2021-07-20 20:39:50 +02:00
|
|
|
-- So that in addition to regex.gsub(pattern, string), we can also do
|
2021-06-02 21:27:00 +02:00
|
|
|
-- pattern:gsub(string).
|
|
|
|
regex.__index = function(table, key) return regex[key]; end
|
|
|
|
|
|
|
|
regex.match = function(pattern_string, string, offset, options)
|
|
|
|
local pattern = type(pattern_string) == "table" and
|
|
|
|
pattern_string or regex.compile(pattern_string)
|
2022-05-28 01:21:41 +02:00
|
|
|
local res = { regex.cmatch(pattern, string, offset or 1, options or 0) }
|
|
|
|
res[2] = res[2] and res[2] - 1
|
|
|
|
return table.unpack(res)
|
2021-06-02 21:27:00 +02:00
|
|
|
end
|
|
|
|
|
2021-07-20 20:39:50 +02:00
|
|
|
-- Will iterate back through any UTF-8 bytes so that we don't replace bits
|
2021-06-02 21:27:00 +02:00
|
|
|
-- mid character.
|
|
|
|
local function previous_character(str, index)
|
|
|
|
local byte
|
|
|
|
repeat
|
|
|
|
index = index - 1
|
|
|
|
byte = string.byte(str, index)
|
|
|
|
until byte < 128 or byte >= 192
|
|
|
|
return index
|
|
|
|
end
|
|
|
|
|
|
|
|
-- Moves to the end of the identified character.
|
|
|
|
local function end_character(str, index)
|
|
|
|
local byte = string.byte(str, index + 1)
|
2021-09-10 14:55:04 +02:00
|
|
|
while byte and byte >= 128 and byte < 192 do
|
2021-06-02 21:27:00 +02:00
|
|
|
index = index + 1
|
|
|
|
byte = string.byte(str, index + 1)
|
|
|
|
end
|
|
|
|
return index
|
|
|
|
end
|
|
|
|
|
|
|
|
-- Build off matching. For now, only support basic replacements, but capture
|
|
|
|
-- groupings should be doable. We can even have custom group replacements and
|
2021-07-20 20:39:50 +02:00
|
|
|
-- transformations and stuff in lua. Currently, this takes group replacements
|
2021-06-02 21:27:00 +02:00
|
|
|
-- as \1 - \9.
|
|
|
|
-- Should work on UTF-8 text.
|
|
|
|
regex.gsub = function(pattern_string, str, replacement)
|
|
|
|
local pattern = type(pattern_string) == "table" and
|
|
|
|
pattern_string or regex.compile(pattern_string)
|
2022-12-03 01:21:05 +01:00
|
|
|
local result, indices = {}
|
2021-06-05 05:58:17 +02:00
|
|
|
local matches, replacements = {}, {}
|
2022-12-03 01:21:05 +01:00
|
|
|
local offset = 0
|
2021-06-02 21:27:00 +02:00
|
|
|
repeat
|
2022-12-03 01:21:05 +01:00
|
|
|
indices = { regex.cmatch(pattern, str, offset) }
|
2021-06-02 21:27:00 +02:00
|
|
|
if #indices > 0 then
|
2021-06-05 05:58:17 +02:00
|
|
|
table.insert(matches, indices)
|
2021-06-02 21:27:00 +02:00
|
|
|
local currentReplacement = replacement
|
|
|
|
if #indices > 2 then
|
|
|
|
for i = 1, (#indices/2 - 1) do
|
|
|
|
currentReplacement = string.gsub(
|
2021-07-20 20:39:50 +02:00
|
|
|
currentReplacement,
|
|
|
|
"\\" .. i,
|
2021-06-02 21:27:00 +02:00
|
|
|
str:sub(indices[i*2+1], end_character(str,indices[i*2+2]-1))
|
|
|
|
)
|
|
|
|
end
|
|
|
|
end
|
|
|
|
currentReplacement = string.gsub(currentReplacement, "\\%d", "")
|
2021-06-05 05:58:17 +02:00
|
|
|
table.insert(replacements, { indices[1], #currentReplacement+indices[1] })
|
2021-06-02 21:27:00 +02:00
|
|
|
if indices[1] > 1 then
|
2022-12-03 01:21:05 +01:00
|
|
|
table.insert(result, str:sub(offset, previous_character(str, indices[1])) .. currentReplacement)
|
2021-06-02 21:27:00 +02:00
|
|
|
else
|
2022-12-03 01:21:05 +01:00
|
|
|
table.insert(result, currentReplacement)
|
2021-06-02 21:27:00 +02:00
|
|
|
end
|
2022-12-03 01:21:05 +01:00
|
|
|
offset = indices[2]
|
2021-06-02 21:27:00 +02:00
|
|
|
end
|
|
|
|
until #indices == 0 or indices[1] == indices[2]
|
2022-12-03 01:21:05 +01:00
|
|
|
return table.concat(result) .. str:sub(offset), matches, replacements
|
2021-06-02 21:27:00 +02:00
|
|
|
end
|
|
|
|
|