(0, pcc[0])
+ eq({"\xc2\x80", 0x80}, test_seq{0xc2, 0x80, 0xcc})
- -- One UTF-8 character
- pcc = to_intp()
- eq(0x20d0, mbyte.utfc_ptr2char_len(to_string({0xe2, 0x83, 0x90}), pcc, 3))
- eq(0, pcc[0])
+ -- One UTF-8 character (composing only)
+ eq({" \xe2\x83\x90", 0x20d0}, test_seq{0xe2, 0x83, 0x90})
end)
itp('4-byte sequences', function()
- local pcc = to_intp()
-- No following combining character
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string({0x7f, 0x7f, 0xcc, 0x80}), pcc, 4))
- eq(0, pcc[0])
+ eq(byte(0x7f), test_seq{0x7f, 0x7f, 0xcc, 0x80})
-- No second UTF-8 character
- pcc = to_intp()
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string({0x7f, 0xc2, 0xcc, 0x80}), pcc, 4))
- eq(0, pcc[0])
+ eq(byte(0x7f), test_seq{0x7f, 0xc2, 0xcc, 0x80})
-- Combining character U+0300
- pcc = to_intp()
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string({0x7f, 0xcc, 0x80, 0xcc}), pcc, 4))
- eq(0x0300, pcc[0])
- eq(0x0000, pcc[1])
+ eq({"\x7f\xcc\x80", 0x7f}, test_seq{0x7f, 0xcc, 0x80, 0xcc})
-- No UTF-8 sequence
- pcc = to_intp()
- eq(0x00c2, mbyte.utfc_ptr2char_len(to_string({0xc2, 0x7f, 0xcc, 0x80}), pcc, 4))
- eq(0, pcc[0])
+ eq({'', 0xc2}, test_seq{0xc2, 0x7f, 0xcc, 0x80})
-- No following UTF-8 character
- pcc = to_intp()
- eq(0x0080, mbyte.utfc_ptr2char_len(to_string({0xc2, 0x80, 0xcc, 0xcc}), pcc, 4))
- eq(0, pcc[0])
+ eq({"\xc2\x80", 0x80}, test_seq{0xc2, 0x80, 0xcc, 0xcc})
-- Combining character U+0301
- pcc = to_intp()
- eq(0x0080, mbyte.utfc_ptr2char_len(to_string({0xc2, 0x80, 0xcc, 0x81}), pcc, 4))
- eq(0x0301, pcc[0])
- eq(0x0000, pcc[1])
+ eq({"\xc2\x80\xcc\x81", 0x80}, test_seq{0xc2, 0x80, 0xcc, 0x81})
-- One UTF-8 character
- pcc = to_intp()
- eq(0x100000, mbyte.utfc_ptr2char_len(to_string({0xf4, 0x80, 0x80, 0x80}), pcc, 4))
- eq(0, pcc[0])
+ eq({"\xf4\x80\x80\x80", 0x100000}, test_seq{0xf4, 0x80, 0x80, 0x80})
end)
itp('5+-byte sequences', function()
- local pcc = to_intp()
-
-- No following combining character
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string({0x7f, 0x7f, 0xcc, 0x80, 0x80}), pcc, 5))
- eq(0, pcc[0])
+ eq(byte(0x7f), test_seq{0x7f, 0x7f, 0xcc, 0x80, 0x80})
-- No second UTF-8 character
- pcc = to_intp()
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string({0x7f, 0xc2, 0xcc, 0x80, 0x80}), pcc, 5))
- eq(0, pcc[0])
+ eq(byte(0x7f), test_seq{0x7f, 0xc2, 0xcc, 0x80, 0x80})
-- Combining character U+0300
- pcc = to_intp()
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string({0x7f, 0xcc, 0x80, 0xcc}), pcc, 5))
- eq(0x0300, pcc[0])
- eq(0x0000, pcc[1])
+ eq({"\x7f\xcc\x80", 0x7f}, test_seq{0x7f, 0xcc, 0x80, 0xcc, 0x00})
-- Combining characters U+0300 and U+0301
