init
This commit is contained in:
@@ -0,0 +1,218 @@
|
||||
--[[
|
||||
An addon for mpv-file-browser which uses the Windows dir command to parse native directories
|
||||
This behaves near identically to the native parser, but IO is done asynchronously.
|
||||
|
||||
Available at: https://github.com/CogentRedTester/mpv-file-browser/tree/master/addons
|
||||
]]--
|
||||
|
||||
local mp = require "mp"
|
||||
local msg = require "mp.msg"
|
||||
local fb = require "file-browser"
|
||||
|
||||
local PLATFORM = fb.get_platform()
|
||||
|
||||
---@param bytes string
|
||||
---@return fun(): number, number
|
||||
local function byte_iterator(bytes)
|
||||
---@async
|
||||
---@return number?
|
||||
local function iter()
|
||||
for i = 1, #bytes do
|
||||
coroutine.yield(bytes:byte(i), i)
|
||||
end
|
||||
error('malformed utf16le string - expected byte but found end of string')
|
||||
end
|
||||
|
||||
return coroutine.wrap(iter)
|
||||
end
|
||||
|
||||
---@param bits number
|
||||
---@param by number
|
||||
---@return number
|
||||
local function lshift(bits, by)
|
||||
return bits * 2^by
|
||||
end
|
||||
|
||||
---@param bits number
|
||||
---@param by number
|
||||
---@return integer
|
||||
local function rshift(bits, by)
|
||||
return math.floor(bits / 2^by)
|
||||
end
|
||||
|
||||
---@param bits number
|
||||
---@param i number
|
||||
---@return number
|
||||
local function bits_below(bits, i)
|
||||
return bits % 2^i
|
||||
end
|
||||
|
||||
---@param bits number
|
||||
---@param i number exclusive
|
||||
---@param j number inclusive
|
||||
---@return integer
|
||||
local function bits_between(bits, i, j)
|
||||
return rshift(bits_below(bits, j), i)
|
||||
end
|
||||
|
||||
---@param bytes string
|
||||
---@return number[]
|
||||
local function utf16le_to_unicode(bytes)
|
||||
msg.trace('converting from utf16-le to unicode codepoints')
|
||||
|
||||
---@type number[]
|
||||
local codepoints = {}
|
||||
|
||||
local get_byte = byte_iterator(bytes)
|
||||
|
||||
while true do
|
||||
-- start of a char
|
||||
local success, little, i = pcall(get_byte)
|
||||
if not success then break end
|
||||
|
||||
local big = get_byte()
|
||||
local codepoint = little + lshift(big, 8)
|
||||
|
||||
if codepoint < 0xd800 or codepoint > 0xdfff then
|
||||
table.insert(codepoints, codepoint)
|
||||
else
|
||||
-- handling surrogate pairs
|
||||
-- grab the next two bytes to grab the low surrogate
|
||||
local high_pair = codepoint
|
||||
local low_pair = get_byte() + lshift(get_byte(), 8)
|
||||
|
||||
if high_pair >= 0xdc00 then
|
||||
error(('malformed utf16le string at byte #%d (0x%04X) - high surrogate pair should be < 0xDC00'):format(i, high_pair))
|
||||
elseif low_pair < 0xdc00 then
|
||||
error(('malformed utf16le string at byte #%d (0x%04X) - low surrogate pair should be >= 0xDC00'):format(i+2, low_pair))
|
||||
end
|
||||
|
||||
-- The last 10 bits of each surrogate are the two halves of the codepoint
|
||||
-- https://en.wikipedia.org/wiki/UTF-16#Code_points_from_U+010000_to_U+10FFFF
|
||||
local high_bits = bits_below(high_pair, 10)
|
||||
local low_bits = bits_below(low_pair, 10)
|
||||
local surrogate_par = (low_bits + lshift(high_bits, 10)) + 0x10000
|
||||
|
||||
table.insert(codepoints, surrogate_par)
|
||||
end
|
||||
end
|
||||
|
||||
return codepoints
|
||||
end
|
||||
|
||||
---@param codepoints number[]
|
||||
---@return string
|
||||
local function unicode_to_utf8(codepoints)
|
||||
---@type number[]
|
||||
local bytes = {}
|
||||
|
||||
-- https://en.wikipedia.org/wiki/UTF-8#Description
|
||||
for i, codepoint in ipairs(codepoints) do
|
||||
if codepoint >= 0xd800 and codepoint <= 0xdfff then
|
||||
