walkero
/
lite-xl
1
0
Fork 0
Code Issues 7 Pull Requests Projects Releases 14 Wiki Activity

tokenizer: fix next utf8 char retrieval bug

This commit is contained in:
jgmdev 2022-05-11 01:05:36 -04:00
parent 2c968073e4
commit 94430bcbd2
1 changed files with 9 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
data/core/tokenizer.lua
View File

@ -169,12 +169,12 @@ function tokenizer.tokenize(incoming_syntax, text, state)
if p.whole_line[p_idx] and next > 1 then if p.whole_line[p_idx] and next > 1 then
return return
end end
-- go to the start of the next utf-8 character
while text:byte(next) and common.is_utf8_cont(text, next) do
next = next + 1
end
res = p.pattern and { text:ufind((at_start or p.whole_line[p_idx]) and "^" .. code or code, next) } res = p.pattern and { text:ufind((at_start or p.whole_line[p_idx]) and "^" .. code or code, next) }
or { regex.match(code, text, next, (at_start or p.whole_line[p_idx]) and regex.ANCHORED or 0) } or { regex.match(code, text, text:ucharpos(next), (at_start or p.whole_line[p_idx]) and regex.ANCHORED or 0) }
if p.regex and #res > 0 then -- set correct utf8 len for regex result
res[2] = res[1] + string.ulen(text:sub(res[1], res[2])) - 1
res[1] = next
end
if res[1] and close and target[3] then if res[1] and close and target[3] then
local count = 0 local count = 0
for i = res[1] - 1, 1, -1 do for i = res[1] - 1, 1, -1 do
@ -189,7 +189,8 @@ function tokenizer.tokenize(incoming_syntax, text, state)
return table.unpack(res) return table.unpack(res)
end end
while i <= #text do local text_len = text:ulen()
while i <= text_len do
-- continue trying to match the end pattern of a pair if we have a state set -- continue trying to match the end pattern of a pair if we have a state set
if current_pattern_idx > 0 then if current_pattern_idx > 0 then
local p = current_syntax.patterns[current_pattern_idx] local p = current_syntax.patterns[current_pattern_idx]
@ -262,13 +263,8 @@ function tokenizer.tokenize(incoming_syntax, text, state)
-- consume character if we didn't match -- consume character if we didn't match
if not matched then if not matched then
local n = 0 push_token(res, "normal", text:usub(i, i))
-- reach the next character i = i + 1
while text:byte(i + n + 1) and common.is_utf8_cont(text, i + n + 1) do
n = n + 1
end
push_token(res, "normal", text:usub(i, i + n))
i = i + n + 1
end end
end end