- pcc = to_intp()
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string({0x7f, 0xcc, 0x80, 0xcc, 0x81}), pcc, 5))
- eq(0x0300, pcc[0])
- eq(0x0301, pcc[1])
- eq(0x0000, pcc[2])
+ eq({"\x7f\xcc\x80\xcc\x81", 0x7f}, test_seq{0x7f, 0xcc, 0x80, 0xcc, 0x81})
-- Combining characters U+0300, U+0301, U+0302
- pcc = to_intp()
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string({0x7f, 0xcc, 0x80, 0xcc, 0x81, 0xcc, 0x82}), pcc, 7))
- eq(0x0300, pcc[0])
- eq(0x0301, pcc[1])
- eq(0x0302, pcc[2])
- eq(0x0000, pcc[3])
+ eq({"\x7f\xcc\x80\xcc\x81\xcc\x82", 0x7f}, test_seq{0x7f, 0xcc, 0x80, 0xcc, 0x81, 0xcc, 0x82})
-- Combining characters U+0300, U+0301, U+0302, U+0303
- pcc = to_intp()
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string({0x7f, 0xcc, 0x80, 0xcc, 0x81, 0xcc, 0x82, 0xcc, 0x83}), pcc, 9))
- eq(0x0300, pcc[0])
- eq(0x0301, pcc[1])
- eq(0x0302, pcc[2])
- eq(0x0303, pcc[3])
- eq(0x0000, pcc[4])
+ eq({"\x7f\xcc\x80\xcc\x81\xcc\x82\xcc\x83", 0x7f}, test_seq{0x7f, 0xcc, 0x80, 0xcc, 0x81, 0xcc, 0x82, 0xcc, 0x83})
-- Combining characters U+0300, U+0301, U+0302, U+0303, U+0304
- pcc = to_intp()
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string(
- {0x7f, 0xcc, 0x80, 0xcc, 0x81, 0xcc, 0x82, 0xcc, 0x83, 0xcc, 0x84}), pcc, 11))
- eq(0x0300, pcc[0])
- eq(0x0301, pcc[1])
- eq(0x0302, pcc[2])
- eq(0x0303, pcc[3])
- eq(0x0304, pcc[4])
- eq(0x0000, pcc[5])
- -- Combining characters U+0300, U+0301, U+0302, U+0303, U+0304,
- -- U+0305
- pcc = to_intp()
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string(
- {0x7f, 0xcc, 0x80, 0xcc, 0x81, 0xcc, 0x82, 0xcc, 0x83, 0xcc, 0x84, 0xcc, 0x85}), pcc, 13))
- eq(0x0300, pcc[0])
- eq(0x0301, pcc[1])
- eq(0x0302, pcc[2])
- eq(0x0303, pcc[3])
- eq(0x0304, pcc[4])
- eq(0x0305, pcc[5])
- eq(1, pcc[6])
-
- -- Combining characters U+0300, U+0301, U+0302, U+0303, U+0304,
- -- U+0305, U+0306, but only save six (= MAX_MCO).
- pcc = to_intp()
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string(
- {0x7f, 0xcc, 0x80, 0xcc, 0x81, 0xcc, 0x82, 0xcc, 0x83, 0xcc, 0x84, 0xcc, 0x85, 0xcc, 0x86}), pcc, 15))
- eq(0x0300, pcc[0])
- eq(0x0301, pcc[1])
- eq(0x0302, pcc[2])
- eq(0x0303, pcc[3])
- eq(0x0304, pcc[4])
- eq(0x0305, pcc[5])
- eq(0x0001, pcc[6])
+ eq({"\x7f\xcc\x80\xcc\x81\xcc\x82\xcc\x83\xcc\x84", 0x7f}, test_seq{0x7f, 0xcc, 0x80, 0xcc, 0x81, 0xcc, 0x82, 0xcc, 0x83, 0xcc, 0x84})
+ -- Combining characters U+0300, U+0301, U+0302, U+0303, U+0304, U+0305
+ eq({"\x7f\xcc\x80\xcc\x81\xcc\x82\xcc\x83\xcc\x84\xcc\x85", 0x7f}, test_seq{0x7f, 0xcc, 0x80, 0xcc, 0x81, 0xcc, 0x82, 0xcc, 0x83, 0xcc, 0x84, 0xcc, 0x85})
- -- Only three following combining characters U+0300, U+0301, U+0302