error(('codepoint %d (U+%05X) is within the reserved surrogate pair range (U+D800-U+DFFF)'):format(i, codepoint))
|
||||
elseif codepoint <= 0x7f then
|
||||
table.insert(bytes, codepoint)
|
||||
elseif codepoint <= 0x7ff then
|
||||
table.insert(bytes, 0xC0 + rshift(codepoint, 6))
|
||||
table.insert(bytes, 0x80 + bits_below(codepoint, 6))
|
||||
elseif codepoint <= 0xffff then
|
||||
table.insert(bytes, 0xE0 + rshift(codepoint, 12))
|
||||
table.insert(bytes, 0x80 + bits_between(codepoint, 6, 12))
|
||||
table.insert(bytes, 0x80 + bits_below(codepoint, 6))
|
||||
elseif codepoint <= 0x10ffff then
|
||||
table.insert(bytes, 0xF0 + rshift(codepoint, 18))
|
||||
table.insert(bytes, 0x80 + bits_between(codepoint, 12, 18))
|
||||
table.insert(bytes, 0x80 + bits_between(codepoint, 6, 12))
|
||||
table.insert(bytes, 0x80 + bits_below(codepoint, 6))
|
||||
else
|
||||
error(('codepoint %d (U+%05X) is larger than U+10FFFF'):format(i, codepoint))
|
||||
end
|
||||
end
|
||||
|
||||
return string.char(table.unpack(bytes))
|
||||
end
|
||||
|
||||
local function utf8(text)
|
||||
return unicode_to_utf8(utf16le_to_unicode(text))
|
||||
end
|
||||
|
||||
---@type ParserConfig
|
||||
local dir = {
|
||||
priority = 109,
|
||||
api_version = "1.9.0",
|
||||
name = "cmd-dir",
|
||||
keybind_name = "file"
|
||||
}
|
||||
|
||||
---@async
|
||||
---@param args string[]
|
||||
---@param parse_state ParseState
|
||||
---@return string|nil
|
||||
local function command(args, parse_state)
|
||||
local async = mp.command_native_async({
|
||||
name = "subprocess",
|
||||
playback_only = false,
|
||||
capture_stdout = true,
|
||||
capture_stderr = true,
|
||||
args = args,
|
||||
}, fb.coroutine.callback(30) )
|
||||
|
||||
---@type boolean, boolean, MPVSubprocessResult
|
||||
local completed, _, cmd = parse_state:yield()
|
||||
if not completed then
|
||||
msg.warn('read timed out for:', table.unpack(args))
|
||||
mp.abort_async_command(async)
|
||||
return nil
|
||||
end
|
||||
|
||||
local success = xpcall(function()
|
||||
cmd.stdout = utf8(cmd.stdout) or ''
|
||||
cmd.stderr = utf8(cmd.stderr) or ''
|
||||
end, fb.traceback)
|
||||
|
||||
if not success then return msg.error('failed to convert utf16-le string to utf8') end
|
||||
|
||||
--dir returns this exact error message if the directory is empty
|
||||
if cmd.status == 1 and cmd.stderr == "File Not Found\r\n" then cmd.status = 0 end
|
||||
if cmd.status ~= 0 then return msg.error(cmd.stderr) end
|
||||
|
||||
return cmd.status == 0 and cmd.stdout or nil
|
||||
end
|
||||
|
||||
function dir:can_parse(directory)
|
||||
if not fb.get_opt('windir_parser') then return false end
|
||||
return PLATFORM == 'windows' and directory ~= '' and not fb.get_protocol(directory)
|
||||
end
|
||||
|
||||
---@async
|
||||
function dir:parse(directory, parse_state)
|
||||
local list = {}
|
||||
|
||||
-- the dir command expects backslashes for our paths
|
||||
directory = string.gsub(directory, "/", "\\")
|
||||
|
||||
local dirs = command({ "cmd", "/U", "/c", "dir", "/b", "/ad", directory }, parse_state)
|
||||
if not dirs then return end
|
||||
|
||||
local files = command({ "cmd", "/U", "/c", "dir", "/b", "/a-d", directory }, parse_state)
|
||||
if not files then return end
|
||||
|
||||
for name in dirs:gmatch("[^\n\r]+") do
|
||||
name = name.."/"
|
||||
if fb.valid_dir(name) then
|
||||
table.insert(list, { name = name, type = "dir" })
|
||||
msg.trace(name)
|
||||
end
|
||||
end
|
||||
|
||||
for name in files:gmatch("[^\n\r]+") do
|
||||
if fb.valid_file(name) then
|
||||
table.insert(list, { name = name, type = "file" })
|
||||
msg.trace(name)
|
||||
end
|
||||
end
|
||||
|
||||
return list, { filtered = true }
|
||||
end
|
||||
|
||||
return dir
|
||||
Reference in New Issue
Block a user