- pcc = to_intp()
- eq(0x007f, mbyte.utfc_ptr2char_len(to_string(
- {0x7f, 0xcc, 0x80, 0xcc, 0x81, 0xcc, 0x82, 0xc2, 0x80, 0xcc, 0x84, 0xcc, 0x85}), pcc, 13))
- eq(0x0300, pcc[0])
- eq(0x0301, pcc[1])
- eq(0x0302, pcc[2])
- eq(0x0000, pcc[3])
+ -- Combining characters U+0300, U+0301, U+0302, U+0303, U+0304, U+0305, U+0306
+ eq({"\x7f\xcc\x80\xcc\x81\xcc\x82\xcc\x83\xcc\x84\xcc\x85\xcc\x86", 0x7f}, test_seq{0x7f, 0xcc, 0x80, 0xcc, 0x81, 0xcc, 0x82, 0xcc, 0x83, 0xcc, 0x84, 0xcc, 0x85, 0xcc, 0x86})
+ -- Only three following combining characters U+0300, U+0301, U+0302
+ eq({"\x7f\xcc\x80\xcc\x81\xcc\x82", 0x7f}, test_seq{0x7f, 0xcc, 0x80, 0xcc, 0x81, 0xcc, 0x82, 0xc2, 0x80, 0xcc, 0x84, 0xcc, 0x85})
-- No UTF-8 sequence
- pcc = to_intp()
- eq(0x00c2, mbyte.utfc_ptr2char_len(to_string({0xc2, 0x7f, 0xcc, 0x80, 0x80}), pcc, 5))
- eq(0, pcc[0])
+ eq({'', 0xc2}, test_seq{0xc2, 0x7f, 0xcc, 0x80, 0x80})
-- No following UTF-8 character
- pcc = to_intp()
- eq(0x0080, mbyte.utfc_ptr2char_len(to_string({0xc2, 0x80, 0xcc, 0xcc, 0x80}), pcc, 5))
- eq(0, pcc[0])
+ eq({"\xc2\x80", 0x80}, test_seq{0xc2, 0x80, 0xcc, 0xcc, 0x80})
-- Combining character U+0301
- pcc = to_intp()
- eq(0x0080, mbyte.utfc_ptr2char_len(to_string({0xc2, 0x80, 0xcc, 0x81, 0x7f}), pcc, 5))
- eq(0x0301, pcc[0])
- eq(0x0000, pcc[1])
+ eq({"\xc2\x80\xcc\x81", 0x80}, test_seq{0xc2, 0x80, 0xcc, 0x81, 0x7f})
-- Combining character U+0301
- pcc = to_intp()
- eq(0x0080, mbyte.utfc_ptr2char_len(to_string({0xc2, 0x80, 0xcc, 0x81, 0xcc}), pcc, 5))
- eq(0x0301, pcc[0])
- eq(0x0000, pcc[1])
+ eq({"\xc2\x80\xcc\x81", 0x80}, test_seq{0xc2, 0x80, 0xcc, 0x81, 0xcc})
-- One UTF-8 character
- pcc = to_intp()
- eq(0x100000, mbyte.utfc_ptr2char_len(to_string({0xf4, 0x80, 0x80, 0x80, 0x7f}), pcc, 5))
- eq(0, pcc[0])
+ eq({"\xf4\x80\x80\x80", 0x100000}, test_seq{0xf4, 0x80, 0x80, 0x80, 0x7f})
-- One UTF-8 character
- pcc = to_intp()
- eq(0x100000, mbyte.utfc_ptr2char_len(to_string({0xf4, 0x80, 0x80, 0x80, 0x80}), pcc, 5))
- eq(0, pcc[0])
+ eq({"\xf4\x80\x80\x80", 0x100000}, test_seq{0xf4, 0x80, 0x80, 0x80, 0x80})
-- One UTF-8 character
- pcc = to_intp()
- eq(0x100000, mbyte.utfc_ptr2char_len(to_string({0xf4, 0x80, 0x80, 0x80, 0xcc}), pcc, 5))
- eq(0, pcc[0])
+ eq({"\xf4\x80\x80\x80", 0x100000}, test_seq{0xf4, 0x80, 0x80, 0x80, 0xcc})
-- Combining characters U+1AB0 and U+0301
- pcc = to_intp()
- eq(0x100000, mbyte.utfc_ptr2char_len(to_string(
- {0xf4, 0x80, 0x80, 0x80, 0xe1, 0xaa, 0xb0, 0xcc, 0x81}), pcc, 9))
- eq(0x1ab0, pcc[0])
- eq(0x0301, pcc[1])
- eq(0x0000, pcc[2])
+ eq({"\xf4\x80\x80\x80\xe1\xaa\xb0\xcc\x81", 0x100000}, test_seq{0xf4, 0x80, 0x80, 0x80, 0xe1, 0xaa, 0xb0, 0xcc, 0x81})
end)
end)